Refrigerator and refrigerator compressor operation control method

Abstract

The refrigerator comprises a refrigerator body, a door body is arranged at an opening of the refrigerator body, and at least one storage chamber is arranged in the refrigerator body and used for storing articles; a compressor for performing compressor and circulation of a refrigerant; the chamber temperature sensor is used for collecting the chamber temperature of the storage chamber; the environment temperature sensor is used for collecting the indoor environment temperature; a controller; the controller is configured to obtain the chamber temperature, the environment temperature and the current set temperature; constructing a feature vector according to the compartment temperature, the environment temperature and the current set temperature; inputting the feature vector into a pre-trained neural network model for calculation to obtain working frequency output; and controlling the compressor to operate at the working frequency. The working frequency of the refrigerator can be accurately calculated, and the energy consumption of household appliance operation is reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of refrigerators, and in particular to a refrigerator and a refrigerator compressor operation control method. Background Art

[0002] Refrigerators provide a good low-temperature environment for food. Refrigerators are powered on 24 hours a day, so the electricity consumed by refrigerators accounts for a large proportion of household electricity consumption. Therefore, users are also very concerned about the energy consumption of refrigerators. The power consumption on the refrigerator energy consumption label is measured under the conditions specified in the standard. The smaller the value, the more energy-saving it is during use. However, the actual state conditions of users are different, and the power consumption marked on the refrigerator cannot truly reflect the actual energy consumption during the operation of the refrigerator.

[0003] In the prior art, usually only one set of control logic is used to control the operating status of the refrigerator compressor throughout the day. However, when the refrigerator is in a balanced state, the required cooling capacity will not change much. If the same compressor control strategy is always used, the compressor speed will be too high and the output cooling capacity / power will be too large, exceeding the required cooling capacity, resulting in high power consumption. Summary of the invention

[0004] The purpose of the embodiments of the present invention is to provide a refrigerator and a refrigerator compressor operation control method, which can accurately calculate the operating frequency of the refrigerator and reduce the energy consumption of the household appliance operation.

[0005] An embodiment of the present invention provides a refrigerator, comprising:

[0006] A box body, wherein at least one storage room is provided in the box body for storing items;

[0007] A compressor, wherein the compressor is used to compress and circulate a refrigerant;

[0008] A compartment temperature sensor, the compartment temperature sensor is used to collect the compartment temperature of the storage room;

[0009] An ambient temperature sensor, wherein the ambient temperature sensor is used to collect the indoor ambient temperature;

[0010] A controller, the controller being configured to:

[0011] Get the compartment temperature, ambient temperature and current set temperature;

[0012] Constructing a feature vector according to the compartment temperature, the ambient temperature and the current set temperature;

[0013] Inputting the feature vector into a pre-trained neural network model for calculation to obtain an operating frequency output;

[0014] The compressor is controlled to operate at the operating frequency.

[0015] Preferably, the controller is further configured to:

[0016] Sequentially obtaining the real-time power of the compressor when it runs at the operating frequency until it stops at a preset time interval;

[0017] Calculate the electric energy consumed by the compressor when it runs at the operating frequency until it stops according to the acquired real-time frequency;

[0018] Obtaining the actual temperature of the compartment and the set temperature of the compartment when the compressor is stopped;

[0019] Calculating the evaluation value of the operating frequency using a preset evaluation value calculation model according to the actual temperature of the compartment, the set temperature of the compartment and the electric energy;

[0020] A data sequence is constructed according to the operating frequency, the feature vector and the evaluation value, and the data sequence is stored in a pre-built experience database as training data for updating the neural network model.

[0021] Furthermore, the controller is also configured to:

[0022] Reading the number of data sequences stored in the experience database;

[0023] When the number of data sequences is greater than M, randomly taking N data sequences from the experience database as training data, training the neural network model, and updating the parameters of the neural network model;

[0024] Wherein, M is a positive integer not less than N.

[0025] As a preferred solution, the evaluation value calculation model is specifically:

[0026] Wherein, R(F) is the evaluation value of the working frequency F, E is the electric energy, t is the cumulative working time of the compressor running at the working frequency until it stops, Tr and Ts are the actual temperature of the compartment and the set temperature of the compartment respectively, α is the temperature difference coefficient, ΔTu is the preset upper limit of the temperature difference.

[0027] Preferably, the neural network model includes 1 input layer, 3 hidden layers and 1 output layer;

[0028] Each hidden layer consists of 64 neurons;

[0029] The output layer includes 22 neurons, and each neuron of the output layer corresponds to a working frequency vector.

[0030] Furthermore, the controller is also configured to:

[0031] Inputting the feature vector into the input layer of the neural network model;

[0032] Calculate the weight of the feature vector of each neuron from the hidden layer to the output layer;

[0033] The operating frequency vector corresponding to the neuron with the largest weight in the output layer is used as the operating frequency output.

[0034] Preferably, the controller is further configured to:

[0035] After the refrigerator is powered on, determining whether it is currently in a user usage time period;

[0036] When the user is currently in the use time period, controlling the compressor to operate at a preset operating frequency;

[0037] When the user is not currently in use, the neural network model is used to calculate the operating frequency of the compressor, and the compressor is controlled to operate at the calculated operating frequency.

[0038] As a preferred solution, the controller is further configured as follows:

[0039] Determine whether all doors of the refrigerator storage compartment are in a closed state within a preset time period before the current moment;

[0040] If yes, determining that the refrigerator is not in the user use time period;

[0041] If not, it is determined that the refrigerator is in the user use time period.

[0042] Preferably, the refrigerator further includes a frequency conversion module:

[0043] The frequency conversion module is used to receive the control instruction of the controller, control the operation of the compressor according to the operating frequency in the control instruction, and calculate the real-time power of the compressor in real time.

[0044] The present invention also provides a refrigerator compressor operation control method, the refrigerator comprising:

[0045] A box body, wherein a door body is provided at the opening of the box body, and at least one storage room is provided in the box body for storing items;

[0046] A compressor, the compressor being used to compress and circulate a refrigerant;

[0047] A compartment temperature sensor, the compartment temperature sensor is used to collect the compartment temperature of the storage room;

[0048] An ambient temperature sensor, wherein the ambient temperature sensor is used to collect the indoor ambient temperature;

[0049] Controller;

[0050] The method comprises:

[0051] Get the compartment temperature, ambient temperature and current set temperature;

[0052] Constructing a feature vector according to the compartment temperature, the ambient temperature and the current set temperature;

[0053] Inputting the feature vector into a pre-trained neural network model for calculation to obtain an operating frequency output;

[0054] The compressor is controlled to operate at the operating frequency.

[0055] Compared with the prior art, the refrigerator and refrigerator compressor operation control method disclosed in the present invention include a box body, a door body is provided at the opening of the box body, and at least one storage room is provided in the box body for storing items; a compressor, the compressor is used to compress and circulate the refrigerant; a compartment temperature sensor, the compartment temperature sensor is used to collect the compartment temperature of the storage room; an ambient temperature sensor, the ambient temperature sensor is used to collect the indoor ambient temperature; a controller; the controller is configured to: obtain the compartment temperature, the ambient temperature and the current set temperature; construct a feature vector according to the compartment temperature, the ambient temperature and the current set temperature; input the feature vector into a pre-trained neural network model for calculation to obtain an operating frequency output; control the compressor to operate at the operating frequency. The present application can accurately calculate the operating frequency of the refrigerator and reduce the energy consumption of the household appliance operation. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] Figure 1 is a structural schematic diagram of a refrigerator provided by an embodiment of the present invention;

[0057] Figure 2 is a schematic structural diagram of a refrigeration system of a refrigerator provided in an embodiment of the present invention;

[0058] Figure 3 is a schematic structural diagram of a refrigerator provided by the present invention in a first embodiment;

[0059] Figure 4 is a flow chart of steps executed by the controller provided by the present invention under the first embodiment;

[0060] Figure 5 is a flow chart of steps executed by the controller provided by the present invention in the second embodiment;

[0061] Figure 6 is a flow chart of steps executed by the controller provided by the present invention in the third embodiment;

[0062] Figure 7 It is a topological structure diagram of a neural network model provided by an embodiment of the present invention;

[0063] Figure 8 is a flow chart of steps executed by the controller provided by the present invention in the fourth embodiment;

[0064] Fig. 9 is a schematic structural diagram of a refrigerator provided by the present invention in a second embodiment;

[0065] Fig.10 It is a schematic structural diagram of the refrigerator provided by the present invention in the third embodiment. DETAILED DESCRIPTION

[0066] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0067] In the description of the present application, it should be understood that the terms "center", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside", etc., indicating orientations or positional relationships, are based on the orientations or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present application and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be understood as a limitation on the present application.

[0068] The terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of the features. In the description of this application, unless otherwise specified, "plurality" means two or more.

[0069] In the description of this application, it should be noted that, unless otherwise clearly specified and limited, the terms "installed", "connected", and "connected" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be indirectly connected through an intermediate medium, or it can be the internal communication of two components. For ordinary technicians in this field, the specific meanings of the above terms in this application can be understood according to specific circumstances.

[0070] See also Figure 1 , is a structural schematic diagram of a refrigerator provided by an embodiment of the present invention. The embodiment of the present invention provides a refrigerator, including a box body 101, in which a storage chamber 102 is provided, including a refrigerating chamber and / or a freezing chamber 103 for storing items with preservation or freezing requirements, and also includes a refrigeration system 104 for performing the refrigeration operation of the refrigerator.

[0071] It should be noted that the refrigerator performs refrigeration operation through the refrigeration system to provide cold energy to the storage chamber so as to maintain the storage chamber at a constant low temperature. Figure 2 , is a structural schematic diagram of a refrigeration system of a refrigerator provided in an embodiment of the present invention, wherein the refrigeration system is composed of a compressor 201, a condenser 202, a drying filter 203, a capillary tube 204 and a first evaporator 205, and the working structure of the refrigeration system includes a compression process, a condensation process, a throttling process and an evaporation process.

[0072] The compression process is as follows: plug in the power cord of the refrigerator, and when the box has a cooling demand, the compressor starts working, and the low-temperature, low-pressure refrigerant is sucked into the compressor, compressed into a high-temperature, high-pressure superheated gas in the compressor cylinder, and then discharged into the condenser; the condensation process is as follows: the high-temperature, high-pressure refrigerant gas dissipates heat through the condenser, the temperature continues to drop, and is gradually cooled to a saturated vapor at room temperature and high pressure, and further cooled to a saturated liquid, and the temperature no longer drops. The temperature at this time is called the condensation temperature. The pressure of the refrigerant is almost unchanged during the entire condensation process; the throttling process is as follows: the condensed refrigerant saturated liquid flows into the capillary after the moisture and impurities are filtered out by the drying filter, and the refrigerant is throttled and depressurized through it, and the refrigerant becomes a wet vapor at room temperature and low pressure; the evaporation process is as follows: it then begins to absorb heat and vaporize in the first evaporator, which not only reduces the temperature of the first evaporator and its surroundings, but also turns the refrigerant into a low-temperature, low-pressure gas. The refrigerant coming out of the first evaporator returns to the compressor again, repeating the above process, transferring the heat in the refrigerator to the air outside the box, and achieving the purpose of refrigeration.

[0073] It should be noted that the above embodiment is only a specific structure and working principle of a conventional refrigerator. The refrigerator disclosed in this embodiment may also have the structure of the above-mentioned existing refrigerator to achieve corresponding functions.

[0074] A refrigerator provided in an embodiment of the present invention comprises: a box body, wherein at least one storage chamber is provided in the box body for storing articles;

[0075] A compressor, wherein the compressor is used to compress and circulate a refrigerant;

[0076] A compartment temperature sensor, the compartment temperature sensor is used to collect the compartment temperature of the storage room;

[0077] An ambient temperature sensor, wherein the ambient temperature sensor is used to collect the indoor ambient temperature;

[0078] Specifically, see Figure 3 , is a schematic structural diagram of a refrigerator provided by the present invention in a first embodiment, wherein the refrigerator 100 comprises: a box body 101, in which a storage chamber 102 is arranged; the number of the storage chamber is at least one, and the storage chamber can specifically be a refrigerating chamber, a freezing chamber and a temperature-changing chamber, which operate under different room temperature conditions and are used to keep different types of items fresh.

[0079] The compressor 201 of the refrigerator is used to compress the refrigerant and circulate the refrigerant in the refrigeration system 104. The storage room in the refrigerator is cooled by the circulation of the refrigerant in the refrigeration system.

[0080] The refrigerator further includes a compartment temperature sensor 301 , which is disposed inside the storage compartment 102 and is used to collect the compartment temperature of the storage compartment 102 .

[0081] The refrigerator further comprises an ambient temperature sensor 302, which is arranged on the refrigerator housing and is used to collect the ambient temperature of the room where the refrigerator is located.

[0082] The refrigerator further includes a controller, and the controller 303 is electrically connected to the compressor 201, the compartment temperature sensor 301 and the ambient temperature sensor 302, and is capable of acquiring data and issuing control.

[0083] See also Figure 4 , is a flow chart of steps performed by the controller provided by the present invention under the first embodiment, wherein the controller performs the following steps:

[0084] Step S401, obtaining the compartment temperature, ambient temperature and current set temperature;

[0085] Step S402, constructing a feature vector according to the compartment temperature, the ambient temperature and the current set temperature;

[0086] Step S403, inputting the feature vector into a pre-trained neural network model for calculation to obtain an operating frequency output;

[0087] Step S404: controlling the compressor to operate at the operating frequency.

[0088] Specifically, after the refrigerator is powered on, the controller is used to obtain operating data of the refrigerator.

[0089] The operation data includes the current set temperature of the storage room set by the user or matched by the system according to the ambient temperature. The current set temperature is the target temperature of the storage room. The temperature of the storage room needs to be maintained at about the current set temperature to achieve the corresponding preservation effect.

[0090] It should be noted that different storage rooms, such as refrigerators, freezers and temperature-controlled chambers, have corresponding set temperatures different to achieve the preservation of different types of items.

[0091] The operation data also includes the compartment temperature obtained by the compartment temperature sensor and the ambient temperature obtained by the ambient temperature sensor; the compartment temperature can represent the temperature of the current compartment of the refrigerator, and the current refrigeration demand of the storage room can be calculated based on the compartment temperature and the current set temperature. The ambient temperature can be used to infer the heat loss of the environment in which the refrigerator is located. When the room is at high temperature, the difference between the refrigerator and the room temperature is large, so the rate of heat conduction from the refrigerator storage room to the room is high, and the heat loss is fast. On the contrary, when the room is at low temperature, the difference between the refrigerator and the room temperature is small, so the rate of heat conduction from the refrigerator storage room to the room is low, and the heat loss is slow.

[0092] Therefore, by including the operating parameters of the compartment temperature, the ambient temperature and the current set temperature, the required operating parameters of the refrigerator can be roughly calculated. Construct a feature vector S according to the compartment temperature, the ambient temperature and the current set temperature;

[0093] Deep learning algorithms have made breakthrough progress in speech recognition, face recognition, autonomous driving, etc. The essence of deep learning methods is to build a multi-hidden nonlinear model based on sufficient data, extract and combine features of input data in high-dimensional space, and improve the efficiency of feature extraction and combination in multiple iterations, so as to finally complete the classification and prediction tasks.

[0094] The feature vector S is input into the pre-trained neural network model Q for calculation to obtain the operating frequency output, the operating frequency F = max(Q(S)).

[0095] The compressor is controlled to operate at the calculated operating frequency.

[0096] The neural network model provided in the present application constructs an algorithm model by utilizing a deep learning algorithm, trains the model through the data generated during the operation of the refrigerator, and obtains a neural network model. The operating frequency is calculated through the neural network model, thereby optimizing the operating parameters, accurately calculating the operating frequency of the refrigerator, and avoiding energy consumption caused by inappropriate operating frequency.

[0097] In another embodiment provided by the present invention, see Figure 5 , is a flow chart of the steps executed by the controller provided by the present invention in the second embodiment. The controller is also used to execute:

[0098] Step S501, sequentially obtaining the real-time power of the compressor when it runs at the operating frequency until it stops at a preset time interval;

[0099] Step S502, calculating the electric energy consumed by the compressor when it runs at the operating frequency until it stops according to the acquired real-time frequency;

[0100] Step S503, obtaining the actual temperature of the compartment and the set temperature of the compartment when the compressor is stopped;

[0101] Step S504, calculating the evaluation value of the operating frequency using a preset evaluation value calculation model according to the actual temperature of the compartment, the set temperature of the compartment and the electric energy;

[0102] Step S505, constructing a data sequence according to the operating frequency, the feature vector and the evaluation value, and storing the data sequence in a pre-built experience database as training data for updating the neural network model.

[0103] For details, see Figure 6 , is a flow chart of the steps executed by the controller provided by the present invention in the third embodiment; the controller executes the following steps:

[0104] Step S601, obtaining the operating data of the refrigerator at the current moment, the operating data including the temperature of the refrigerator compartment, the ambient temperature and the current set temperature;

[0105] Step S602: construct a feature vector S according to the acquired operation data, wherein the feature vector S includes the ambient temperature, the current set temperature, and the actual temperature of the compartment.

[0106] Step S603, determining whether the refrigerator needs refrigeration;

[0107] If not, return to step S603;

[0108] If yes, execute step S604;

[0109] Step S604, input the feature vector S into the neural network model Q to calculate the operating frequency F of the compressor, where the operating frequency F = max(Q(S)).

[0110] Step S605, operating the compressor at the operating frequency F until it stops, and counting the cumulative operating time t in seconds;

[0111] Step S606: During the operation of the compressor, the real-time power P of the compressor is obtained according to a preset time interval Δt. i , in watts;

[0112] Step S607, calculating the electric energy consumed by the compressor, that is, calculating the electric energy consumed by the compressor when it runs at the operating frequency until it stops according to the real-time power. The unit is kilowatt-hour;

[0113] Among them, n is the real-time power P obtained i The number of The calculation result is rounded.

[0114] Step S608, obtaining the actual compartment temperature Tr and the set compartment temperature Ts when the compressor is stopped;

[0115] Step S609, calculating the evaluation value of the operating frequency, that is, calculating the evaluation value R(F) of the operating frequency using a preset evaluation value calculation model according to the actual temperature of the compartment, the set temperature of the compartment and the electric energy.

[0116] Step S610, constructing a data sequence and storing the data sequence in a pre-built empirical database, that is, constructing a data sequence according to the operating frequency, the feature vector and the evaluation value, and storing the data sequence in a pre-built empirical database as training data for updating the neural network model.

[0117] The operating frequency and the calculated evaluation value obtained by the neural network model each time are stored as a data sequence. The neural network model is updated using the data sequence, which can optimize the operating parameters through self-learning and automatic training, improve the model accuracy, and reduce the loss during the use of the refrigerator.

[0118] In another embodiment provided by the present invention, the controller is further configured to:

[0119] Reading the number of data sequences stored in the experience database;

[0120] When the number of data sequences is greater than M, randomly taking N data sequences from the experience database as training data, training the neural network model, and updating the parameters of the neural network model;

[0121] Wherein, M is a positive integer not less than N.

[0122] When this embodiment is implemented, see Figure 6 , the controller is further configured to perform the following steps:

[0123] Step S611, reading the number D of data sequences stored in the experience database;

[0124] Step S612, determine whether D>M holds; that is, determine whether the number of data sequences stored in the experience database exceeds a preset value.

[0125] If not, return to step S601;

[0126] If yes, execute step S613;

[0127] Step S613, extracting training data from the experience database, updating the parameters of the neural network model, and returning to step S601. That is, randomly taking N data sequences from the experience database as training data, training the neural network model, and updating the parameters of the neural network model.

[0128] It should be noted that, in the embodiment provided by the present invention, the total number of training times is M = 1600, and the sampling scale of the memory database is N = 32. In other embodiments, M and N can be selected from other values ​​or adjusted according to actual conditions.

[0129] By using the data sequence of the refrigerator during actual operation as training data, the model can be modified according to the actual operation of the refrigerator. The impact of other factors on power consumption during the actual operation of the refrigerator, such as the aging of the refrigerator, can be considered, and the model can be updated in a targeted manner to improve the model accuracy, improve the accuracy of the operating frequency output, and reduce power consumption.

[0130] In another embodiment provided by the present invention, the evaluation value calculation model is specifically:

[0131] Wherein, R(F) is the evaluation value of the working frequency F, E is the electric energy, t is the cumulative working time of the compressor running at the working frequency until it stops, Tr and Ts are the actual temperature of the compartment and the set temperature of the compartment respectively, α is the temperature difference coefficient, ΔTu is the preset upper limit of the temperature difference.

[0132] In this embodiment, the evaluation value calculation model used when calculating the evaluation value is specifically:

[0133] Wherein, R(F) is the evaluation value of the working frequency F, E is the electric energy consumed by the compressor from running at the working frequency to stopping, t is the cumulative working time of the compressor from running at the working frequency to stopping, Tr and Ts are the actual temperature of the compartment and the set temperature of the compartment, respectively.

[0134] α is the temperature difference coefficient, When the difference between the actual temperature of the compartment and the set temperature of the compartment is less than the upper limit of the temperature difference ΔTu, it indicates that when the refrigerator is running to shutdown, the actual temperature of the compartment can be controlled within the set temperature range of the compartment, and the corresponding temperature difference coefficient is 0.01. When the difference between the actual temperature of the compartment and the set temperature of the compartment is not less than the upper limit of the temperature difference ΔTu, it indicates that when the refrigerator is running to shutdown, the actual temperature of the compartment cannot be controlled within the set temperature range of the compartment, and the temperature difference coefficient is 1.

[0135] In this embodiment, the upper limit of the difference between the actual temperature of the compartment and the set temperature of the compartment is ΔTu = 5. In other embodiments, the upper limit of the difference ΔTu can be set or adjusted according to actual conditions, which will not be described in detail here.

[0136] The evaluation value calculation model can be used to evaluate the operating frequency and operating effect according to the actual temperature of the refrigerator room from the time the refrigerator runs to the time it stops and the set temperature of the compartment. Therefore, the operating effect can be evaluated in the obtained data sequence, and then the neural network model can be corrected according to the operating effect of the refrigerator.

[0137] In another embodiment provided by the present invention, the neural network model includes 1 input layer, 3 hidden layers and 1 output layer;

[0138] Each hidden layer consists of 64 neurons;

[0139] The output layer includes 22 neurons, and each neuron of the output layer corresponds to a working frequency vector.

[0140] When this embodiment is implemented, see Figure 7 , is a topological structure diagram of a neural network model provided by an embodiment of the present invention. The neural network includes an input layer (vector A [0] ), 3 hidden layers (vector A [1] ~A [3] ), an output layer (vector A [4] );

[0141] The input layer inputs the feature vector S, which includes three input variables: the compartment temperature, the ambient temperature, and the current set temperature; each hidden layer contains 64 neurons; the vector A of the first hidden layer [1] It consists of 64 neurons, namely a [1] 1~a [1]64 ; The vector A of the second hidden layer [2] It consists of 64 neurons, namely a [2] 1~a [2] 64 ; The vector A of the third hidden layer [3] It consists of 64 neurons, namely a [3] 1~a [3] 64 ; The output layer contains 22 neurons, namely a [4] 1~a [4] 22 , the neurons correspond to the operating frequency vector of the refrigerator compressor one by one. By inputting the eigenvector S into the input layer, each input variable of the eigenvector is sequentially input into the three hidden layers for calculation, and then input into the 22 neurons of the output layer for calculation after the three hidden layers, and finally the optimal operating frequency vector output is obtained.

[0142] In another embodiment provided by the present invention, the controller is further configured to:

[0143] Inputting the feature vector into the input layer of the neural network model;

[0144] Calculate the evaluation value of the feature vector of each neuron from the input layer to the output layer;

[0145] The operating frequency vector corresponding to the neuron with the largest evaluation value in the output layer is output as the operating frequency.

[0146] In the specific implementation of this embodiment, the operating frequency vector of the refrigerator compressor is {[40, 45, 50, 55, 60, ... 120, 125, 130, 135, 140, 145]}, which contains 22 values;

[0147] By inputting the feature vector S into the input layer, each input variable of the feature vector is sequentially input into the three hidden layers for calculation, and then input into the 22 neurons of the output layer for calculation after passing through the three hidden layers, and the weight of the feature vector of each neuron from the hidden layer to the output layer is calculated; the neuron max(A) with the largest weight in the output layer vector of the neural network model Q is converted into [4] ) corresponds to the output value of the neural network, denoted as max(Q(S)).

[0148] In another embodiment provided by the present invention, the controller is further configured to:

[0149] After the refrigerator is powered on, determining whether it is currently in a user usage time period;

[0150] When the user is currently in the use time period, controlling the compressor to operate at a preset operating frequency;

[0151] When the user is not currently in use, the neural network model is used to calculate the operating frequency of the compressor, and the compressor is controlled to operate at the calculated operating frequency.

[0152] When this embodiment is implemented, see Figure 8 , is a flow chart of the steps executed by the controller provided by the present invention in the fourth embodiment. The controller also executes the following steps when performing compressor operation control:

[0153] Step S801, the refrigerator is powered on and operated;

[0154] Step S802, determining whether the current time period is a user usage time period;

[0155] If not, execute step S803;

[0156] If yes, execute step S804;

[0157] Step S803, controlling the compressor to run at a preset frequency, that is, when the user uses the refrigerator, the operating frequency is not calculated through the neural network model, and the refrigerator is directly controlled to run at the preset frequency to avoid repeated changes in the operating frequency of the refrigerator due to repeated opening of the refrigerator door by the user.

[0158] Step S804, using the neural network model to calculate the operating frequency of the compressor, and controlling the operation of the compressor with the calculated operating frequency. That is, when the user is not in the use time period, the operating frequency of the refrigerator compressor is calculated according to the neural network model, and then the refrigerator compressor is controlled.

[0159] In another embodiment provided by the present invention, the controller is further configured to:

[0160] Determine whether all doors of the refrigerator storage compartment are in a closed state within a preset time period before the current moment;

[0161] If yes, determining that the refrigerator is not in the user use time period;

[0162] If not, it is determined that the refrigerator is in the user use time period.

[0163] In the specific implementation of this embodiment, the controller determines whether the refrigerator is in the user's use time period by judging the state of the door of the refrigerator storage chamber. When the door of the refrigerator storage chamber is in an open state, it indicates that the user is using it.

[0164] In specific implementation, the state of the door of the refrigerator storage room within 10 minutes before the current time T is obtained;

[0165] Determine whether the door of the refrigerator storage room is in a closed state within 10 minutes before the current time T;

[0166] If yes, that is, all doors were in a closed state 10 minutes before the current moment, it means that the user has not opened the refrigerator within 10 minutes, and the refrigerator is not in the user's use time period.

[0167] If not, that is, the door was not in a closed state 10 minutes before the current moment, it means that the user used the refrigerator within 10 minutes and opened the door of the storage room, and the refrigerator is in the user's use time period.

[0168] It should be noted that, in this embodiment, the preset duration is set to 10 minutes. In other embodiments, the preset duration can be set or adjusted according to actual conditions, which will not be elaborated here.

[0169] By detecting the state of the refrigerator door, the refrigerator is placed in the user's use state within ten minutes after the refrigerator door is used, so as to prevent the user from repeatedly opening the refrigerator in a short period of time, causing the temperature in the refrigerator to fluctuate frequently in a short period of time, and thus frequently calculating the operating frequency through the neural network model, repeatedly changing the operating frequency, and affecting the service life of the compressor. It can also prevent the refrigerator from frequently changing abnormally due to the user frequently opening the door, thereby calculating a large number of operating frequencies as training data to affect the model accuracy.

[0170] In another embodiment of the present invention, the refrigerator further includes a frequency conversion module:

[0171] The frequency conversion module is used to receive the control instruction of the controller, control the frequency of the compressor according to the control parameters in the control instruction, and calculate the real-time power of the compressor in real time.

[0172] When this embodiment is implemented, see Fig. 9 , is a schematic structural diagram of a refrigerator provided by the present invention in a second embodiment;

[0173] The refrigerator also includes a frequency conversion module 901; the frequency conversion module 901 is used to receive control instructions from the controller, control the operation of the compressor according to the operating frequency in the control instructions, calculate the real-time power of the variable frequency compressor according to the actual operation of the compressor, and upload the real-time power to the controller.

[0174] See also Fig.10 , is a schematic structural diagram of a refrigerator provided by the present invention in a third embodiment, wherein the refrigerator comprises a controller 303, a compartment temperature sensor 301, an ambient temperature sensor 302, a variable frequency compressor 201, and a frequency conversion module 901;

[0175] The compartment temperature sensor 301 and the ambient temperature sensor 302 are both electrically connected to the controller 303 and can transmit the collected temperature data to the controller. The controller is used to run the neural network model according to the operating data and calculate the operating frequency.

[0176] The compressor 201 is connected to the controller via the frequency conversion module 901. The controller 303 sends a control instruction including the operating frequency to the frequency conversion module 901, and the frequency conversion module 901 controls the operation of the compressor 201 according to the control instruction. The real-time power is calculated according to the operation of the compressor, and the real-time power is uploaded to the controller 303, so that the controller 303 calculates according to the real-time power and the operation data to obtain a data sequence for updating the neural network model.

[0177] The controller calculates the operating frequency based on the refrigerator operating data, and then controls the operation through the operating frequency to obtain the data sequence, update the neural network model, and continuously iterate and update the neural network model through self-learning to improve the model accuracy, ensure the accuracy of the compressor control process, and reduce useless power consumption losses.

[0178] Another embodiment of the present invention provides a refrigerator compressor operation control method, the refrigerator comprising:

[0179] A box body, wherein a door body is provided at the opening of the box body, and at least one storage room is provided in the box body for storing items;

[0180] A compressor, the compressor being used to compress and circulate a refrigerant;

[0181] A compartment temperature sensor, the compartment temperature sensor is used to collect the compartment temperature of the storage room;

[0182] An ambient temperature sensor, wherein the ambient temperature sensor is used to collect the indoor ambient temperature;

[0183] Controller;

[0184] The method comprises:

[0185] Get the compartment temperature, ambient temperature and current set temperature;

[0186] Constructing a feature vector according to the compartment temperature, the ambient temperature and the current set temperature;

[0187] Inputting the feature vector into a pre-trained neural network model for calculation to obtain an operating frequency output;

[0188] The compressor is controlled to operate at the operating frequency.

[0189] It should be noted that the refrigerator compressor operation control method provided in the embodiment of the present invention is the same as all the process steps executed by the refrigerator controller in the above embodiment, see Figure 4 The working principles and beneficial effects of the two correspond one to one. For the specific workflow diagram, see Figure 5 , Figure 6 and Figure 8 , so I will not go into details.

[0190] A person skilled in the art can understand that all or part of the processes in the above-mentioned embodiments can be implemented by instructing the relevant hardware through a computer program, and the program can be stored in a computer-readable storage medium, and when the program is executed, it can include the processes of the embodiments of the above-mentioned methods. The storage medium can be a disk, an optical disk, a read-only memory (ROM) or a random access memory (RAM), etc.

[0191] The above is a preferred embodiment of the present invention. It should be pointed out that a person skilled in the art can make several improvements and modifications without departing from the principle of the present invention. These improvements and modifications are also considered to be within the scope of protection of the present invention.

Claims

1. A refrigerator, characterized in that: include: A box body, wherein at least one storage room is provided in the box body for storing items; A compressor, wherein the compressor is used to compress and circulate a refrigerant; A compartment temperature sensor, the compartment temperature sensor is used to collect the compartment temperature of the storage room; An ambient temperature sensor, wherein the ambient temperature sensor is used to collect the indoor ambient temperature; A controller, the controller being configured to: Get the compartment temperature, ambient temperature and current set temperature; Constructing a feature vector according to the compartment temperature, the ambient temperature and the current set temperature; Inputting the feature vector into a pre-trained neural network model for calculation to obtain an operating frequency output; The compressor is controlled to operate at the operating frequency.

2. The refrigerator according to claim 1, characterized in that: The controller is also configured to: Sequentially obtaining the real-time power of the compressor when it runs at the operating frequency until it stops at a preset time interval; Calculate the electric energy consumed by the compressor when it runs at the operating frequency until it stops according to the acquired real-time frequency; Obtaining the actual temperature of the compartment and the set temperature of the compartment when the compressor is stopped; Calculating the evaluation value of the operating frequency using a preset evaluation value calculation model according to the actual temperature of the compartment, the set temperature of the compartment and the electric energy; A data sequence is constructed according to the operating frequency, the feature vector and the evaluation value, and the data sequence is stored in a pre-built experience database as training data for updating the neural network model.

3. The refrigerator according to claim 2, characterized in that: The controller is also configured to: Reading the number of data sequences stored in the experience database; When the number of data sequences is greater than M, randomly taking N data sequences from the experience database as training data, training the neural network model, and updating the parameters of the neural network model; Wherein, M is a positive integer not less than N.

4. The refrigerator according to claim 2, characterized in that: The evaluation value calculation model is specifically: Wherein, R(F) is the evaluation value of the working frequency F, E is the electric energy, t is the cumulative working time of the compressor running at the working frequency until it stops, Tr and Ts are the actual temperature of the compartment and the set temperature of the compartment respectively, α is the temperature difference coefficient, ΔTu is the preset upper limit of the temperature difference.

5. The refrigerator according to claim 1, characterized in that: The neural network model includes 1 input layer, 3 hidden layers and 1 output layer; Each hidden layer consists of 64 neurons; The output layer includes 22 neurons, and each neuron of the output layer corresponds to a working frequency vector.

6. The refrigerator according to claim 5, characterized in that: The controller is also configured to: Inputting the feature vector into the input layer of the neural network model; Calculate the weight of the feature vector of each neuron from the hidden layer to the output layer; The operating frequency vector corresponding to the neuron with the largest weight in the output layer is used as the operating frequency output.

7. The refrigerator according to claim 1, characterized in that: The controller is also configured to: After the refrigerator is powered on, determining whether it is currently in a user usage time period; When the user is currently in the use time period, controlling the compressor to operate at a preset operating frequency; When the user is not currently in use, the neural network model is used to calculate the operating frequency of the compressor, and the compressor is controlled to operate at the calculated operating frequency.

8. The refrigerator according to claim 7, characterized in that: The controller is also configured to: Determine whether all doors of the refrigerator storage compartment are in a closed state within a preset time period before the current moment; If yes, determining that the refrigerator is not in the user use time period; If not, it is determined that the refrigerator is in the user use time period.

9. The refrigerator according to claim 2, characterized in that: The refrigerator also includes a frequency conversion module: The frequency conversion module is used to receive the control instruction of the controller, control the operation of the compressor according to the operating frequency in the control instruction, and calculate the real-time power of the compressor in real time.

10. A refrigerator compressor operation control method, characterized in that: The refrigerator comprises: A box body, wherein a door body is provided at the opening of the box body, and at least one storage room is provided in the box body for storing items; A compressor, the compressor being used for compressing and circulating a refrigerant; A compartment temperature sensor, the compartment temperature sensor is used to collect the compartment temperature of the storage room; An ambient temperature sensor, wherein the ambient temperature sensor is used to collect the indoor ambient temperature; Controller; The method comprises: Get the compartment temperature, ambient temperature and current set temperature; Constructing a feature vector according to the compartment temperature, the ambient temperature and the current set temperature; Inputting the feature vector into a pre-trained neural network model for calculation to obtain an operating frequency output; The compressor is controlled to operate at the operating frequency.

Images