Air conditioner and defrosting control method and device thereof, storage medium and program product

Abstract

The invention provides an air conditioner and a defrosting control method and device thereof, a storage medium and a program product, and the method comprises the steps: collecting the current operation state data of the air conditioner, and obtaining the last defrosting time and historical operation time of the air conditioner and the operation state data of the air conditioner in the past first preset time; a pre-trained frosting state prediction model of the air conditioner is called, and the frosting state of the air conditioner is predicted through the collected current operation state data of the air conditioner and the obtained operation state data of the air conditioner within the past preset time; and if it is determined that defrosting treatment needs to be carried out on the air conditioner according to the prediction result of the frosting state, after defrosting treatment is carried out on the air conditioner, the frosting state prediction model is updated. According to the scheme provided by the invention, the influence of air conditioner aging and environmental factors on the frosting detection result can be reduced, and the frosting detection accuracy is improved.

Description

Technical Field

[0001] The present invention relates to the field of control, and in particular to an air conditioner and a defrosting control method, device, storage medium and program product thereof. Background Art

[0002] The compression refrigeration system of a general air conditioner is composed of a compressor, a condenser, an evaporator and a throttling element, etc., and is connected by pipes in a certain order to form a closed system. The closed system is filled with an appropriate amount of refrigerant. Under the action of the compressor, the refrigerant comes out of the compressor and returns to the compressor through the condenser, evaporator and throttling structure. The heat moves along with the movement of the refrigerant, realizing the refrigeration process from absorbing heat from the evaporator to releasing heat from the condenser. When the air conditioning system is heating, the system sends heat from the outdoors to the indoors, and the outdoor heat exchanger acts as an evaporator. If the temperature of the outdoor air is low and the humidity is high, the water vapor in the air will condense and frost on the surface of the outdoor heat exchanger. The frost on the outdoor heat exchanger increases the heat transfer resistance between the surface of the outdoor heat exchanger and the air, and increases the flow resistance when the airflow passes through the outdoor heat exchanger, so that the air flow through the outdoor heat exchanger is reduced, and the heat exchange efficiency is significantly reduced, which not only affects the indoor comfort, but also causes the system energy efficiency to decrease. In order to prevent this phenomenon, when the amount of frost on the outdoor heat exchanger exceeds a certain amount, it is necessary to melt the frost on the outdoor heat exchanger through defrosting operation, so that it flows down and is discharged outside the machine. In the related art, the defrost control of the outdoor heat exchanger is mostly fixed-time defrosting. When the ambient temperature is low, the outdoor heat exchanger tube temperature is used to determine whether frosting occurs. However, in actual low-temperature conditions, the outdoor heat exchanger may not necessarily frost. If it is determined based on the tube temperature, it will cause misjudgment and increase the number of defrosting times. During defrosting, it is usually necessary to switch the heating cycle to a refrigeration cycle, using the compressor as a heat source, so that the high-temperature gas refrigerant from the compressor flows into the outdoor heat exchanger for defrosting, which causes a significant drop in indoor temperature and a longer time required for the room temperature to reach the set temperature, greatly reducing the user's comfort. Summary of the invention

[0003] The main purpose of the present invention is to overcome the defects of the above-mentioned related technologies and to provide an air conditioner and its defrosting control method, device, storage medium and program product to solve the problem of inaccurate judgment of air conditioner frosting in the related technologies.

[0004] On the one hand, the present invention provides a defrost control method for an air conditioner, comprising: collecting current operating status data of the air conditioner, and obtaining the time when the air conditioner was last defrosted and the historical operating time, as well as the operating status data of the air conditioner within a first preset time in the past; calling a pre-trained frost state prediction model of the air conditioner, and using the collected current operating status data of the air conditioner and the operating status data of the air conditioner within a first preset time in the past to predict the frost state of the air conditioner; updating the frost state prediction model according to the prediction result of the frost state and the actual frost state of the air conditioner.

[0005] Optionally, the pre-trained air conditioner frost state prediction model includes: a pre-trained model and a user personal model; the pre-trained model is a frost state prediction model of the air conditioner pre-trained based on experimental test data; and / or the user personal model is a frost state prediction model of the air conditioner trained based on historical operating status data of the air conditioner; wherein, when the current target user uses the air conditioner for the first time, the pre-trained model is used to predict the frost state of the air conditioner, and when the current target user is not using the air conditioner for the first time, the user personal model of the current target user is used to predict the frost state of the air conditioner.

[0006] Optionally, updating the frost state prediction model includes: if the current target user is using the air conditioner for the first time, using the historical operating status data of the air conditioner to update the pre-trained model to form a user personal model of the current target user; if the current target user is not using the air conditioner for the first time, using the historical operating status data of the air conditioner to update the user personal model of the current target user.

[0007] Optionally, the frosting state prediction model is updated according to the prediction result of the frosting state and the actual frosting state of the air conditioner, including: if the prediction result is inconsistent with the actual frosting state of the air conditioner, updating the frosting state model.

[0008] Optionally, calling a pre-trained frost state prediction model of the air conditioner includes: issuing a call request to a cloud platform to call the frost state prediction model of the air conditioner stored on the cloud platform.

[0009] On the other hand, the present invention provides a defrost control device for an air conditioner, comprising: a collection unit, used to collect current operating status data of the air conditioner, and obtain the time when the air conditioner was last defrosted and the historical operating time, as well as the operating status data of the air conditioner within a first preset time in the past; a prediction unit, used to call a pre-trained frost state prediction model of the air conditioner, and use the collected current operating status data of the air conditioner and the operating status data of the air conditioner within a first preset time in the past to predict the frost state of the air conditioner; an updating unit, used to update the frost state prediction model after the air conditioner is defrosted if it is determined that the air conditioner needs to be defrosted based on the prediction result of the frosting state.

[0010] Optionally, the pre-trained air conditioner frost state prediction model includes: a pre-trained model and a user personal model; the pre-trained model is a frost state prediction model of the air conditioner pre-trained based on experimental test data; and / or the user personal model is a frost state prediction model of the air conditioner trained based on historical operating status data of the air conditioner; wherein, when the current target user uses the air conditioner for the first time, the pre-trained model is used to predict the frost state of the air conditioner, and when the current target user is not using the air conditioner for the first time, the user personal model of the current target user is used to predict the frost state of the air conditioner.

[0011] Optionally, the updating unit updates the frost state prediction model, including: if the current target user is using the air conditioner for the first time, using the historical operating status data of the air conditioner to update the pre-trained model to form a user personal model of the current target user; if the current target user is not using the air conditioner for the first time, using the historical operating status data of the air conditioner to update the user personal model of the current target user.

[0012] Optionally, the updating unit updates the frosting state prediction model according to the prediction result of the frosting state and the actual frosting state of the air conditioner, including: if the prediction result is inconsistent with the actual frosting state of the air conditioner, updating the frosting state model.

[0013] Optionally, the prediction unit calls the pre-trained frost state prediction model of the air conditioner, including: issuing a call request to the cloud platform to call the frost state prediction model of the air conditioner stored on the cloud platform.

[0014] Another aspect of the present invention provides a storage medium having a computer program stored thereon, wherein the program implements the steps of any of the aforementioned methods when executed by a processor.

[0015] In another aspect, the present invention provides an air conditioner, comprising a processor, a memory, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of any of the aforementioned methods when executing the program.

[0016] In another aspect, the present invention provides an air conditioner, comprising any of the above-mentioned defrost control devices.

[0017] In yet another aspect, the present invention provides a computer program product, comprising a computer program, wherein when the computer program is executed by a processor, the steps of any of the aforementioned methods are implemented.

[0018] According to the technical solution of the present invention, the operating data of the air conditioner is collected to call the frost state prediction model to predict the frost state of the air conditioner, and the LSTM model is automatically updated according to the operating conditions of the air conditioner, thereby reducing the impact of air conditioner aging and environmental factors on the detection results and improving the accuracy of detection.

[0019] According to the technical solution of the present invention, the frost state prediction model can be updated based on the operating status data of the user's personal air conditioner as the air conditioner is used, so as to adapt to the air conditioner's own state and avoid inaccurate detection of the frost state due to environmental factors and air conditioner aging problems. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] The drawings described herein are used to provide a further understanding of the present invention and constitute a part of the present invention. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention. In the drawings:

[0021] Figure 1 It is a method schematic diagram of an embodiment of the defrost control method of an air conditioner provided by the present invention;

[0022] Figure 2 It is a method schematic diagram of a specific embodiment of the defrosting control method of an air conditioner provided by the present invention;

[0023] Figure 3 shows a system structure diagram of an air conditioning system according to the present invention;

[0024] Figure 4 It is a structural block diagram of an embodiment of the defrost control device for an air conditioner provided by the present invention. DETAILED DESCRIPTION

[0025] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the specific embodiments of the present invention and the corresponding drawings. 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 making creative work are within the scope of protection of the present invention.

[0026] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0027] The invention provides a defrosting control method for an air conditioner.

[0028] Figure 1 It is a method schematic diagram of an embodiment of the defrost control method of an air conditioner provided by the present invention.

[0029] like Figure 1 As shown, according to one embodiment of the present invention, the defrost control method of the air conditioner at least includes step S110, step S120 and step S130.

[0030] Step S110, collecting the current operating status data of the air conditioner, and obtaining the operating status data of the air conditioner within a first preset time in the past.

[0031] The current operating status data of the air conditioner may specifically include: current environmental parameters and the current operating parameters of the air conditioner; the operating status data of the air conditioner within the past first preset time may specifically include: environmental parameters within the past first preset time and the operating parameters of the air conditioner. The operating status data of the air conditioner is acquired once at a preset interval within the past first preset time.

[0032] The environmental parameters may specifically include: at least one of outdoor ambient temperature and outdoor ambient humidity. The operating parameters may specifically include: at least one of outdoor heat exchanger temperature (preferably outdoor heat exchanger surface temperature), indoor heat exchanger temperature, operating mode, outdoor fan speed and compressor power, and the last defrosting time and historical operating time of the air conditioner.

[0033] For example, the temperature and humidity data can be collected through corresponding sensors and corresponding temperature sensing packages, and the operation mode of the air conditioner, the speed of the outdoor fan and the power of the compressor can be obtained through the controller.

[0034] Step S120, calling a pre-trained frost state prediction model of the air conditioner, and using the collected current operating state data of the air conditioner and the operating state data of the air conditioner within a first preset time in the past to predict the frost state of the air conditioner.

[0035] The pre-trained air conditioner frost state prediction model may specifically include: a pre-trained model and a user personal model. When the current target user uses the air conditioner for the first time, the pre-trained model is used to predict the frost state of the air conditioner; when the current target user does not use the air conditioner for the first time, the user personal model of the current target user is used to predict the frost state of the air conditioner.

[0036] The pre-trained model is a frost state prediction model of the air conditioner pre-trained based on experimental test data. For example, an experimental test is performed in a laboratory environment to obtain the frost state of the air conditioner under different operating states, and the operating state data and frost state of the different operating states are recorded as sample data for model training, and the model is trained using the sample data to obtain a pre-trained model.

[0037] The user personal model is a frost state prediction model of the air conditioner trained based on the historical operating state data of the air conditioner, that is, a personal model of the current target user of the air conditioner. Specifically, the user personal model is obtained by updating the pre-trained model using the historical operating state data of the air conditioner.

[0038] In actual applications, the pre-trained model is used to predict the frosting state when the air conditioner is first run, and then the user's personal model is used, that is, the pre-trained model is updated using the operating status data of the air conditioner to form a user's personal model for the target user's personal use of the air conditioner. Each user can obtain his or her own personal LSTM model according to his or her ID number.

[0039] By calling the pre-trained frost state prediction model of the air conditioner, and inputting the collected current operating state data of the air conditioner and the operating state data of the air conditioner within the first preset time in the past into the frost state prediction model, the future frost state of the air conditioner can be predicted. That is, the collected current environmental parameters and the current operating parameters of the air conditioner, as well as the environmental parameters and the operating parameters of the air conditioner within the first preset time in the past are input into the pre-trained frost state prediction model of the air conditioner, and the frost state of the air conditioner within the second preset time in the future is predicted. The frost state can specifically include: a frosted state and an unfrosted state. Optionally, it can also include an approaching frost state.

[0040] The frosting state prediction model can be specifically an LSTM model (Long Short-Term Memory Artificial Neural Network Model). Therefore, the data for model training is time series data. The collected operating state data of the air conditioner at different times are processed into time series data and then used for model training.

[0041] The current environmental parameters, the operating parameters of the air conditioner, the time when the air conditioner was last defrosted and the historical operating time, and multiple groups of operating status data collected over a period of time in the past, including the environmental parameters, the operating parameters of the air conditioner, the time when the air conditioner was last defrosted and the historical operating time, are processed into an operating status sequence as the input of a frosting status prediction model to predict and output the frosting status of the air conditioner.

[0042] For example, the current vector xt and the operating state sequence ht-1 in the past period of time are used as input (when t=0, ht-1=0) to predict the frosting state ht of the air conditioner in the future period of time.

[0043] it=σ(Wi[ht-1,xt]+bi)

[0044] ft=σ(Wf[ht-1,xt]+bf)

[0045] ct=ftct-1+ittanh(Wc[ht-1,xt]+bc)

[0046] ot=σ(Wo[ht-1,xt]+bo)

[0047] ht=ottanh(ct)

[0048] Among them, mathematical σ is used as the standard sigmoid function; i, f, o and c are the input gate, forget gate, output gate and memory unit respectively; bi, bf, bo and bc are the bias vectors of the input gate, forget gate, output gate and memory unit respectively; W is the weight matrix between each unit and the gate vector.

[0049] Preferably, the frost state prediction model of the air conditioner can be stored in a cloud platform, and a call request is sent to the cloud platform to call the frost state prediction model of the air conditioner stored on the cloud platform. Specifically, a call request is sent to the cloud platform, and the cloud platform calls the model to predict the frost state of the air conditioner.

[0050] Step S130: updating the frosting state prediction model according to the prediction result of the frosting state and the actual frosting state of the air conditioner.

[0051] If it is predicted that the frosting state of the air conditioner within the second time in the future is frosting, the air conditioner can be defrosted, or a reminder message for defrosting can be issued to remind the user to perform defrosting (for example, notifying the user to perform defrosting through a bound client APP). The defrosting process can switch the heating operation to the cooling operation, using the compressor as the heat source, so that the high-temperature gas refrigerant from the compressor flows into the outdoor heat exchanger for defrosting. According to the prediction result of the frosting state and the actual frosting state of the air conditioner, the frosting state prediction model is updated. Specifically, if the prediction result is inconsistent with the actual frosting state of the air conditioner, the frosting state model is updated. For example, if it is predicted that the frosting state of the air conditioner within the second time in the future is frosting, but it is judged that the air conditioner is not frosted based on the air conditioner operation status data collected within the second preset time, the frosting state prediction model is updated.

[0052] Among them, if the current target user is using the air conditioner for the first time, the historical operating status data of the air conditioner is used to update the pre-trained model to form a user personal model of the current target user. If the current target user is not using the air conditioner for the first time, the historical operating status data of the air conditioner is used to update the user personal model of the current target user.

[0053] That is, if the current user is using the air conditioner for the first time, the frost state prediction model used for predicting the frost state of the air conditioner is a pre-trained model, and the historical operating state data of the air conditioner (specifically, it may include the operating state data of the air conditioner collected from the last time the frost state prediction model was updated to the current moment) is used to update the pre-trained model. If the current user is not using the air conditioner for the first time, the frost state prediction model used for predicting the frost state of the air conditioner is a user personal model, and the historical operating state data of the air conditioner (specifically, it may include the operating state data of the air conditioner collected from the last time the frost state prediction model was updated to the current moment) is used to update the user personal model.

[0054] Specifically, the n operating state vectors x = (x1, x2, .., xn) collected after the last model training are used to form a data set (for example, outdoor ambient temperature, outdoor ambient humidity, outdoor heat exchanger temperature, indoor heat exchanger temperature, operating mode, last defrosting time and historical operating time), and the LSTM model is trained and updated to adapt to the air conditioner's own state and weather factors, and the impact of the air conditioner frost problem on the length of the time series is divided:

[0055] Short-term time series: Changes in a short period of time will have a greater impact on the frosting state of the air conditioner, and the changed values ​​need to be memorized or forgotten in a short period of time, such as humidity, temperature, and air conditioner operation mode;

[0056] Long-term time series: Changes in values ​​over a long period of time will affect the frosting of the air conditioner. There is no need to memorize or forget the changes in a short period of time, such as the total use time of the air conditioner, the time of the last defrosting process, temperature and humidity, etc.

[0057] Under normal circumstances, short-term time series and long-term time series do not need to be manually divided and labeled. The LSTM model can automatically remember and forget the impact of the characteristics of this factor on the degree of air conditioning frost.

[0058] Running the information vector x through the LSTM model outputs the result y:

[0059] {y1,y2,...,yn}=LSTM{x1,x2,..,xn}

[0060] Loss function:

[0061] Among them, y i represents the target value in a training. The prediction result of the representative model is expressed as the prediction of the frosting state. The model calibrated and recursively updated by the LSTM model is used as the user's personal model, which can more accurately detect the frosting state of the user's air conditioner.

[0062] The present invention can control the air conditioning system to call and update the LSTM model through the client APP on the mobile terminal (such as a mobile phone) communicating with the cloud platform. The call calculation and update of the LSTM model can be performed on the cloud platform, and the user obtains the calculation frosting state result of the LSTM model through the smart device APP. When the user receives the frost reminder and performs defrosting, the user feeds back the frosting processing status (such as frosted or not frosted, defrosted or not defrosted) on the APP to the cloud platform to update the LSTM model.

[0063] In order to clearly illustrate the technical solution of the present invention, the execution process of the defrost control method of the air conditioner provided by the present invention is described below with reference to a specific embodiment.

[0064] Figure 2 FIG. 1 is a schematic diagram of a specific embodiment of the air conditioner defrosting control method provided by the present invention. Figure 2 As shown, the air-conditioning operation status data is collected through the information collection module to determine whether the LSTM model has been updated. If not, the pre-trained model is used to predict the frosting status. If the LSTM model has been updated, the user's personal model is used to predict the frosting status. According to the model prediction result, it is determined whether defrosting is required. If necessary, defrosting is performed. After defrosting, the LSTM model is trained and updated. If defrosting is not required, the frosting status of the air conditioner is fed back.

[0065] Figure 3 1 shows a system structure diagram of an air conditioning system according to the present invention. Figure 3 As shown, the air conditioning system mainly includes a compressor 1, an indoor unit 2, an outdoor unit 3 (including an outdoor unit heat exchanger), an outdoor fan 4, a throttling device 5 and a sensor 6 (including a temperature sensor and a humidity sensor).

[0066] The invention also provides a defrosting control device for an air conditioner.

[0067] Figure 4 FIG. 1 is a structural block diagram of an embodiment of the defrost control device for an air conditioner provided by the present invention. Figure 4 As shown, the defrost control device 100 includes: a collection unit 110 , a prediction unit 120 and an update unit 130 .

[0068] The collecting unit 110 is used to collect the current running status data of the air conditioner, and obtain the time when the air conditioner last defrosted and the historical running time, as well as the running status data of the air conditioner within a first preset time in the past.

[0069] The current operating status data of the air conditioner may specifically include: current environmental parameters and the current operating parameters of the air conditioner; the operating status data of the air conditioner within the past first preset time may specifically include: environmental parameters within the past first preset time and the operating parameters of the air conditioner. The operating status data of the air conditioner is acquired once at a preset interval within the past first preset time.

[0070] The environmental parameters may specifically include: at least one of outdoor ambient temperature and outdoor ambient humidity. The operating parameters may specifically include: at least one of outdoor heat exchanger temperature (preferably outdoor heat exchanger surface temperature), indoor heat exchanger temperature, operating mode, outdoor fan speed and compressor power, and the last defrosting time and historical operating time of the air conditioner.

[0071] For example, the temperature and humidity data can be collected through corresponding sensors and corresponding temperature sensing packages, and the operation mode of the air conditioner, the speed of the outdoor fan and the power of the compressor can be obtained through the controller.

[0072] The prediction unit 120 is used to call a pre-trained frost state prediction model of the air conditioner, and use the collected current operating state data of the air conditioner and the operating state data of the air conditioner within a first preset time in the past to predict the frost state of the air conditioner.

[0073] The pre-trained air conditioner frost state prediction model may specifically include: a pre-trained model and a user personal model. When the current target user uses the air conditioner for the first time, the pre-trained model is used to predict the frost state of the air conditioner; when the current target user does not use the air conditioner for the first time, the user personal model of the current target user is used to predict the frost state of the air conditioner.

[0074] The pre-trained model is a frost state prediction model of the air conditioner pre-trained based on experimental test data. For example, an experimental test is performed in a laboratory environment to obtain the frost state of the air conditioner under different operating states, and the operating state data and frost state of the different operating states are recorded as sample data for model training, and the model is trained using the sample data to obtain a pre-trained model.

[0075] The user personal model is a frost state prediction model of the air conditioner trained based on the historical operating state data of the air conditioner, that is, a personal model of the current target user of the air conditioner. Specifically, the user personal model is obtained by updating the pre-trained model using the historical operating state data of the air conditioner.

[0076] In actual applications, the pre-trained model is used to predict the frosting state when the air conditioner is first run, and then the user's personal model is used, that is, the pre-trained model is updated using the operating status data of the air conditioner to form a user's personal model for the target user's personal use of the air conditioner. Each user can obtain his or her own personal LSTM model according to his or her ID number.

[0077] By calling the pre-trained frost state prediction model of the air conditioner, and inputting the collected current operating state data of the air conditioner and the operating state data of the air conditioner within the first preset time in the past into the frost state prediction model, the frost state of the air conditioner within the second preset time in the future can be predicted. That is, the collected current environmental parameters and the current operating parameters of the air conditioner, as well as the environmental parameters and the operating parameters of the air conditioner within the first preset time in the past are input into the pre-trained frost state prediction model of the air conditioner to predict the future frost state of the air conditioner. The frost state can specifically include: a frosted state and an unfrosted state. Optionally, it can also include an approaching frosted state.

[0078] The frosting state prediction model can be specifically an LSTM model (Long Short-Term Memory Artificial Neural Network Model). Therefore, the data for model training is time series data. The collected operating state data of the air conditioner at different times are processed into time series data and then used for model training.

[0079] The current environmental parameters, the operating parameters of the air conditioner, the time when the air conditioner was last defrosted and the historical operating time, and multiple groups of operating status data collected over a period of time in the past, including the environmental parameters, the operating parameters of the air conditioner, the time when the air conditioner was last defrosted and the historical operating time, are processed into an operating status sequence as the input of a frosting status prediction model to predict and output the frosting status of the air conditioner.

[0080] For example, the current vector xt and the operating state sequence ht-1 in the past period of time are used as input (when t=0, ht-1=0) to predict the frosting state ht of the air conditioner in the future period of time.

[0081] it=σ(Wi[ht-1,xt]+bi)

[0082] ft=σ(Wf[ht-1,xt]+bf)

[0083] ct=ftct-1+ittanh(Wc[ht-1,xt]+bc)

[0084] ot=σ(Wo[ht-1,xt]+bo)

[0085] ht=ottanh(ct)

[0086] Among them, mathematical σ is used as the standard sigmoid function; i, f, o and c are the input gate, forget gate, output gate and memory unit respectively; bi, bf, bo and bc are the bias vectors of the input gate, forget gate, output gate and memory unit respectively; W is the weight matrix between each unit and the gate vector.

[0087] Preferably, the frost state prediction model of the air conditioner can be stored in a cloud platform, and a call request is sent to the cloud platform to call the frost state prediction model of the air conditioner stored on the cloud platform. Specifically, a call request is sent to the cloud platform, and the cloud platform calls the model to predict the frost state of the air conditioner.

[0088] The updating unit 130 is used to update the frosting state prediction model according to the prediction result of the frosting state and the actual frosting state of the air conditioner.

[0089] If it is predicted that the frosting state of the air conditioner within the second time in the future is frosting, the air conditioner can be defrosted, or a reminder message for defrosting can be issued to remind the user to perform defrosting (for example, notifying the user to perform defrosting through a bound client APP). The defrosting process can switch the heating operation to the cooling operation, using the compressor as the heat source, so that the high-temperature gas refrigerant from the compressor flows into the outdoor heat exchanger for defrosting. According to the prediction result of the frosting state and the actual frosting state of the air conditioner, the frosting state prediction model is updated. Specifically, if the prediction result is inconsistent with the actual frosting state of the air conditioner, the frosting state model is updated. For example, if it is predicted that the frosting state of the air conditioner within the second time in the future is frosting, but it is judged that the air conditioner is not frosted based on the air conditioner operation status data collected within the second preset time, the frosting state prediction model is updated.

[0090] Among them, if the current target user is using the air conditioner for the first time, the historical operating status data of the air conditioner is used to update the pre-trained model to form a user personal model of the current target user. If the current target user is not using the air conditioner for the first time, the historical operating status data of the air conditioner is used to update the user personal model of the current target user.

[0091] That is, if the current user is using the air conditioner for the first time, the frost state prediction model used for predicting the frost state of the air conditioner is a pre-trained model, and the historical operating state data of the air conditioner (specifically, it may include the operating state data of the air conditioner collected from the last time the frost state prediction model was updated to the current moment) is used to update the pre-trained model. If the current user is not using the air conditioner for the first time, the frost state prediction model used for predicting the frost state of the air conditioner is a user personal model, and the historical operating state data of the air conditioner (specifically, it may include the operating state data of the air conditioner collected from the last time the frost state prediction model was updated to the current moment) is used to update the user personal model.

[0092] Specifically, the n operation information vectors x = (x1, x2, .., xn) after the last model training are used to form a data set (for example, outdoor ambient temperature, outdoor ambient humidity, outdoor heat exchanger temperature, indoor heat exchanger temperature, operation mode, last defrosting time and historical operation time), and the LSTM model is trained and updated to adapt to the air conditioner's own state and weather factors, and the impact of the air conditioner frost problem is divided based on the length of the time series:

[0093] Short-term time series: Changes in a short period of time will have a greater impact on the frosting state of the air conditioner, and the changed values ​​need to be memorized or forgotten in a short period of time, such as humidity, temperature, and air conditioner operation mode;

[0094] Long-term time series: Changes in values ​​over a long period of time will affect the frosting of the air conditioner. There is no need to memorize or forget the changes in a short period of time, such as the total use time of the air conditioner, the time of the last defrosting process, temperature and humidity, etc.

[0095] Under normal circumstances, short-term time series and long-term time series do not need to be manually divided and labeled. The LSTM model can automatically remember and forget the impact of the characteristics of this factor on the degree of air conditioning frost.

[0096] Running the information vector x through the LSTM model outputs the result y:

[0097] {y1,y2,...,yn}=LSTM{x1,x2,..,xn}

[0098] Loss function:

[0099] Among them, y i represents the target value in a training. The prediction result of the representative model is expressed as the prediction of the frosting state. The model calibrated and recursively updated by the LSTM model is used as the user's personal model, which can more accurately detect the frosting state of the user's air conditioner.

[0100] The present invention can control the air conditioning system to call and update the LSTM model through the client APP on the mobile terminal (such as a mobile phone) communicating with the cloud platform. The call calculation and update of the LSTM model can be performed on the cloud platform, and the user obtains the calculation frosting state result of the LSTM model through the smart device APP. When the user receives the frost reminder and performs defrosting, the user feeds back the frosting processing status (such as frosted or not frosted, defrosted or not defrosted) on the APP to the cloud platform to update the LSTM model.

[0101] The present invention also provides a storage medium corresponding to the defrost control method of the air conditioner, on which a computer program is stored, and when the program is executed by a processor, the steps of any of the above methods are implemented.

[0102] The present invention also provides an air conditioner corresponding to the defrost control method of the air conditioner, comprising a processor, a memory, and a computer program stored in the memory and executable on the processor, wherein the steps of any of the aforementioned methods are implemented when the processor executes the program.

[0103] The present invention also provides an air conditioner corresponding to the defrost control device of the air conditioner, comprising any of the defrost control devices of the air conditioner described above.

[0104] The present invention also provides a computer program product corresponding to the defrost control method of the air conditioner, comprising a computer program, and when the computer program is executed by a processor, the steps of any of the aforementioned methods are implemented.

[0105] Based on this, the solution provided by the present invention collects the operating data of the air conditioner and calls the frost state prediction model to predict the frost state of the air conditioner, and automatically updates the LSTM model according to the operating conditions of the air conditioner, thereby reducing the impact of air conditioner aging and environmental factors on the detection results and improving the accuracy of the detection.

[0106] According to the technical solution of the present invention, the frost state prediction model can be updated based on the operating status data of the user's personal air conditioner as the air conditioner is used, so as to adapt to the air conditioner's own state and avoid inaccurate detection of the frost state due to environmental factors and air conditioner aging problems.

[0107] The functions described herein may be implemented in hardware, software executed by a processor, firmware, or any combination thereof. If implemented in software executed by a processor, the functions may be stored as one or more instructions or codes on a computer-readable medium or transmitted via a computer-readable medium. Other examples and implementations are within the scope and spirit of the present invention and the appended claims. For example, due to the nature of software, the functions described above may be implemented using software executed by a processor, hardware, firmware, hard wiring, or a combination of any of these. In addition, each functional unit may be integrated into a processing unit, each unit may exist physically separately, or two or more units may be integrated into one unit.

[0108] In the several embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways. Among them, the device embodiments described above are only schematic. For example, the division of the units can be a logical function division. There may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of units or modules, which can be electrical or other forms.

[0109] The units described as separate components may or may not be physically separated, and the components of the control device may or may not be physical units, that is, they may be located in one place or distributed in multiple units. Some or all of the units may be selected according to actual needs to achieve the purpose of the present embodiment.

[0110] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the relevant technology or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions to enable a computer device (which can be a personal computer, a server or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), mobile hard disk, disk or optical disk and other media that can store program codes.

[0111] The above description is only an embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent substitution, improvement, etc. made within the spirit and principle of the present invention shall be included in the scope of the claims of the present invention.

Claims

1. A defrosting control method for an air conditioner, characterized in that: include: Collecting the current operating status data of the air conditioner, and obtaining the last defrosting time and historical operating time of the air conditioner and the operating status data of the air conditioner within a preset time in the past; Calling a pre-trained frost state prediction model of the air conditioner, and using the collected current operating state data of the air conditioner and the operating state data of the air conditioner within a first preset time in the past to predict the frost state of the air conditioner; The frosting state prediction model is updated according to the prediction result of the frosting state and the actual frosting state of the air conditioner.

2. The method according to claim 1, characterized in that Pre-trained air conditioner frost state prediction model, including: pre-trained model and user personal model; The pre-trained model is a frost state prediction model of the air conditioner pre-trained based on experimental test data; and / or, The user personal model is a frosting state prediction model of the air conditioner trained based on historical operating state data of the air conditioner; Among them, when the current target user uses the air conditioner for the first time, the pre-trained model is used to predict the frosting state of the air conditioner. When the current target user does not use the air conditioner for the first time, the user personal model of the current target user is used to predict the frosting state of the air conditioner.

3. The method according to claim 2, characterized in that Updating the frosting state prediction model includes: If the current target user is using the air conditioner for the first time, the pre-trained model is updated using the historical operating status data of the air conditioner to form a user personal model of the current target user; If the current target user is not using the air conditioner for the first time, the historical operating status data of the air conditioner is used to update the user personal model of the current target user.

4. The method according to any one of claims 1 to 3, characterized in that: The method includes updating the frosting state prediction model according to the prediction result of the frosting state and the actual frosting state of the air conditioner, including: If the prediction result is inconsistent with the actual frosting state of the air conditioner, the frosting state model is updated.

5. The method according to any one of claims 1 to 3, characterized in that: Calling a pre-trained frost state prediction model of the air conditioner, including: A call request is issued to the cloud platform to call the frost state prediction model of the air conditioner stored on the cloud platform.

6. A defrost control device for an air conditioner, characterized in that: include: A collection unit, used to collect the current running status data of the air conditioner, and obtain the time when the air conditioner last defrosted and the historical running time, as well as the running status data of the air conditioner within a first preset time in the past; A prediction unit, configured to call a pre-trained prediction model for the frosting state of the air conditioner, and predict the frosting state of the air conditioner by using the collected current operating state data of the air conditioner and the operating state data of the air conditioner acquired within a first preset time in the past; An updating unit is used to update the frosting state prediction model according to the prediction result of the frosting state and the actual frosting state of the air conditioner.

7. The device according to claim 6, characterized in that Pre-trained air conditioner frost state prediction model, including: pre-trained model and user personal model; The pre-trained model is a frost state prediction model of the air conditioner pre-trained based on experimental test data; and / or, The user personal model is a frosting state prediction model of the air conditioner trained based on historical operating state data of the air conditioner; Among them, when the current target user uses the air conditioner for the first time, the pre-trained model is used to predict the frosting state of the air conditioner. When the current target user does not use the air conditioner for the first time, the user personal model of the current target user is used to predict the frosting state of the air conditioner.

8. A storage medium, characterized in that: A computer program is stored thereon, and when the program is executed by a processor, the steps of any method described in claims 1-5 are implemented.

9. An air conditioner, characterized in that: The method comprises a processor, a memory and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of any method described in claims 1-5 when executing the program, or comprises a defrost control device described in any one of claims 6-7.

10. A computer program product, characterized in that The invention comprises a computer program, which implements the steps of any one of the methods of claims 1 to 5 when executed by a processor.

Images