A method and system for predicting cleaning time of refrigeration unit

By analyzing the historical operation data of the refrigeration unit and establishing a three-dimensional model, identifying the dirt accumulation mode and evaluating the difficulty of cleaning, the problem of poor cleaning effect in the existing technology is solved, and an efficient and customized cleaning solution is achieved, and the equipment is extended. Lifespan and optimize resource utilization.

CN118657257BActive Publication Date: 2024-11-12SHUZHONG IND EQUIP QIDONG CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411138757.7
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-08-19
Publication Date
2024-11-12
Estimated Expiration
2044-08-19

AI Technical Summary

Technical Problem

The prior art is difficult to accurately predict the accumulation of dirt in various components of the refrigeration unit, resulting in poor cleaning results, increasing cleaning time and cost, and affecting equipment performance and life.

Method used

By obtaining the historical operation data of the refrigeration unit, identifying the dirt accumulation pattern, analyzing heat transfer efficiency and pressure loss, establishing a three-dimensional model, evaluating the difficulty of cleaning, matching the most suitable cleaning method, and predicting the cleaning completion time.

Benefits of technology

Customized cleaning is realized, cleaning efficiency and equipment life are improved, maintenance costs and production stagnation risks are reduced, and the rational use of resources and the scientific nature of cleaning plans are ensured.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118657257B_ABST
    Figure CN118657257B_ABST
Patent Text Reader

Abstract

The present invention discloses a method and system for predicting the cleaning time of a refrigeration unit, and relates to the technical field of refrigeration unit cleaning. The method for predicting the cleaning time of a refrigeration unit includes the following steps: identifying the dirt accumulation mode of each key component in the refrigeration unit based on historical operation data; analyzing the heat transfer efficiency and pressure loss of each key component based on the dirt accumulation mode of each key component, and judging the dirt accumulation degree of each key component based on the analysis result; establishing a three-dimensional model of the refrigeration unit, and comprehensively evaluating the cleaning difficulty of each key component in combination with the structural characteristics and dirt accumulation degree of each key component; matching the corresponding cleaning method for each key component based on the cleaning difficulty of each key component, and predicting the cleaning completion time of the refrigeration unit. The present invention matches the cleaning method by considering the cleaning difficulty of each key component, thereby ensuring that the most suitable cleaning strategy is adopted for each key component, and improving the cleaning efficiency.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of refrigeration unit cleaning, and in particular to a method and system for predicting the cleaning time of a refrigeration unit. Background Art

[0002] A refrigeration unit is a device that achieves a cooling effect by transferring heat from a low-temperature area to a high-temperature area through the coordinated work of key components such as a compressor, condenser, expansion valve, evaporator, and control system. It is commonly used in air conditioning, cold storage, cold chain transportation, and other occasions that require temperature control. The circulating refrigerant absorbs heat in a closed system and releases it to the external environment to maintain the required low-temperature environment.

[0003] After a refrigeration unit has been used for a long time, dust, dirt and oil stains will accumulate on the components of the refrigeration unit. Dust and dirt will affect the heat exchange efficiency of the refrigeration unit, reduce the refrigeration effect and increase energy consumption. Cleaning the refrigeration unit can restore the design performance of the refrigeration unit, reduce energy consumption and extend the life of the equipment. Regular cleaning can also detect potential faults or signs of wear, such as leakage and corrosion, so that timely repairs or replacements can be carried out.

[0004] During the cleaning process of the refrigeration unit, accurate cleaning time prediction can help managers arrange the cleaning work reasonably to avoid unnecessary impact on production. It also helps to control the cost of manpower, equipment and materials and avoid excessive or insufficient resource allocation. However, since the types and accumulation of dirt on different components in the refrigeration unit may be different, if they cannot be analyzed separately, it is impossible to use the cleaning method that is most suitable for each component, which will lead to poor cleaning results, increase cleaning time and cost, and affect the predicted time of the refrigeration unit.

[0005] Currently, no effective solution has been proposed for the problems in the related technologies. Summary of the invention

[0006] In view of the problems in the related art, the present invention proposes a method and system for predicting the cleaning time of a refrigeration unit to overcome the above technical problems existing in the existing related art.

[0007] To this end, the specific technical solution adopted by the present invention is as follows:

[0008] According to one aspect of the present invention, a method for predicting the cleaning time of a refrigeration unit is provided, and the method for predicting the cleaning time of a refrigeration unit comprises the following steps:

[0009] S1. Obtain historical operation data of the refrigeration unit, and identify the dirt accumulation pattern of each key component in the refrigeration unit based on the historical operation data;

[0010] S2. analyzing the heat transfer efficiency and pressure loss of each key component based on the fouling accumulation pattern of each key component, and determining the fouling accumulation degree of each key component based on the analysis results;

[0011] S3. Establish a three-dimensional model of the refrigeration unit, and comprehensively evaluate the cleaning difficulty of each key component based on the structural characteristics and dirt accumulation degree of each key component;

[0012] S4. Match a corresponding cleaning method to each key component based on the cleaning difficulty of each key component, and predict the cleaning completion time of the refrigeration unit.

[0013] Preferably, the historical operation data of the refrigeration unit includes historical ambient temperature data, historical pressure data, historical energy consumption data, historical operation time, historical operation cycle and historical cleaning records;

[0014] The key components in the refrigeration unit include at least the condenser, evaporator, pipes and filters.

[0015] Preferably, analyzing the heat transfer efficiency and pressure loss of each key component based on the fouling accumulation mode of each key component, and judging the fouling accumulation degree of each key component based on the analysis result comprises the following steps:

[0016] S21, extracting the dirt accumulation pattern of the condenser from step S1, identifying and dividing the dirt accumulation prone area of ​​the condenser;

[0017] S22, using a heat transfer model to calculate the heat transfer temperature of each easily polluted area of ​​the condenser, and comprehensively determine the heat transfer efficiency of the condenser based on the heat transfer temperature of each easily polluted area;

[0018] S23. Evaluate the degree of fouling accumulation of the condenser based on the heat transfer efficiency of the condenser;

[0019] S24. Analyze the pressure loss inside the pipeline when air flow is introduced by using the pressure difference monitoring method, and judge the degree of dirt accumulation in the pipeline and the filter screen inside the pipeline based on the pressure loss inside the pipeline.

[0020] Preferably, using the heat transfer model to calculate the heat transfer temperature of each easily polluted area of ​​the condenser respectively, and comprehensively determining the heat transfer efficiency of the condenser based on the heat transfer temperature of each easily polluted area includes the following steps:

[0021] S221, obtaining the wall heat flux and the hourly outlet temperature of each area prone to contamination in the condenser, and analyzing the internal and external temperature difference of the area prone to contamination based on the wall heat flux and the hourly outlet temperature of the heat flux;

[0022] S222, analyzing and calculating the wall temperature transfer function and heat flow transfer function of the condenser within a time period based on the internal and external temperature difference of the area prone to contamination;

[0023] S223, constructing a heat transfer model using the wall temperature transfer function and the heat flow transfer function, outputting the temperature distribution of the refrigerant inside the condenser through the heat transfer model, and assigning a value to the temperature distribution as the heat transfer temperature at the condenser outlet;

[0024] S224, continuously calculating the heat transfer temperature of each easily polluted area of ​​the condenser based on the heat transfer temperature at the condenser outlet, and determining a reasonable value of the heat transfer temperature of each easily polluted area based on the law of conservation of energy;

[0025] S225. Utilize a heat transfer model to calculate the temperature distribution of the refrigerant along the path when a reasonable value is reached, and comprehensively determine the heat transfer efficiency of the condenser based on the temperature distribution along the path.

[0026] Preferably, using the pressure difference monitoring method to analyze the pressure loss inside the pipeline when the airflow is introduced, and judging the degree of dirt accumulation in the pipeline and the filter screen inside the pipeline based on the pressure loss inside the pipeline respectively includes the following steps:

[0027] S241, extracting the dirt accumulation pattern of the pipeline from step S1, and identifying the airflow pressure and airflow velocity when airflow is introduced into the pipeline;

[0028] S242, establishing a fluid kinematics model, and inputting the airflow pressure and airflow velocity inside the pipeline into the fluid kinematics model, and outputting the pressure test result inside the pipeline through the fluid kinematics model;

[0029] S243. Combined with the influence of pipeline roughness change on pressure test results, the pressure loss of the pipeline under actual airflow operating conditions is predicted by introducing a correction coefficient;

[0030] S244. Compare the pressure loss under the actual airflow operating conditions with the preset pressure difference value. If the pressure loss is greater than the preset threshold, it means that the degree of dirt accumulation in the pipeline and the filter inside the pipeline is high. If the pressure loss is less than the preset threshold, it means that the degree of dirt accumulation in the pipeline and the filter inside the pipeline is low.

[0031] Preferably, matching a corresponding cleaning method to each key component based on the cleaning difficulty of each key component, and predicting the cleaning completion time of the refrigeration unit includes the following steps:

[0032] S41, matching the cleaning methods corresponding to the key components in the refrigeration unit based on the cleaning difficulty, and generating a cleaning team combination based on the cleaning method characteristics;

[0033] S42, optimizing the cleaning allocation among the cleaning team combinations by using a cleaning allocation algorithm;

[0034] S43. Based on the cleaning allocation model, historical cleaning records are extracted to establish a time prediction model to predict the cleaning completion time of each key component in the refrigeration unit.

[0035] Preferably, the cleaning allocation optimization between the cleaning team combinations using the cleaning allocation algorithm comprises the following steps:

[0036] S421, define and quantify the cleaning workload of each cleaning team combination, and use the cleaning workload as the allocation basis when solving the cleaning allocation optimization;

[0037] S422, setting the number of backtracking times, and judging whether the cleaning workload of each cleaning team combination meets the standard according to the set number of backtracking times;

[0038] S423, using a greedy algorithm to sort the cleaning workload of each cleaning team combination, one by one assigning the cleaning team combination with the largest cleaning workload to the cleaning team combination with the smallest cleaning workload, and recording the current allocation method as allocation method A;

[0039] S424, calculating the cleaning throughput of allocation method A. If the throughput is less than a preset threshold, allocation method A is considered infeasible, and the allocation basis for cleaning allocation optimization is adjusted, and returning to step S422;

[0040] S425. If the throughput of allocation method A is greater than a preset threshold, the number of backtracking times is adjusted, and a sub-allocation method is re-formulated to replace allocation method A according to the preset rules of the cleaning allocation algorithm, so as to optimize the cleaning allocation between the cleaning team combinations.

[0041] Preferably, based on the cleaning distribution model, extracting historical cleaning records to establish a time prediction model to predict the cleaning completion time of each key component in the refrigeration unit includes the following steps:

[0042] S431, extracting historical cleaning records as a cleaning time sample set, and constructing a cleaning time attribute variable based on the cleaning team combination using an attribute reduction algorithm;

[0043] S432, dividing the cleaning time sample set, establishing a sequence pair between the cleaning team combination and the cleaning completion time as an approximate space, and defining an upper approximate set and a lower approximate set for the approximate space, thereby simplifying the cleaning time attribute variable;

[0044] S433, presetting the cleaning duration of each cleaning team combination, and dividing the preset cleaning duration into several levels, and generating a cleaning time attribute variable combination corresponding to the cleaning team combination;

[0045] S434, using the relative generalization principle of equivalence relations to establish a cleaning time attribute variable test equation, and using the cleaning time attribute variable test equation to test the cleaning time attribute variable combination;

[0046] S435. Establish a multivariate time decision tree based on the verified cleaning time attribute variable combination, analyze the cleaning time attribute variables on the leaf nodes of the multivariate time decision tree, and predict the cleaning completion time of the cleaning team combination.

[0047] Preferably, the expression of the cleaning time attribute variable test equation is:

[0048] ;

[0049] In the formula, C represents the cleaning time attribute variable test;

[0050] P Represents the core attribute set after the cleaning time attribute variable is simplified;

[0051] K Indicates the number of core attribute sets.

[0052] According to another aspect of the present invention, a prediction system for cleaning time of a refrigeration unit is also provided, the prediction system for cleaning time of a refrigeration unit comprises a dirt accumulation pattern recognition module, a dirt accumulation degree judgment module, a cleaning degree evaluation module and a cleaning time prediction module;

[0053] The dirt accumulation pattern recognition module is sequentially connected to the dirt accumulation degree judgment module, the cleaning degree evaluation module and the cleaning time prediction module;

[0054] A dirt accumulation pattern recognition module is used to obtain historical operation data of the refrigeration unit and identify the dirt accumulation pattern of each key component in the refrigeration unit based on the historical operation data;

[0055] A fouling accumulation degree judgment module is used to analyze the heat transfer efficiency and pressure loss of each key component based on the fouling accumulation mode of each key component, and judge the fouling accumulation degree of each key component based on the analysis result;

[0056] The cleaning degree assessment module is used to establish a three-dimensional model of the refrigeration unit and comprehensively assess the cleaning difficulty of each key component based on the structural characteristics and dirt accumulation of each key component;

[0057] The cleaning time prediction module is used to match the corresponding cleaning method to each key component based on the cleaning difficulty of each key component, and predict the cleaning completion time of the refrigeration unit.

[0058] The beneficial effects of the present invention are:

[0059] 1. The present invention can accurately identify the dirt accumulation pattern of key components by acquiring and analyzing the historical operation data of the refrigeration unit, and can match the most appropriate cleaning method for each component by comprehensively evaluating the dirt accumulation degree and cleaning difficulty of key components such as the condenser, evaporator, pipes and filters. This customized cleaning method not only improves the cleaning efficiency, but also helps to extend the service life of the equipment. At the same time, by predicting the cleaning time, it helps to avoid production stagnation and increased maintenance costs caused by sudden failures. Timely maintenance can reduce unexpected failures and thus control operating costs.

[0060] 2. The present invention matches the cleaning method by considering the difficulty of cleaning each key component, ensuring that the most appropriate cleaning strategy is adopted for each key component. This not only improves the cleaning efficiency, but also helps to protect the equipment and extend its service life. The cleaning allocation algorithm is used to optimize the allocation of cleaning teams to ensure the rational use of resources. By quantifying the workload and defining the correlation between teams, and using a greedy algorithm for optimization, these measures help to improve the efficiency and effectiveness of the entire cleaning process. At the same time, the introduction of sequence analysis technology and the use of historical cleaning records to establish a time prediction model can accurately predict the cleaning completion time of each key component, which in turn helps to better plan the cleaning operation and improve production efficiency.

[0061] 3. The present invention can dynamically optimize the cleaning distribution by evaluating the throughput of the distribution method and adjusting the number of backtracking times to ensure the efficiency and effectiveness of the cleaning process. This dynamic adjustment mechanism makes the cleaning plan more flexible and able to cope with emergencies. The established multivariate time decision tree provides strong decision support for the cleaning of the refrigeration unit, making the cleaning plan more scientific and reasonable. By analyzing the decision tree to predict the cleaning completion time, managers can more accurately predict and plan cleaning operations, thereby improving the quality and efficiency of decision-making. BRIEF DESCRIPTION OF THE DRAWINGS

[0062] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0063] Figure 1 is a flow chart of a method for predicting cleaning time of a refrigeration unit according to an embodiment of the present invention;

[0064] Figure 2 The present invention is a block diagram of a system for predicting the cleaning time of a refrigeration unit according to an embodiment of the present invention.

[0065] In the figure:

[0066] 1. Dirt accumulation pattern recognition module; 2. Dirt accumulation degree judgment module; 3. Cleaning degree assessment module; 4. Cleaning time prediction module. DETAILED DESCRIPTION

[0067] To further illustrate each embodiment, the present invention provides drawings, which are part of the disclosure of the present invention and are mainly used to illustrate the embodiments and can be used in conjunction with the relevant descriptions in the specification to explain the operating principles of the embodiments. With reference to these contents, ordinary technicians in the field should be able to understand other possible implementations and advantages of the present invention. The components in the figures are not drawn to scale, and similar component symbols are generally used to represent similar components.

[0068] According to an embodiment of the present invention, a method and system for predicting cleaning time of a refrigeration unit are provided.

[0069] The present invention is further described with reference to the accompanying drawings and specific embodiments. Figure 1 As shown, according to the method for predicting the cleaning time of a refrigeration unit according to an embodiment of the present invention, the method for predicting the cleaning time of a refrigeration unit comprises the following steps:

[0070] S1. Obtain historical operation data of the refrigeration unit, and identify the dirt accumulation pattern of each key component in the refrigeration unit based on the historical operation data.

[0071] Among them, the historical operation data of the refrigeration unit includes historical ambient temperature data, historical pressure data, historical energy consumption data, historical operation time, historical operation cycle and historical cleaning records;

[0072] The key components in the refrigeration unit include at least the condenser, evaporator, pipes and filters.

[0073] It should be noted that obtaining the historical operation data of the refrigeration unit and identifying the dirt accumulation mode of each key component in the refrigeration unit based on the historical operation data includes the following steps:

[0074] Analyze historical operating data and use trend analysis technology to find signs of dirt accumulation patterns of key components; extract features from historical operating data, such as temperature changes, pressure changes, energy consumption changes, etc. of each component, as well as the relationship between these changes and time, workload, environmental conditions, etc.; use machine learning methods to identify the dirt accumulation pattern of key components (air conditioners, compressors, pipes and internal filters of pipes, etc.). The dirt accumulation pattern includes periodic dirt accumulation, temperature gradient changes, increased energy consumption and other characteristics, explain the identified dirt accumulation pattern, and analyze its possible causes, such as the impact of environmental conditions, refrigerant properties, operating methods and other factors on dirt accumulation; the dirt accumulation pattern may be different for different key components. For example, the compressor may have internal scaling, resulting in reduced energy efficiency, the condenser may have reduced heat dissipation due to external dirt adhesion, and the evaporator may have frosting affecting the heat transfer efficiency.

[0075] It should be noted that for the identification of fouling accumulation patterns, machine learning methods include clustering algorithms (such as K-means, DBSCAN), time series analysis (such as ARIMA), classification algorithms (such as decision trees, random forests), etc., and the identification of fouling accumulation patterns of condensers mainly includes the following aspects:

[0076] Cyclic changes: Fouling may accumulate over a specific time period, such as seasonal changes or accumulation after long-term operation.

[0077] Types of accumulated dirt: including scale, oil, rust, etc.

[0078] Non-linear trends: The performance of the condenser may show a non-linear decline over time, which may be related to fouling accumulation.

[0079] Anomaly Detection: Detects abnormal points in operating data related to dirt accumulation.

[0080] Accumulation mode: includes accumulation speed, location and density.

[0081] S2. Analyze the heat transfer efficiency and pressure loss of each key component based on the fouling accumulation pattern of each key component, and determine the degree of fouling accumulation of each key component based on the analysis results.

[0082] The steps of analyzing the heat transfer efficiency and pressure loss of each key component based on the fouling accumulation mode of each key component and judging the fouling accumulation degree of each key component based on the analysis results include the following steps:

[0083] S21, extracting the dirt accumulation pattern of the condenser from step S1, identifying and dividing the dirt accumulation-prone areas of the condenser.

[0084] It should be noted that areas prone to fouling are areas in the condenser where temperature changes are large, pressure losses are significant, or energy consumption increases. The fouling accumulation pattern is analyzed from the operating data of the condenser, and areas prone to fouling are identified, so that effective preventive measures and cleaning strategies can be taken to improve the operating efficiency of the condenser and extend its service life.

[0085] S22. Calculate the heat transfer temperature of each area of ​​the condenser that is prone to contamination using a heat transfer model, and comprehensively determine the heat transfer efficiency of the condenser based on the heat transfer temperature of each area that is prone to contamination.

[0086] The heat transfer model is used to calculate the heat transfer temperature of each easily polluted area of ​​the condenser, and the heat transfer efficiency of the condenser is comprehensively determined based on the heat transfer temperature of each easily polluted area, including the following steps:

[0087] S221. Obtain the wall heat flux and the hourly outlet temperature of each area prone to contamination in the condenser, and analyze the internal and external temperature difference of the area prone to contamination based on the wall heat flux and the hourly outlet temperature of the heat flux.

[0088] It should be noted that in obtaining the wall heat flux and the hourly outlet temperature of each area prone to fouling in the condenser, and analyzing the internal and external temperature difference of the area prone to fouling based on the wall heat flux and the hourly outlet temperature of the heat flux, the purpose of this step is to obtain the wall heat flux and the hourly outlet temperature of each area prone to fouling in the condenser. These data are the basis for analyzing the heat transfer efficiency, which specifically include:

[0089] Real-time data of each area of ​​the condenser that is prone to fouling is collected by using devices such as temperature sensors, flow meters and heat flow sensors. The collected data is processed to calculate the wall heat flow and the hourly outlet temperature of the heat flow. Based on the wall heat flow and the hourly outlet temperature of the heat flow, the internal and external temperature difference of the area prone to fouling is analyzed, and the temperature difference is used as an indicator of fouling accumulation.

[0090] It should be noted that the calculation of the wall heat flux and the hourly outlet temperature of the heat flux includes the following steps:

[0091] Define the system boundary of the condenser, including all energy inflow and outflow points; calculate the internal and external temperature difference of the condenser wall, that is, the difference between the wall temperature and the ambient temperature; determine the heat transfer coefficient based on the wall material, fluid type and flow state; calculate the wall heat flow using Fourier's law in existing technology.

[0092] S222. Analyze and calculate the wall temperature transfer function and heat flow transfer function of the condenser within a time period based on the internal and external temperature difference of the area prone to contamination.

[0093] It should be noted that the wall temperature transfer function and the heat flow transfer function describe the relationship between the wall temperature and the heat flow. These functions are crucial to understanding the heat transfer process. The wall temperature transfer function and the heat flow transfer function can be in a variety of functional forms. The specific selection depends on factors such as the characteristics of the heat transfer process, the geometric structure of the system, and the operating conditions.

[0094] The wall temperature transfer function often takes the form of an exponential function, such as exponential decay or exponential growth. This function form can better describe the trend of wall temperature changes over time. For the convective heat transfer of fluid on the wall, Newton's cooling law is used to describe the relationship between heat flow and temperature difference.

[0095] S223, constructing a heat transfer model using the wall temperature transfer function and the heat flow transfer function, outputting the temperature distribution of the refrigerant inside the condenser through the heat transfer model, and assigning a value to the temperature distribution as the heat transfer temperature at the condenser outlet;

[0096] S224. Continuously calculate the heat transfer temperature of each area of ​​the condenser that is prone to fouling based on the heat transfer temperature at the condenser outlet, and determine a reasonable value of the heat transfer temperature of each area that is prone to fouling based on the law of conservation of energy.

[0097] It should be noted that the law of conservation of energy expresses the balance between energy input and output, and is used here to calculate the heat transfer temperature of each area. The heat transfer term in the law of conservation of energy is used in combination with the characteristics and operating parameters of the condenser to calculate the heat transfer temperature of each area prone to fouling. If fouling has an impact on heat transfer, the thermal resistance of the fouling layer needs to be considered in the calculation to accurately reflect the impact of fouling accumulation on the heat transfer temperature. If the changes in heat transfer temperature at multiple time points or cycles need to be considered, iterative calculations are performed to gradually optimize the heat transfer temperature of each area prone to fouling.

[0098] It should be noted that the law of conservation of energy is one of the basic principles in the process of heat transfer. It expresses the relationship of energy conservation in the process of energy transfer and conversion within the system. In the process of heat transfer, energy input includes the energy input from the heat source, energy output is the energy released by the system, and energy accumulation is the change in energy stored in the system. Energy input is the sum of energy output and energy accumulation.

[0099] S225. Utilize a heat transfer model to calculate the temperature distribution of the refrigerant along the path when a reasonable value is reached, and comprehensively determine the heat transfer efficiency of the condenser based on the temperature distribution along the path.

[0100] It should be noted that the heat transfer model is used to calculate the temperature distribution of the refrigerant along the path when a reasonable value is reached, and the heat transfer efficiency of the condenser is comprehensively determined based on the temperature distribution along the path, including the following steps:

[0101] Determine the boundary conditions of the model, including the refrigerant temperature at the inlet, the heat transfer temperature at the outlet, the temperature distribution on the wall, etc.; set initial conditions, such as the initial temperature distribution of the refrigerant, and any known initial dirt layer thickness and thermal resistance, etc.; use the finite difference method in the prior art to solve the heat transfer model to obtain the temperature distribution of the refrigerant along the way; analyze the solved temperature distribution to determine the temperature change of the refrigerant inside the condenser; calculate the heat transfer efficiency of the condenser based on the temperature distribution; identify possible heat transfer bottlenecks or inefficient areas based on the calculation results, and consider optimization measures.

[0102] S23. Evaluate the degree of fouling accumulation of the condenser based on the heat transfer efficiency of the condenser;

[0103] S24. Analyze the pressure loss inside the pipeline when air flow is introduced by using the pressure difference monitoring method, and judge the degree of dirt accumulation in the pipeline and the filter screen inside the pipeline based on the pressure loss inside the pipeline.

[0104] Among them, using the pressure difference monitoring method to analyze the pressure loss inside the pipeline when the air flow is introduced, and judging the degree of dirt accumulation in the pipeline and the filter screen inside the pipeline based on the pressure loss inside the pipeline respectively includes the following steps:

[0105] S241, extracting the dirt accumulation pattern of the pipeline from step S1, and identifying the airflow pressure and airflow velocity when airflow is introduced into the pipeline.

[0106] It should be noted that compressed air is introduced into the pipeline system using an air compressor, usually starting from one end of the pipeline, and pressure gauges, flow meters, thermometers and other detection equipment are used to measure and record the airflow pressure and airflow velocity in real time, analyze the measurement data, compare the data at different time points and different positions in the pipeline, and identify the pressure and flow velocity distribution in the pipeline.

[0107] S242, establishing a fluid kinematics model, and inputting the airflow pressure and airflow velocity inside the pipeline into the fluid kinematics model, and outputting the pressure test result inside the pipeline through the fluid kinematics model.

[0108] It should be noted that establishing a fluid kinematics model, inputting the airflow pressure and airflow velocity inside the pipeline into the fluid kinematics model, and outputting the pressure test results inside the pipeline through the fluid kinematics model include the following steps:

[0109] Determine the parameters in the Bernoulli model based on the type of fluid in the pipeline (such as gas or liquid), flow characteristics (such as laminar flow or turbulent flow), and complexity of the flow; set boundary conditions according to the actual conditions of the pipeline, including the flow rate and pressure at the inlet and outlet; extract the airflow pressure and airflow velocity data inside the pipeline, and use the finite difference method in the above content to solve the fluid kinematic model; verify the pressure distribution obtained to ensure that it conforms to physical laws and actual engineering requirements; convert the pressure distribution data output by the model into a readable form, such as a chart or report, to facilitate analysis and decision-making.

[0110] S243. Considering the influence of pipeline roughness change on pressure test results, a correction factor is introduced to predict the pressure loss of the pipeline under actual airflow operating conditions.

[0111] It should be noted that, in combination with the influence of pipeline roughness change on pressure test results, by introducing a correction coefficient, the prediction of the pressure loss of the pipeline under actual airflow operating conditions includes the following steps:

[0112] Analyze the influence of pipeline roughness on airflow movement. Roughness is a parameter that describes the roughness of the inner wall of the pipeline, which affects the friction loss and pressure distribution of the fluid; determine the correction coefficient of roughness change on pressure loss based on fluid dynamics theory to determine the relationship between roughness change and pressure loss; establish a correction model for pipeline pressure loss based on the correction coefficient. This model will consider the influence of roughness change on pressure loss, and can predict pressure loss under different roughness conditions, simulate actual airflow operating conditions, including flow rate, temperature, humidity and other factors, and input these conditions into the correction model; use the correction model to predict pipeline pressure loss under actual airflow operating conditions.

[0113] S244. Compare the pressure loss under the actual airflow operating conditions with the preset pressure difference value. If the pressure loss is greater than the preset threshold, it means that the degree of dirt accumulation in the pipeline and the filter inside the pipeline is high. If the pressure loss is less than the preset threshold, it means that the degree of dirt accumulation in the pipeline and the filter inside the pipeline is low.

[0114] S3. Establish a three-dimensional model of the refrigeration unit, and comprehensively evaluate the cleaning difficulty of each key component based on its structural characteristics and the degree of dirt accumulation.

[0115] It should be noted that the following steps are involved in establishing a three-dimensional model of the refrigeration unit and comprehensively evaluating the cleaning difficulty of each key component based on the structural characteristics of each key component and the degree of dirt accumulation:

[0116] Collect the design drawings, technical specifications and dimensional data of the actual components of the refrigeration unit, and select 3D modeling software, such as SolidWorks, AutoCAD, Inventor, etc., to build a 3D model of the refrigeration unit in the software to ensure the accuracy of the model.

[0117] Classify the key components of the refrigeration unit, such as compressor, condenser, evaporator, throttling device, etc., analyze the structural characteristics of each component, such as shape, size, connection method and material, and evaluate the impact of these structural characteristics on the difficulty of cleaning.

[0118] Different cleaning methods are adopted based on different key components, such as mechanical cleaning, chemical cleaning, high-pressure water jet cleaning, etc., and evaluation criteria are formulated, including cleaning time, labor requirements, equipment costs, safety risks, etc.

[0119] Combined with the structural characteristics of the components and the degree of dirt accumulation, a comprehensive assessment is made on the cleaning difficulty of each key component, and a detailed assessment report is prepared, including the cleaning difficulty of each component and the recommended cleaning method.

[0120] S4. Match a corresponding cleaning method to each key component based on the cleaning difficulty of each key component, and predict the cleaning completion time of the refrigeration unit.

[0121] Among them, matching corresponding cleaning methods for each key component based on the cleaning difficulty of each key component, and predicting the cleaning completion time of the refrigeration unit includes the following steps:

[0122] S41. Based on the cleaning difficulty of each key component in the refrigeration unit, a cleaning method corresponding to each key component is matched, and a cleaning team combination is generated based on the characteristics of the cleaning method.

[0123] It should be noted that the cleaning team consists of:

[0124] Professional cleaning personnel: personnel with specific cleaning skills, such as high-pressure water jet cleaning, chemical cleaning, etc., responsible for performing specific cleaning tasks, such as cleaning key components such as evaporators and condensers.

[0125] Cleaning equipment: Choose appropriate cleaning equipment according to the needs of the cleaning task, such as high-pressure cleaners, vacuum cleaners, steam cleaners, etc.

[0126] Equipped with necessary cleaning tools, such as spray guns, brushes, scrapers, cleaning cloths, etc., choose appropriate tools and accessories according to different cleaning tasks.

[0127] Cleaning processes and procedures: Develop detailed cleaning processes and operating procedures to ensure that cleaning work is carried out in an orderly manner.

[0128] S42. Optimize the cleaning allocation among the cleaning team combinations using a cleaning allocation algorithm.

[0129] Among them, using the cleaning allocation algorithm to optimize the cleaning allocation between the cleaning team combinations includes the following steps:

[0130] S421, define and quantify the cleaning workload of each cleaning team combination, and use the cleaning workload as the allocation basis when solving the cleaning allocation optimization;

[0131] S422. Set the number of backtracking times, and determine whether the cleaning workload of each cleaning team combination meets the standard based on the set number of backtracking times.

[0132] It should be noted that setting the number of backtracking times and judging whether the cleaning workload of each cleaning team combination meets the standard according to the set number of backtracking times includes the following steps:

[0133] The number of backtracking is determined according to the complexity and importance of the cleaning task. The more backtracking times, the more reliable the evaluation result is, but it also requires more computing and time costs.

[0134] For each cleaning team combination, its cleaning workload is calculated based on the specific requirements of the cleaning task and the workload standard. The workload includes indicators such as the area to be cleaned, the number of equipment, and the cleaning time.

[0135] Start backtracking calculation, starting from the first cleaning, gradually increase the number of backtracking times, and recalculate the cleaning workload of each cleaning team combination.

[0136] After each backtracking, the cleaning workload of each cleaning team combination is evaluated. If the cleaning workload of a combination meets the standard within the set number of backtrackings, the combination is judged to meet the cleaning requirements.

[0137] Summarize the cleaning workload of each cleaning team combination under different backtracking times to form statistical data on whether the cleaning workload meets the standards. According to the results of the backtracking calculation, optimize and adjust the cleaning team combination to select a combination with a reasonable workload and that meets the standards.

[0138] S423, using a greedy algorithm to sort the cleaning workload of each cleaning team combination, and assigning the cleaning team combination with the largest cleaning workload to the cleaning team combination with the smallest cleaning workload one by one, and recording the current allocation method as allocation method A.

[0139] It should be noted that the greedy algorithm is an algorithm that takes the best or optimal choice in the current state in each step, hoping to lead to the best or optimal result in the world. When the cleaning workload of the cleaning team combinations is sorted and allocated in the order of the largest cleaning workload to the smallest cleaning workload, the specific content of the greedy algorithm is as follows:

[0140] Collect the cleaning workload data of all cleaning team combinations, and sort all cleaning team combinations from large to small according to the cleaning workload; select the cleaning team combination with the largest cleaning workload from the sorted list; assign the cleaning team combination to the cleaning team combination with the smallest current cleaning workload. This step can ensure that each allocation is reducing the imbalance of the total workload; after the allocation, update the cleaning workload of the assigned cleaning team combination, which usually means adding the workload of the two combinations. Since the workload has been updated, the cleaning team combinations need to be re-sorted to ensure the accuracy of the next selection; repeat the above steps until all cleaning team combinations are allocated; record the cleaning team combination allocated each time to form allocation method A.

[0141] S424, calculating the cleaning throughput of allocation method A. If the throughput is less than a preset threshold, allocation method A is considered infeasible, and the allocation basis for cleaning allocation optimization is adjusted, and returning to step S422;

[0142] S425. If the throughput of allocation method A is greater than a preset threshold, the number of backtracking times is adjusted, and a sub-allocation method is re-formulated to replace allocation method A according to the preset rules of the cleaning allocation algorithm, so as to optimize the cleaning allocation between the cleaning team combinations.

[0143] It should be noted that the cleaning throughput is calculated based on the collected data. In the process of cleaning task allocation and optimization, the cleaning throughput refers to the total amount of cleaning work that can be completed by the cleaning team or system within a certain period of time. This is a key indicator for measuring cleaning efficiency, usually expressed in terms of the area cleaned per unit time, the number of equipment, or other units of workload. For example, if an area of ​​100 square meters is cleaned in 2 hours, the cleaning throughput is 50 square meters / hour.

[0144] S43. Based on the cleaning allocation model, historical cleaning records are extracted to establish a time prediction model to predict the cleaning completion time of each key component in the refrigeration unit.

[0145] Among them, based on the cleaning allocation model, extracting historical cleaning records to establish a time prediction model to predict the cleaning completion time of each key component in the refrigeration unit includes the following steps:

[0146] S431, extracting historical cleaning records as a cleaning time sample set, and constructing a cleaning time attribute variable based on the cleaning team combination using an attribute reduction algorithm;

[0147] S432, dividing the cleaning time sample set, establishing the sequence pairs between the cleaning team combination and the cleaning completion time as the approximate space, and defining the upper approximate set and the lower approximate set for the approximate space to achieve simplification of the cleaning time attribute variables.

[0148] It should be noted that in the approximate space, according to the attribute variables of the cleaning completion time, an upper approximate set and a lower approximate set can be defined. The upper approximate set contains sample points with shorter cleaning completion time, while the lower approximate set contains sample points with longer cleaning completion time. By comparing the differences between the upper approximate set and the lower approximate set, the main attribute variables affecting the cleaning time can be identified. These attribute variables may include factors such as the composition of the cleaning team, the use of cleaning equipment, and the size of the cleaning area. Based on the identified main attribute variables, the attributes of the cleaning time are simplified, and the attribute simplification is achieved by excluding variables that have little impact on the cleaning time or merging related variables.

[0149] S433, presetting the cleaning duration of each cleaning team combination, and dividing the preset cleaning duration into several levels, and generating a cleaning time attribute variable combination corresponding to the cleaning team combination;

[0150] S434. Establish a cleaning time attribute variable test equation using the relative generalization principle of equivalence relations, and use the cleaning time attribute variable test equation to test the cleaning time attribute variable combination.

[0151] It should be noted that the principle of relative generalization of equivalence relations is a concept in set theory. It is used to define equivalence relations. Equivalence relations are a binary relation on a set. It has three basic properties: reflexivity, symmetry, and transitivity. As long as a binary relation satisfies reflexivity, symmetry, and transitivity, it is an equivalence relation.

[0152] Among them, the expression of the cleaning time attribute variable test equation is:

[0153] ;

[0154] In the formula, C represents the cleaning time attribute variable test;

[0155] P Represents the core attribute set after the cleaning time attribute variable is simplified;

[0156] K Indicates the number of core attribute sets.

[0157] S435. Establish a multivariate time decision tree based on the verified cleaning time attribute variable combination, analyze the cleaning time attribute variables on the leaf nodes of the multivariate time decision tree, and predict the cleaning completion time of the cleaning team combination.

[0158] It should be noted that the multivariate time decision tree is established based on the verified cleaning time attribute variable combination, the cleaning time attribute variables on the leaf nodes of the multivariate time decision tree are analyzed, and the cleaning completion time of the cleaning team combination is predicted, which includes the following steps:

[0159] A multivariate time decision tree model is established. The time decision tree is a special decision tree that takes into account the dynamic characteristics of time series data. A part of the data (training set) is used to train the decision tree model. Another part of the data (validation set) is used to verify the accuracy of the model. The cleaning time attribute variables on the leaf nodes of the decision tree are analyzed. The trained decision tree model is used to predict the cleaning completion time of the new cleaning team combination, and the cleaning completion time of all key components in the refrigeration unit is generated based on the cleaning completion time of the cleaning team combination.

[0160] It should be noted that the multivariate time decision tree (a decision tree model, which is specially designed to process and analyze time series data with multiple feature variables, and these feature variables change over time. The characteristics of the multivariate time decision tree mainly include:

[0161] Handling multivariate data: MTDT can consider multiple feature variables simultaneously, which can be synchronized or asynchronous.

[0162] Time series features: MTDT is able to process time series data, which means it can take into account time order and trends.

[0163] Nonparametric methods: Decision trees are usually nonparametric, meaning that they do not require making assumptions about the distribution of the data.

[0164] Easy to understand: The decision tree structure generated by MTDT is intuitive and easy to understand, which helps to explain the decision-making process of the model.

[0165] Steps to build a multivariate time decision tree: Collect multivariate time series data and ensure the accuracy and completeness of the data; Select features related to the target variable, which should be able to provide useful information about the target variable; Select a decision tree algorithm, such as CART (Classification and Regression Tree), ID3 (Iterative Dichotomiser 3), and use the training data set to train the decision tree model. During the training process, the algorithm will automatically select the best features and split points. In order to prevent overfitting, the decision tree needs to be pruned, and the validation data set is used to evaluate the performance of the model.

[0166] like Figure 2 As shown, according to another embodiment of the present invention, a prediction system for cleaning time of a refrigeration unit is also provided, and the prediction system for cleaning time of a refrigeration unit comprises a dirt accumulation pattern recognition module 1, a dirt accumulation degree judgment module 2, a cleaning degree evaluation module 3 and a cleaning time prediction module 4;

[0167] The dirt accumulation pattern recognition module 1 is connected in sequence to the dirt accumulation degree judgment module 2, the cleaning degree evaluation module 3 and the cleaning time prediction module 4;

[0168] The dirt accumulation pattern recognition module 1 is used to obtain the historical operation data of the refrigeration unit and recognize the dirt accumulation pattern of each key component in the refrigeration unit based on the historical operation data;

[0169] A fouling accumulation degree judgment module 2 is used to analyze the heat transfer efficiency and pressure loss of each key component based on the fouling accumulation mode of each key component, and judge the fouling accumulation degree of each key component based on the analysis result;

[0170] Cleaning degree assessment module 3 is used to establish a three-dimensional model of the refrigeration unit and comprehensively assess the cleaning difficulty of each key component based on the structural characteristics and dirt accumulation degree of each key component;

[0171] The cleaning time prediction module 4 is used to match the corresponding cleaning method to each key component based on the cleaning difficulty of each key component, and predict the cleaning completion time of the refrigeration unit.

[0172] In summary, with the help of the above-mentioned technical scheme of the present invention, the present invention can accurately identify the dirt accumulation pattern of key components by acquiring and analyzing the historical operation data of the refrigeration unit, and can match the most appropriate cleaning method for each component by comprehensively evaluating the dirt accumulation degree and cleaning difficulty of key components such as condensers, evaporators, pipes and filters. This customized cleaning method not only improves the cleaning efficiency, but also helps to extend the service life of the equipment. At the same time, by predicting the cleaning time, it helps to avoid production stagnation and increased maintenance costs caused by sudden failures. Timely maintenance can reduce unexpected failures and thus control operating costs. The present invention matches the cleaning method by considering the cleaning difficulty of each key component, ensuring that each key component adopts the most appropriate cleaning strategy, which not only improves the cleaning efficiency, but also helps to protect the equipment and extend its service life. The cleaning team is optimized using a cleaning allocation algorithm. Allocation ensures the rational use of resources. By quantifying the workload and defining the correlation between teams, and optimizing with a greedy algorithm, these measures help to improve the efficiency and effectiveness of the entire cleaning process. At the same time, the introduction of sequence analysis technology and the use of historical cleaning records to establish a time prediction model can accurately predict the cleaning completion time of each key component, thereby helping to better plan the cleaning operation and improve production efficiency. The present invention can dynamically optimize the cleaning allocation by evaluating the throughput of the allocation method and adjusting the number of backtracking times to ensure the efficiency and effectiveness of the cleaning process. This dynamic adjustment mechanism makes the cleaning plan more flexible and able to respond to emergencies. The established multivariate time decision tree provides strong decision support for the cleaning of the refrigeration unit, making the cleaning plan more scientific and reasonable. By analyzing the decision tree to predict the cleaning completion time, managers can more accurately predict and plan cleaning operations, thereby improving the quality and efficiency of decision-making.

[0173] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for predicting the cleaning time of a refrigeration unit, characterized in that: The method for predicting the cleaning time of the refrigeration unit comprises the following steps: S1. Obtain historical operation data of the refrigeration unit, and identify the dirt accumulation pattern of each key component in the refrigeration unit based on the historical operation data; S2. analyzing the heat transfer efficiency and pressure loss of each key component based on the fouling accumulation pattern of each key component, and determining the fouling accumulation degree of each key component based on the analysis results; S3. Establish a three-dimensional model of the refrigeration unit, and comprehensively evaluate the cleaning difficulty of each key component based on the structural characteristics and dirt accumulation degree of each key component; S4. Matching a corresponding cleaning method to each key component based on the cleaning difficulty of each key component, and predicting the cleaning completion time of the refrigeration unit; The historical operation data of the refrigeration unit includes historical ambient temperature data, historical pressure data, historical energy consumption data, historical operation time, historical operation cycle and historical cleaning records; The key components of the refrigeration unit include at least a condenser, an evaporator, a pipeline and a filter; The method of matching a corresponding cleaning method to each key component based on the cleaning difficulty of each key component and predicting the cleaning completion time of the refrigeration unit includes the following steps: S41, matching the cleaning methods corresponding to the key components in the refrigeration unit based on the cleaning difficulty, and generating a cleaning team combination based on the cleaning method characteristics; S42, optimizing the cleaning allocation among the cleaning team combinations by using a cleaning allocation algorithm; S43, based on the cleaning distribution model, extract historical cleaning records to establish a time prediction model to predict the cleaning completion time of each key component in the refrigeration unit; The use of a cleaning allocation algorithm to optimize cleaning allocation between cleaning team combinations comprises the following steps: S421, define and quantify the cleaning workload of each cleaning team combination, and use the cleaning workload as the allocation basis when solving the cleaning allocation optimization; S422, setting the number of backtracking times, and judging whether the cleaning workload of each cleaning team combination meets the standard according to the set number of backtracking times; S423, using a greedy algorithm to sort the cleaning workload of each cleaning team combination, one by one assigning the cleaning team combination with the largest cleaning workload to the cleaning team combination with the smallest cleaning workload, and recording the current allocation method as allocation method A; S424, calculating the cleaning throughput of allocation method A. If the throughput is less than a preset threshold, allocation method A is considered infeasible, and the allocation basis for cleaning allocation optimization is adjusted, and returning to step S422; S425. If the throughput of allocation method A is greater than a preset threshold, the number of backtracking times is adjusted, and a sub-allocation method is re-formulated to replace allocation method A according to the preset rules of the cleaning allocation algorithm, so as to optimize the cleaning allocation between the cleaning team combinations; The method of extracting historical cleaning records to establish a time prediction model to predict the cleaning completion time of each key component in the refrigeration unit based on the cleaning distribution model includes the following steps: S431, extracting historical cleaning records as a cleaning time sample set, and constructing a cleaning time attribute variable based on the cleaning team combination using an attribute reduction algorithm; S432, dividing the cleaning time sample set, establishing a sequence pair between the cleaning team combination and the cleaning completion time as an approximate space, and defining an upper approximate set and a lower approximate set for the approximate space, thereby simplifying the cleaning time attribute variable; S433, presetting the cleaning duration of each cleaning team combination, and dividing the preset cleaning duration into several levels, and generating a cleaning time attribute variable combination corresponding to the cleaning team combination; S434, using the relative generalization principle of equivalence relations to establish a cleaning time attribute variable test equation, and using the cleaning time attribute variable test equation to test the cleaning time attribute variable combination; S435. Establish a multivariate time decision tree based on the verified cleaning time attribute variable combination, analyze the cleaning time attribute variables on the leaf nodes of the multivariate time decision tree, and predict the cleaning completion time of the cleaning team combination.

2. A method for predicting the cleaning time of a refrigeration unit according to claim 1, characterized in that: The method of analyzing the heat transfer efficiency and pressure loss of each key component based on the fouling accumulation mode of each key component and determining the fouling accumulation degree of each key component based on the analysis result comprises the following steps: S21, extracting the dirt accumulation pattern of the condenser from step S1, identifying and dividing the dirt accumulation prone area of ​​the condenser; S22, using a heat transfer model to calculate the heat transfer temperature of each easily polluted area of ​​the condenser, and comprehensively determine the heat transfer efficiency of the condenser based on the heat transfer temperature of each easily polluted area; S23. Evaluate the degree of fouling accumulation of the condenser based on the heat transfer efficiency of the condenser; S24. Analyze the pressure loss inside the pipeline when air flow is introduced by using the pressure difference monitoring method, and judge the degree of dirt accumulation in the pipeline and the filter screen inside the pipeline based on the pressure loss inside the pipeline.

3. A method for predicting the cleaning time of a refrigeration unit according to claim 2, characterized in that: The method of using the heat transfer model to calculate the heat transfer temperature of each easily polluted area of ​​the condenser, and comprehensively determining the heat transfer efficiency of the condenser based on the heat transfer temperature of each easily polluted area includes the following steps: S221, obtaining the wall heat flux and the hourly outlet temperature of each area prone to contamination in the condenser, and analyzing the internal and external temperature difference of the area prone to contamination based on the wall heat flux and the hourly outlet temperature of the heat flux; S222, analyzing and calculating the wall temperature transfer function and heat flow transfer function of the condenser within a time period based on the internal and external temperature difference of the area prone to contamination; S223, constructing a heat transfer model using the wall temperature transfer function and the heat flow transfer function, outputting the temperature distribution of the refrigerant inside the condenser through the heat transfer model, and assigning a value to the temperature distribution as the heat transfer temperature at the condenser outlet; S224, continuously calculating the heat transfer temperature of each easily polluted area of ​​the condenser based on the heat transfer temperature at the condenser outlet, and determining a reasonable value of the heat transfer temperature of each easily polluted area based on the law of conservation of energy; S225. Utilize a heat transfer model to calculate the temperature distribution of the refrigerant along the path when a reasonable value is reached, and comprehensively determine the heat transfer efficiency of the condenser based on the temperature distribution along the path.

4. A method for predicting the cleaning time of a refrigeration unit according to claim 3, characterized in that: The method of using the pressure difference monitoring method to analyze the pressure loss inside the pipeline when the airflow is introduced, and judging the degree of dirt accumulation in the pipeline and the filter screen inside the pipeline based on the pressure loss inside the pipeline, respectively, includes the following steps: S241, extracting the dirt accumulation pattern of the pipeline from step S1, and identifying the airflow pressure and airflow velocity when airflow is introduced into the pipeline; S242, establishing a fluid kinematics model, and inputting the airflow pressure and airflow velocity inside the pipeline into the fluid kinematics model, and outputting the pressure test result inside the pipeline through the fluid kinematics model; S243. Combined with the influence of pipeline roughness change on pressure test results, the pressure loss of pipeline under actual airflow operating conditions is predicted by introducing correction coefficient; S244. Compare the pressure loss under the actual airflow operating conditions with the preset pressure difference value. If the pressure loss is greater than the preset threshold, it means that the degree of dirt accumulation in the pipeline and the filter inside the pipeline is high. If the pressure loss is less than the preset threshold, it means that the degree of dirt accumulation in the pipeline and the filter inside the pipeline is low.

5. A method for predicting the cleaning time of a refrigeration unit according to claim 4, characterized in that: The expression of the cleaning time attribute variable test equation is: ; In the formula, C represents the cleaning time attribute variable test; P Represents the core attribute set after the cleaning time attribute variable is simplified; K Indicates the number of core attribute sets.

6. A refrigeration unit cleaning time prediction system, used to implement the refrigeration unit cleaning time prediction method according to any one of claims 1 to 5, characterized in that: The prediction system for the cleaning time of the refrigeration unit includes a dirt accumulation pattern recognition module, a dirt accumulation degree judgment module, a cleaning degree evaluation module and a cleaning time prediction module; The dirt accumulation pattern recognition module is sequentially connected to the dirt accumulation degree judgment module, the cleaning degree evaluation module and the cleaning time prediction module; The dirt accumulation pattern recognition module is used to obtain historical operation data of the refrigeration unit and identify the dirt accumulation pattern of each key component in the refrigeration unit based on the historical operation data; The fouling accumulation degree judgment module is used to analyze the heat transfer efficiency and pressure loss of each key component based on the fouling accumulation mode of each key component, and judge the fouling accumulation degree of each key component based on the analysis result; The cleaning degree assessment module is used to establish a three-dimensional model of the refrigeration unit and comprehensively assess the cleaning difficulty of each key component based on the structural characteristics and dirt accumulation degree of each key component; The cleaning time prediction module is used to match a corresponding cleaning method to each key component based on the cleaning difficulty of each key component, and predict the cleaning completion time of the refrigeration unit.