CN115389380B - Particle size prediction method and system for adult milk powder atomization process - Google Patents
Particle size prediction method and system for adult milk powder atomization process Download PDFInfo
- Publication number
- CN115389380B CN115389380B CN202211114055.6A CN202211114055A CN115389380B CN 115389380 B CN115389380 B CN 115389380B CN 202211114055 A CN202211114055 A CN 202211114055A CN 115389380 B CN115389380 B CN 115389380B
- Authority
- CN
- China
- Prior art keywords
- atomization
- particle
- particle size
- milk powder
- atomization process
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 239000002245 particle Substances 0.000 title claims abstract description 111
- 238000000034 method Methods 0.000 title claims abstract description 94
- 238000000889 atomisation Methods 0.000 title claims abstract description 79
- 230000008569 process Effects 0.000 title claims abstract description 60
- 235000013336 milk Nutrition 0.000 title claims abstract description 51
- 239000008267 milk Substances 0.000 title claims abstract description 51
- 210000004080 milk Anatomy 0.000 title claims abstract description 51
- 239000000843 powder Substances 0.000 title claims abstract description 35
- 238000009826 distribution Methods 0.000 claims abstract description 28
- 239000007788 liquid Substances 0.000 claims abstract description 25
- 238000001514 detection method Methods 0.000 claims abstract description 22
- 238000005259 measurement Methods 0.000 claims abstract description 6
- 238000012545 processing Methods 0.000 claims description 19
- 238000006243 chemical reaction Methods 0.000 claims description 9
- 238000004458 analytical method Methods 0.000 claims description 6
- 238000012549 training Methods 0.000 claims description 5
- 238000004364 calculation method Methods 0.000 claims description 3
- 238000011056 performance test Methods 0.000 claims description 3
- 230000009897 systematic effect Effects 0.000 claims description 3
- 238000012546 transfer Methods 0.000 claims description 3
- 239000000839 emulsion Substances 0.000 abstract description 7
- 238000002360 preparation method Methods 0.000 abstract description 3
- 230000003044 adaptive effect Effects 0.000 description 16
- 238000012360 testing method Methods 0.000 description 12
- 238000001035 drying Methods 0.000 description 10
- 238000005315 distribution function Methods 0.000 description 5
- 238000005516 engineering process Methods 0.000 description 5
- 230000006978 adaptation Effects 0.000 description 4
- 238000010586 diagram Methods 0.000 description 4
- 239000011159 matrix material Substances 0.000 description 4
- 239000000047 product Substances 0.000 description 4
- 238000005507 spraying Methods 0.000 description 4
- 238000011049 filling Methods 0.000 description 3
- 238000005286 illumination Methods 0.000 description 3
- 238000004519 manufacturing process Methods 0.000 description 3
- 239000003595 mist Substances 0.000 description 3
- 238000010606 normalization Methods 0.000 description 3
- 238000004806 packaging method and process Methods 0.000 description 3
- 230000009466 transformation Effects 0.000 description 3
- 230000000694 effects Effects 0.000 description 2
- 238000004134 energy conservation Methods 0.000 description 2
- 238000011156 evaluation Methods 0.000 description 2
- 238000000265 homogenisation Methods 0.000 description 2
- 230000006872 improvement Effects 0.000 description 2
- 239000000203 mixture Substances 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 235000015097 nutrients Nutrition 0.000 description 2
- 235000016709 nutrition Nutrition 0.000 description 2
- 238000007781 pre-processing Methods 0.000 description 2
- 230000004044 response Effects 0.000 description 2
- 238000003860 storage Methods 0.000 description 2
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 2
- JLPULHDHAOZNQI-ZTIMHPMXSA-N 1-hexadecanoyl-2-(9Z,12Z-octadecadienoyl)-sn-glycero-3-phosphocholine Chemical compound CCCCCCCCCCCCCCCC(=O)OC[C@H](COP([O-])(=O)OCC[N+](C)(C)C)OC(=O)CCCCCCC\C=C/C\C=C/CCCCC JLPULHDHAOZNQI-ZTIMHPMXSA-N 0.000 description 1
- 235000020244 animal milk Nutrition 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 150000001720 carbohydrates Chemical class 0.000 description 1
- 238000004140 cleaning Methods 0.000 description 1
- 230000000052 comparative effect Effects 0.000 description 1
- 238000001816 cooling Methods 0.000 description 1
- 238000012937 correction Methods 0.000 description 1
- 201000010099 disease Diseases 0.000 description 1
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 description 1
- 238000010981 drying operation Methods 0.000 description 1
- 230000002708 enhancing effect Effects 0.000 description 1
- 238000001704 evaporation Methods 0.000 description 1
- 230000008020 evaporation Effects 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 238000003384 imaging method Methods 0.000 description 1
- 230000036039 immunity Effects 0.000 description 1
- 239000004615 ingredient Substances 0.000 description 1
- 229910052500 inorganic mineral Inorganic materials 0.000 description 1
- 239000011707 mineral Substances 0.000 description 1
- 235000010755 mineral Nutrition 0.000 description 1
- 230000035764 nutrition Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 102000004169 proteins and genes Human genes 0.000 description 1
- 108090000623 proteins and genes Proteins 0.000 description 1
- 239000002994 raw material Substances 0.000 description 1
- 239000010865 sewage Substances 0.000 description 1
- 229940083466 soybean lecithin Drugs 0.000 description 1
- 230000001954 sterilising effect Effects 0.000 description 1
- 239000013589 supplement Substances 0.000 description 1
- 235000013343 vitamin Nutrition 0.000 description 1
- 229940088594 vitamin Drugs 0.000 description 1
- 229930003231 vitamin Natural products 0.000 description 1
- 239000011782 vitamin Substances 0.000 description 1
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N15/00—Investigating characteristics of particles; Investigating permeability, pore-volume or surface-area of porous materials
- G01N15/02—Investigating particle size or size distribution
- G01N15/0205—Investigating particle size or size distribution by optical means
- G01N15/0227—Investigating particle size or size distribution by optical means using imaging; using holography
Landscapes
- Chemical & Material Sciences (AREA)
- Dispersion Chemistry (AREA)
- Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Analytical Chemistry (AREA)
- Biochemistry (AREA)
- General Health & Medical Sciences (AREA)
- General Physics & Mathematics (AREA)
- Immunology (AREA)
- Pathology (AREA)
- Image Processing (AREA)
Abstract
The invention discloses a particle size prediction method, belongs to the technical field of adult milk powder preparation processes, and particularly relates to a particle size prediction method and a particle size prediction system for an adult milk powder atomization process, wherein basic information of atomization equipment is input, and a prediction model is established; secondly, inputting a predictive model according to the volume of the input atomized concentrated milk, and simulating the distribution condition of concentrated milk drops; further carrying out an atomization process and detecting the distribution condition of the liquid drops by using a detection module; finally, collecting measurement error data and carrying out an improved prediction model; therefore, the invention can carry out distribution prediction on concentrated emulsion liquid drops when carrying out atomization process each time, detect atomization results after atomization, update an atomization model according to the detection results, thereby greatly reducing uneven distribution condition during atomization work and improving the quality of milk powder.
Description
Technical Field
The invention discloses a particle size prediction method, belongs to the technical field of adult milk powder preparation processes, and particularly relates to a particle size prediction method and a particle size prediction system for an adult milk powder atomization process.
Background
The nutrition value of the milk powder is very high, so that not only children need to drink the milk powder, but also the sales volume of a lot of adult milk powder on the market is very high, and the adult milk powder is very popular; the adult milk powder is prepared by removing water from animal milk, contains rich nutrients such as protein, fat, saccharide, vitamins, minerals, soybean lecithin, etc., and can supplement nutrients required by human body after being eaten, so as to achieve the effects of providing energy for organism, enhancing immunity and preventing diseases.
The processing technology of milk powder is mainly divided into two major types, namely a dry process and a wet process, the two processing technologies are applied at present, the two processing technologies have the advantages of the two processing technologies, the production process flow of milk powder generally comprises raw material acceptance, pretreatment and standardization, concentration, atomization drying, cooling storage, packaging and finished product, the wet process technology is complex, more equipment is needed in the implementation process, a milk cleaning machine, a homogenizer, a sterilizing machine, concentrating equipment, drying equipment, packaging equipment, boilers, sewage treatment and the like are needed in the needed equipment. The milk powder produced by the wet process has good uniformity and stable physicochemical index. The dry process is relatively simple in production process, and the required equipment mainly comprises a mixer, a packaging machine and the like. However, the uniformity of the product produced by the dry process is not equal to that of the product produced by the wet process, but the thermosensitive nutritional ingredients are easy to add. The dry process has high requirements on workshops, and a CMP workshop is generally adopted.
In the milk powder production process, the atomization drying operation of the milk powder is needed, and the atomization drying adopts mechanical force, so that the concentrated milk is sprayed into very fine mist emulsion drops in a drying chamber through an atomizer, the surface area of the emulsion drops is increased, and the water evaporation rate is accelerated. Once the atomized emulsion drops are contacted with hot air blown in at the same time, the moisture is evaporated and removed instantaneously, so that the fine emulsion drops are dried into emulsion powder particles. The atomization drying is a close combination of atomization and drying, and concentrated milk can be dried into milk powder by a single process, and the two aspects affect the quality of the product at the same time.
However, in the prior art, when the milk powder atomizing process is carried out, concentrated milk is sprayed into very fine mist milk drops through an atomizer, if the process is used for a long time, the mist milk drops become larger or smaller, so that the milk drops are unevenly distributed in the falling process, and the quality of the milk powder is reduced.
Disclosure of Invention
The invention aims to: a particle size prediction method and a particle size prediction system for an adult milk powder atomization process are provided, and the problems are solved.
The technical scheme is as follows: a particle size prediction method and system for an adult milk powder atomization process comprises the following steps: basic information of the current input atomizing equipment is obtained, and a prediction model is established;
inputting a predictive model according to the volume of the input atomized concentrated milk quantity, and simulating the distribution condition of concentrated milk liquid drops;
performing an atomization process, and detecting the distribution condition of the liquid drops by using a detection module;
and collecting measurement error data and performing an improved prediction model.
In a further embodiment, a two-dimensional axisymmetric coordinate system is established based on the atomizing outlet as the origin, the axis is downward Z along the atomizing axis, the axis is radial R along the atomizing axis, a series of suitable physical constraints are established, and then the distribution when the system reaches maximum entropy under the constraints is found.
In a further embodiment, when the atomization process is performed, the external device records the process of the whole atomization process by using the detection device and detects the process;
firstly, a detection unit in the detection device collects image information, light signals are changed into electric signals through exposing discrete pixels, then electric quantity transfer is carried out, and finally, the electric quantity is converted into digital quantity by an A/D conversion module, namely, image gray values which can be identified by a computer:
in the method, in the process of the invention,is the gray value of the image generated per unit time, of->Is the image gray value;
secondly, analyzing the area size of the particle image, namely the number of pixels occupied by the particles, through image processing, and obtaining the real projection area of the particles through the size of the pixel elements of the camera and the calibrated magnification, thereby obtaining the equivalent projection area diameter of the particles;
finally, the particle size distribution of the liquid drops under different working conditions is obtained through statistics of a large number of particle images.
In a further embodiment, according to statistics of particle images, obtaining droplet size distribution of different working conditions, performing error analysis on droplet distribution, firstly selecting one or more particle droplets, calculating the number of pixels after processing the particle images, and obtaining actual particle size and actual size of the pixels;
secondly, carrying out processing analysis by using standard particle liquid drops and selected particle liquid drops, calculating the equivalent circular area diameter of shot particles, and respectively counting the number average diameters of two samples;
and finally, measuring errors including systematic errors, calibration error experimental errors, calculation errors and random errors.
In a further embodiment, the improved prediction model is a proposed improved model based on the measured error and the prediction model is updated based on the improved model; the method comprises the following specific steps:
step 1, performing an atomization performance test of an atomization device; repeatedly testing and measuring the diameter of the fog drops by changing the spraying parameters, and calculating the volume diameter, the number diameter, the uniformity and the standard deviation of the diameter of the fog drops which characterize the atomization quality, so as to record the corresponding relation between the spraying parameters and the parameters which characterize the atomization quality;
step 2, forming a training sample set; constructing a model training sample set using raw test dataFor test number, wherein the parameter +.>Is a multidimensional vector and is composed of main parameters affecting the diameter of fog drops; output parameter->Is thatParameters of the number test representing atomization quality;
step 3, initializing a particle swarm; setting particle swarm parameters in a definition spaceRandom generation of->Individual grainsSonComposition of the initial population->The method comprises the steps of carrying out a first treatment on the surface of the Randomly generating the initial velocity of each particle>Composing a velocity matrix->The method comprises the steps of carrying out a first treatment on the surface of the Individual optimal solution for each particle->Initial value is->Is set to an initial value of (1);
step 4, evaluating the fitness function of each particle; the fitness function can be calculated according to the mean square error of the residual error in the solving of a specific matrix equationThe smaller the adaptation capacity is, the stronger the adaptation capacity is;
step 5, comparing the current fitness of all particles of the population with the fitness of the best position of the population;
step 6, generating an improved model; firstly checking end conditions, if the conditions are met, ending optimizing and returning to the current optimal individual as a result, otherwise, turning to the step 2, and setting the end conditions to the maximum iteration times of optimizing or the evaluation value is smaller than the given precision; secondly, outputting optimal parameters, and bringing the parameters into a prediction model to establish an intelligent prediction improvement model based on the LS-SVM; process parameters of the test numberInputting the model response +.>The predicted value of the parameters representing the atomization quality of the test number is obtained; and (3) actual measured value->The comparison can be performed to obtain a prediction error, thereby improving the model.
In the above step 5, the specific steps of comparing the current fitness of all particles of the population with the fitness of the best position of the population are as follows:
step 51, assigning particles to the model in sequenceAnd->The two parameters are used for running a prediction model, and the optimal position of the particles and the optimal position of the whole particle swarm are determined according to the returned adaptive value;
step 52, next, comparing the adaptive value of each particle with the adaptive value of the optimal position experienced by the particle, and if the adaptive value is good, taking the optimal position as the current optimal position;
step 53, comparing the adaptive value of each particle with the adaptive value of the optimal position experienced by the whole particle swarm, and if the adaptive value is good, taking the adaptive value as the optimal position of the whole particle swarm at present;
step 54, updating the velocity and position of the particles
Step 55, if the termination condition is not satisfied, returning to step 52; otherwise, the algorithm is exited to obtain the optimal solution.
In a second aspect, a particle size prediction system for an adult milk powder atomization process, comprising:
an input unit for inputting basic information of the atomizing device;
an atomization unit performing atomization operation through an external atomization device;
the acquisition and detection unit is used for carrying out acquisition and image processing of an atomization process;
and the control unit is used for predicting, detecting and updating the particle size of the atomization process.
In a further embodiment, the acquisition detection unit is formed by integrating an analog-to-digital converter, a photosensitive cell array, a row array driver, a column timing control logic module, a DATA bus output interface, a control interface and the like on the same chip.
The beneficial effects are that: the invention discloses a particle size prediction method, belongs to the technical field of adult milk powder preparation processes, and particularly relates to a particle size prediction method and a particle size prediction system for an adult milk powder atomization process, wherein basic information of atomization equipment is input, and a prediction model is established; secondly, inputting a predictive model according to the volume of the input atomized concentrated milk, and simulating the distribution condition of concentrated milk drops; further carrying out an atomization process and detecting the distribution condition of the liquid drops by using a detection module; finally, collecting measurement error data and carrying out an improved prediction model; therefore, the invention can carry out distribution prediction on concentrated emulsion liquid drops when carrying out atomization process each time, detect atomization results after atomization, update an atomization model according to the detection results, thereby greatly reducing uneven distribution condition during atomization work and improving the quality of milk powder.
Drawings
FIG. 1 is a schematic representation of the process of the present invention.
Fig. 2 is a schematic illustration of the atomization of the present invention.
Fig. 3 is a schematic view of the image processing of the present invention.
Fig. 4 is a schematic diagram of the improved modeling method of the present invention.
Fig. 5 is a schematic diagram of a comparative particle of the present invention.
Fig. 6 is a system configuration diagram of the present invention.
Fig. 7 is a schematic diagram of an analog-to-digital converter of the present invention.
Detailed Description
The following description of the embodiments of the present invention will be made apparent and fully in view of the accompanying drawings, in which some, but not all embodiments of the invention are shown. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
A particle size prediction method and system for an adult milk powder atomization process comprises the following steps: basic information of the current input atomizing equipment is obtained, and a prediction model is established;
inputting a predictive model according to the volume of the input atomized concentrated milk quantity, and simulating the distribution condition of concentrated milk liquid drops;
performing an atomization process, and detecting the distribution condition of the liquid drops by using a detection module;
and collecting measurement error data and performing an improved prediction model.
In one embodiment, a two-dimensional axisymmetric coordinate system is established according to an atomization outlet as an origin, a Z axis is downwards used as an axis along an atomization axis, a R axis is used as an axis along an atomization radial direction, a series of proper physical constraint conditions are established, and then distribution when the system reaches maximum entropy under the constraint conditions is found;
specifically, firstly, selecting probability distribution of the number of liquid drops, wherein the characteristic diameter is a dimensionless diameter:
wherein,,representing the actual diameter of the droplet +.>Is of diameter +.>The number of drops of>Represents volume average diameter>The method comprises the steps of carrying out a first treatment on the surface of the Furthermore, the quantitative probability distribution can be +.>Conversion to a volume probability distribution->The method comprises the steps of carrying out a first treatment on the surface of the Namely:
the discrete state is adopted to replace the continuous state, 3 constraint conditions are used, namely a normalization condition, a specific surface area constraint condition and a large liquid drop constraint condition, and the normalization condition is as follows:
the following can be derived using the conservation of surface energy, conservation of mass, and surface energy division as constraints:
the entropy S of the system is maximized, and a Lagrange multiplier method is used for obtaining a quantity distribution function:
in the middle ofThe number distribution function is converted into a corresponding volume distribution function respectively as Lagrange multipliers to be determined:
the method comprises the steps of obtaining a quantity distribution function by taking normalization, surface energy conservation, mass conservation and partial surface energy conservation as constraint conditions without considering the speed factors of liquid drops:
thereby obtaining a prediction model according to the quantity distribution function.
In one embodiment, when the atomization process is performed, the external device records the process of the whole atomization process by using the detection device and detects the process;
firstly, a detection unit in the detection device collects image information, light signals are changed into electric signals through exposing discrete pixels, then electric quantity transfer is carried out, and finally, the electric quantity is converted into digital quantity by an A/D conversion module, namely, image gray values which can be identified by a computer:
in the method, in the process of the invention,is the gray value of the image generated per unit time, of->Is the image gray value;
secondly, analyzing the area size of the particle image, namely the number of pixels occupied by the particles, through image processing, and obtaining the real projection area of the particles through the size of the pixel elements of the camera and the calibrated magnification, thereby obtaining the equivalent projection area diameter of the particles;
finally, the particle size distribution of the liquid drops under different working conditions is obtained through statistics of a large number of particle images.
Specifically, the positions of the particles of the liquid drops are random in the working process, all the particles which are shot cannot be guaranteed to be just positioned on the focusing plane of the lens, when the concentration of the liquid drops is high, the phenomena of adhesion, overlapping and the like are easy to occur to the imaging of the particles, and the information of the particles is difficult to directly analyze and extract from the image, so that the image preprocessing is carried out before the area of the particles is calculated, and the preprocessing process mainly comprises the steps of image background homogenization, image enhancement, binarization, denoising, filling, incomplete particle removal at the edge, tiny and extremely large area particle removal and the like, and the method comprises the following steps of:
firstly, the acquired image has uneven background brightness due to uneven illumination and other factors, and local target information is likely to be lost when the image is directly subjected to binarization processing; the non-uniformity of background illumination can be corrected through top cap transformation or bottom cap transformation in a mathematical morphology method, a picture with uniform background brightness is obtained through bottom cap transformation, the contrast of an image after illumination correction is low, and in order to effectively divide and identify the image, the contrast between a target and the background and between the target and the target is improved through an image enhancement method;
secondly, binarizing the image; setting a certain threshold value, judging that all pixels with gray levels larger than the threshold value belong to a specific object, wherein the gray level value is 255, otherwise, the pixels are excluded from the object area, the gray level value is 0, and the background or exceptional object area is represented, so that the purposes of reducing the data amount of an image matrix and simplifying the salient particle outline of the image matrix are achieved; after background homogenization treatment, a better binarization effect can be obtained; in addition, in the image acquisition process, image noise can be generated by factors such as signal conversion, transmission, storage and the like, and the noise can be mistakenly processed into tiny target particles during image processing
Finally, image filling is carried out, bright spots are generated in the middle of partial liquid drop images due to the optical effect generated by the special shape of the liquid drops, holes are formed in the binarization process, difficulty is brought to subsequent feature extraction, the finally calculated particle projection area is influenced, and filling is needed.
In one embodiment, according to statistics of particle images, obtaining droplet size distribution of different working conditions, performing error analysis on droplet distribution, firstly selecting one or more particle droplets, calculating the number of pixels after particle image processing, and obtaining actual particle size and actual size of pixels;
secondly, carrying out processing analysis by using standard particle liquid drops and selected particle liquid drops, calculating the equivalent circular area diameter of shot particles, and respectively counting the number average diameters of two samples;
and finally, measuring errors including systematic errors, calibration error experimental errors, calculation errors and random errors.
In one embodiment, the improved prediction model is an improved prediction model suggested based on the measured error, and the prediction model is updated based on the improved model; the method comprises the following specific steps:
step 1, performing an atomization performance test of an atomization device; repeatedly testing and measuring the diameter of the fog drops by changing the spraying parameters, and calculating the volume diameter, the number diameter, the uniformity and the standard deviation of the diameter of the fog drops which characterize the atomization quality, so as to record the corresponding relation between the spraying parameters and the parameters which characterize the atomization quality;
step 2, forming a training sample set; constructing a model training sample set using raw test dataFor test number, wherein the parameter +.>Is a multidimensional vector and is composed of main parameters affecting the diameter of fog drops; output parameter->Is thatParameters of the number test representing atomization quality;
step 3, initializing a particle swarm; setting particle swarm parameters in a definition spaceRandom generation of->Individual particlesComposition of the initial population->The method comprises the steps of carrying out a first treatment on the surface of the Randomly generating the initial velocity of each particle>Composing a velocity matrix->The method comprises the steps of carrying out a first treatment on the surface of the Individual optimal solution for each particle->Initial value is->Is set to an initial value of (1);
step 4, evaluating the fitness function of each particle; the fitness function can be calculated according to the mean square error of the residual error in the solving of a specific matrix equationThe smaller the adaptation capacity is, the stronger the adaptation capacity is;
step 5, comparing the current fitness of all particles of the population with the fitness of the best position of the population;
step 6, generating an improved model; firstly checking end conditions, if the conditions are met, ending optimizing and returning to the current optimal individual as a result, otherwise, turning to the step 2, and setting the end conditions to the maximum iteration times of optimizing or the evaluation value is smaller than the given precision; secondly, outputting optimal parameters, and bringing the parameters into a prediction model to establish an intelligent prediction improvement model based on the LS-SVM; process parameters of the test numberInputting the model response +.>The predicted value of the parameters representing the atomization quality of the test number is obtained; and (3) actual measured value->The comparison can be performed to obtain a prediction error, thereby improving the model.
In the above step 5, the specific steps of comparing the current fitness of all particles of the population with the fitness of the best position of the population are as follows:
step 51, assigning particles to the model in sequenceAnd->The two parameters are used for running a prediction model, and the optimal position of the particles and the optimal position of the whole particle swarm are determined according to the returned adaptive value;
step 52, next, comparing the adaptive value of each particle with the adaptive value of the optimal position experienced by the particle, and if the adaptive value is good, taking the optimal position as the current optimal position;
step 53, comparing the adaptive value of each particle with the adaptive value of the optimal position experienced by the whole particle swarm, and if the adaptive value is good, taking the adaptive value as the optimal position of the whole particle swarm at present;
step 54, updating the speed and position of the particles;
step 55, if the termination condition is not satisfied, returning to step 52; otherwise, the algorithm is exited to obtain the optimal solution.
An adult milk powder atomizing process particle size prediction system comprising:
an input unit for inputting basic information of the atomizing device;
an atomization unit performing atomization operation through an external atomization device;
the acquisition and detection unit is used for carrying out acquisition and image processing of an atomization process;
and the control unit is used for predicting, detecting and updating the particle size of the atomization process.
In one embodiment, the acquisition detection unit is formed by integrating an analog-to-digital converter, a photosensitive cell array, a row array driver, a column timing control logic module, a DATA bus output interface, a control interface and the like on the same chip.
In one embodiment, the analog-to-digital converter is comprised of a comparator, a register, and a converter; the input end of the comparator inputs an analog signal, and is connected with the register and the converter through the comparator to be converted into a digital signal; the column timing control logic module is connected with the register and the converter.
In the above description, the analog-to-digital converter is implemented by a column timing control logic module according to the bisection principle, and the conversion process is as follows: after the conversion is started, the column sequence control logic module firstly obtains a voltage value which is about half of the full-scale output of the highest position 1 and the other positions 0 of the register after the content of the register passes through the converter, the voltage value is compared with an input signal in the comparator, the output of the comparator is fed back to the converter and corrected before the next comparison, the register is continuously subjected to comparison and shift operation under the clock drive of the logic control circuit until the conversion of the least significant bit is completed, and all bit values of the register are determined at the moment, so that the conversion is completed and the digital signal is output.
It is apparent that the above examples are given by way of illustration only and are not limiting of the embodiments. Other variations or modifications of the above teachings will be apparent to those of ordinary skill in the art. It is not necessary here nor is it exhaustive of all embodiments. While still being apparent from variations or modifications that may be made by those skilled in the art are within the scope of the invention.
Claims (6)
1. The particle size prediction method for the adult milk powder atomization process is characterized by comprising the following steps of:
basic information of the current input atomizing equipment is obtained, and a prediction model is established;
inputting the volume of the input atomized concentrated milk into a prediction model to simulate the distribution condition of concentrated milk drops;
performing an atomization process, and detecting the distribution condition of the liquid drops by using a detection module;
collecting measurement error data and performing an improved prediction model;
the prediction model is characterized in that an atomization outlet is taken as an origin, a two-dimensional axisymmetric coordinate system is established, a Z axis is downwards taken along an atomization axis, an R axis is taken along an atomization radial direction, a series of proper physical constraint conditions are established, and then distribution when the system reaches maximum entropy under the constraint conditions is found.
2. The method for predicting the particle size of an adult milk powder atomization process according to claim 1, wherein the external device records the whole atomization process by using the detection device and detects the process when the atomization process is performed;
firstly, a detection unit in the detection device collects image information, light signals are changed into electric signals through exposing discrete pixels, then electric quantity transfer is carried out, and finally, the electric quantity is converted into digital quantity by an A/D conversion module, namely, image gray values which can be identified by a computer:
in the method, in the process of the invention,is the gray value of the image generated per unit time, of->Is the image gray value;
secondly, analyzing the area size of the particle image, namely the number of pixels occupied by the particles, through image processing, and obtaining the real projection area of the particles through the size of the pixel elements of the camera and the calibrated magnification, thereby obtaining the equivalent projection area diameter of the particles;
finally, the particle size distribution of the liquid drops under different working conditions is obtained through statistics of a large number of particle images.
3. The method for predicting the particle size of the adult milk powder atomization process according to claim 2, wherein the error analysis of the droplet distribution is performed by obtaining the droplet size distribution of different working conditions according to statistics of the particle images, one or more particle droplets are selected first, the number of pixels after the particle image processing is calculated, and the actual particle size and the actual size of the pixels are obtained;
secondly, carrying out processing analysis by using standard particle liquid drops and selected particle liquid drops, calculating the equivalent circular area diameter of shot particles, and respectively counting the number average diameters of two samples;
and finally, measuring errors including systematic errors, calibration error experimental errors, calculation errors and random errors.
4. The method for predicting the particle size of an adult milk powder atomization process according to claim 1, wherein the improved prediction model is a proposed improved model according to measurement errors, and the improved prediction model is updated according to the improved model; the method comprises the following specific steps:
step 1, performing an atomization performance test of an atomization device;
step 2, forming a training sample set;
step 3, initializing a particle swarm;
step 4, evaluating the fitness function of each particle;
step 5, comparing the current fitness of all particles of the population with the fitness of the best position of the population;
and 6, generating an improved model.
5. An adult milk powder atomization process particle size prediction system, which adopts the adult milk powder atomization process particle size prediction method according to any one of claims 1-4, comprising:
an input unit for inputting basic information of the atomizing device;
an atomization unit performing atomization operation through an external atomization device;
the acquisition and detection unit is used for carrying out acquisition and image processing of an atomization process;
and the control unit is used for predicting, detecting and updating the particle size of the atomization process.
6. The system of claim 5, wherein the acquisition and detection unit is composed of an analog-to-digital converter, a photosensitive cell array, a row array driver, a column timing control logic module, a DATA bus output interface and a control interface integrated on the same chip.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211114055.6A CN115389380B (en) | 2022-09-14 | 2022-09-14 | Particle size prediction method and system for adult milk powder atomization process |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211114055.6A CN115389380B (en) | 2022-09-14 | 2022-09-14 | Particle size prediction method and system for adult milk powder atomization process |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115389380A CN115389380A (en) | 2022-11-25 |
| CN115389380B true CN115389380B (en) | 2023-10-27 |
Family
ID=84125837
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211114055.6A Active CN115389380B (en) | 2022-09-14 | 2022-09-14 | Particle size prediction method and system for adult milk powder atomization process |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115389380B (en) |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2003185559A (en) * | 2001-12-18 | 2003-07-03 | National Institute Of Advanced Industrial & Technology | Sampling probe for collecting droplet and method and apparatus for measuring distribution of grain size of atomized droplet |
| JP2006170902A (en) * | 2004-12-17 | 2006-06-29 | Japan Aerospace Exploration Agency | Particle size distribution measuring instrument |
| JP2010185691A (en) * | 2009-02-10 | 2010-08-26 | Doshisha | Sprayed particle imaging analysis system and analysis method |
| CN111402319A (en) * | 2020-02-21 | 2020-07-10 | 南通大学 | Atomized particle size measuring method and system |
| CN212658575U (en) * | 2020-08-07 | 2021-03-05 | 中国矿业大学(北京) | Device for testing particle sizes of different atomized liquid drops |
| CN113109215A (en) * | 2021-03-12 | 2021-07-13 | 南京航空航天大学 | Spray droplet particle size detection device and detection method |
| CN114925616A (en) * | 2022-05-31 | 2022-08-19 | 江苏大学 | Test device and method for testing heterogeneous system particle collision of spray fluidized bed |
-
2022
- 2022-09-14 CN CN202211114055.6A patent/CN115389380B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2003185559A (en) * | 2001-12-18 | 2003-07-03 | National Institute Of Advanced Industrial & Technology | Sampling probe for collecting droplet and method and apparatus for measuring distribution of grain size of atomized droplet |
| JP2006170902A (en) * | 2004-12-17 | 2006-06-29 | Japan Aerospace Exploration Agency | Particle size distribution measuring instrument |
| JP2010185691A (en) * | 2009-02-10 | 2010-08-26 | Doshisha | Sprayed particle imaging analysis system and analysis method |
| CN111402319A (en) * | 2020-02-21 | 2020-07-10 | 南通大学 | Atomized particle size measuring method and system |
| CN212658575U (en) * | 2020-08-07 | 2021-03-05 | 中国矿业大学(北京) | Device for testing particle sizes of different atomized liquid drops |
| CN113109215A (en) * | 2021-03-12 | 2021-07-13 | 南京航空航天大学 | Spray droplet particle size detection device and detection method |
| CN114925616A (en) * | 2022-05-31 | 2022-08-19 | 江苏大学 | Test device and method for testing heterogeneous system particle collision of spray fluidized bed |
Non-Patent Citations (3)
| Title |
|---|
| 可调式喷嘴雾化液滴粒径分布预测;刘妍卿;仇性启;王超;;工业加热(第04期) * |
| 空气雾化喷嘴的液滴雾化性能实验研究;刘丽艳;杨静;孔庆森;王初琛;谭蔚;;化学工业与工程(第03期);全文 * |
| 静电喷头雾化特性预测模型;刘春景;郑加强;王科元;;农业机械学报(第04期);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115389380A (en) | 2022-11-25 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN116188475B (en) | Intelligent control method, system and medium for automatic optical detection of appearance defects | |
| CN109377487B (en) | Fruit surface defect detection method based on deep learning segmentation | |
| CN116416250B (en) | Finished product detecting system of fast food canned product production line | |
| CN109800619B (en) | Image recognition method for citrus fruits in mature period | |
| Alipasandi et al. | Classification of three varieties of peach fruit using artificial neural network assisted with image processing techniques. | |
| Chalidabhongse et al. | 2D/3D vision-based mango's feature extraction and sorting | |
| CN102495068A (en) | Method and system for detecting chalky rice on line | |
| Gao et al. | Detection of wheat unsound kernels based on improved ResNet | |
| CN115389380B (en) | Particle size prediction method and system for adult milk powder atomization process | |
| CN111380813A (en) | Portable wheat seed multi-quality nondestructive testing device and testing method | |
| CN113326925A (en) | Density graph regression-based flower amount statistical method, system, equipment and medium for convolutional neural network | |
| CN113012244A (en) | Egg freshness detection method based on inclusion module and Attention mechanism | |
| CN110929787B (en) | Apple objective grading system based on image | |
| CN116559111A (en) | Sorghum variety identification method based on hyperspectral imaging technology | |
| CN113255255A (en) | Integrated circuit thermal vacuum test system | |
| CN103955693A (en) | Nine-ball computer-assisted detection identification method | |
| AU2020449440B2 (en) | Overlooking-type golf ball detection method, system and storage medium | |
| Gao et al. | Design of an efficient multi-objective recognition approach for 8-ball billiards vision system | |
| CN109034172B (en) | Product appearance defect detection method based on fuzzy relaxation constraint multi-core learning | |
| CN114612549A (en) | Method and device for predicting optimal fruiting picking time | |
| CN111294588B (en) | Device and method for analyzing white pixel distribution in CMOS image sensor | |
| CN106404779A (en) | Auxiliary detection method for corn seed purity based on form analysis | |
| Gan et al. | Design and testing of a machine-vision-based air-blow sorting platform for famous tea fresh leaves production | |
| Belbachir et al. | Real-time vision using a smart sensor system | |
| Chen et al. | Nondestructive testing of runny salted egg yolk based on improved ConvNeXt‐T |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| CB02 | Change of applicant information | ||
| CB02 | Change of applicant information |
Address after: 801-207-1, Hongqiao Road, Binhu District, Wuxi City, Jiangsu Province, 214000 Applicant after: Wuxi Wufan Food Co.,Ltd. Address before: 801-207-1, Hongqiao Road, Binhu District, Wuxi City, Jiangsu Province, 214000 Applicant before: Jiangsu Wufan Food Co.,Ltd. |
|
| GR01 | Patent grant | ||
| GR01 | Patent grant |