CN116989510A - Intelligent refrigeration method combining frosting detection and hot gas defrosting - Google Patents
Intelligent refrigeration method combining frosting detection and hot gas defrosting Download PDFInfo
- Publication number
- CN116989510A CN116989510A CN202311265396.8A CN202311265396A CN116989510A CN 116989510 A CN116989510 A CN 116989510A CN 202311265396 A CN202311265396 A CN 202311265396A CN 116989510 A CN116989510 A CN 116989510A
- Authority
- CN
- China
- Prior art keywords
- image
- hot gas
- electromagnetic valve
- pipe
- valve
- 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.)
- Pending
Links
- 238000010257 thawing Methods 0.000 title claims abstract description 71
- 238000005057 refrigeration Methods 0.000 title claims abstract description 41
- 238000000034 method Methods 0.000 title claims abstract description 39
- 238000001514 detection method Methods 0.000 title claims abstract description 27
- 238000012544 monitoring process Methods 0.000 claims abstract description 16
- 238000012549 training Methods 0.000 claims abstract description 14
- 238000007781 pre-processing Methods 0.000 claims abstract description 11
- 239000007788 liquid Substances 0.000 claims description 51
- 230000008020 evaporation Effects 0.000 claims description 48
- 238000001704 evaporation Methods 0.000 claims description 48
- 238000013527 convolutional neural network Methods 0.000 claims description 31
- 230000004913 activation Effects 0.000 claims description 6
- 238000010224 classification analysis Methods 0.000 claims description 6
- 238000010276 construction Methods 0.000 claims description 6
- 238000012423 maintenance Methods 0.000 claims description 5
- 238000012545 processing Methods 0.000 claims description 4
- 230000003044 adaptive effect Effects 0.000 claims description 3
- 238000011176 pooling Methods 0.000 claims description 3
- 230000008569 process Effects 0.000 claims description 3
- 239000003507 refrigerant Substances 0.000 claims description 3
- 230000011218 segmentation Effects 0.000 claims description 3
- 230000001105 regulatory effect Effects 0.000 claims description 2
- 239000003570 air Substances 0.000 description 36
- 238000010586 diagram Methods 0.000 description 4
- 238000005265 energy consumption Methods 0.000 description 4
- 230000015572 biosynthetic process Effects 0.000 description 3
- 230000009286 beneficial effect Effects 0.000 description 2
- 230000001276 controlling effect Effects 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 239000012080 ambient air Substances 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 230000005494 condensation Effects 0.000 description 1
- 238000009833 condensation Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 239000000155 melt Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 238000001931 thermography Methods 0.000 description 1
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Chemical compound O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/50—Context or environment of the image
- G06V20/52—Surveillance or monitoring of activities, e.g. for recognising suspicious objects
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F25—REFRIGERATION OR COOLING; COMBINED HEATING AND REFRIGERATION SYSTEMS; HEAT PUMP SYSTEMS; MANUFACTURE OR STORAGE OF ICE; LIQUEFACTION SOLIDIFICATION OF GASES
- F25B—REFRIGERATION MACHINES, PLANTS OR SYSTEMS; COMBINED HEATING AND REFRIGERATION SYSTEMS; HEAT PUMP SYSTEMS
- F25B47/00—Arrangements for preventing or removing deposits or corrosion, not provided for in another subclass
- F25B47/02—Defrosting cycles
- F25B47/022—Defrosting cycles hot gas defrosting
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F25—REFRIGERATION OR COOLING; COMBINED HEATING AND REFRIGERATION SYSTEMS; HEAT PUMP SYSTEMS; MANUFACTURE OR STORAGE OF ICE; LIQUEFACTION SOLIDIFICATION OF GASES
- F25B—REFRIGERATION MACHINES, PLANTS OR SYSTEMS; COMBINED HEATING AND REFRIGERATION SYSTEMS; HEAT PUMP SYSTEMS
- F25B49/00—Arrangement or mounting of control or safety devices
- F25B49/02—Arrangement or mounting of control or safety devices for compression type machines, plants or systems
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/77—Processing image or video features in feature spaces; using data integration or data reduction, e.g. principal component analysis [PCA] or independent component analysis [ICA] or self-organising maps [SOM]; Blind source separation
- G06V10/774—Generating sets of training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/82—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using neural networks
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F25—REFRIGERATION OR COOLING; COMBINED HEATING AND REFRIGERATION SYSTEMS; HEAT PUMP SYSTEMS; MANUFACTURE OR STORAGE OF ICE; LIQUEFACTION SOLIDIFICATION OF GASES
- F25B—REFRIGERATION MACHINES, PLANTS OR SYSTEMS; COMBINED HEATING AND REFRIGERATION SYSTEMS; HEAT PUMP SYSTEMS
- F25B2700/00—Sensing or detecting of parameters; Sensors therefor
- F25B2700/11—Sensor to detect if defrost is necessary
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Evolutionary Computation (AREA)
- Computing Systems (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Artificial Intelligence (AREA)
- Databases & Information Systems (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Software Systems (AREA)
- Mechanical Engineering (AREA)
- Thermal Sciences (AREA)
- General Engineering & Computer Science (AREA)
- Defrosting Systems (AREA)
Abstract
The invention provides an intelligent refrigeration method combining frosting detection and hot gas defrosting, which is applicable to the field of refrigeration systems. The method comprises the steps of arranging monitoring points, acquiring a monitoring image, preprocessing the image, constructing a label image data set, constructing an identification model, training the identification model and applying the identification model; the intelligent refrigeration method combining frosting detection and hot gas defrosting provided by the invention is suitable for the field of refrigeration systems, and has the advantages of improving energy efficiency, saving energy, reducing emission, protecting equipment, improving user experience, being widely applicable and the like.
Description
Technical Field
The invention relates to an intelligent refrigeration method combining frosting detection and hot gas defrosting, which is applicable to the field of refrigeration systems.
Background
Conventional refrigeration systems often suffer from frost formation during operation due to the fact that the evaporator surface temperature in the refrigeration cycle is below the dew point temperature in the ambient air, causing condensation of water vapor to form frost. Frosting not only reduces the efficiency of the refrigeration system, but also increases energy consumption and may cause damage to the equipment. Therefore, solving the problem of frosting becomes an important link in the design of an intelligent refrigerating system.
In the past studies, many methods have been proposed for frost detection and treatment, such as a temperature sensor-based method, a pressure change-based method, and the like. However, these methods have some limitations, such as inability to monitor frost formation in real time, large errors, or the need for additional hardware equipment support. Therefore, there is a need for a more accurate and reliable method of detecting frost formation. On the other hand, hot gas defrosting technology is widely used as a common frosting treatment means. The method rapidly melts the frost and discharges the frost outside the system by supplying hot air to the frosting surface. However, the existing hot gas defrosting method has the problems of high energy consumption and low defrosting efficiency. Therefore, there is a need for a more efficient and energy efficient hot gas defrosting technique.
In order to solve the problems, the study combines frosting detection and hot gas defrosting, and an intelligent refrigerating system is provided. The system monitors the frosting condition in real time by utilizing an advanced frosting detection technology, and realizes rapid frosting by precisely controlling hot gas supply. Through optimization and adjustment of an intelligent algorithm, the efficiency of the refrigerating system can be improved, the energy consumption is reduced, and comfortable use experience of a user is ensured.
Disclosure of Invention
The invention aims to provide an intelligent refrigeration method combining frosting detection and hot gas defrosting, which is suitable for the field of refrigeration systems and has the advantages of improving energy efficiency, saving energy, reducing emission, protecting equipment, improving user experience, being widely applicable and the like.
The aim of the invention can be achieved by adopting the following technical scheme:
s101, monitoring the arrangement of the point positions,
the arrangement of the monitoring points comprises the steps that an infrared image acquisition device is arranged at a specified position of the refrigeration equipment, the specified position is a unified refrigeration equipment frosting observation position, the infrared image acquisition device is a high-definition infrared thermal imager, and the fixed size of image acquisition is 12 multiplied by 12cm;
s102, acquiring a monitoring image,
the acquisition of the monitoring image comprises the steps of at a certain time interval t in the whole using process of the refrigeration equipment 1 Collecting infrared images, and finally obtaining a large number of images;
s103, preprocessing the image, wherein,
the image preprocessing comprises image enhancement and image denoising of an infrared image, wherein the image enhancement adopts a histogram equalization method, the image denoising adopts a Gaussian filter, the size range of the Gaussian filter is 5 multiplied by 5, the standard deviation of the Gaussian filter is 1.5, a processed image is finally obtained, and an image data set A { m } is constructed according to the processed image 1 ，m 2 ,...m n }；
S104, constructing a label image data set,
the construction of the label image data set comprises the steps of carrying out image classification analysis on each image to obtain a label 0 or a label 1, and carrying out image data set A { m } according to the labels 1 ，m 2 ,...m n Data set A0{ m } split into tag 0 1 ，m 2 ,...m i Data set A1{ m } and tag 1 1 ，m 2 ,...m j The step of image classification analysis is as follows: a) Selecting image dataset A { m 1 ，m 2 ,...m n One image m in } k B) applying an adaptive thresholding method to the image m k Performing frosting region segmentation to obtain a segmented image m k1 Calculate image m k Number of pixels N k Calculating a segmented image m k1 Number of pixels N k1 C) calculating the pixel ratio f, if f is less than or equal to 10%, giving the image m k Tag 0, if f>10% gives the image m k Label 1, the formula of the pixel ratio f is formula (1),
（1）
d) Steps a-c are cyclically performed until the image dataset a { m 1 ，m 2 ,...m n Each image in the image gets a label;
s105, constructing an identification model,
the construction of the identification model comprises the steps of constructing a convolutional neural network image identification model, wherein the convolutional neural network image identification model comprises 4 convolutional layers, 4 pooling layers, convolution kernels with characteristics extracted by 5 multiplied by 5 convolutional layers and 2 full-connection layers, the convolutional neural network image identification model adopts a piecewise linear activation function as an activation function of a convolutional neural network, and the convolutional neural network image identification model uses a cross entropy loss function to quantify the accuracy of the convolutional neural network image identification model;
s106, training the recognition model,
training of the recognition model, including tag 0-based dataset A0{ m 1 ，m 2 ,...m i Data set A1{ m } and tag 1 1 ，m 2 ,...m j Training the convolutional neural network image recognition model, wherein the training iteration number is set to 5000, and finally the trained convolutional neural network image recognition model is obtained;
s107, the application of the recognition model,
the application of the recognition model comprises the step of applying the trained convolutional neural network image recognition model to frost detection of refrigeration equipment, wherein the application steps are as follows:
a) The refrigerating device is operated at fixed time intervals t 1 An infrared image acquisition device is adopted to acquire an infrared image, and the infrared image is preprocessed to acquire a preprocessed image;
b) Inputting the preprocessed image into a trained convolutional neural network image recognition model to obtain a label 0 or a label 1;
c) When the tag 0 is obtained, no processing is carried out to continue refrigeration, and the next step is not carried out;
d) When the tag 1 is obtained, a hot gas defrosting system is started, the hot gas defrosting system comprises an evaporation device, a liquid supply electromagnetic valve, a pilot control valve, a hot gas electromagnetic valve, an electromagnetic control valve, a pneumatic electromagnetic valve, a bypass electromagnetic valve, a control pipe, a liquid supply main pipe and a return air main pipe, and the hot gas defrosting system is started by the following steps:
d1 Closing the liquid supply electromagnetic valve and enabling the evaporation device to continue to operate so as to evacuate the refrigerant in the evaporation device, wherein the evaporation device continues to operate for 2-10 minutes;
d2 Opening a pilot control valve, a hot gas electromagnetic valve and an electromagnetic control valve, wherein part of hot gas is introduced into a control pipe to close a pneumatic electromagnetic valve, and part of hot gas defrosts the evaporation device;
d3 When the evaporation device reaches a set temperature, closing the pilot control valve, the hot gas electromagnetic valve and the electromagnetic control valve, keeping the closed state of the liquid supply electromagnetic valve and the pneumatic electromagnetic valve, and opening a bypass electromagnetic valve; releasing the pressure in the evaporation device through the bypass electromagnetic valve until the pressure in the evaporation device is reduced to be within the pressure range of a low-pressure system, connecting a liquid supply main pipe and a return air main pipe of the hot gas defrosting system to a low-pressure circulation barrel, and completing defrosting when the pressure difference between the pressure in the evaporation device and the pressure in the low-pressure circulation barrel is lower than 1.25 bar;
e) Re-opening the liquid supply electromagnetic valve and the pneumatic electromagnetic valve, and closing the bypass electromagnetic valve to enable the evaporation device to continue refrigerating;
f) And c, acquiring an infrared image by adopting an infrared image acquisition device, preprocessing the infrared image to acquire a preprocessed image, and circularly executing the steps b-f.
In the step S107, the hot gas defrosting system includes a plurality of parallel-connected evaporation devices, the hot gas defrosting system includes a liquid supply main pipe and a return air main pipe, and a plurality of liquid supply branch pipes are split from the liquid supply main pipe and correspondingly connected with the plurality of parallel-connected evaporation devices; each evaporation device is connected with an air return branch pipe, all the air return branch pipes are converged on the air return main pipe, a liquid supply electromagnetic valve and a first check valve are arranged on each liquid supply branch pipe, a normally open pneumatic electromagnetic valve is arranged on each air return branch pipe, a bypass electromagnetic valve with the path smaller than that of the pneumatic electromagnetic valve is connected in parallel on each pneumatic electromagnetic valve, the hot air defrosting system further comprises a hot air main pipe, a pilot control valve is arranged on each hot air main pipe, a plurality of hot air branch pipes corresponding to a plurality of air return branch pipes are branched on each hot air main pipe, one end of each hot air branch pipe is connected to the hot air main pipe, the other end of each hot air branch pipe is connected to the air return branch pipe between each pneumatic electromagnetic valve and each evaporation device, and the hot air branch pipe is provided with a hot air electromagnetic valve and a second check valve; a control pipe is branched from the hot gas branch pipe and connected to the pneumatic electromagnetic valve, and an electromagnetic control valve is arranged on the control pipe; the hot gas defrosting system further comprises a return pipe, one end of the return pipe is connected to the air return branch pipe behind the pneumatic electromagnetic valve, the other end of the return pipe is connected to the liquid supply branch pipe between the first check valve and each evaporation device, and the return pipe is provided with an overflow valve;
the hot gas defrosting system is further characterized in that the liquid supply branch pipe is further provided with a first maintenance valve, a first filter and a manual regulating valve;
the hot gas defrosting system is further characterized in that a stop valve is arranged on the hot gas branch pipe, and hot gas on the hot gas main pipe enters the control pipe again through the stop valve;
the hot gas defrosting system is further characterized in that a second maintenance valve and a second filter are arranged on the control pipe;
the hot gas defrosting system is characterized in that a liquid supply main pipe and a return air main pipe of the hot gas defrosting system are connected to a low-pressure circulating barrel, the low-pressure circulating barrel is sequentially connected with a compressor, an oil separator, a condenser and a liquid storage device through pipelines, and the liquid storage device is connected with the low-pressure circulating barrel again to form circulation; the oil separator delivers hot gas obtained from the compressor to the condenser and the hot gas main pipe, respectively.
The beneficial effects of the invention are as follows:
the beneficial effects of the invention are as follows: according to the invention, the convolutional neural network image recognition model is trained based on the thermal imaging picture in the aspect of frosting detection, and the system pressure fluctuation caused by defrosting is reduced by controlling a plurality of pressures of the system in the aspect of a hot gas defrosting method, so that the influence on the refrigerating capacity of the system is avoided, each evaporation device can be independently defrosted without causing great influence on other evaporation devices, the current situation that the accuracy and reliability of the existing frosting detection method are insufficient is solved, and the problems of high energy consumption and low defrosting efficiency of the existing hot gas defrosting method are solved. The intelligent refrigerating method combining frosting detection and hot gas defrosting has the advantages of improving energy efficiency, saving energy, reducing emission, protecting equipment, improving user experience, being widely applicable and the like.
Drawings
Fig. 1: the invention relates to a flow chart of an intelligent refrigeration method combining frosting detection and hot gas defrosting.
Fig. 2: the invention relates to an overall structure schematic diagram of a hot gas defrosting system of an intelligent refrigeration method combining frosting detection and hot gas defrosting.
Fig. 3: and the refrigerating operation schematic diagram of the hot gas defrosting system is shown.
Fig. 4: is one of defrosting operation schematic diagrams of the hot gas defrosting system.
Fig. 5: and the second defrosting operation schematic diagram of the hot gas defrosting system.
In the figure: 10. an evaporation device; 20. a liquid supply main pipe; 21. a liquid supply branch pipe; 211. a first service valve; 212. a first filter; 213. a liquid supply electromagnetic valve; 214. a first check valve; 215. a manual adjustment valve; 30. a return air main pipe; 31. an air return branch pipe; 311. a pneumatic solenoid valve; 312. a bypass solenoid valve; 32. a return pipe; 321. an overflow valve; 40. a hot gas main pipe; 41. a hot gas branch pipe; 411. a stop valve; 412. a hot gas solenoid valve; 413. a second check valve; 42. a control tube; 421. a second service valve; 422. a second filter; 423. an electromagnetic control valve; 43. a pilot control valve; 50. a low pressure circulation tank; 60. a compressor; 70. an oil separator; 80. a condenser; 90. a reservoir.
Detailed Description
Specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings; it should be understood that the particular embodiments presented herein are illustrative and explanatory only and are not restrictive of the invention.
The following is a specific embodiment of an intelligent refrigeration method combining frost detection and hot gas defrosting.
As shown in fig. 1, a flow chart of an intelligent refrigeration method combining frosting detection and hot gas defrosting according to the invention is shown.
S101, monitoring the arrangement of the point positions,
the arrangement of the monitoring points comprises the steps that an infrared image acquisition device is arranged at a specified position of the refrigeration equipment, the specified position is a unified refrigeration equipment frosting observation position, the infrared image acquisition device is a high-definition infrared thermal imager, and the fixed size of image acquisition is 12 multiplied by 12cm;
s102, acquiring a monitoring image,
the acquisition of the monitoring image comprises the steps of at a certain time interval t in the whole using process of the refrigeration equipment 1 Collecting infrared images, and finally obtaining a large number of images;
s103, preprocessing the image, wherein,
the image preprocessing comprises image enhancement and image denoising of an infrared image, wherein the image enhancement adopts a histogram equalization method, the image denoising adopts a Gaussian filter, the size range of the Gaussian filter is 5 multiplied by 5, the standard deviation of the Gaussian filter is 1.5, a processed image is finally obtained, and an image data set A { m } is constructed according to the processed image 1 ，m 2 ,...m n }；
S104, constructing a label image data set,
the construction of the label image data set comprises the steps of carrying out image classification analysis on each image to obtain a label 0 or a label 1, and carrying out image data set A { m } according to the labels 1 ，m 2 ,...m n Data set A0{ m } split into tag 0 1 ，m 2 ,...m i Data set A1{ m } and tag 1 1 ，m 2 ,...m j The step of image classification analysis is as follows: a) Selecting image dataset A { m 1 ，m 2 ,...m n One image m in } k B) applying an adaptive thresholding method to the image m k Performing frosting region segmentation to obtain a segmented image m k1 Calculate image m k Number of pixels N k Calculating a segmented image m k1 Number of pixels N k1 C) calculating the pixel ratio f, if f is less than or equal to 10%, giving the image m k Tag 0, if f>10% gives the image m k Label 1, the formula of the pixel ratio f is formula (1),
（1）
d) Steps a-c are cyclically performed until the image dataset a { m 1 ，m 2 ,...m n Each image in the image gets a label;
s105, constructing an identification model,
the construction of the identification model comprises the steps of constructing a convolutional neural network image identification model, wherein the convolutional neural network image identification model comprises 4 convolutional layers, 4 pooling layers, convolution kernels with characteristics extracted by 5 multiplied by 5 convolutional layers and 2 full-connection layers, the convolutional neural network image identification model adopts a piecewise linear activation function as an activation function of a convolutional neural network, and the convolutional neural network image identification model uses a cross entropy loss function to quantify the accuracy of the convolutional neural network image identification model;
s106, training the recognition model,
training of the recognition model, including tag 0-based dataset A0{ m 1 ，m 2 ,...m i Data set A1{ m } and tag 1 1 ，m 2 ,...m j Training the convolutional neural network image recognition model, wherein the training iteration number is set to 5000, and finally the trained convolutional neural network image recognition model is obtained;
s107, the application of the recognition model,
the application of the recognition model comprises the step of applying the trained convolutional neural network image recognition model to frost detection of refrigeration equipment, wherein the application steps are as follows:
a) The refrigerating device is operated at fixed time intervals t 1 An infrared image acquisition device is adopted to acquire an infrared image, and the infrared image is preprocessed to acquire a preprocessed image;
b) Inputting the preprocessed image into a trained convolutional neural network image recognition model to obtain a label 0 or a label 1;
c) When the tag 0 is obtained, no processing is carried out to continue refrigeration, and the next step is not carried out;
d) When the tag 1 is obtained, a hot gas defrosting system is started, the hot gas defrosting system comprises an evaporation device, a liquid supply electromagnetic valve, a pilot control valve, a hot gas electromagnetic valve, an electromagnetic control valve, a pneumatic electromagnetic valve, a bypass electromagnetic valve, a control pipe, a liquid supply main pipe and a return air main pipe, and the hot gas defrosting system is started by the following steps:
d1 Closing the liquid supply electromagnetic valve and enabling the evaporation device to continue to operate so as to evacuate the refrigerant in the evaporation device, wherein the evaporation device continues to operate for 2-10 minutes;
d2 Opening a pilot control valve, a hot gas electromagnetic valve and an electromagnetic control valve, wherein part of hot gas is introduced into a control pipe to close a pneumatic electromagnetic valve, and part of hot gas defrosts the evaporation device;
d3 When the evaporation device reaches a set temperature, closing the pilot control valve, the hot gas electromagnetic valve and the electromagnetic control valve, keeping the closed state of the liquid supply electromagnetic valve and the pneumatic electromagnetic valve, and opening a bypass electromagnetic valve; releasing the pressure in the evaporation device through the bypass electromagnetic valve until the pressure in the evaporation device is reduced to be within the pressure range of a low-pressure system, connecting a liquid supply main pipe and a return air main pipe of the hot gas defrosting system to a low-pressure circulation barrel, and completing defrosting when the pressure difference between the pressure in the evaporation device and the pressure in the low-pressure circulation barrel is lower than 1.25 bar;
e) Re-opening the liquid supply electromagnetic valve and the pneumatic electromagnetic valve, and closing the bypass electromagnetic valve to enable the evaporation device to continue refrigerating;
f) And c, acquiring an infrared image by adopting an infrared image acquisition device, preprocessing the infrared image to acquire a preprocessed image, and circularly executing the steps b-f.
In the step S107, the hot air defrosting system includes a plurality of evaporation devices 10 connected in parallel, and the hot air defrosting system includes a liquid supply main pipe 20 and a return air main pipe 30, and a plurality of liquid supply branch pipes 21 are branched from the liquid supply main pipe 20 and connected to the plurality of evaporation devices 10; each evaporation device 10 is connected with an air return branch pipe 31, all the air return branch pipes 31 are converged on the air return main pipe 30, the liquid supply branch pipe 21 is provided with a liquid supply electromagnetic valve 213 and a first check valve 214, the air return branch pipe 31 is provided with a normally open pneumatic electromagnetic valve 311, and the pneumatic electromagnetic valve 311 is connected with a bypass electromagnetic valve 312 with a smaller drift diameter than the pneumatic electromagnetic valve 311 in parallel;
the hot gas defrosting system further comprises a hot gas main pipe 40, a pilot control valve 43 is arranged on the hot gas main pipe 40, a plurality of hot gas branch pipes 41 corresponding to the plurality of air return branch pipes 31 are branched from the hot gas main pipe 40, one end of each hot gas branch pipe 41 is connected with the hot gas main pipe 40, the other end of each hot gas branch pipe 41 is connected with the air return branch pipe 31 between the pneumatic electromagnetic valve 311 and the evaporation device 10, and the hot gas branch pipe 41 is provided with a hot gas electromagnetic valve 412 and a second check valve 413; a control pipe 42 is branched from the hot gas branch pipe 41 and connected to the pneumatic electromagnetic valve 311, and an electromagnetic control valve 423 is arranged on the control pipe 42; comprises a return pipe 32, one end of the return pipe 32 is connected to the air return branch pipe 31 behind the pneumatic electromagnetic valve 311, the other end is connected to the liquid supply branch pipe 21 between the first check valve 214 and the evaporation device 10, and the return pipe 32 is provided with an overflow valve 321;
a further hot gas defrosting system is characterized in that the liquid supply branch pipe is also provided with a first maintenance valve 211, a first filter 212 and a manual adjusting valve 215;
a further hot gas defrosting system is characterized in that a stop valve 411 is arranged on the hot gas branch pipe, and the hot gas on the hot gas main pipe 40 reenters the control pipe 42 through the stop valve 411;
a further hot gas defrosting system is characterized in that the control pipe is provided with a second maintenance valve 421 and a second filter 422;
a hot gas defrosting system as described above is further characterized in that the main liquid supply pipe 20 and the main return pipe 30 of the hot gas defrosting system are connected to a low pressure circulation tank 50, the low pressure circulation tank 50 is sequentially connected to a compressor 60, an oil separator 70, a condenser 80 and a liquid storage device 90 through pipes, and the liquid storage device 90 is connected to the low pressure circulation tank 50 again to form a circulation; the oil separator 70 delivers hot gas obtained from the compressor to the condenser 80 and the hot gas main pipe 40, respectively.
In the embodiment, the invention discloses an intelligent refrigeration method combining frosting detection and hot gas defrosting, which comprises the steps of arranging monitoring points, acquiring monitoring images, preprocessing images, constructing a label image data set, constructing an identification model, training the identification model and applying the identification model; the intelligent refrigeration method combining frosting detection and hot gas defrosting provided by the invention is suitable for the field of refrigeration systems, and has the advantages of improving energy efficiency, saving energy, reducing emission, protecting equipment, improving user experience, being widely applicable and the like.
The foregoing description of the preferred embodiments of the invention is not intended to limit the invention to the particular embodiments disclosed, but on the contrary, the intention is to cover all modifications, equivalents, and alternatives falling within the spirit and scope of the invention.
Claims (6)
1. An intelligent refrigeration method combining frosting detection and hot gas defrosting is characterized by comprising the following steps:
1) Monitoring the arrangement of the point positions;
2) Monitoring the acquisition of an image;
3) Preprocessing an image;
4) Constructing a label image dataset;
5) Building an identification model;
6) Training of an identification model;
7) Identifying an application of the model;
the arrangement of the monitoring points comprises the steps that an infrared image acquisition device is arranged at a specified position of the refrigeration equipment, the specified position is a unified refrigeration equipment frosting observation position, the infrared image acquisition device is a high-definition infrared thermal imager, and the fixed size of image acquisition is 12 multiplied by 12cm;
the acquisition of the monitoring image comprises the steps of at a certain time interval t in the whole using process of the refrigeration equipment 1 Collecting infrared images, and finally obtaining a large number of images;
the image preprocessing comprises image enhancement and image denoising of an infrared image, wherein the image enhancement adopts a histogram equalization method, the image denoising adopts a Gaussian filter, the size range of the Gaussian filter is 5 multiplied by 5, the standard deviation of the Gaussian filter is 1.5, a processed image is finally obtained, and an image data set A { m } is constructed according to the processed image 1 ，m 2 ,...m n }；
The construction of the label image data set comprises the steps of carrying out image classification analysis on each image to obtain a label 0 or a label 1, and carrying out image data set A { m } according to the labels 1 ，m 2 ,...m n Data set A0{ m } split into tag 0 1 ，m 2 ,...m i Data set A1{ m } and tag 1 1 ，m 2 ,...m j The step of image classification analysis is as follows:
a) Selecting image dataset A { m 1 ，m 2 ,...m n One image m in } k ；
b) Image m is processed by adopting an adaptive threshold method k Performing frosting region segmentation to obtain a segmented image m k1 Calculate image m k Number of pixels N k Calculating a segmented image m k1 Number of pixels N k1 ；
c) Calculating the pixel ratio f, and giving an image m if f is less than or equal to 10 percent k Tag 0, if f>10% gives the image m k A label 1, wherein the formula of the pixel ratio f is shown as formula (1);
（1）
d) Steps a-c are cyclically performed until the image dataset a { m 1 ，m 2 ,...m n Each image in the image gets a label;
the construction of the identification model comprises the steps of constructing a convolutional neural network image identification model, wherein the convolutional neural network image identification model comprises 4 convolutional layers, 4 pooling layers, convolution kernels with characteristics extracted by 5 multiplied by 5 convolutional layers and 2 full-connection layers, the convolutional neural network image identification model adopts a piecewise linear activation function as an activation function of a convolutional neural network, and the convolutional neural network image identification model uses a cross entropy loss function to quantify the accuracy of the convolutional neural network image identification model;
training of the recognition model, including tag 0-based dataset A0{ m 1 ，m 2 ,...m i Data set A1{ m } and tag 1 1 ，m 2 ,...m j Training the convolutional neural network image recognition model, wherein the training iteration number is set to 5000, and finally the trained convolutional neural network image recognition model is obtained;
the application of the identification model comprises the step of applying the trained convolutional neural network image identification model to frost detection and processing of refrigeration equipment, wherein the application steps of the identification model are as follows:
a) The refrigerating device is operated at fixed time intervals t 1 An infrared image acquisition device is adopted to acquire an infrared image, and the infrared image is preprocessed to acquire a preprocessed image;
b) Inputting the preprocessed image into a trained convolutional neural network image recognition model to obtain a label 0 or a label 1;
c) When the tag 0 is obtained, no processing is carried out to continue refrigeration, and the next step is not carried out;
d) When the tag 1 is obtained, a hot gas defrosting system is started, the hot gas defrosting system comprises an evaporation device, a liquid supply electromagnetic valve, a pilot control valve, a hot gas electromagnetic valve, an electromagnetic control valve, a pneumatic electromagnetic valve, a bypass electromagnetic valve, a control pipe, a liquid supply main pipe and a return air main pipe, and the hot gas defrosting system is started by the following steps:
d1 Closing the liquid supply electromagnetic valve and enabling the evaporation device to continue to operate so as to evacuate the refrigerant in the evaporation device, wherein the evaporation device continues to operate for 2-10 minutes;
d2 Opening a pilot control valve, a hot gas electromagnetic valve and an electromagnetic control valve, wherein part of hot gas is introduced into a control pipe to close a pneumatic electromagnetic valve, and part of hot gas defrosts the evaporation device;
d3 When the evaporation device reaches a set temperature, closing the pilot control valve, the hot gas electromagnetic valve and the electromagnetic control valve, keeping the closed state of the liquid supply electromagnetic valve and the pneumatic electromagnetic valve, and opening a bypass electromagnetic valve; releasing the pressure in the evaporation device through the bypass electromagnetic valve until the pressure in the evaporation device is reduced to be within the pressure range of a low-pressure system, connecting a liquid supply main pipe and a return air main pipe of the hot gas defrosting system to a low-pressure circulation barrel, and completing defrosting when the pressure difference between the pressure in the evaporation device and the pressure in the low-pressure circulation barrel is lower than 1.25 bar;
e) Re-opening the liquid supply electromagnetic valve and the pneumatic electromagnetic valve, and closing the bypass electromagnetic valve to enable the evaporation device to continue refrigerating;
f) And c, acquiring an infrared image by adopting an infrared image acquisition device, preprocessing the infrared image to acquire a preprocessed image, and circularly executing the steps b-f.
2. The intelligent refrigeration method combining frost detection and hot gas defrosting according to claim 1, wherein in step 7), the hot gas defrosting system comprises a plurality of evaporation devices used in parallel, the hot gas defrosting system comprises a liquid supply main pipe and a return air main pipe, and a plurality of liquid supply branch pipes are branched from the liquid supply main pipe and correspondingly connected with the plurality of evaporation devices used in parallel; each evaporation device is connected with an air return branch pipe, all the air return branch pipes are converged on the air return main pipe, a liquid supply electromagnetic valve and a first check valve are arranged on each liquid supply branch pipe, a normally open pneumatic electromagnetic valve is arranged on each air return branch pipe, a bypass electromagnetic valve with the path smaller than that of the pneumatic electromagnetic valve is connected in parallel on each pneumatic electromagnetic valve, the hot air defrosting system further comprises a hot air main pipe, a pilot control valve is arranged on each hot air main pipe, a plurality of hot air branch pipes corresponding to a plurality of air return branch pipes are branched on each hot air main pipe, one end of each hot air branch pipe is connected to the hot air main pipe, the other end of each hot air branch pipe is connected to the air return branch pipe between each pneumatic electromagnetic valve and each evaporation device, and the hot air branch pipe is provided with a hot air electromagnetic valve and a second check valve; a control pipe is branched from the hot gas branch pipe and connected to the pneumatic electromagnetic valve, and an electromagnetic control valve is arranged on the control pipe; the hot gas defrosting system further comprises a return pipe, one end of the return pipe is connected to the air return branch pipe behind the pneumatic electromagnetic valve, the other end of the return pipe is connected to the liquid supply branch pipe between the first check valve and each evaporation device, and the return pipe is provided with an overflow valve.
3. The intelligent refrigeration method combining frosting detection and hot gas defrosting according to claim 2, wherein the liquid supply branch pipe is further provided with a first maintenance valve, a first filter and a manual regulating valve.
4. The intelligent refrigeration method combining frosting detection and hot gas defrosting according to claim 2, wherein a stop valve is arranged on the hot gas branch pipe, and hot gas on the hot gas main pipe reenters the control pipe through the stop valve.
5. The intelligent refrigeration method combining frost detection and hot gas defrosting as claimed in claim 2, wherein the control tube is provided with a second service valve and a second filter.
6. The intelligent refrigeration method combining frosting detection and hot gas defrosting according to claim 2, wherein a liquid supply main pipe and a return air main pipe of the hot gas defrosting system are connected to a low-pressure circulation barrel, the low-pressure circulation barrel is sequentially connected with a compressor, an oil separator, a condenser and a liquid storage device through pipelines, and the liquid storage device is connected with the low-pressure circulation barrel again to form circulation; the oil separator delivers hot gas obtained from the compressor to the condenser and the hot gas main pipe, respectively.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311265396.8A CN116989510A (en) | 2023-09-28 | 2023-09-28 | Intelligent refrigeration method combining frosting detection and hot gas defrosting |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311265396.8A CN116989510A (en) | 2023-09-28 | 2023-09-28 | Intelligent refrigeration method combining frosting detection and hot gas defrosting |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116989510A true CN116989510A (en) | 2023-11-03 |
Family
ID=88527067
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311265396.8A Pending CN116989510A (en) | 2023-09-28 | 2023-09-28 | Intelligent refrigeration method combining frosting detection and hot gas defrosting |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116989510A (en) |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107860099A (en) * | 2017-09-15 | 2018-03-30 | 珠海格力电器股份有限公司 | Air conditioner frosting detection method and device, storage medium and equipment |
| CN109146000A (en) * | 2018-09-07 | 2019-01-04 | 电子科技大学 | A kind of method and device for improving convolutional neural networks based on frost weight |
| CN110793246A (en) * | 2019-11-29 | 2020-02-14 | 广州冰泉制冷设备有限责任公司 | Hot gas defrosting system and hot gas defrosting method |
| CN211424781U (en) * | 2019-11-29 | 2020-09-04 | 广州冰泉制冷设备有限责任公司 | Hot gas defrosting system |
| CN113536989A (en) * | 2021-06-29 | 2021-10-22 | 广州博通信息技术有限公司 | Refrigerator frosting monitoring method and system based on camera video frame-by-frame analysis |
| JP6980958B1 (en) * | 2021-06-23 | 2021-12-15 | 中国科学院西北生態環境資源研究院 | Rural area classification garbage identification method based on deep learning |
| CN115690668A (en) * | 2021-07-23 | 2023-02-03 | 菅政 | Damage identification technology based on image semantic segmentation algorithm |
| CN116229053A (en) * | 2022-12-07 | 2023-06-06 | 安徽唯嵩光电科技有限公司 | Dynamic adjustment method for red date machine striker plate based on neural network |
-
2023
- 2023-09-28 CN CN202311265396.8A patent/CN116989510A/en active Pending
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107860099A (en) * | 2017-09-15 | 2018-03-30 | 珠海格力电器股份有限公司 | Air conditioner frosting detection method and device, storage medium and equipment |
| CN109146000A (en) * | 2018-09-07 | 2019-01-04 | 电子科技大学 | A kind of method and device for improving convolutional neural networks based on frost weight |
| CN110793246A (en) * | 2019-11-29 | 2020-02-14 | 广州冰泉制冷设备有限责任公司 | Hot gas defrosting system and hot gas defrosting method |
| CN211424781U (en) * | 2019-11-29 | 2020-09-04 | 广州冰泉制冷设备有限责任公司 | Hot gas defrosting system |
| JP6980958B1 (en) * | 2021-06-23 | 2021-12-15 | 中国科学院西北生態環境資源研究院 | Rural area classification garbage identification method based on deep learning |
| CN113536989A (en) * | 2021-06-29 | 2021-10-22 | 广州博通信息技术有限公司 | Refrigerator frosting monitoring method and system based on camera video frame-by-frame analysis |
| CN115690668A (en) * | 2021-07-23 | 2023-02-03 | 菅政 | Damage identification technology based on image semantic segmentation algorithm |
| CN116229053A (en) * | 2022-12-07 | 2023-06-06 | 安徽唯嵩光电科技有限公司 | Dynamic adjustment method for red date machine striker plate based on neural network |
Non-Patent Citations (1)
| Title |
|---|
| 王鑫 等: "基于融合边缘变化信息全卷积神经网络的遥感图像变化检测", 《电子与信息学报》, vol. 44, no. 5, pages 1695 - 1703 * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109612027B (en) | Method and control system for analyzing frosting of air source heat pump based on microscopic photographing | |
| CN106152644B (en) | Defrosting control method and system for heat pump air conditioning unit | |
| CN105157292B (en) | The control method for electronic expansion valve and device of air-cooled screw unit | |
| CN110425112B (en) | Compressor liquid impact prevention air conditioner and compressor liquid impact prevention control method | |
| CN110030665B (en) | Defrosting control method and device and air conditioner | |
| CN101929755A (en) | Image recognition technology-based air source heat pump defrosting system and control method thereof | |
| CN111156657A (en) | Air conditioner frosting state determining method and device | |
| CN108981081A (en) | A kind of intelligent defrosting method | |
| CN203949402U (en) | A kind of cooling device with hot gas defrosting function | |
| KR20130102219A (en) | Heat pump chiller system by non-frosting continuous operating the heat exchanger and defrost method | |
| Wang et al. | A novel defrosting initiation strategy based on convolutional neural network for air-source heat pump | |
| CN116989510A (en) | Intelligent refrigeration method combining frosting detection and hot gas defrosting | |
| CN111928339A (en) | Outdoor unit heat exchanger, control method and control unit thereof and air conditioning system | |
| Chen et al. | Applying deep learning-based regional feature recognition from macro-scale image to assist energy saving and emission reduction in industrial energy systems | |
| CN110410938A (en) | Defrosting control method for air conditioner | |
| CN110260467B (en) | Air conditioner and anti-freezing protection control method and control device thereof | |
| CN113536989B (en) | Refrigerator frosting monitoring method and system based on frame-by-frame analysis of camera video | |
| US20210164711A1 (en) | Cooling system for low temperature storage | |
| CN207622306U (en) | One kind detecting air source heat pump defrosting control system based on infrared thermal imaging | |
| CN108981241A (en) | A kind of fin top row bundled tube freezer evaporator and its defrosting water collecting device | |
| CN112197455B (en) | Air conditioning unit capable of recycling refrigerant and control method thereof | |
| CN212274093U (en) | Outer quick-witted heat exchanger and air conditioning system | |
| CN211424781U (en) | Hot gas defrosting system | |
| CN114001505A (en) | Device and method for early warning and processing low pressure in refrigerator after door closing of ultralow temperature refrigerator | |
| WO2020150889A1 (en) | Control system for sequential defrosting of moist-air heat exchanger |
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 |