Intelligent control method and control system for small refrigeration equipment

Abstract

The invention belongs to the technical field of refrigeration, and discloses an intelligent control method and system for small refrigeration equipment. On one hand, working condition parameters and operation parameters are monitored through various sensors, and a multi-dimensional database is established; establishing an agent model of the refrigeration equipment based on the multi-dimensional database, and performing intelligent matching control on the operation of a compressor and a fan in combination with real-time monitoring data; and on the other hand, a bid evaluation index of the period COPcyc is established to comprehensively measure energy loss in the refrigeration period and the defrosting period, the indicated defrosting amount is determined, and defrosting is started when the accumulated frosting amount reaches the indicated defrosting amount. On the basis that the refrigeration requirement of the refrigeration space is met, the temperature control precision in the refrigeration space is improved by optimizing part operation, inhibiting frost layer growth and starting defrosting according to needs, the operation energy consumption of the refrigeration equipment is reduced, and a new thought and method are provided for precise temperature control and energy conservation of the refrigeration equipment.

Description

Technical Field

[0001] The present invention belongs to the field of refrigeration technology, and in particular relates to an intelligent control method and control system for small refrigeration equipment such as refrigerators, freezers, and display cabinets. Background Art

[0002] With the continuous improvement of people's living standards and the enhancement of environmental awareness, the development trend of intelligent and energy-saving household appliances has become increasingly obvious. Refrigeration equipment such as refrigerators, freezers, display cabinets, wine cabinets, etc. have become commonly used refrigeration equipment in homes and businesses. Their performance and energy efficiency are directly related to the household's electricity burden and environmental impact. Therefore, how to effectively reduce energy consumption and optimize system operation while ensuring the refrigeration effect has become an important topic in the research and development of refrigeration technology. Traditional refrigeration systems usually use compressors and fans with fixed operating strategies. This method is difficult to adapt to the changing use environment and working conditions, resulting in high energy consumption. In addition, the frosting problem on the evaporator surface is also one of the key factors affecting equipment performance and energy efficiency. Frosting reduces the refrigeration efficiency of the equipment, blocks the air duct, causes temperature fluctuations in the refrigeration space, increases the operating burden of the compressor, and thus increases energy consumption. In order to maintain the good working condition of the evaporator and improve the control accuracy in the room, frosting should be suppressed, and the accumulated frost on the evaporator surface should be defrosted on demand. The commonly used method to inhibit frosting is to optimize the evaporator design. In terms of defrost control optimization, some researchers have proposed using photoelectric sensors, image acquisition, acoustic signal acquisition and other methods to monitor the degree of frost accumulation. However, small refrigeration equipment often has a small size and compact structure, which limits the installation of frost layer measurement and acquisition equipment. In addition, the frost layer on the surface of the evaporator is usually unevenly distributed, and the installation position of the equipment affects the accuracy of monitoring. At present, most equipment still uses timed defrosting for defrosting. In recent years, with the rapid development of sensor technology, data analysis and artificial intelligence algorithms, new solutions have been provided for the precise temperature control and energy-saving control of small refrigeration systems. However, most small refrigeration equipment products on the market are still in the early stages in terms of intelligent and adaptive control, and have not yet fully utilized the technical advantages of modern optimization algorithms and data-driven. Summary of the invention

[0003] The present invention provides an intelligent control method and control system for small refrigeration equipment, which can dynamically adjust the operating status of the compressor and the fan, dynamically adjust the defrost frequency, achieve the goals of reducing refrigeration power, reducing frosting rate, and starting defrost on demand, thereby providing users with a more accurate, efficient and intelligent use experience.

[0004] In order to achieve the above-mentioned object of the invention, the present invention is implemented by the following technical solutions:

[0005] According to one aspect of the present invention, there is provided an intelligent control method for a small refrigeration device, including a refrigeration operation control method and a defrosting control method;

[0006] Wherein, the refrigeration operation control method comprises:

[0007] Monitor the working parameters and operation parameters of the refrigeration equipment through multiple sensors;

[0008] Building a multidimensional database based on historical data of operating parameters and operation parameters, and building an agent model of the refrigeration equipment based on the multidimensional database;

[0009] Based on the real-time monitoring data of the operating parameters and the operation parameters and the agent model, a multi-objective optimization model with the goals of reducing the operating energy consumption and slowing down the frosting rate is constructed, and a Pareto solution set that meets the constraint conditions is generated;

[0010] Decisions are made by analyzing the solutions in the Pareto solution set, taking the frosting rate and the whole machine operating power as decision variables, and selecting the best solution from the Pareto solution set according to a multi-objective decision criterion as a refrigeration operation optimization solution;

[0011] adjusting the operation of the compressor and the fan according to the optimal solution;

[0012] Among them, the first kind of the defrost control method includes:

[0013] Based on the multidimensional database, the cycle COP of the refrigeration-defrosting cycle is calculated cyc , the cycle COP cyc The accumulated frost amount at the maximum value is used as the indicated defrost amount;

[0014] Adopting the correlation analysis technology, calculating the correlation index between each parameter in the multidimensional database and the accumulated frost amount, and selecting the parameters whose correlation index exceeds the first set threshold as candidate characteristic parameters; calculating the correlation index between the candidate characteristic parameters, selecting the parameters whose correlation index value is lower than the second set threshold, and then using the selected characteristic parameters as input variables to establish the accumulated frost amount prediction model;

[0015] Based on the real-time monitoring data of the operating parameters and the operation parameters and the cumulative frost amount prediction model, the cumulative frost amount on the surface of the evaporator is predicted;

[0016] When the accumulated frost amount is greater than or equal to the indicated defrost amount, defrost is started;

[0017] The second defrost control method includes:

[0018] Based on the multidimensional database, the cycle COP of the refrigeration-defrosting cycle is calculated cyc , the cycle COPcyc The accumulated frost amount at the maximum value is used as the indicated defrost amount;

[0019] Based on the real-time monitoring data of the operating parameters and the operation parameters and the agent model, the real-time frosting rate is predicted and cumulative calculation is performed to obtain the cumulative frost amount on the evaporator surface;

[0020] When the accumulated frost amount is greater than or equal to the indicated defrost amount, defrost is started; when defrost is started, the accumulated frost amount is reset to zero.

[0021] Furthermore, the operating parameters are selected from the ambient temperature, ambient humidity, compartment setting temperature, and compartment humidity; the operating parameters include directly measured data and data calculated from the directly measured data; wherein the directly measured data are selected from the actual temperature in the compartment, the compressor speed, the fan speed, the evaporation pressure, the condensation pressure, the flow rate of the refrigerant in the refrigeration system, the evaporator pipe inlet temperature, the evaporator pipe outlet temperature, the evaporator heat exchange air duct inlet temperature, the evaporator heat exchange air duct outlet temperature, the evaporator heat exchange air duct inlet humidity, the evaporation The outlet humidity of the heat exchange air duct of the evaporator, the cold surface temperature of the evaporator, the wind speed of the heat exchange air duct of the evaporator, the wind speed at the air outlet of the compartment, the inlet and outlet pressure drop of the evaporator, the compressor speed, the compressor operating power, the fan operating power, the whole machine operating power, the compressor operating current, and the fan operating current; wherein, the data calculated after the directly measured data are selected from the temperature difference between the inlet and outlet of the evaporator pipeline, the temperature difference between the inlet and outlet of the evaporator heat exchange air duct, the difference between the cold surface of the evaporator and the indoor temperature of the compartment, the frosting rate, the accumulated frosting amount, the heat exchange, the frosting heat exchange, and the cooling heat exchange;

[0022] The frosting heat exchange amount refers to the portion of the heat exchange amount used for condensing water vapor into frost, and the cooling heat exchange amount refers to the portion of the heat exchange amount used for cooling.

[0023] Furthermore, the proxy model is implemented using machine learning and / or deep learning methods, the input parameters of the proxy model are selected from ambient temperature, ambient humidity, compartment humidity, compartment set temperature, compressor speed, and fan speed, and the output parameters of the proxy model are selected from whole machine operating power, cooling heat exchange, and frosting rate.

[0024] Furthermore, the constraints of the multi-objective optimization model are selected from ambient temperature, ambient humidity, compartment setting temperature, compartment humidity, cooling heat exchange capacity, compressor speed, and fan speed; the objective equation is the whole machine operating power and frosting rate, and the optimization goal is to minimize the whole machine operating power and frosting rate.

[0025] Furthermore, the multi-objective decision criterion adopts weighted sum method, ideal point method or TOPSIS to make decision.

[0026] Furthermore, the start and end of the refrigeration-defrosting cycle are respectively the first start-up time of the compressor after two adjacent defrosting operations are completed.

[0027] Furthermore, the cycle COP cyc It is defined as the ratio of the cumulative effective cooling capacity of the cycle to the cumulative energy consumption of the cycle, that is: Among them, Q eff is the cumulative effective cooling capacity of the cycle, E cyc Cumulative energy consumption for the cycle;

[0028] The cumulative effective cooling capacity of the cycle is the difference between the cumulative heat exchange of the cycle and the heat generated by the defrosting system, that is: Q eff =QQ df ; where Q is the cumulative heat exchange during the period, Q df Generates heat for the defrost system.

[0029] Furthermore, the correlation index is a Pearson correlation coefficient or a Spearman correlation coefficient.

[0030] Furthermore, a cumulative frost amount prediction model is established by adopting a machine learning and / or deep learning method, with the characteristic parameters as input and the cumulative frost amount as output.

[0031] According to another aspect of the present invention, there is provided an intelligent control method for executing the above-mentioned small refrigeration equipment; comprising a data acquisition module, a multidimensional database, an analysis and decision algorithm module, and a control module;

[0032] The data acquisition module includes a plurality of sensors for monitoring the working parameters and operating parameters of the refrigeration equipment;

[0033] The multidimensional database is constructed based on historical data of operating parameters and operation parameters;

[0034] The analysis and decision algorithm module includes a refrigeration operation optimization module and a defrost start analysis module;

[0035] The refrigeration operation optimization module is used to first build a proxy model of the refrigeration equipment according to the multidimensional database; then, based on the real-time monitoring data of the operating parameters and the operating parameters and the proxy model, build a multi-objective optimization model with the goals of reducing the operating energy consumption and slowing down the frosting rate, and generate a Pareto solution set that meets the constraint conditions; then, make a decision by analyzing the solutions in the Pareto solution set, taking the frosting rate and the whole machine operating power as decision variables, and selecting the optimal solution from the Pareto solution set according to the multi-objective decision criterion as the refrigeration operation optimization solution;

[0036] The defrost start analysis module is used to first calculate the cycle COP of the refrigeration-defrost cycle according to the multidimensional database.cyc , the cycle COP cyc The accumulated frost amount at the time of the maximum value is used as the indicated defrost amount; then, the correlation analysis technology is used to calculate the correlation index between each parameter in the multidimensional database and the accumulated frost amount, and the parameters whose correlation index exceeds the first set threshold are selected as candidate feature parameters; the correlation index between the candidate feature parameters is calculated, and the parameters whose correlation index value is lower than the second set threshold are selected, and then the selected feature parameters are used as input variables to establish a cumulative frost amount prediction model; then, based on the real-time monitoring data of the operating parameters and the operating parameters and the accumulated frost amount prediction model, the accumulated frost amount on the evaporator surface is predicted;

[0037] Alternatively, the defrost start analysis module is used to first calculate the cycle COP of the refrigeration-defrost cycle according to the multidimensional database. cyc , the cycle COP cyc The accumulated frost amount at the time of the maximum value is used as the indicative defrost amount; then based on the real-time monitoring data of the operating parameters and the operation parameters and the agent model, the real-time frost rate is predicted and accumulated calculation is performed to obtain the accumulated frost amount on the evaporator surface;

[0038] The control module is used to adjust the operation of the compressor and the fan according to the refrigeration operation optimization plan, and start defrosting when it is determined that the accumulated frosting amount is greater than or equal to the indicated defrosting amount.

[0039] Furthermore, the proxy model is implemented using machine learning and / or deep learning methods, the input parameters of the proxy model are selected from ambient temperature, ambient humidity, compartment humidity, compartment set temperature, compressor speed, and fan speed, and the output parameters of the proxy model are selected from whole machine operating power, cooling heat exchange, and frosting rate.

[0040] Furthermore, the constraints of the multi-objective optimization model are selected from ambient temperature, ambient humidity, compartment setting temperature, compartment humidity, cooling heat exchange capacity, compressor speed, and fan speed; the objective equation is the whole machine operating power and frosting rate, and the optimization goal is to minimize the whole machine operating power and frosting rate.

[0041] Furthermore, the multi-objective decision criterion adopts weighted sum method, ideal point method or TOPSIS to make decision.

[0042] Furthermore, the start and end of the refrigeration-defrosting cycle are respectively the first start-up time of the compressor after two adjacent defrosting operations are completed.

[0043] Furthermore, the COP cyc It is defined as the ratio of the cumulative effective cooling capacity of the cycle to the cumulative energy consumption of the cycle, that is: Among them, Q effis the cumulative effective cooling capacity of the cycle, E cyc Cumulative energy consumption for the cycle;

[0044] The cumulative effective cooling capacity of the cycle is the difference between the cumulative heat exchange of the cycle and the heat generated by the defrosting system, that is: Q eff -QQ df ; where Q is the cumulative heat exchange during the period, Q df Generates heat for the defrost system.

[0045] Furthermore, the correlation index is a Pearson correlation coefficient or a Spearman correlation coefficient.

[0046] Furthermore, a cumulative frost amount prediction model is established by adopting a machine learning and / or deep learning method, with the characteristic parameters as input and the cumulative frost amount as output.

[0047] The beneficial effects of the present invention are:

[0048] The present invention comprehensively considers multiple characteristic parameters, controls defrosting based on the changing law of operating characteristics during frost accumulation of small refrigeration equipment, and has high prediction accuracy and reliability.

[0049] Aiming at the periodic characteristics of the operation of small refrigeration equipment, the present invention proposes an evaluation index of periodic COP, so that defrosting can be started according to actual needs, avoiding more cooling loss and achieving a better balance between cooling and defrosting.

[0050] The present invention comprehensively considers the two factors of the whole machine operating power and frosting, can more truly reflect the operating law of the equipment, and performs collaborative optimization with the goal of reducing the whole machine operating power and reducing the frosting rate.

[0051] In summary, the present invention proposes a full-cycle optimization method for the refrigeration-defrosting operation of a small refrigeration equipment. By optimizing the operation of components, inhibiting the growth of frost layer, and starting defrosting on demand, the temperature control accuracy in the refrigeration space is improved and the operating energy consumption of the refrigeration equipment is reduced, providing new ideas and methods for precise temperature control and energy saving and consumption reduction of refrigeration equipment. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] Figure 1 is a schematic diagram of the operation cycle of a refrigerator in an embodiment of the present invention;

[0053] Figure 2 It is a flow chart of the intelligent control method according to an embodiment of the present invention. DETAILED DESCRIPTION

[0054] The present invention is further described in detail below through specific examples. The following examples can enable those skilled in the art to more fully understand the present invention, but do not limit the present invention in any way.

[0055] This embodiment provides an intelligent control method and control system for a household air-cooled refrigerator, which can reduce the temperature fluctuation of the refrigeration space and reduce the operating energy consumption of the equipment through intelligent control.

[0056] The operation of the refrigerator is cyclical. The starting time of the refrigeration-defrost cycle is when the compressor is started for the first time after the defrost is completed, and the end time of the refrigeration-defrost cycle and the starting time of the next refrigeration-defrost cycle are when the compressor is started after the next defrost is completed.

[0057] The start and end of the refrigeration-defrosting cycle are respectively the first start-up time of the compressor after two adjacent defrosting ends.

[0058] like Figure 1 As shown, A i -A i+1 A complete refrigeration-defrosting cycle can be divided into two stages, namely the refrigeration stage (A i -C i ) and defrosting stage (C i -A i+1 The cooling stage includes the rapid cooling stage after defrosting (A i -B i ) and stable cooling stage (B i -C i ), in A i -B i In this stage, the compressor and the fan usually run at a set high speed, so that the indoor temperature of the refrigerator compartment quickly drops to the set temperature. After the temperature drops to the set temperature, it enters B i -C i The refrigerator system operates in this stage most of the time.

[0059] like Figure 2 As shown, the intelligent control method of the household air-cooled refrigerator of this embodiment includes a refrigeration operation control method and a defrost control method.

[0060] The refrigeration operation control method can dynamically adjust the operating speed of the compressor and the fan according to the operating status of the refrigeration equipment, thereby reducing the energy consumption of the whole machine and effectively slowing down the frosting rate of the evaporator.

[0061] Multiple sensors are arranged in the refrigerator refrigeration system, air duct, compartment and other locations to monitor the operating parameters and operation parameters of the refrigerator during operation.

[0062] Among them, the operating parameters can be: ambient temperature, ambient humidity, compartment setting temperature, and compartment humidity.

[0063] The operating parameters include directly measured data and data calculated from the directly measured data.

[0064] Furthermore, the directly measured data may be: actual temperature in the compartment, compressor speed, fan speed, evaporation pressure, condensation pressure, flow rate of refrigerant in the refrigeration system, evaporator pipe inlet temperature, evaporator pipe outlet temperature, inlet air temperature of the evaporator heat exchange air duct, outlet air temperature of the evaporator heat exchange air duct, inlet air humidity of the evaporator heat exchange air duct, outlet air humidity of the evaporator heat exchange air duct, evaporator cold surface temperature, wind speed of the evaporator heat exchange air duct, wind speed at the compartment air outlet, evaporator inlet and outlet pressure drop, compressor speed, compressor operating power, fan operating power, whole machine operating power, compressor operating current, and fan operating current.

[0065] Furthermore, the directly measured data can be calculated to obtain: the temperature difference between the inlet and outlet of the evaporator pipeline, the temperature difference between the inlet and outlet of the evaporator heat exchange air duct, the difference between the cold surface of the evaporator and the indoor temperature, the frosting rate, the accumulated frost amount, the heat exchange, the frosting heat exchange, and the cooling heat exchange.

[0066] Among them, the frosting heat exchange refers to the part of the heat exchange used for condensing water vapor into frost, and the cooling heat exchange refers to the part of the heat exchange used for cooling.

[0067] A multidimensional database is constructed based on the historical data of operating parameters and operation parameters.

[0068] Based on a multidimensional database, an agent model of the refrigerator is established using machine learning and deep learning methods.

[0069] For example, a neural network method can be used, with ambient temperature, ambient humidity, compartment humidity, compartment set temperature, compressor speed, and fan speed as the input of the proxy model, and the whole machine operating power, cooling heat exchange, and frosting rate as the output of the proxy model.

[0070] Based on the real-time monitoring data of operating parameters and operation parameters and the agent model, a multi-objective optimization model is constructed with the goals of reducing operating energy consumption and slowing down the frosting rate.

[0071] The constraints of the multi-objective optimization model can be ambient temperature, ambient humidity, compartment humidity, compartment setting temperature, cooling heat exchange, compressor speed, and fan speed. Among them, ambient temperature, ambient humidity, compartment humidity, and compartment setting temperature are all operating parameters and are set to fixed values; cooling heat exchange, compressor speed, and fan speed are set to the operating range of the components.

[0072] The objective equation of the multi-objective optimization model is the whole machine operating power and frosting rate, and the operating objective is to minimize the whole machine operating power and frosting rate.

[0073] The multi-objective optimization model can be run to obtain a Pareto solution set, each solution represents a trade-off between different objectives. In this solution set, no solution can be surpassed by other solutions on all objectives at the same time. In order to select the best solution from these solutions, it is necessary to analyze and make decisions on the solutions in the Pareto solution set. According to the multi-objective decision-making criterion, for example, the TOPSIS method can be used, with the frosting rate and the whole machine operating power as decision variables, and the goal is to minimize the frosting rate and the whole machine operating power. Equal weights are assigned to these two goals (each with a weight of 0.5), that is, the balance between the frosting rate and energy efficiency is considered to be equally important. The TOPSIS method can be used to score different compressor speed and fan speed matching schemes. Based on the TOPSIS scoring results, the optimal solution is selected as the final solution. This optimal solution is used to guide the refrigerator in B under the operating parameters set in the multi-objective optimization model. i -C i In the cooling stage, the surface of the evaporator maintains a relatively low temperature. When the air in the room exchanges heat with the evaporator, the water vapor it carries condenses into frost and adheres to the surface of the evaporator. As the frost accumulates on the surface of the evaporator, it will block the heat exchange air duct and increase the heat exchange thermal resistance. Therefore, defrosting is required in time. It is very important to evaluate the degree of frost accumulation on the evaporator surface and start defrosting at the best time.

[0074] Based on the multidimensional database, the cycle COP of the refrigeration-defrosting cycle is calculated cyc , the cycle COP cyc The accumulated frost amount at the maximum value is used as the indicated defrost amount;

[0075] The time to start defrosting needs to balance the cooling capacity loss in the refrigeration stage and the defrosting stage. Based on the multidimensional database, the cycle COP when running under the same working conditions and starting defrosting at different accumulated frost amounts is calculated. cyc , COP cyc It is defined as the ratio of the cumulative effective cooling capacity of the cycle to the cumulative energy consumption of the cycle, that is: Among them, Q eff E is the cumulative effective cooling capacity of the cycle, cyc Cumulative energy consumption for the cycle.

[0076] The cumulative effective cooling capacity of a cycle is the difference between the cumulative heat exchange of the cycle and the heat generated by the defrosting system, that is: Q eff =QQ df ; where Q is the cumulative heat exchange during the period, Q df Generates heat for the defrost system.

[0077] Comparison cycle COPcyc The value of COP cyc The accumulated frost amount at the maximum value is taken as the indicated defrost amount. The indicated defrost amount is used to determine the defrost time.

[0078] There are two methods to obtain the cumulative frost amount. The first method is to establish a cumulative frost amount prediction model.

[0079] By using correlation analysis technology, the correlation index between each parameter in the multidimensional database and the accumulated frost amount, such as the Pearson correlation coefficient and the Spearman correlation coefficient, is first calculated, and the parameters whose correlation index exceeds the first set threshold are selected as candidate characteristic parameters. Then, the correlation index between the candidate characteristic parameters is calculated, and the parameters whose correlation index value is lower than the second set threshold are retained, and the selected parameters are used as characteristic parameters.

[0080] The first set threshold and the second set threshold are determined according to specific application requirements and can be adjusted through experimental data or model performance verification to ensure that the selected parameters can reflect the main influencing factors of the accumulated frost amount and reduce the redundancy between parameters. For example, the first set threshold is 0.8 and the second set threshold is 0.9.

[0081] Machine learning and deep learning methods, such as vector machines and random forests, are used to establish a cumulative frost prediction model with feature parameters as input and cumulative frost as output.

[0082] Based on the real-time monitoring data of operating parameters and operation parameters and the cumulative frost amount prediction model, the cumulative frost amount on the evaporator surface is predicted, and defrosting is started when the cumulative frost amount is greater than or equal to the indicated defrost amount.

[0083] The output parameters of the proxy model include the frosting rate. The second method to obtain the accumulated frosting amount is to predict the real-time frosting rate and perform time accumulation calculation based on the real-time monitoring data of the operating parameters and operation parameters and the proxy model to obtain the accumulated frosting amount on the evaporator surface. When the accumulated frosting amount is greater than or equal to the indicated defrosting amount, defrosting is started and the accumulated frosting amount is cleared. When defrosting is completed, the compressor restarts and enters the refrigeration stage. At this time, the frosting rate of the new cycle is predicted and accumulated.

[0084] Accordingly, the intelligent control system of the household air-cooled refrigerator of this embodiment includes a data acquisition module, a multidimensional database, an analysis and decision algorithm module, and a control module.

[0085] The data acquisition module includes multiple sensors, which are arranged in the refrigerator refrigeration system, air duct, indoor and other locations to monitor the working parameters and operation parameters of the refrigeration equipment.

[0086] Among them, the operating parameters can be: ambient temperature, ambient humidity, compartment setting temperature, and compartment humidity.

[0087] The operating parameters include directly measured data and data calculated from the directly measured data.

[0088] Furthermore, the directly measured data may be: actual temperature in the compartment, compressor speed, fan speed, evaporation pressure, condensation pressure, flow rate of refrigerant in the refrigeration system, evaporator pipe inlet temperature, evaporator pipe outlet temperature, inlet air temperature of the evaporator heat exchange air duct, outlet air temperature of the evaporator heat exchange air duct, inlet air humidity of the evaporator heat exchange air duct, outlet air humidity of the evaporator heat exchange air duct, evaporator cold surface temperature, wind speed of the evaporator heat exchange air duct, wind speed at the compartment air outlet, evaporator inlet and outlet pressure drop, compressor speed, compressor operating power, fan operating power, whole machine operating power, compressor operating current, and fan operating current.

[0089] Furthermore, the directly measured data can be calculated to obtain: the temperature difference between the inlet and outlet of the evaporator pipeline, the temperature difference between the inlet and outlet of the evaporator heat exchange air duct, the difference between the cold surface of the evaporator and the indoor temperature, the frosting rate, the accumulated frost amount, the heat exchange, the frosting heat exchange, and the cooling heat exchange.

[0090] Among them, the frosting heat exchange refers to the part of the heat exchange used for condensing water vapor into frost, and the cooling heat exchange refers to the part of the heat exchange used for cooling.

[0091] The multidimensional database is constructed based on the historical data of operating parameters and operation parameters.

[0092] The analysis and decision-making algorithm module includes a refrigeration operation optimization module and a defrost start-up analysis module.

[0093] The refrigeration operation optimization module is used to first establish a proxy model of the refrigerator based on a multidimensional database using machine learning and deep learning methods; then, based on the real-time monitoring data of operating parameters and operating parameters and the proxy model, a multi-objective optimization model with the goals of reducing operating energy consumption and slowing down the frosting rate is constructed, and a Pareto solution set that meets the constraints is generated; then, a decision is made by analyzing the solutions in the Pareto solution set, with the frosting rate and the whole machine operating power as decision variables, and according to the multi-objective decision criterion, the optimal solution is selected from the Pareto solution set. The optimal solution is used to guide the refrigerator in B under the operating parameters set in the multi-objective optimization model. i -C i Segment operation.

[0094] For example, a neural network method can be used, with ambient temperature, ambient humidity, compartment humidity, compartment set temperature, compressor speed, and fan speed as the input of the proxy model, and the whole machine operating power, cooling heat exchange, and frosting rate as the output of the proxy model.

[0095] Among them, the constraints of the multi-objective optimization model can be ambient temperature, ambient humidity, compartment humidity, compartment setting temperature, cooling heat exchange, compressor speed, and fan speed. Among them, ambient temperature, ambient humidity, compartment humidity, and compartment setting temperature are all operating parameters and are set to fixed values; cooling heat exchange, compressor speed, and fan speed are set to the operating range of the components.

[0096] Among them, the objective equation of the multi-objective optimization model is the whole machine operating power and frosting rate, and the operating goal is to minimize the whole machine operating power and frosting rate.

[0097] The defrost start analysis module can be selected: first calculate the cycle COP of the refrigeration-defrost cycle based on the multidimensional database cyc , the cycle COP cyc The accumulated frost amount at the moment of the maximum value is used as the indicative defrost amount; then, the correlation analysis technology is used to calculate the correlation index between each parameter in the multidimensional database and the accumulated frost amount, and the parameters whose correlation index exceeds the first set threshold are screened as candidate feature parameters; the correlation index between the candidate feature parameters is calculated, and the parameters whose correlation index value is lower than the second set threshold are screened, and then the screened feature parameters are used as input variables to establish a cumulative frost amount prediction model; then, based on the real-time monitoring data of the operating parameters and the cumulative frost amount prediction model, the accumulated frost amount on the evaporator surface is predicted.

[0098] Among them, the correlation index is the Pearson correlation coefficient or the Spearman correlation coefficient.

[0099] The first set threshold and the second set threshold are determined according to specific application requirements and can be adjusted through experimental data or model performance verification to ensure that the selected parameters can reflect the main influencing factors of the accumulated frost amount and reduce the redundancy between parameters. For example, the first set threshold is 0.8 and the second set threshold is 0.9.

[0100] Among them, the cumulative frost amount prediction model adopts machine learning and deep learning methods, such as vector machine, random forest and other methods, with feature parameters as input and cumulative frost amount as output.

[0101] The defrost start analysis module can also choose to: first calculate the cycle COP of the refrigeration-defrost cycle based on the multidimensional database cyc , the cycle COP cyc The accumulated frost amount at the moment of the maximum value is used as the indicative defrost amount. Then, based on the real-time monitoring data of the operating parameters and the operation parameters and the proxy model, the real-time frost rate is predicted and accumulated calculation is performed to obtain the accumulated frost amount on the evaporator surface.

[0102] When the accumulated frost amount is greater than or equal to the indicated defrost amount, defrosting is started and the accumulated frost amount is reset to zero. When the defrost is completed, the compressor restarts and enters the refrigeration stage. At this time, the frost rate of the new cycle is predicted and accumulated.

[0103] The control module is used to adjust the operation of the compressor and the fan according to the refrigeration operation optimization plan, and start defrosting when it is judged that the accumulated frost amount is greater than or equal to the indicated defrost amount.

[0104] It can be seen that the present invention improves the temperature control accuracy in the refrigeration space and reduces the operating energy consumption of the refrigeration equipment on the basis of meeting the refrigeration demand of the refrigeration space by optimizing the operation of components, inhibiting the growth of frost layer, and starting defrosting on demand. It provides new ideas and methods for precise temperature control and energy saving of refrigeration equipment, has strong application prospects, and can be widely used in the fields of low temperature and refrigeration.

[0105] Although the preferred embodiments of the present invention have been described above in conjunction with the accompanying drawings, the present invention is not limited to the above-mentioned specific embodiments, which are merely illustrative and not restrictive. Under the guidance of the present invention, ordinary technicians in this field can also make many forms of specific changes without departing from the scope of protection of the invention and the claims, all of which fall within the scope of protection of the present invention.

Claims

1. An intelligent control method for small refrigeration equipment, characterized in that: Including refrigeration operation control method and defrost control method; Wherein, the refrigeration operation control method comprises: Monitor the working parameters and operation parameters of the refrigeration equipment through multiple sensors; Building a multidimensional database based on historical data of operating parameters and operation parameters, and building an agent model of the refrigeration equipment based on the multidimensional database; Based on the real-time monitoring data of the operating parameters and the operation parameters and the agent model, a multi-objective optimization model with the goals of reducing the operating energy consumption and slowing down the frosting rate is constructed, and a Pareto solution set that meets the constraint conditions is generated; Decisions are made by analyzing the solutions in the Pareto solution set, taking the frosting rate and the whole machine operating power as decision variables, and selecting the best solution from the Pareto solution set according to a multi-objective decision criterion as a refrigeration operation optimization solution; adjusting the operation of the compressor and the fan according to the optimal solution; Among them, the first kind of the defrost control method includes: Based on the multidimensional database, the cycle COP of the refrigeration-defrosting cycle is calculated cyc , the cycle COP cyc The accumulated frost amount at the maximum value is used as the indicated defrost amount; Adopting the correlation analysis technology, calculating the correlation index between each parameter in the multidimensional database and the accumulated frost amount, and selecting the parameters whose correlation index exceeds the first set threshold as candidate characteristic parameters; calculating the correlation index between the candidate characteristic parameters, selecting the parameters whose correlation index value is lower than the second set threshold, and then using the selected characteristic parameters as input variables to establish the accumulated frost amount prediction model; Based on the real-time monitoring data of the operating parameters and the operation parameters and the cumulative frost amount prediction model, the cumulative frost amount on the surface of the evaporator is predicted; When the accumulated frost amount is greater than or equal to the indicated defrost amount, defrost is started; The second defrost control method includes: Based on the multidimensional database, the cycle COP of the refrigeration-defrosting cycle is calculated cyc , the cycle COP cyc The accumulated frost amount at the maximum value is used as the indicated defrost amount; Based on the real-time monitoring data of the operating parameters and the operation parameters and the agent model, the real-time frosting rate is predicted and cumulative calculation is performed to obtain the cumulative frost amount on the evaporator surface; When the accumulated frost amount is greater than or equal to the indicated defrost amount, defrost is started; when defrost is started, the accumulated frost amount is reset to zero.

2. The intelligent control method for small refrigeration equipment according to claim 1 is characterized in that: The working condition parameters are selected from the ambient temperature, ambient humidity, compartment setting temperature, and compartment humidity; the operating parameters include directly measured data and data calculated from the directly measured data; wherein the directly measured data are selected from the actual temperature in the compartment, the compressor speed, the fan speed, the evaporation pressure, the condensation pressure, the flow rate of the refrigerant in the refrigeration system, the evaporator pipe inlet temperature, the evaporator pipe outlet temperature, the evaporator heat exchange air duct inlet temperature, the evaporator heat exchange air duct outlet temperature, the evaporator heat exchange air duct inlet humidity, the evaporator heat exchange air duct outlet temperature, the evaporator heat exchange air duct inlet humidity, the evaporator heat exchange air duct outlet temperature, the evaporator heat exchange air duct outlet ... The outlet humidity of the hot air duct, the cold surface temperature of the evaporator, the wind speed of the evaporator heat exchange duct, the wind speed at the air outlet of the compartment, the inlet and outlet pressure drop of the evaporator, the compressor speed, the compressor operating power, the fan operating power, the whole machine operating power, the compressor operating current, and the fan operating current; wherein, the data calculated from the directly measured data are selected from the temperature difference between the inlet and outlet of the evaporator pipeline, the temperature difference between the inlet and outlet of the evaporator heat exchange duct, the difference between the cold surface of the evaporator and the temperature in the compartment, the frosting rate, the accumulated frosting amount, the heat exchange, the frosting heat exchange, and the cooling heat exchange; The frosting heat exchange amount refers to the portion of the heat exchange amount used for condensing water vapor into frost, and the cooling heat exchange amount refers to the portion of the heat exchange amount used for cooling.

3. The intelligent control method for small refrigeration equipment according to claim 1 is characterized in that: The proxy model is implemented using machine learning and / or deep learning methods, the input parameters of the proxy model are selected from ambient temperature, ambient humidity, compartment humidity, compartment set temperature, compressor speed, and fan speed, and the output parameters of the proxy model are selected from whole machine operating power, cooling heat exchange, and frosting rate.

4. The intelligent control method for small refrigeration equipment according to claim 1 is characterized in that: The constraints of the multi-objective optimization model are selected from ambient temperature, ambient humidity, compartment setting temperature, compartment humidity, cooling heat exchange, compressor speed, and fan speed; the objective equation is the whole machine operating power and frosting rate, and the optimization goal is to minimize the whole machine operating power and frosting rate.

5. The intelligent control method for small refrigeration equipment according to claim 1, characterized in that: The multi-objective decision criterion adopts weighted sum method, ideal point method or TOPSIS to make decision.

6. The intelligent control method for small refrigeration equipment according to claim 1, characterized in that: The start and end of the refrigeration-defrosting cycle are respectively the first start-up time of the compressor after two adjacent defrosting ends.

7. The intelligent control method for small refrigeration equipment according to claim 1, characterized in that: The cycle COP cyc It is defined as the ratio of the cumulative effective cooling capacity of the cycle to the cumulative energy consumption of the cycle, that is: Among them, Q eff E is the cumulative effective cooling capacity of the cycle, cyc Cumulative energy consumption for the cycle; The cumulative effective cooling capacity of the cycle is the difference between the cumulative heat exchange of the cycle and the heat generated by the defrosting system, that is: Q eff =QQ df ; where Q is the cumulative heat exchange during the period, Q df Generates heat for the defrost system.

8. The intelligent control method for small refrigeration equipment according to claim 1, characterized in that: The correlation index is the Pearson correlation coefficient or the Spearman correlation coefficient.

9. The intelligent control method for small refrigeration equipment according to claim 1, characterized in that: The cumulative frost amount prediction model is established by adopting machine learning and / or deep learning methods, with the characteristic parameters as input and the cumulative frost amount as output.

10. An intelligent control system for a small refrigeration equipment, characterized in that: Used to execute the intelligent control method of a small refrigeration device as claimed in any one of claims 1 to 9; It includes data acquisition module, multi-dimensional database, analysis and decision algorithm module, and control module; The data acquisition module includes a plurality of sensors for monitoring the working parameters and operating parameters of the refrigeration equipment; The multidimensional database is constructed based on historical data of operating parameters and operation parameters; The analysis and decision algorithm module includes a refrigeration operation optimization module and a defrost start analysis module; The refrigeration operation optimization module is used to first build a proxy model of the refrigeration equipment according to the multidimensional database; then, based on the real-time monitoring data of the operating parameters and the operating parameters and the proxy model, build a multi-objective optimization model with the goals of reducing the operating energy consumption and slowing down the frosting rate, and generate a Pareto solution set that meets the constraint conditions; then, make a decision by analyzing the solutions in the Pareto solution set, taking the frosting rate and the whole machine operating power as decision variables, and selecting the optimal solution from the Pareto solution set according to the multi-objective decision criterion as the refrigeration operation optimization solution; The defrost start analysis module is used to first calculate the cycle COP of the refrigeration-defrost cycle according to the multidimensional database. cyc , the cycle COP cyc The accumulated frost amount at the time of the maximum value is used as the indicated defrosting amount; then, the correlation analysis technology is used to calculate the correlation index between each parameter in the multidimensional database and the accumulated frost amount, and the parameters whose correlation index exceeds the first set threshold are selected as candidate characteristic parameters; calculate Correlation indexes between candidate characteristic parameters, screening parameters whose correlation index values ​​are lower than a second set threshold, and then using the screened characteristic parameters as input variables to establish a cumulative frost amount prediction model; and then predicting the cumulative frost amount on the evaporator surface based on the real-time monitoring data of the operating parameters and the operating parameters and the cumulative frost amount prediction model; Alternatively, the defrost start analysis module is used to first calculate the cycle COP of the refrigeration-defrost cycle according to the multidimensional database. cyc , the cycle COP cyc The accumulated frost amount at the time of the maximum value is used as the indicative defrost amount; then based on the real-time monitoring data of the operating parameters and the operation parameters and the agent model, the real-time frost rate is predicted and accumulated calculation is performed to obtain the accumulated frost amount on the evaporator surface; The control module is used to adjust the operation of the compressor and the fan according to the refrigeration operation optimization plan, and start defrosting when it is determined that the accumulated frosting amount is greater than or equal to the indicated defrosting amount.

Images