Medical refrigerator compressor cooling management system and method based on image recognition

Abstract

The present invention relates to the field of refrigeration control technology, specifically to a medical refrigerator compressor cooling management system and method based on image recognition, the system includes: a hot zone identification module, a temperature partition calculation module, a wind valve control module, an air duct optimization module, and an energy consumption evaluation module. In the present invention, the temperature value is extracted by infrared images, and the pixel temperature difference and temperature gradient analysis are combined to accurately identify the local heat concentration area, improve the heat source distribution perception ability, divide the temperature zone based on the temperature gradient and the arrangement of the heat dissipation fins, and generate temperature partition data, so that the cooling control dynamically matches the temperature difference, enhances the heat dissipation uniformity, combines the high temperature partition with the air duct outlet angle calculation, dynamically adjusts the air valve, realizes the accurate distribution of air volume, avoids insufficient or excessive cooling, and optimizes the air duct in combination with the air flow pressure gradient calculation to adjust the guide vane to make the cooling path more reasonable, calculates the energy consumption by the fan wind speed change rate, and optimizes the fan strategy in combination with historical data to reduce energy consumption.

Description

Technical Field

[0001] The present invention relates to the technical field of refrigeration control, and in particular to a medical refrigerator compressor cooling management system and method based on image recognition. Background Art

[0002] The field of refrigeration control technology includes various control methods and devices for regulating and maintaining low-temperature environments, involving temperature regulation of cooling systems, compressor operation management, heat exchange optimization, and refrigerant flow control. The core content of this technical field includes precise control of the operating status of components such as compressors, condensers, and evaporators in the refrigeration cycle system to ensure stable and efficient operation of the system. Its overall technical field covers mechanical refrigeration systems, electronically controlled temperature management, refrigerant flow regulation, and energy efficiency optimization strategies. Sensors are used to detect key parameters such as temperature, pressure, and current, and the operating status of the refrigeration system is adjusted in combination with control logic to meet cooling needs in different environments while reducing energy consumption and equipment losses.

[0003] Among them, the medical refrigerator compressor cooling management system refers to a system that uses a specific heat dissipation structure and cooling control method to reduce the temperature of the compressor inside the medical refrigerator due to the temperature rise problem caused by long-term operation to ensure its stable operation. The system covers the use of heat pipes or heat sinks to directly dissipate heat on the surface of the compressor, the use of air-cooling structures to form air circulation in a specific area to remove accumulated heat, and the use of independent cooling circuits to transport cold air to the area around the compressor to reduce the temperature. The specific methods include setting heat-conducting materials on the compressor casing to improve the heat dissipation efficiency, or arranging fans near the compressor to enhance air flow, and even using a low-temperature fluid circulation cooling system to indirectly cool the compressor to ensure that the temperature of the compressor is maintained within a reasonable range under high-load operating conditions.

[0004] The patent application with publication number CN118158985A provides a liquid-cooled server cluster and a data center liquid cooling system, which specifically relates to the field of temperature control technology. By designing a server cluster architecture, deploying a liquid-cooled server cluster, and installing power meters and temperature sensors at key locations of the liquid-cooled server cluster, the power consumption and temperature parameters are monitored, and the data center is transmitted to the data center for storage through the network, and the temperature data is connected to the data center to obtain temperature data. According to the obtained temperature data, the temperature gradient at different locations in the server cluster is evaluated. Based on the evaluation results, the heat dissipation layout of the server cluster is optimized. The feedback control method is used to continuously adjust the cooling water flow and temperature so that the operating temperature of the server cluster is maintained within the set stable range. According to the scale of the deployed liquid-cooled server cluster, flow scheduling is performed, and the data flow flowing into and out of the server cluster is monitored. Combined with flow scheduling and data flow monitoring, the energy consumption of the liquid cooling system is calculated.

[0005] Existing technologies rely on a small number of temperature sensors, which make it difficult to fully sense heat distribution, resulting in difficulty in timely detection of local temperature rise, affecting the accuracy of heat dissipation regulation. Heat dissipation methods mostly rely on fixed air duct design and single wind speed control, and do not dynamically adjust the air volume according to temperature changes in different areas, which may lead to uneven local heat dissipation and affect the long-term stable operation of the compressor. Energy consumption assessment methods are usually based on a single fan parameter, without combining comprehensive analysis of wind speed changes and pressure gradients, making it difficult for energy consumption management to adapt to different working conditions, resulting in energy waste. During operation, due to the lack of intelligent optimization strategies based on temperature data, it is difficult for existing cooling systems to effectively improve cooling efficiency, resulting in insufficient heat dissipation capacity or excessive energy consumption. Summary of the invention

[0006] The purpose of the present invention is to solve the problems existing in the prior art and to propose a medical refrigerator compressor cooling management system and method based on image recognition.

[0007] In order to achieve the above-mentioned purpose, the present invention adopts the following technical scheme: a medical refrigerator compressor cooling management system based on image recognition, comprising:

[0008] The hot zone recognition module obtains infrared image data of the medical refrigerator compressor, extracts the temperature values ​​of the compressor cylinder, condenser cooling fins, and exhaust pipe, calculates the pixel temperature difference, calls the boundary pixels, analyzes the temperature gradient, screens the mutation points, and establishes the local heat concentration area of ​​the compressor;

[0009] The temperature partition calculation module calls the local heat concentration area of ​​the compressor, extracts pixel points according to the temperature, calculates the temperature gradient according to the arrangement of the condenser heat dissipation fins, divides the temperature zone, and generates the compressor temperature partition data;

[0010] The air valve control module calls the compressor temperature partition data, extracts the high temperature partition, calculates the opening angle of the air duct outlet, adjusts the air valve, and records the air valve adjustment parameters;

[0011] The air duct optimization module calls the air valve adjustment parameters, records the air volume change, calculates the cooling air flow pressure gradient, adjusts the air duct steering guide vane, and calculates and obtains the optimized air duct pressure gradient;

[0012] The energy consumption evaluation module calls the optimized air duct pressure gradient, obtains the fan power consumption, calculates the fan speed change rate to calculate the current energy consumption, and compares the previous operation energy consumption data to calculate the energy consumption change, and generates the cooling mechanism energy consumption evaluation value.

[0013] As a further scheme of the present invention, the local heat concentration area of ​​the compressor includes the high temperature area of ​​the compressor cylinder, the high temperature area of ​​the condenser cooling fins, and the high temperature area of ​​the exhaust duct. The compressor temperature partition data specifically includes temperature gradient partition, condenser cooling fin arrangement partition, and pixel temperature partition. The high temperature partition specifically refers to the temperature abnormality area, the heat dissipation lag area, and the heat accumulation area. The air valve adjustment parameters include the opening angle of the air duct outlet, the air valve adjustment angle, and the air valve opening record. The optimized air duct pressure gradient includes the cooling air flow pressure change, the air duct steering guide vane adjustment amount, and the air flow distribution in the air duct. The cooling mechanism energy consumption evaluation value specifically includes the fan power consumption, the fan wind speed change rate, and the energy consumption change.

[0014] As a further solution of the present invention, the hot zone identification module includes:

[0015] The infrared image data extraction submodule obtains the infrared image data of the medical refrigerator compressor, extracts the compressor cylinder temperature value, the condenser heat sink fin temperature value, and the exhaust pipe temperature value, and at the same time calls the boundary pixels and screens their temperature values ​​to obtain key temperature data;

[0016] The temperature gradient calculation submodule calculates the pixel temperature difference between the temperature values ​​of multiple parts based on the key temperature data, and analyzes the temperature gradient distribution using the formula:

[0017] ;

[0018] Calculate and obtain the temperature gradient value, select the mutation point, and obtain the temperature gradient mutation range;

[0019] in, Represents the temperature gradient value, Representative The temperature value of each pixel, Represents the temperature value of the previous pixel. Represents the horizontal position coordinates of the pixel point. Represents the vertical position coordinate of the pixel. Represents the total number of pixels;

[0020] The local heat concentration analysis submodule calls the temperature gradient mutation interval, analyzes the distribution characteristics of local temperature anomaly points, calculates the heat distribution density, and establishes the local heat concentration area of ​​the compressor.

[0021] As a further solution of the present invention, the temperature zone calculation module includes:

[0022] The temperature pixel extraction submodule calls the local heat concentration area of ​​the compressor, extracts the temperature distribution pixel points in the area, screens the effective temperature pixel values ​​according to the compressor surface thermal image data, and removes abnormal temperature points to generate effective temperature pixel data;

[0023] The temperature gradient analysis submodule calculates the temperature difference between pixels based on the effective temperature pixel data and the arrangement of the condenser fins to obtain the multi-point temperature difference using the formula:

[0024] ;

[0025] Calculate and obtain comprehensive temperature difference indicators and temperature gradient distribution data;

[0026] in, Represents the comprehensive index of temperature gradient, Representative The temperature value of each temperature pixel, Representative The position weight of the temperature pixel, Representative The temperature difference adjustment amount of each temperature pixel, Represents the total number of temperature pixels, Representative The temperature value of each temperature pixel;

[0027] The temperature zone division submodule calls the temperature gradient distribution data, classifies the temperature areas according to the gradient change range, sets the gradient threshold, distinguishes between high gradient and low gradient areas, and divides multiple temperature intervals based on the temperature difference distribution to generate compressor temperature partition data.

[0028] As a further solution of the present invention, the air valve control module includes:

[0029] The high temperature partition extraction submodule calls the compressor temperature partition data, screens the area whose temperature value exceeds the set threshold, divides the area according to the partition boundary characteristics, calibrates the high temperature area range, and obtains the high temperature partition data;

[0030] The air duct angle calculation submodule calculates the air duct outlet angle based on the high temperature partition data and obtains the air valve adjustment angle range using the formula:

[0031] ;

[0032] Calculate and obtain the duct angle parameters;

[0033] in, Represents the opening angle of the air duct outlet. Represents the high temperature zone The temperature value of each pixel, represents the mean temperature of the high temperature zone, Representative The distance from the pixel to the center of the wind duct, Represents the angle compensation value, Represents the total number of pixels in the high temperature partition, represents the inverse tangent operator;

[0034] The air valve execution control submodule calls the air duct angle parameter, adjusts the air valve opening and wind direction, sets the air valve control parameters, records the adjustment data, and obtains the air valve adjustment parameters.

[0035] As a further solution of the present invention, the air duct optimization module includes:

[0036] The air valve adjustment submodule calls the air valve adjustment parameters, screens the differential air valve openings, monitors the air volume change trend, and calculates the change rate between the air valve opening and the air volume, and simultaneously obtains the air volume data under the differential air valve openings, determines whether the air volume change rate meets the air duct optimization requirements, and obtains the air volume data after the air valve is adjusted;

[0037] The cooling airflow calculation submodule calculates the pressure change at different positions in the air duct based on the air volume data adjusted by the air valve and the cooling airflow pressure distribution, using the formula:

[0038] ;

[0039] Calculate and obtain the cooling airflow pressure gradient, determine whether there is a mutation point in the pressure gradient, and obtain the current cooling airflow pressure gradient distribution;

[0040] in, represents the cooling air flow pressure gradient, Represents the air volume at the first position of the air duct. The cross-sectional area of ​​the air duct at the representative location, The drag coefficient at the representative position, Represents the air volume at the start of the air duct. Represents the cross-sectional area of ​​the duct at the starting position of the duct, Represents the drag coefficient at the start of the wind duct, Represents the total number of air duct calculation points;

[0041] The air duct structure adjustment submodule calls the current cooling air flow pressure gradient distribution, calculates the pressure imbalance area at the differentiated position in the air duct, adjusts the air duct steering guide vane, recalculates the adjusted pressure gradient, and obtains the optimized air duct pressure gradient.

[0042] As a further solution of the present invention, the energy consumption assessment module includes:

[0043] The air duct pressure gradient optimization submodule calculates the pressure change values ​​of multiple nodes of the air duct based on the optimized air duct pressure gradient data, obtains the pressure gradient by combining the inlet and outlet pressures of the fan, analyzes the pressure distribution characteristics of the differentiated positions inside the air duct, and calculates the total pressure loss of the air duct to obtain the air duct pressure loss;

[0044] The fan power consumption calculation submodule calls the air duct pressure loss to calculate the required compensation power of the fan, combines the current operating state of the fan, obtains the fan power consumption value, and calculates the fan speed change rate;

[0045] The energy consumption trend analysis submodule calls the wind speed change rate of the fan to calculate the current fan energy consumption, combines the fan operation data of multiple time periods, and calculates the energy consumption change using the formula:

[0046] ;

[0047] Obtain the fan energy consumption trend by calculation, compare the energy consumption data in different time periods, and establish the energy consumption evaluation value of the cooling mechanism;

[0048] in, Represents the change in energy consumption, Represents the current fan power consumption value, Represents the fan power consumption value in the previous time period, Represents the current wind speed of the fan. Represents the wind speed of the fan in the previous time period, Representative The fan power consumption value at a time point is Representative The wind speed of the fan at a certain time point, Represents the number of time points selected for calculation.

[0049] The medical refrigerator compressor cooling management method based on image recognition is executed according to the above-mentioned medical refrigerator compressor cooling management system based on image recognition, and includes the following steps:

[0050] S1: Obtain infrared image data of the compressor of a medical refrigerator, extract the temperature values ​​of the compressor cylinder, condenser cooling fins, and exhaust pipe, calculate the temperature difference of the pixel points, call the boundary pixel points, analyze the temperature gradient, screen the mutation points, and establish the local heat concentration area of ​​the compressor;

[0051] S2: Based on the local heat concentration area of ​​the compressor, call the pixel points of the local temperature abnormal area, calculate the temperature change gradient according to the arrangement of the condenser heat dissipation fins, call the pixel sets in different temperature ranges, judge the continuity of the temperature change, screen the temperature difference exceeding the limit area, divide the high temperature area, medium temperature area and low temperature area according to the temperature distribution, and establish the compressor temperature partition data;

[0052] S3: calling the compressor temperature partition data, extracting the high temperature zone range, calculating the air duct outlet position corresponding to the high temperature zone, calculating the outlet opening angle according to the temperature distribution, adjusting the air valve opening and closing degree, and recording the air valve adjustment parameters;

[0053] S4: calling the air valve adjustment parameter, recording the influence of the air valve adjustment on the air volume, calculating the pressure gradient change of the cooling airflow in the air duct, adjusting the angle of the air duct steering guide vane, calculating the adjusted air duct pressure distribution, and obtaining the optimized air duct pressure gradient;

[0054] S5: Call the optimized air duct pressure gradient, obtain the fan power consumption data, calculate the fan wind speed change rate, calculate the current energy consumption based on the wind speed change rate, compare the fan power consumption data before adjustment, calculate the energy consumption change, and obtain the cooling mechanism energy consumption evaluation value.

[0055] Compared with the prior art, the advantages and beneficial effects of the present invention are:

[0056] In the present invention, temperature values ​​are extracted through infrared image data, and combined with pixel temperature difference and temperature gradient analysis, local heat concentration areas are accurately identified to improve the perception of heat source distribution. Based on the temperature gradient and the arrangement of the condenser heat dissipation fins, the temperature zones are divided and temperature partition data is generated, so that the cooling control can be dynamically adjusted according to the temperature differences in different areas to enhance the uniformity of heat dissipation. Combined with the high-temperature partition identification and the calculation of the air duct outlet angle, the air valve is dynamically adjusted to achieve accurate air volume distribution and avoid local insufficient cooling or excessive cooling problems. The air duct optimization is combined with the calculation of the cooling air flow pressure gradient to adjust the guide vane to make the cooling air flow path more reasonable and improve the heat dissipation capacity. The energy consumption is calculated by the fan speed change rate, and the energy consumption changes are analyzed in combination with historical data to optimize the fan operation strategy, reduce unnecessary energy consumption, and make the overall cooling management more energy-saving and efficient. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] Figure 1 is a system flow chart of the present invention;

[0058] Figure 2 This is a flow chart of the hot zone identification module of the present invention;

[0059] Figure 3 This is a flow chart of the temperature zoning calculation module of the present invention;

[0060] Figure 4 This is a flow chart of the air valve control module of the present invention;

[0061] Figure 5 This is a flow chart of the air duct optimization module of the present invention;

[0062] Figure 6 This is a flow chart of the energy consumption assessment module of the present invention. DETAILED DESCRIPTION

[0063] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0064] See also Figure 1 The present invention provides a technical solution: a medical refrigerator compressor cooling management system based on image recognition includes:

[0065] The hot zone recognition module obtains infrared image data of the medical refrigerator compressor, extracts the temperature values ​​of the compressor cylinder, condenser cooling fins, and exhaust pipe, calculates the pixel temperature difference, calls the boundary pixels, analyzes the temperature gradient, screens the mutation points, and establishes the local heat concentration area of ​​the compressor;

[0066] The temperature partition calculation module calls the local heat concentration area of ​​the compressor, extracts pixel points according to the temperature, calculates the temperature gradient according to the arrangement of the condenser heat dissipation fins, divides the temperature zone, and generates the compressor temperature partition data;

[0067] The air valve control module calls the compressor temperature partition data, extracts the high-temperature partition, calculates the opening angle of the air duct outlet, adjusts the air valve, and records the air valve adjustment parameters;

[0068] The air duct optimization module calls the air valve to adjust the parameters, records the air volume changes, calculates the cooling air flow pressure gradient, adjusts the air duct steering guide vanes, and calculates and obtains the optimized air duct pressure gradient;

[0069] The energy consumption assessment module calls the optimized air duct pressure gradient, obtains the fan power consumption, calculates the fan speed change rate to calculate the current energy consumption, and compares the previous operating energy consumption data to calculate the energy consumption change, and generates the cooling mechanism energy consumption assessment value.

[0070] The local heat concentration areas of the compressor include the high-temperature area of ​​the compressor cylinder, the high-temperature area of ​​the condenser cooling fins, and the high-temperature area of ​​the exhaust duct. The compressor temperature zoning data specifically includes temperature gradient zoning, condenser cooling fin arrangement zoning, and pixel temperature zoning. The high-temperature zoning specifically refers to the temperature abnormality area, the heat dissipation lag area, and the heat accumulation area. The air valve adjustment parameters include the opening angle of the air duct outlet, the air valve adjustment angle, and the air valve opening record. The optimized air duct pressure gradient includes the cooling air flow pressure change, the air duct steering guide vane adjustment amount, and the air flow distribution in the air duct. The energy consumption assessment value of the cooling mechanism specifically includes the fan power consumption, the fan speed change rate, and the energy consumption change.

[0071] See also Figure 2 , the hot zone identification module includes:

[0072] The infrared image data extraction submodule obtains the infrared image data of the medical refrigerator compressor, extracts the compressor cylinder temperature value, the condenser heat sink fin temperature value, and the exhaust pipe temperature value, and at the same time calls the boundary pixels and screens their temperature values ​​to obtain key temperature data;

[0073] First, the external temperature of the compressor is measured at multiple points, including the cylinder surface, condenser cooling fins, exhaust pipes and other key parts. At the same time, the boundary pixel data is called to obtain the overall temperature distribution. When obtaining temperature data, the grayscale value of the pixel point needs to be converted into a temperature value, and the calibration formula is used to correct the error. The actual collected temperature data needs to be converted into standard units (°C or K). Specifically, the blackbody radiation equation can be used for correction. During the data extraction process, high temperature abnormal points need to be preliminarily screened and the temperature threshold set. , for example, setting , when the temperature of a pixel exceeds When the temperature difference between adjacent pixels exceeds the set threshold, it is marked as an abnormal point and the temperature gradient change of the surrounding pixels is analyzed. ,For example , then further calculate the average temperature of the area where it is located and confirm the abnormal range. To avoid noise interference, the median filtering algorithm is used to remove isolated high-temperature points when screening boundary pixels. The specific method is to sort the temperature values ​​in the 3×3 window and take the middle value as the final temperature value. For example, the temperature of a pixel is , the temperatures of the eight adjacent pixels are , sort and take the median As the final temperature value, after completing the preliminary data extraction, the temperature values ​​of each key point need to be stored and expressed in matrix form, as shown in Table 1.

[0074] Table 1: Key temperature data table for infrared image acquisition

[0075]

[0076] As shown in Table 1, the key point temperature data have been successfully extracted and stored, providing input data for subsequent analysis.

[0077] The temperature gradient calculation submodule calculates the pixel temperature difference between the temperature values ​​of multiple parts based on the key temperature data, and analyzes the temperature gradient distribution using the formula:

[0078] ;

[0079] Calculate and obtain the temperature gradient value, select the mutation point, and obtain the temperature gradient mutation range;

[0080] in, Represents the temperature gradient value, Representative The temperature value of each pixel, Represents the temperature value of the previous pixel. Represents the horizontal position coordinates of the pixel point. Represents the vertical position coordinate of the pixel. Represents the total number of pixels;

[0081] First, the temperature data obtained in the previous stage is extracted, and the horizontal and vertical pixel spacing is calculated based on the pixel coordinates. By analyzing the arrangement structure of the pixels, the and ,in Represents the pixel offset in the horizontal direction. Represents the pixel offset in the vertical direction. Taking a 3×3 pixel area as an example, assuming that the temperature values ​​of three adjacent pixels are , , , calculate the temperature difference , assuming the pixel spacing is 0.1 mm, calculate the temperature gradient for: ;

[0082] Enter the value:

[0083] ;

[0084] After calculating the local temperature gradient, it is necessary to screen the temperature gradient mutation point and set the threshold , if the gradient value of a pixel exceeds , it is marked as a temperature mutation point, the positions of all mutation points are counted, and their coordinates are recorded to form a mutation interval list. After the mutation point screening is completed, the next step is to enter the local heat concentration analysis.

[0085] The local heat concentration analysis submodule calls the temperature gradient mutation interval, analyzes the distribution characteristics of local temperature anomaly points, calculates the heat distribution density, and establishes the local heat concentration area of ​​the compressor.

[0086] First, the coordinate data of the mutation point is extracted and the local heat distribution density is calculated. Assuming that the mutation interval contains 50 abnormal pixels, the temperature value of each pixel is , then calculate the average temperature of the area:

[0087] ;

[0088] Assuming that after calculation , set the temperature concentration threshold ,like Higher than , then the area is determined to be a local heat concentration area, and the heat distribution density is calculated at the same time As a basis for judgment:

[0089] ;

[0090] Bring in data:

[0091] ;

[0092] Assuming that after calculation ,if This indicates that the heat distribution is uneven. Further analyze the difference between the local high temperature point and the surrounding temperature and set the local temperature difference threshold. If the temperature of a certain pixel is 5°C higher than the surrounding temperature, it is determined to be a high-temperature gathering point, and the coordinates of all high-temperature gathering points are output, ultimately forming the analysis results of the local heat concentration area.

[0093] See also Figure 3 , the temperature zone calculation module includes:

[0094] The temperature pixel extraction submodule calls the local heat concentration area of ​​the compressor, extracts the temperature distribution pixel points in the area, screens the effective temperature pixel values ​​according to the compressor surface thermal image data, and removes abnormal temperature points to generate effective temperature pixel data;

[0095] In the extraction process, it is first necessary to obtain infrared thermal imaging data of the compressor surface from the thermal imager. The thermal imaging data is stored in the form of a temperature matrix. Each pixel corresponds to a temperature value. It is necessary to select the local heat concentration area of ​​the compressor, such as around the compressor exhaust port, and determine the boundary of the area. Taking a 100×100 pixel area as an example, each pixel corresponds to the actual temperature measurement value. When extracting temperature distribution pixels, it is necessary to exclude pixel data with low signal-to-noise ratio or abnormal values. To this end, the temperature mean of the area is first calculated. and standard deviation , set the screening threshold to , that is, remove or higher For example, assuming that the mean temperature of the selected area is 80°C and the standard deviation is 5°C, the effective temperature pixel value range is 70°C to 90°C. Pixels outside this range are considered abnormal values ​​and are removed, and finally a valid temperature pixel data matrix is ​​formed, in which all pixels fall within the above range to ensure the credibility of the data. This result shows that the temperature pixel data has been screened and optimized to ensure that the data input for subsequent temperature gradient calculations are all valid values, avoid abnormal data from causing deviations in the calculation results, and ensure that the calculated temperature gradient is more accurate.

[0096] The temperature gradient analysis submodule calculates the temperature difference between pixels based on the effective temperature pixel data and the arrangement of the condenser fins to obtain the multi-point temperature difference using the formula:

[0097] ;

[0098] Calculate and obtain comprehensive temperature difference indicators and temperature gradient distribution data;

[0099] in, Represents the comprehensive index of temperature gradient, Representative The temperature value of each temperature pixel, Representative The position weight of the temperature pixel, Representative The temperature difference adjustment amount of each temperature pixel, Represents the total number of temperature pixels, Representative The temperature value of each temperature pixel;

[0100] In the calculation process, the fin arrangement direction is first selected as the X-axis direction, and each column of pixels is calculated. Assuming that the temperature value of a certain column of pixels is (Unit: °C), then calculate the temperature difference between adjacent pixels , then according to the formula:

[0101] ;

[0102] The temperature difference is comprehensively calculated, where For the The weight of the temperature pixel points, set the weight matrix , is the temperature difference adjustment amount, set , bring into calculation:

[0103] ;

[0104] Take the square root and divide by the total number of points ;

[0105] ;

[0106] Finally, the comprehensive index of temperature gradient is obtained The result shows that the temperature difference in this area is relatively uniform, and there is no obvious temperature mutation phenomenon. According to the set gradient threshold, this area can be divided into a medium gradient area. This numerical result will serve as an important basis for subsequent temperature zone division to determine whether the area needs additional temperature control measures or whether there is potential heat dissipation anomaly, to ensure the rationality of the temperature zone division.

[0107] The temperature zone division submodule calls the temperature gradient distribution data, classifies the temperature areas according to the gradient change range, sets the gradient threshold, distinguishes between high-gradient and low-gradient areas, and divides multiple temperature intervals based on the temperature difference distribution to generate compressor temperature partition data.

[0108] First, calculate the gradient distribution of all pixels and set the gradient threshold range. For example, if the gradient value is greater than 0.8, it is considered a high gradient area, and if it is 0.4≤ ≤0.8 is considered as medium gradient area. <0.4 is considered as a low gradient area. For example, this area belongs to the medium gradient area. Next, the temperature intervals are further divided according to the temperature difference distribution. Assuming that the temperature distribution range is 50°C to 120°C, it is divided into 4 intervals:

[0109] 50°C~70°C (low temperature zone);

[0110] 70°C~90°C (medium and low temperature zone);

[0111] 90°C~110°C (medium and high temperature range);

[0112] 110°C~120°C (high temperature area);

[0113] The temperature data in the area are classified according to the above intervals, and finally the compressor temperature partition data is generated, as shown in Table 2.

[0114] Table 2: Compressor temperature zone data

[0115]

[0116] As shown in Table 2, the distribution of the number of pixels in different temperature ranges has been statistically completed. The results show that the temperature distribution in this area is in line with expectations. The temperature is mainly concentrated in the medium-low temperature zone and the medium-high temperature zone. A small part of the area is in the low temperature zone or the high temperature zone. The numerical results further illustrate that the overall temperature gradient in this area is within an acceptable range and can be used for subsequent heat dissipation optimization or abnormal temperature monitoring. If the number of pixels in the high temperature zone accounts for too high a proportion, the heat dissipation design may need to be further optimized. If the number of low temperature zones increases abnormally, there may be abnormal local cooling of the compressor. Therefore, the data results are of great reference value for compressor temperature control optimization.

[0117] See also Figure 4 , the air valve control module includes:

[0118] The high-temperature partition extraction submodule calls the compressor temperature partition data, screens the areas where the temperature value exceeds the set threshold, divides the area according to the partition boundary characteristics, calibrates the high-temperature area range, and obtains the high-temperature partition data;

[0119] First, analyze the temperature distribution in the data, traverse the temperature values ​​of each area, and determine whether the temperature of each area exceeds the set threshold in turn. The temperature value is screened according to the set threshold range. For example, if the threshold is set to 80°C, the area with a temperature value greater than or equal to 80°C is extracted during the traversal process, and the boundary features of the area are recorded. Assume that a temperature distribution data matrix is:

[0120] ;

[0121] In this matrix, the values ​​exceeding 80°C include 82, 85, 80, 83, 81, 88, 81, 85, 84, 90, 82, 86, etc. The pixels corresponding to these data will be marked as high temperature areas, and the areas will be divided according to the boundary features. For example, adjacent high temperature pixels are connected to form a continuous area, and the coordinates of the area boundary are recorded to form high temperature area data. For example, the high temperature area can be represented as a set of coordinate points. Finally, the output high temperature partition data includes the regional coordinate set and the corresponding temperature value information.

[0122] The results show that the filtered high-temperature area data is used for subsequent duct angle calculations, providing accurate temperature distribution information for duct adjustment, ensuring that subsequent air valve control is optimized based on the actual high-temperature area position, thereby improving control accuracy.

[0123] The air duct angle calculation submodule calculates the air duct outlet angle based on the high temperature partition data and obtains the air valve adjustment angle range using the formula:

[0124] ;

[0125] Calculate and obtain the duct angle parameters;

[0126] in, Represents the opening angle of the air duct outlet. Represents the high temperature zone The temperature value of each pixel, represents the mean temperature of the high temperature zone, Representative The distance from the pixel to the center of the wind duct, Represents the angle compensation value, Represents the total number of pixels in the high temperature partition, represents the inverse tangent operator;

[0127] First, calculate the mean temperature of the high temperature zone ,Right now:

[0128] ;

[0129] In the above high temperature data set, take the temperature value of the high temperature pixel , then calculate the mean temperature:

[0130] ;

[0131] Then, calculate the difference between the temperature value of each pixel and the mean value multiplied by the distance from the pixel to the center of the air duct , assuming that the center of the wind duct is located at , calculate the Euclidean distance of a pixel, for example Distance of points:

[0132] ;

[0133] Calculate all high-temperature pixels and substitute them into the formula:

[0134] ;

[0135] Setting the angle compensation value , calculated:

[0136] ;

[0137] After calculation, assuming that the final The calculated result is 15°, so the duct angle parameter is obtained. .

[0138] The results show that: It is used to determine the optimal opening angle of the air duct outlet. This angle directly determines the adjustment direction of the air valve, so that the wind direction can be adjusted more accurately to the high-temperature area, thereby optimizing the cooling effect and improving the responsiveness of the control system.

[0139] The air valve execution control submodule calls the air duct angle parameters, adjusts the air valve opening and wind direction, sets the air valve control parameters, records the adjustment data, and obtains the air valve adjustment parameters.

[0140] First call the current opening value of the air valve , assuming that the current air valve opening is 30%, set the opening adjustment coefficient to , then the new opening value of the air valve is calculated as follows:

[0141] ;

[0142] Bring in data:

[0143] ;

[0144] After the air valve is adjusted, the opening adjustment data is recorded, and the opening 30% before adjustment and the opening 37.5% after adjustment are stored in the air valve adjustment parameter table, and finally the air valve adjustment parameters are output.

[0145] The results show that the adjusted air valve opening of 37.5% is consistent with the calculated Correspondingly, ensure that the air valve adjusts the air volume and direction within a reasonable range so that the airflow can effectively cover the high-temperature area and improve the cooling efficiency, while avoiding excessive adjustment that causes increased energy consumption or insufficient local cooling.

[0146] See also Figure 5 , the air duct optimization module includes:

[0147] The air valve adjustment submodule calls the air valve adjustment parameters, screens the differential air valve opening, monitors the air volume change trend, and calculates the change rate between the air valve opening and the air volume. At the same time, it obtains the air volume data under the differential air valve opening, determines whether the air volume change rate meets the air duct optimization requirements, and obtains the air volume data after the air valve is adjusted;

[0148] First, call the air valve adjustment parameters, set the air valve opening in stages, set the opening range to 0% to 100%, and divide it into multiple adjustment gears (e.g. 10%, 20%, 30%…100%). At each gear, obtain the air volume data through the air volume sensor, and gradually adjust the air valve opening according to the set step size (e.g. 10%). At the same time, record the air volume changes at each gear, and store the data in Table 2. Then, calculate the change rate between the air valve opening change and the air volume. For each adjustment gear, the calculation method is as follows:

[0149] ;

[0150] in, Representative Air volume under the gear air valve opening, Represents the corresponding air valve opening value, Indicates the rate of change of air volume. If the rate of change of a certain gear is lower than the set threshold (such as 0.5 ), the adjustment effect of this gear is not obvious, and it is necessary to re-screen the appropriate opening range and adjust the sampling step size to further obtain a more detailed air volume change trend. Finally, the air valve opening that meets the air duct optimization requirements is extracted, and the air volume data generated after the air valve opening adjustment is selected as the input for subsequent calculations, as shown in Table 3.

[0151] Table 3: Air valve adjustment data table

[0152]

[0153] As shown in Table 3, within the range of 10% to 50% of the air valve opening, the air volume gradually increases, but its rate of change gradually decreases. Therefore, the air volume adjustment is more significant in the range of 30%-40%, which can be used as the key range for air valve optimization adjustment to determine the optimal air valve opening setting.

[0154] The cooling airflow calculation submodule calculates the pressure change at different positions in the air duct based on the air volume data adjusted by the air valve and the cooling airflow pressure distribution, using the formula:

[0155] ;

[0156] Calculate and obtain the cooling airflow pressure gradient, determine whether there is a mutation point in the pressure gradient, and obtain the current cooling airflow pressure gradient distribution;

[0157] in, represents the cooling air flow pressure gradient, Representative of the wind Air volume at the location, The cross-sectional area of ​​the air duct at the representative location, The drag coefficient at the representative position, Represents the air volume at the start of the air duct. Represents the cross-sectional area of ​​the duct at the starting position of the duct, Represents the drag coefficient at the start of the wind duct, Represents the total number of air duct calculation points;

[0158] Set multiple calculation points inside the air duct, each measuring point Air volume The data is obtained through the adjustment of the air valve and based on the cross-sectional area. and drag coefficient Calculate the local pressure loss at this point:

[0159] ;

[0160] The total pressure gradient for all measuring points is then calculated :

[0161] ;

[0162] in, is the air volume at the starting position of the air duct, is the cross-sectional area at the starting position of the air duct, is the drag coefficient at the start of the duct. Set the following parameters: , , , (Set 3 calculation points), , , - , , - , , .

[0163] Calculate the pressure value:

[0164] ;

[0165] ;

[0166] ;

[0167] ;

[0168] ;

[0169] Calculate the pressure gradient of the cooling airflow It is 9.66. If this value exceeds the set threshold, it indicates that there may be resistance mutation points in certain areas of the air duct, and the air duct structure needs to be adjusted.

[0170] The air duct structure adjustment submodule calls the current cooling air flow pressure gradient distribution, calculates the pressure imbalance area at the differentiated position in the air duct, adjusts the air duct steering guide vane, recalculates the adjusted pressure gradient, and obtains the optimized air duct pressure gradient.

[0171] Assuming that the pressure change at measurement point 2 is large and there is a high pressure difference compared to the surrounding area, the guide vane angle at this point needs to be adjusted to optimize the airflow distribution. Set the current guide vane angle to 30°, change it to 45°, and recalculate the adjusted pressure gradient:

[0172] ;

[0173] Calculate the total pressure gradient after adjustment:

[0174] ;

[0175] After adjusting the guide vane angle, the calculation results show that the pressure gradient drops from 9.66 to 8.94, reaching the optimization standard. Therefore, this air duct structure adjustment scheme can effectively reduce the local resistance mutation and improve the uniformity of the cooling airflow.

[0176] See also Figure 6 , the energy consumption assessment module includes:

[0177] The duct pressure gradient optimization submodule calculates the pressure change values ​​of multiple nodes in the duct based on the optimized duct pressure gradient data, obtains the pressure gradient by combining the fan inlet and outlet pressures, analyzes the pressure distribution characteristics of differentiated positions inside the duct, and calculates the total pressure loss of the duct to obtain the duct pressure loss;

[0178] First, it is necessary to obtain the initial pressure value of each monitoring point in the air duct, set multiple measurement points, such as installing pressure sensors at the air duct entrance, key inflection point, end, etc., to measure their pressure data. The specific measurement point data can be found in Table 3. After obtaining the pressure of each monitoring point, a data sequence is formed according to the measurement time sequence, and the pressure change value between adjacent time points is calculated. The pressure change value of each measurement point can be expressed as:

[0179] ;

[0180] in, and are the pressure values ​​at the current time point and the previous time point respectively. Then, the pressure data is extracted at the inlet and outlet of the fan to calculate the inlet pressure respectively. With outlet pressure The difference between the fan pressure difference and the fan pressure difference is:

[0181] ;

[0182] When analyzing the pressure distribution characteristics of each point inside the air duct, compare the pressure values ​​at different positions of the air duct, identify the key points with large pressure drops, and set judgment criteria. For example, if the pressure change value at a certain point is If the pressure drop exceeds 20% of the average change value, it can be considered that there is a significant pressure loss at this point. At the same time, when calculating the total pressure loss of the air duct, the pressure change values ​​of all monitoring points need to be accumulated to obtain the total pressure loss of the air duct:

[0183] ;

[0184] in, is the number of measurement points in the duct, and the pressure loss can be used for subsequent fan power consumption calculation. After the duct pressure loss is finally obtained, the next module is entered for calculation. The result shows that the pressure change trend in the duct has been quantified, and the key pressure loss points and overall pressure loss have been determined, which is of great significance for the subsequent fan power consumption calculation and can provide necessary data support for optimizing fan operation, as shown in Table 4.

[0185] Table 4: Initial pressure table of air duct monitoring points

[0186]

[0187] As shown in Table 4, the initial pressure value of each monitoring point in the air duct can be used to calculate the subsequent air duct pressure loss.

[0188] The fan power consumption calculation submodule calls the duct pressure loss to calculate the required compensation power of the fan, combines the current operating status of the fan, obtains the fan power consumption value, and calculates the fan speed change rate;

[0189] First, according to the air duct pressure loss calculated above , combined with fan flow Calculate the theoretical compensation power of the fan :

[0190] ;

[0191] in, is the fan efficiency, assuming the fan efficiency is 85% (i.e. ), fan flow Set to 5m 3 / s, the compensation power can be calculated:

[0192] ;

[0193] Then, combined with the current operating status of the fan, the current power consumption value of the fan is extracted , and calculate the wind speed change rate of the fan, which can be expressed as:

[0194] ;

[0195] Assuming that the fan power at the previous time point was 2800W, the fan power at the current time point is 3000W, the fan speed at the previous time point was 12m / s, and the fan speed at the current time point is 14m / s, calculate the wind speed change rate:

[0196] ;

[0197] The results show that the wind speed change rate of the fan has been quantified. The data is used to analyze the operating status of the fan and provide basic data for subsequent energy consumption evaluation, further determining the energy consumption trend of the fan operation and optimizing the adjustment strategy.

[0198] The energy consumption trend analysis submodule calls the fan speed change rate to calculate the current fan energy consumption, combines the fan operation data of multiple time periods, and calculates the energy consumption change using the formula:

[0199] ;

[0200] Obtain the fan energy consumption trend by calculation, compare the energy consumption data in different time periods, and establish the energy consumption evaluation value of the cooling mechanism;

[0201] in, Represents the change in energy consumption, Represents the current fan power consumption value, Represents the fan power consumption value in the previous time period, Represents the current wind speed of the fan. Represents the wind speed of the fan in the previous time period, Representative The fan power consumption value at a time point is Representative The wind speed of the fan at a certain time point, Represents the number of time points selected for calculation.

[0202] First, call the wind speed change rate of the fan calculated above , set multiple operating time point data of the fan and calculate the energy consumption change :

[0203] ;

[0204] Assume that the wind turbine power and wind speed data at multiple time points are shown in Table 5 below:

[0205] Table 5: Fan operation data table

[0206]

[0207] Calculate the summation part:

[0208] ;

[0209] Calculate the change in energy consumption:

[0210] ;

[0211] The results show that the energy consumption trend of the fan has been calculated. This value is used to determine the energy consumption characteristics of the fan under different operating conditions, and can be used to further optimize the fan operation strategy to reduce energy consumption to a reasonable range. It can also be used as reference data for subsequent fan energy-saving optimization control strategies.

[0212] The medical refrigerator compressor cooling management method based on image recognition is executed according to the above-mentioned medical refrigerator compressor cooling management system based on image recognition, and includes the following steps:

[0213] S1: Obtain infrared image data of the compressor of a medical refrigerator, extract the temperature values ​​of the compressor cylinder, condenser cooling fins, and exhaust pipe, calculate the temperature difference of the pixel points, call the boundary pixel points, analyze the temperature gradient, screen the mutation points, and establish the local heat concentration area of ​​the compressor;

[0214] S2: Based on the local heat concentration area of ​​the compressor, call the pixel points of the local temperature abnormal area, calculate the temperature change gradient according to the arrangement of the condenser heat dissipation fins, call the pixel sets in different temperature ranges, judge the continuity of temperature change, screen the temperature difference exceeding the limit area, divide the high temperature area, medium temperature area and low temperature area according to the temperature distribution, and establish the compressor temperature partition data;

[0215] S3: Call the compressor temperature partition data, extract the high temperature zone range, calculate the air duct outlet position corresponding to the high temperature zone, calculate the outlet opening angle according to the temperature distribution, adjust the air valve opening and closing degree, and record the air valve adjustment parameters;

[0216] S4: calling the air valve adjustment parameters, recording the influence of the air valve adjustment on the air volume, calculating the pressure gradient change of the cooling air flow in the air duct, adjusting the angle of the air duct steering guide vane, calculating the adjusted air duct pressure distribution, and obtaining the optimized air duct pressure gradient;

[0217] S5: Call the optimized air duct pressure gradient, obtain the fan power consumption data, calculate the fan wind speed change rate, calculate the current energy consumption based on the wind speed change rate, compare the fan power consumption data before adjustment, calculate the energy consumption change, and obtain the cooling mechanism energy consumption evaluation value.

[0218] The above are only preferred embodiments of the present invention and are not intended to limit the present invention in other forms. Any technician familiar with the profession may use the technical contents disclosed above to change or modify them into equivalent embodiments with equivalent changes and apply them to other fields. However, any simple modification, equivalent change and modification made to the above embodiments based on the technical essence of the present invention without departing from the technical solution of the present invention still falls within the protection scope of the technical solution of the present invention.

Claims

1. A medical refrigerator compressor cooling management system based on image recognition, characterized in that: The system comprises: The hot zone recognition module obtains infrared image data of the medical refrigerator compressor, extracts the temperature values ​​of the compressor cylinder, condenser cooling fins, and exhaust pipe, calculates the pixel temperature difference, calls the boundary pixels, analyzes the temperature gradient, screens the mutation points, and establishes the local heat concentration area of ​​the compressor; The temperature partition calculation module calls the local heat concentration area of ​​the compressor, extracts pixel points according to the temperature, calculates the temperature gradient according to the arrangement of the condenser heat dissipation fins, divides the temperature zone, and generates the compressor temperature partition data; The air valve control module calls the compressor temperature partition data, extracts the high temperature partition, calculates the opening angle of the air duct outlet, adjusts the air valve, and records the air valve adjustment parameters; The air duct optimization module calls the air valve adjustment parameters, records the air volume change, calculates the cooling air flow pressure gradient, adjusts the air duct steering guide vane, and calculates and obtains the optimized air duct pressure gradient; The energy consumption evaluation module calls the optimized air duct pressure gradient, obtains the fan power consumption, calculates the fan wind speed change rate to calculate the current energy consumption, and calls the previous operation energy consumption data to calculate the energy consumption change, and generates the cooling mechanism energy consumption evaluation value.

2. The medical refrigerator compressor cooling management system based on image recognition according to claim 1, characterized in that: The local heat concentration area of ​​the compressor includes the high temperature area of ​​the compressor cylinder, the high temperature area of ​​the condenser cooling fins, and the high temperature area of ​​the exhaust duct. The compressor temperature partition data specifically includes temperature gradient partition, condenser cooling fin arrangement partition, and pixel temperature partition. The high temperature partition specifically refers to the temperature abnormality area, the heat dissipation lag area, and the heat accumulation area. The air valve adjustment parameters include the opening angle of the air duct outlet, the air valve adjustment angle, and the air valve opening record. The optimized air duct pressure gradient includes the cooling air flow pressure change, the air duct steering guide vane adjustment amount, and the air flow distribution in the air duct. The cooling mechanism energy consumption assessment value specifically includes the fan power consumption, the fan wind speed change rate, and the energy consumption change.

3. The medical refrigerator compressor cooling management system based on image recognition according to claim 2, characterized in that: The hot zone identification module includes: The infrared image data extraction submodule obtains the infrared image data of the medical refrigerator compressor, extracts the compressor cylinder temperature value, the condenser heat sink fin temperature value, and the exhaust pipe temperature value, and at the same time calls the boundary pixels and screens their temperature values ​​to obtain key temperature data; The temperature gradient calculation submodule calculates the pixel temperature difference between the temperature values ​​of multiple parts based on the key temperature data, and analyzes the temperature gradient distribution using the formula: ； Calculate and obtain the temperature gradient value, select the mutation point, and obtain the temperature gradient mutation range; in, Represents the temperature gradient value, Representative The temperature value of each pixel, Represents the temperature value of the previous pixel. Represents the horizontal position coordinates of the pixel point. Represents the vertical position coordinate of the pixel. Represents the total number of pixels; The local heat concentration analysis submodule calls the temperature gradient mutation interval, analyzes the distribution characteristics of local temperature anomaly points, calculates the heat distribution density, and establishes the local heat concentration area of ​​the compressor.

4. The medical refrigerator compressor cooling management system based on image recognition according to claim 3 is characterized in that: The temperature partition calculation module includes: The temperature pixel extraction submodule calls the local heat concentration area of ​​the compressor, extracts the temperature distribution pixel points in the area, screens the effective temperature pixel values ​​according to the compressor surface thermal image data, and removes abnormal temperature points to generate effective temperature pixel data; The temperature gradient analysis submodule calculates the temperature difference between pixels based on the effective temperature pixel data and the arrangement of the condenser fins to obtain the multi-point temperature difference using the formula: ； Calculate and obtain comprehensive temperature difference indicators and temperature gradient distribution data; in, Represents the comprehensive index of temperature gradient, Representative The temperature value of each temperature pixel, Representative The position weight of the temperature pixel, Representative The temperature difference adjustment amount of each temperature pixel, Represents the total number of temperature pixels, Representative The temperature value of each temperature pixel; The temperature zone division submodule calls the temperature gradient distribution data, classifies the temperature areas according to the gradient change range, sets the gradient threshold, distinguishes between high gradient and low gradient areas, and divides multiple temperature intervals based on the temperature difference distribution to generate compressor temperature partition data.

5. The medical refrigerator compressor cooling management system based on image recognition according to claim 4, characterized in that: The air valve control module includes: The high temperature partition extraction submodule calls the compressor temperature partition data, screens the area whose temperature value exceeds the set threshold, divides the area according to the partition boundary characteristics, calibrates the high temperature area range, and obtains the high temperature partition data; The air duct angle calculation submodule calculates the air duct outlet angle based on the high temperature partition data and obtains the air valve adjustment angle range using the formula: ； Calculate and obtain the duct angle parameters; in, Represents the opening angle of the air duct outlet. Represents the high temperature zone The temperature value of each pixel, represents the mean temperature of the high temperature zone, Representative The distance from the pixel to the center of the wind duct, Represents the angle compensation value, Represents the total number of pixels in the high temperature partition, represents the inverse tangent operator; The air valve execution control submodule calls the air duct angle parameter, adjusts the air valve opening and wind direction, sets the air valve control parameters, records the adjustment data, and obtains the air valve adjustment parameters.

6. The medical refrigerator compressor cooling management system based on image recognition according to claim 5, characterized in that: The air duct optimization module includes: The air valve adjustment submodule calls the air valve adjustment parameters, screens the differential air valve openings, monitors the air volume change trend, and calculates the change rate between the air valve opening and the air volume, and simultaneously obtains the air volume data under the differential air valve openings, determines whether the air volume change rate meets the air duct optimization requirements, and obtains the air volume data after the air valve is adjusted; The cooling airflow calculation submodule calculates the pressure change at different positions in the air duct based on the air volume data adjusted by the air valve and the cooling airflow pressure distribution, using the formula: ； Calculate and obtain the cooling airflow pressure gradient, determine whether there is a mutation point in the pressure gradient, and obtain the current cooling airflow pressure gradient distribution; in, represents the cooling air flow pressure gradient, Representative of the wind Air volume at the location, The cross-sectional area of ​​the air duct at the representative location, The drag coefficient at the representative position, Represents the air volume at the start of the air duct. Represents the cross-sectional area of ​​the duct at the starting position of the duct, Represents the drag coefficient at the start of the wind duct, Represents the total number of air duct calculation points; The air duct structure adjustment submodule calls the current cooling air flow pressure gradient distribution, calculates the pressure imbalance area at the differentiated position in the air duct, adjusts the air duct steering guide vane, recalculates the adjusted pressure gradient, and obtains the optimized air duct pressure gradient.

7. The medical refrigerator compressor cooling management system based on image recognition according to claim 6, characterized in that: The energy consumption assessment module comprises: The air duct pressure gradient optimization submodule calculates the pressure change values ​​of multiple nodes of the air duct based on the optimized air duct pressure gradient data, obtains the pressure gradient by combining the inlet and outlet pressures of the fan, analyzes the pressure distribution characteristics of the differentiated positions inside the air duct, and calculates the total pressure loss of the air duct to obtain the air duct pressure loss; The fan power consumption calculation submodule calls the air duct pressure loss to calculate the required compensation power of the fan, combines the current operating state of the fan, obtains the fan power consumption value, and calculates the fan speed change rate; The energy consumption trend analysis submodule calls the wind speed change rate of the fan to calculate the current fan energy consumption, combines the fan operation data of multiple time periods, and calculates the energy consumption change using the formula: ； Obtain the fan energy consumption trend by calculation, compare the energy consumption data in different time periods, and establish the energy consumption evaluation value of the cooling mechanism; in, Represents the change in energy consumption, Represents the current fan power consumption value, Represents the fan power consumption value in the previous time period, Represents the current wind speed of the fan. Represents the wind speed of the fan in the previous time period, Representative The fan power consumption value at a time point is Representative The wind speed of the fan at a certain time point, Represents the number of time points selected for calculation.

8. A method for managing the cooling of a compressor of a medical refrigerator based on image recognition, which is implemented according to the medical refrigerator compressor cooling management system based on image recognition according to any one of claims 1 to 7, characterized in that: The following steps are involved: S1: Obtain infrared image data of the compressor of a medical refrigerator, extract the temperature values ​​of the compressor cylinder, condenser cooling fins, and exhaust pipe, calculate the temperature difference of the pixel points, call the boundary pixel points, analyze the temperature gradient, screen the mutation points, and establish the local heat concentration area of ​​the compressor; S2: Based on the local heat concentration area of ​​the compressor, call the pixel points of the local temperature abnormal area, calculate the temperature change gradient according to the arrangement of the condenser heat dissipation fins, call the pixel sets in different temperature ranges, judge the continuity of the temperature change, screen the temperature difference exceeding the limit area, divide the high temperature area, medium temperature area and low temperature area according to the temperature distribution, and establish the compressor temperature partition data; S3: calling the compressor temperature partition data, extracting the high temperature zone range, calculating the air duct outlet position corresponding to the high temperature zone, calculating the outlet opening angle according to the temperature distribution, adjusting the air valve opening and closing degree, and recording the air valve adjustment parameters; S4: calling the air valve adjustment parameter, recording the influence of the air valve adjustment on the air volume, calculating the pressure gradient change of the cooling air flow in the air duct, adjusting the angle of the air duct steering guide vane, calculating the adjusted air duct pressure distribution, and obtaining the optimized air duct pressure gradient; S5: Call the optimized air duct pressure gradient, obtain the fan power consumption data, calculate the fan wind speed change rate, calculate the current energy consumption based on the wind speed change rate, compare the fan power consumption data before adjustment, calculate the energy consumption change, and obtain the cooling mechanism energy consumption evaluation value.

Images