EP4663835A1

EP4663835A1 - Method and apparatus for measuring a moisture level of fabric in fabric processing appliance, fabric processing appliance, and computer program product

Info

Publication number: EP4663835A1
Application number: EP25181517.1A
Authority: EP
Inventors: Zhigang Sun; Li Zhang; Chichen BAO; Lei BAO; Hui Kuang; Xuli CUI; Lijun Zhang; Maoyin Li
Original assignee: BSH Hausgeraete GmbH
Current assignee: BSH Hausgeraete GmbH
Priority date: 2024-06-12
Filing date: 2025-06-06
Publication date: 2025-12-17
Also published as: CN121110319A

Abstract

Embodiments of this application provide a method and an apparatus for measuring a moisture level of a fabric in a fabric processing appliance. The method includes the following steps: collecting, during a drying operation performed by a fabric processing appliance, an air temperature parameter related to air circulating in the fabric processing appliance during the drying operation and a drain characteristic parameter related to liquid water extracted from a fabric in the fabric processing appliance through the drying operation that are of the fabric processing appliance; and determining a moisture level of the fabric in the fabric processing appliance based on at least the air temperature parameter and the drain characteristic parameter. This application further relates to a fabric processing appliance and a computer program product.

Description

TECHNICAL FIELD

This application relates to the field of household appliances, and in particular, to a method for measuring a moisture level of a fabric in a fabric processing appliance, an apparatus for measuring a moisture level of a fabric in a fabric processing appliance, a fabric processing appliance including the apparatus according to this application, and a computer program product configured to at least assistantly implement steps of the method according to this application.

BACKGROUND

During running of a fabric processing appliance with a drying function (such as a clothes dryer or a washing and drying machine), a moisture level of a fabric in the fabric processing appliance is an important control factor for adjusting drying time. Currently, in a market, a dryer that can automatically adjust drying time usually monitors a change of a water content of a fabric in real time by using a moisture sensor arranged in a drum. In this solution, the moisture level is measured mainly in dependence on contact between a measurement electrode of the moisture sensor and the fabric. However, because the moisture sensor cannot be directly placed on the fabric, the water content in the fabric is difficult to be sensed, and a conductive metal material (such as a zipper) carried by the fabric or an excessively small amount of fabric causes a deviation in a reading of the moisture sensor. Therefore, a moisture sensing result obtained by using the moisture sensor cannot accurately reflect an actual moisture condition of the fabric. In addition, such a moisture sensor has high costs and is difficult to maintain, which further restricts application of the moisture sensor to the fabric processing appliance.
Therefore, how to measure the moisture level of the fabric in the fabric processing appliance accurately and at low costs becomes a technical problem that needs to be resolved currently.

SUMMARY

An objective of this application is to provide a method for measuring a moisture level of a fabric in a fabric processing appliance, an apparatus for measuring a moisture level of a fabric in a fabric processing appliance, a fabric processing appliance including the apparatus according to this application, and a computer program product, to solve the problem in the prior art.
According to a first aspect of this application, a method for measuring a moisture level of a fabric in a fabric processing appliance is provided. The fabric processing appliance is configured to perform a drying operation on a fabric, and the method includes the following steps:

collecting an air temperature parameter and a drain characteristic parameter of the fabric processing appliance during the drying operation performed by the fabric processing appliance, where the air temperature parameter is related to air circulating in the fabric processing appliance during the drying operation, and the drain characteristic parameter is related to liquid water extracted from the fabric in the fabric processing appliance through the drying operation; and
determining a moisture level of the fabric in the fabric processing appliance based on at least the air temperature parameter and the drain characteristic parameter.

A core concept of this application lies in: Currently, it is recognized that limited by a mounting position of a sensor and a contact detection principle, it is very difficult to accurately reflect an actual moisture level of the fabric by using detected ambient moisture information in a drum. In this application, by analyzing the air temperature parameter related to the air circulating in the fabric processing appliance and the drain characteristic parameter related to the liquid water extracted from the fabric that are collected during the drying operation, a real-time moisture level of the fabric can be determined without being affected by a piled form and moisture distribution of the fabric, which is beneficial to improving accuracy of measuring the moisture level of the fabric, and avoiding using a moisture sensor with high costs and limited measurement conditions, thereby achieving a technical objective of measuring the moisture level of the fabric in the fabric processing appliance with high precision and low costs.
Optionally, the air temperature parameter may include an air temperature of at least one position from an outlet of a drum to an inlet of a dehumidifier of an air circulation loop of the fabric processing appliance during the drying operation and/or a time change characteristic parameter and the like of the air temperature. The drain characteristic parameter may include a mass flow, a volume flow, and a flow rate of the liquid water extracted from the fabric in the fabric processing appliance in a drain pipe of the fabric processing appliance, and/or time change characteristic parameters and the like of the mass flow, the volume flow, and the flow rate. These air temperature parameters and drain characteristic parameters are introduced as influencing factors for determining the moisture level of the fabric, so that the real-time moisture level of the fabric can be determined more accurately, and a foundation is laid for accurate control of a drying process of the fabric processing appliance.
Optionally, water pressure information of a determined position in the drain pipe of the fabric processing appliance is measured by using a pressure sensor, and the drain characteristic parameter is obtained based on the water pressure information. Therefore, a low-cost pressure sensor may be used to replace an expensive flow meter to measure the drain characteristic parameter, so that hardware overheads are saved, and measurement accuracy can be ensured even when an amount of extracted water is small or the flow rate is slow.
Optionally, the determined position includes at least a section of the drain pipe of the fabric processing appliance that extends along a vertical direction. This helps convert an amount of extracted liquid water into a pressure or a height for measurement, thereby ensuring measurement accuracy even when the amount of extracted water is small or a drain flow rate is slow.
Optionally, the moisture level of the fabric in the fabric processing appliance may be determined based on at least the air temperature parameter and the drain characteristic parameter by using a trained first machine learning model. The first machine learning model includes, for example, a distributed gradient boosting library model and/or a support vector machine model. By introducing targeted analysis of machine learning models, the real-time moisture level of the fabric can be inferred more accurately and quickly.
Optionally, the first machine learning model may be trained in the following manner: collecting first sample data during the drying operation performed by the fabric processing appliance on the fabric in a historical time period, where the first sample data includes the air temperature parameter and the drain characteristic parameter; marking the first sample data in terms of the moisture level of the fabric; and adjusting a parameter of the first machine learning model based on marked first sample data until a performance evaluation indicator of the first machine learning model satisfies a preset condition and/or until a preset quantity of training steps is reached. Therefore, a large amount of accumulated prior knowledge can be fully used, so that the first machine learning model self-learns relationships among the air temperature parameter, the drain characteristic parameter, and the moisture level of the fabric, thereby improving training efficiency, and optimizing a model generalization capability.
Optionally, the time change characteristic parameter of the air temperature of the at least one position from the outlet of the drum to the inlet of the dehumidifier during the drying operation may be collected and determined, and the moisture level of the fabric in the fabric processing appliance is determined with reference to the time change characteristic parameter of the air temperature. It is considered that an initial weight of the fabric can be determined based on the time change characteristic parameter of the air temperature, and the initial weight of the fabric not only can represent a load weight of a clothes dryer, but also is closely correlated to a time change process of the moisture level of the fabric, the time change characteristic parameter of the air temperature is introduced as one of important influencing factors for determining the moisture level of the fabric in the fabric processing appliance, so that accuracy of evaluating the real-time moisture level of the fabric can be effectively improved.
Optionally, a step of determining the time change characteristic parameter of the air temperature may include: determining whether the air temperature of the at least one position from the outlet of the drum to the inlet of the dehumidifier reaches a first temperature threshold T₁; determining, if the air temperature reaches the first temperature threshold T₁, whether first duration Δt₁ through which the air temperature reaches the first temperature threshold T₁ exceeds a preset time threshold t_w; determining, if the first duration Δt₁ does not exceed the preset time threshold t_w, the first duration Δt₁ as the time change characteristic parameter of the air temperature; and determining, if the first duration Δt₁ exceeds the preset time threshold t_w, second duration Δt₂ through which the air temperature reaches a second temperature threshold T₂ as the time change characteristic parameter of the air temperature, where the second temperature threshold T₂ is greater than the first temperature threshold T₁. Whether a weight of the fabric falls within a light-load operating range or a heavy-load operating range can be distinguished through a duration through which the air temperature reaches a temperature threshold, to avoid a problem that the fabric is excessively dry in the drying process, thereby effectively protecting the fabric. By adjusting a value of the temperature threshold, a duration through which the fabric reaches the temperature threshold may also be adjusted, thereby more precisely determining the initial weight of the fabric, and laying the foundation for optimizing subsequent fabric moisture determination and timing information obtaining.
Optionally, the method may further include: obtaining, based on at least the determined moisture level of the fabric and/or the determined time change characteristic parameter of the air temperature, timing information for enabling the fabric in the fabric processing appliance to reach a predetermined drying degree. Therefore, precisely controlling the drying process according to real-time parameters collected in the drying process helps to improve accuracy of estimating clothes drying time, thereby improving user experience.
Optionally, the timing information for enabling the fabric in the fabric processing appliance to reach the predetermined drying degree may be obtained based on at least the determined moisture level of the fabric and/or the determined time change characteristic parameter of the air temperature by using a trained second machine learning model. The second machine learning model includes, for example, a distributed gradient boosting library model and/or a support vector machine model. By introducing targeted analysis of machine learning models, the timing information can be more reliably inferred.
Optionally, the second machine learning model may be trained in the following manner: collecting second sample data during the drying operation performed by the fabric processing appliance on the fabric in a historical time period, where the second sample data includes the moisture level of the fabric and/or the time change characteristic parameter of the air temperature; marking the second sample data in terms of the timing information for enabling the fabric to reach the predetermined drying degree; and adjusting a parameter of the second machine learning model based on marked second sample data until a performance evaluation indicator of the second machine learning model satisfies a preset condition and/or until a preset quantity of training steps is reached. Therefore, a large amount of accumulated prior knowledge can be fully used, so that the second machine learning model self-learns a relationship between a fabric moisture level and/or the time change characteristic parameter of the air temperature and an end timing of a drying program, thereby improving training efficiency and optimizing a model generalization capability.
Optionally, the second machine learning model is retrained based on at least feedback behavior information of the timing information obtained by a user from the trained second machine learning model. Therefore, based on a basic model framework established according to experience or experimental data, an initial model may be further optimized according to a specific operation habit and a drying degree preference of a user, to accelerate a convergence process of a training algorithm and ensure to satisfy a personalized user requirement.
Optionally, the method may further include: controlling running of the fabric processing appliance based on the obtained timing information, where the fabric processing appliance is controlled to terminate the drying operation when the timing information is satisfied, and/or the fabric processing appliance is controlled to display the determined moisture level of the fabric, the collected air temperature parameter, estimated drying time and/or remaining drying time needed by the fabric, and the like to reach the predetermined drying degree. Therefore, the drying process can be dynamically adjusted according to a change of the moisture level of the fabric during the drying operation, thereby improving energy efficiency. In addition, time is displayed to help the user to know a drying progress, thereby improving user satisfaction.
According to a second aspect of this application, an apparatus for measuring the moisture level of a fabric in a fabric processing appliance is provided. The apparatus may include the following components:

air temperature collecting modules, configured to collect an air temperature parameter of the fabric processing appliance during a drying operation performed by the fabric processing appliance, where the air temperature parameter is related to air circulating in the fabric processing appliance during the drying operation;
a drain characteristic collecting module, configured to collect a drain characteristic parameter of the fabric processing appliance during the drying operation performed by the fabric processing appliance, where the drain characteristic parameter is related to liquid water extracted from a fabric in the fabric processing appliance through the drying operation; and
a control module, configured to perform the method according to this application.

According to a third aspect of this application, a fabric processing appliance is provided. The fabric processing appliance may include:

an air circulation loop, air in the fabric processing appliance circulating along the air circulation loop during a drying operation;
a drain pipe, liquid water extracted from a fabric in the fabric processing appliance through the drying operation being led to an external environment along the drain pipe; and
the apparatus according to this application.

Optionally, the air temperature collecting modules may include a first temperature sensor arranged at the outlet of the drum in the air circulation loop of the fabric processing appliance and/or a second temperature sensor arranged at the inlet of the dehumidifier in the air circulation loop of the fabric processing appliance.
Optionally, the drain characteristic collecting module may include a pressure sensor, and the pressure sensor is arranged in the drain pipe and is configured to measure the water pressure information of the determined position in the drain pipe, where the drain characteristic parameter is obtained based on the water pressure information.
According to a fourth aspect of this application, a computer program product is provided, such as a computer readable program carrier, including computer program instructions, the computer program instructions, when executed by a processor, at least assistantly implementing steps of the method according to this application.

BRIEF DESCRIPTION OF THE DRAWINGS

Principles, features, and advantages of this application can be better understood through descriptions of this application in more detail below with reference to accompanying drawings. The accompanying drawings include:

FIG. 1: is a schematic structural diagram of a fabric processing appliance according to an exemplary embodiment of this application;
FIG. 2: is a schematic block diagram of an apparatus for measuringthe moisture level of a fabric in a fabric processing appliance according to an exemplary embodiment of this application;
FIG. 3: is a flowchart of a method for measuring the moisture level of a fabric in a fabric processing appliance according to an exemplary embodiment of this application;
FIG. 4: is a curve graph of an air temperature changing with time during drying of fabrics of different initial weights;
FIG. 5: is a flowchart of step S1 according to another exemplary embodiment of this application;
FIG. 6: is a flowchart of step S1 according to another exemplary embodiment of this application;
FIG. 7: is a schematic principle diagram of determining a level of a fabric in a fabric processing appliance by using a first machine learning model;
FIG. 8: is a flowchart of a method for measuring the moisture level of a fabric in a fabric processing appliance according to another exemplary embodiment of this application;
FIG. 9: is a flowchart of a method for measuring the moisture level of a fabric in a fabric processing appliance according to another exemplary embodiment of this application;
FIG. 10: is a schematic principle diagram of obtaining, by using a second machine learning model, timing information for enabling a fabric in a fabric processing appliance to reach a predetermined drying degree;
FIG. 11: is a flowchart of a method for measuring the moisture level of a fabric in a fabric processing appliance according to another exemplary embodiment of this application; and
FIG. 12: is a flowchart of a method for measuring the moisture level of a fabric in a fabric processing appliance according to another exemplary embodiment of this application.

DETAILED DESCRIPTION

To make the technical problem to be resolved by this application, the technical solutions, and the beneficial technical effects of this application clearer and more comprehensible, the following further describes this application in detail with reference to accompanying drawings and a plurality of exemplary embodiments. It should be understood that the specific embodiments described herein are only intended to explain this application and are not intended to limit the protection scope of this application.
FIG. 1 is a schematic structural diagram of a fabric processing appliance 1 according to an exemplary embodiment of this application. The fabric processing appliance 1 may be a clothes dryer of various types having only a drying function, or may be a washing and drying machine. FIG. 1 exemplarily shows a heat pump clothes dryer. The heat pump clothes dryer includes, for example, a drum 2 configured to accommodate a load, a motor 4, a heat pump system 6, an air channel 5 configured to guide circulating air, and a fan unit 3 configured to cause the circulating air to directionally flow in the air channel 5. The drum 2 and the fan unit 3 are coaxially driven by the motor 4. During a drying program of the fabric processing appliance 1, the drum 2 is driven by the motor 4 to perform a rotational drying operation, and the fan unit 3 is driven by the motor 4 to provide flow power for the circulating air, so that dry air in the air channel 5 can enter the drum 2, and humid air discharged from the drum 2 is sent into the air channel 5 again.
The heat pump system 6 of the heat pump clothes dryer shown in FIG. 1 includes, for example, an evaporator 7, a heat pump compressor 12, a condenser 8, and an expansion valve 13 that are sequentially connected. In the heat pump system 6, a refrigerant (coolant) is compressed from a low-pressure/low-temperature gas state to a high-pressure/high-temperature gas state by the heat pump compressor 12. At the high pressure, the refrigerant condenses into a liquid in the condenser 8 at or near a boiling point of the refrigerant at this pressure to release heat into a surrounding air flow, thereby heating the air flow. Then, the refrigerant reaches the expansion valve 13 and expands to a low-pressure level at the expansion valve 13, thereby achieving a two-phase state. Finally, the refrigerant cools the air flow in the evaporator 7, causing water vapor in the air flow to condense, absorb heat from the air flow, and reach the gas state again, thereby realizing cyclic evaporation of the refrigerant.
During the drying operation, the air in the fabric processing appliance 1 circulates along an air circulation loop of the fabric processing appliance 1. As shown in FIG. 1, the air circulation loop of the heat pump clothes dryer may include the drum 2, the air channel 5, the fan unit 3, the evaporator 7, the heat pump compressor 12, the expansion valve 13, and the condenser 8 that are sequentially connected. During the drying operation, warm and humid air flowing out from an outlet of the drum 2 enters the evaporator 7 under guiding of the air channel 5, and is therefore cooled and separated into dry and cold air. After entering the condenser 8, the dry and cold air is heated to become dry and hot air, then is blown into the drum 2 by the fan unit 3, absorbs moisture when passing through a wet fabric load to become warm and humid air, and finally, enters the evaporator 7 again, thereby achieving a drying effect of a fabric.
Considering that a moisture level of the fabric in the fabric processing appliance 1 is an important controlling factor in a drying process, an apparatus 20 for measuring the moisture level of a fabric in a fabric processing appliance 1 may be arranged in the fabric processing appliance 1. FIG. 2 is a schematic block diagram of the apparatus 20 according to an exemplary embodiment of this application. The apparatus 20 may include air temperature collecting modules 211 and 212, a drain characteristic collecting module 22, and a control module 23.
The air temperature collecting modules of the apparatus 20 are configured to collect an air temperature parameter of the fabric processing appliance 1 during the drying operation performed by the fabric processing appliance 1, where the air temperature parameter is related to air circulating in the fabric processing appliance 1 during the drying operation. For example, the air temperature collecting modules include a first temperature sensor 211 arranged at the outlet of the drum 2 in the air circulation loop of the fabric processing appliance 1, and/or a second temperature sensor 212 arranged at an inlet of a dehumidifier 7 in the air circulation loop of the fabric processing appliance 1, to collect the air temperature parameter of at least one position from the outlet of the drum 2 to the inlet of the dehumidifier 7 during the drying operation.
The control module 23 of the apparatus 20 may be connected to the first temperature sensor 211 and/or the second temperature sensor 212, and obtain, based on the air temperature parameter collected by the temperature sensor, not only an air temperature of the at least one position from the outlet of the drum 2 to the inlet of the dehumidifier 7, but also a time change characteristic parameter of the air temperature of the at least one position, for example, duration through which the air temperature reaches a preset temperature threshold, or a rate of change of the air temperature with time.
It should be noted that the first temperature sensor 211 and the second temperature sensor 212 shown in FIG. 1 are merely examples. It may also be expected that one or more temperature sensors may also be arranged at other positions from the outlet of the drum 2 to the inlet of the dehumidifier 7. Types, a quantity, and mounting positions of the temperature sensors are not intended to be limited herein.
In the heat pump clothes dryer shown in FIG. 1, the dehumidifier configured to condense the water vapor in the circulating air flow is the evaporator 7. For other types of clothes dryers not shown, such as water-cooled clothes dryers or air-cooled clothes dryers that use a condensation drying principle, the dehumidifier configured to condense the water vapor in the circulating air flow is a water-cooled heat exchanger or an air-cooled heat exchanger, which is also referred to as a condenser.
In addition, the fabric processing appliance 1 further includes a water collector 91, a drain pipe 9, a drain pump 10, and a pressure sensor 221 that are arranged below the evaporator 7. Condensed water formed by converting through condensation at the evaporator 7 enters, in a form of water drops, the water collector 91 arranged below the evaporator 7, and is collected at the water collector 91. The water collector 91 is, for example, a part of the drain pipe 9 configured to guide the condensed water and extends in a horizontal direction. The water collector 91 is connected to one end of the drain pump 10 through a section 92 of the drain pipe 9 extending in a vertical direction. The drain pump 10 is, for example, configured as a blade-type electrically-driven pumping apparatus. The other end of the drain pump 10 is connected to an inlet side of a sewer, so that under an action of the drain pump 10, the liquid water separated from the fabric in the fabric processing appliance 1 through the drying operation can be regularly discharged to an external environment along the drain pipe 9.
The drain characteristic collecting module 22 of the apparatus 20 is configured to collect, during the drying operation performed by the fabric processing appliance 1, a drain characteristic parameter of the fabric processing appliance 1, where the drain characteristic parameter is related to the liquid water extracted from the fabric in the fabric processing appliance 1 through the drying operation. The drain characteristic collecting module 22, for example, includes the pressure sensor 221 arranged in the section 92 of the drain pipe 9 extending in the vertical direction. Water pressure information of a determined position in the drain pipe 9 can be measured by using the pressure sensor 221. The "determined position" may refer to, for example, a bottom of the section 92 (which is briefly referred to as a vertical section 92 for ease of description below) of the drain pipe extending in the vertical direction, but may alternatively refer to a plurality of positions arranged at intervals along the vertical section 92. It should be noted that although only one pressure sensor 221 is shown at the bottom of the vertical section 92 of the drain pipe 9 with reference to FIG. 1, it may also be expected that one or more pressure sensors are arranged at other positions in the drain pipe 9. Types, a quantity, and mounting positions of the pressure sensors are not intended to be limited herein.
Optionally, it may be further considered that other sensors (not shown in FIG. 1) such as a flow rate meter, a flow meter, and/or a liquid level sensor is arranged in the drain pipe 9, to regularly detect a mass flow, a volume flow, a flow rate, and/or a liquid level value of the liquid water extracted from the fabric.
As shown in FIG. 1, the control module 23 of the apparatus 20 may be connected to the pressure sensor 221. The control module 23 can obtain, based on the water pressure information measured by the pressure sensor 221, the drain characteristic parameter related to the liquid water extracted from the fabric through the drying operation. The drain characteristic parameter includes, for example, the mass flow, the volume flow, and the flow rate of the liquid water extracted from the fabric in the fabric processing appliance 1 in the drain pipe 9 of the fabric processing appliance 1, and/or time change characteristic parameters of the mass flow, the volume flow, and the flow rate. The time change characteristic parameter is used for representing a change process of the mass flow, the volume flow, and/or the flow rate with time. Optionally, the control module 23 of the apparatus 20 may alternatively be connected to another sensor, such as the flow rate meter, the flow meter, and/or the liquid level sensor not shown in FIG. 1.
A first machine learning model 231 may be set in the control module 23, and the first machine learning model 231 is configured to determine the moisture level of the fabric in the fabric processing appliance 1 based on at least the air temperature parameter and the drain characteristic parameter. A second machine learning model 232 may further be set in the control module 23, and the second machine learning model 232 is configured to obtain, based on at least the determined moisture level of the fabric and/or the determined time change characteristic parameter of the air temperature, timing information for enabling the fabric in the fabric processing appliance 1 to reach a predetermined drying degree.
In addition, the control module 23 may be further configured to control running of the fabric processing appliance 1 based on the obtained timing information. As shown in FIG. 1, the control module 23 may be connected to the motor 4, to output an instruction about starting and stopping of the motor 4 to the motor 4. In a case that it is determined by using the control module 23 that the timing information is satisfied, the instruction sent by the control module 23 may stop running of the motor 4, to terminate the drying operation.
Optionally, the control module 23 may further be connected to a display 11 of the fabric processing appliance 1. The display 11 can not only display the determined moisture level of the fabric and/or the collected air temperature parameter, but also display estimated drying time and/or remaining drying time of the fabric processing appliance 1.
It should be further noted that, although the drying process of the fabric processing appliance is introduced by using a heat pump drying principle as an example in FIG. 1, the fabric processing appliance in this application can also complete the drying operation by using the condensation drying principle (such as the water-cooled clothes dryer or the air-cooled clothes dryer) or in other possible heat exchange manners, and implementation of an operation control solution involved in the specification is not limited to a specific type of drying principle.
FIG. 3 is a flowchart of a method for measuring the moisture level of a fabric in a fabric processing appliance according to an exemplary embodiment of this application. The method exemplarily includes steps S1 and S2, and may be implemented, for example, when the apparatus 20 shown in FIG. 2 is used.
In step S1, the air temperature parameter and the drain characteristic parameter of the fabric processing appliance 1 are collected during the drying operation performed by the fabric processing appliance 1, where the air temperature parameter is related to the air circulating in the fabric processing appliance 1 during the drying operation, and the drain characteristic parameter is related to the liquid water extracted from the fabric in the fabric processing appliance 1 through the drying operation.
The temperature sensor cannot be directly placed inside the to-be-dried fabric, in a case that there are many to-be-dried fabrics, uneven temperature distribution easily occurs in the fabrics, and a temperature of the air circulating in the fabric processing appliance 1 has a certain correlation with a temperature of the fabric during the drying operation. Therefore, in this embodiment of this application, the air temperature parameter related to the circulating air is collected to replace the temperature of the fabric. As shown in FIG. 1, the air circulation loop of the heat pump clothes dryer may include the drum 2, the air channel 5, the fan unit 3, the evaporator 7, the heat pump compressor 12, the expansion valve 13, and the condenser 8 that are sequentially connected.
It is considered that there is a certain correlation between a temperature of the warm and humid air passing through the fabric in the drum 2 and flowing out from the drum 2 and a temperature of the fabric, and after entering the dehumidifier 7 (namely, the evaporator 7 of the heat pump clothes dryer in FIG. 1) under guiding of the air channel 5, the warm and humid air is cooled and separated into the dry and cold air. Therefore, in this embodiment of this application, an air temperature of at least one position from the outlet of the drum 2 to the inlet of the dehumidifier 7 in the air circulation loop of the fabric processing appliance 1 during the drying operation is collected as the air temperature parameter. The air temperature may be, for example, collected by using the first temperature sensor 211 arranged at the outlet of the drum 2 of the fabric processing appliance 1 and/or the second temperature sensor 212 arranged at the inlet of the dehumidifier 7 of the fabric processing appliance 1.
In another embodiment, the air temperature parameter may further include the time change characteristic parameter of the air temperature of the at least one position from the outlet of the drum 2 to the inlet of the dehumidifier 7 in the air circulation loop of the fabric processing appliance 1 during the drying operation. FIG. 4 is a curve graph of an air temperature changing with time during drying of fabrics of different initial weights. As shown in FIG. 4, in a process of drying fabrics of different initial weights w₁, w₂, w₃, and w₄, time change trends of air temperatures at the at least one position from the outlet of the drum 2 to the inlet of the dehumidifier 7 are approximately the same, where w₁ < w₂ < w₃ < w₄. The initial weight of the fabric especially refers to a weight of the fabric in a dry state, that is, a water content of the fabric is zero, which can represent a load degree of the clothes dryer. It may be observed from FIG. 4 that, a greater initial weight of the fabric corresponds to a gentler time change process of the air temperature, that is, a slower rising speed.
Duration through which the air temperatures of the fabrics of different initial weights w₁, w₂, w₃, and w₄ reach a certain temperature (for example, a first temperature threshold T₁ or a second temperature threshold T₂) is different, and final air temperatures of the fabrics of different initial weights w₁, w₂, w₃, and w₄ when the dry state (that is, the moisture level is zero) is reached are also different. Therefore, the time change characteristic parameter of the air temperature may be used as one of important influencing factors for determining the moisture level of the fabric in the fabric processing appliance 1.
The time change characteristic parameter of the air temperature may be determined based on the collected air temperature of the at least one position from the outlet of the drum 2 to the inlet of the dehumidifier 7. A process of determining the time change characteristic parameter of the air temperature is described in detail below with reference to a flowchart of step S1 shown in FIG. 5 according to another exemplary embodiment of this application, and the process may include steps S101 to S105.
First, in step S101, whether a preheat stage of the fabric processing appliance 1 ends is determined. In the preheat stage, the heat pump system 6 is turned on and operating parameters of components of the heat pump system 6 gradually reach stability. To make the collected air temperature parameter more intuitively reflect a progress of a formal drying stage, the air temperature parameter is not collected in the preheat stage, or the air temperature parameter collected in the preheat stage is not used for subsequent fabric moisture level determining.
If it is determined in step S101 that the preheat phase does not end, such determining continues to be performed in step S101. If it is determined in step S101 that the preheat stage ends, that is, a starting point corresponding to an initial drying moment t₀ and an initial drying temperature T₀ in FIG. 4, a fabric drying stage starts to be entered.
In step S102, whether the air temperature of the at least one position from the outlet of the drum 2 to the inlet of the dehumidifier 7 reaches the first temperature threshold T₁ is determined. As shown in FIG. 4, in an early stage of the drying process of the fabrics of different initial weights w₁, w₂, w₃, and w₄, air temperature change curves corresponding to the fabrics of different initial weights have a large overlapping area. Therefore, it is difficult to determine the initial weights of the fabrics by using the time change characteristic parameter of the air temperature in the early stage of the drying process. The first temperature threshold T₁ (for example, 36 °C) may be set. In a case that the fabrics of different initial weights w₁, w₂, w₃, and w₄ are heated to the first temperature threshold T₁, there is no overlapping area between the air temperature change curves corresponding to the fabrics of different initial weights w₁, w₂, w₃, and w₄, as shown in FIG. 4, so that the initial weights of the fabrics may be determined based on the duration through which the air temperature reaches the temperature threshold.
If the air temperature reaches the first temperature threshold T₁, in step S103, whether first duration Δt₁ through which the air temperature reaches the first temperature threshold T₁ exceeds a preset time threshold t_w is determined, where the first duration Δt₁ is a time period from the initial drying moment t₀ to a moment t₁ at which the air temperature reaches the first temperature threshold T₁. By setting the preset time threshold t_w, it may be identified whether an initial weight of a current to-be-dried fabric falls within a light-load operating range or a heavy-load operating range.
If the first duration Δt₁ does not exceed the preset time threshold t_w, that is, the weight (for example, w₁, w₂, or w₃) of the fabric falls within the light-load operating range, the first duration Δt₁ is determined as the time change characteristic parameter of the air temperature in step S104, so that a light-load operating mode of the fabric processing appliance 1 may be determined within the first duration Δt₁, to avoid a problem that the fabric is excessively dry during the drying process, thereby effectively protecting the fabric.
If the first duration Δt₁ exceeds the preset time threshold t_w, that is, the weight (for example, w₄) of the fabric falls within the heavy-load operating range, second duration Δt₂ through which the air temperature reaches the second temperature threshold T₂ is determined as the time change characteristic parameter of the air temperature in step S105, where the second temperature threshold T₂ (for example, 44 °C) is greater than the first temperature threshold T₁, and the second duration Δt₂ is a time period from the initial drying moment t₀ to a moment t₂ at which the air temperature reaches the second temperature threshold T₂.
In a case that the fabrics of different initial weights w₁, w₂, w₃, and w₄ are heated to the second temperature threshold T₂, there is also no overlapping area between the air temperature change curves corresponding to the fabrics of different initial weights w₁, w₂, w₃, and w₄ as shown in FIG. 4, so that the initial weights of the fabrics may be determined based on the duration through which the air temperature reaches the temperature threshold. Considering that a difference between the air temperature time change curves corresponding to the fabrics of different weights increases with an increase of drying time, the duration through which the fabrics reach the temperature threshold can be prolonged by increasing the temperature threshold, so that the initial weights of the fabrics are more accurately determined based on prolonged second duration Δt₂, laying a foundation for optimizing subsequent fabric moisture level determining and timing information obtaining.
A collection process of the air temperature parameter is described above, and a collection process of the drain characteristic parameter is described below. In this context, the "drain characteristic parameter related to the liquid water extracted from the fabric in the fabric processing appliance 1 through the drying operation" may be understood as follows: A "drain characteristic" does not generally refer to an emission characteristic of a liquid guided by the drain pipe when the fabric processing appliance performs washing, rinsing, or other fabric processing operations, and does not relate to a flow characteristic of the coolant in a coolant circulation loop of the fabric processing appliance, but has a clear correspondence only with drain of the liquid water extracted from the fabric in the drying stage. To some extent, such a drain characteristic parameter can reflect an amount of wet load carried by the to-be-dried fabric and changes of the wet load with time.
The drain characteristic parameter may exist in a plurality of forms, and for example, the drain characteristic parameter includes: a mass flow, a volume flow, and a flow rate of the liquid water extracted from the fabric in the fabric processing appliance in the drain pipe 9 of the fabric processing appliance, and/or time change characteristic parameters of the mass flow, the volume flow, and the flow rate. The flow characteristic of the liquid water extracted from the fabric may be directly measured by using a sensor such as a flow meter or a flow rate meter arranged in the drain pipe 9, and is used as the drain characteristic parameter.
In another embodiment, it is considered that the drying process of the fabric is a relatively slow process, moisture extracted from the fabric generally drops to the water collector 91 in a form of water drops. Therefore, even if the water collector 91, as shown in FIG. 1, is designed to have a slope to promote liquid flow, the liquid collected therein usually flows at only a very slow rate. Only a low flow rate meter or a flow rate meter that has high costs can achieve accuracy of the measurement rate. Therefore, it is particularly advantageous to measure the drain characteristic parameter until the liquid water enters the section 92 of the drain pipe extending in the vertical direction, so that the flow characteristic of the liquid water is converted into a pressure characteristic. Further, the water pressure information of the determined position in the drain pipe 9 may be measured by using the pressure sensor 221 that has relatively low costs, and then the drain characteristic parameter is obtained based on the water pressure information.
The following describes in detail a process of determining the drain characteristic parameter with reference to a flowchart of step S1 according to another exemplary embodiment of this application shown in FIG. 6, and the process includes, for example, steps S111 to S118.
First, in step S111, whether a preheat stage of the fabric processing appliance ends is determined. In the preheat stage, the heat pump system 6 is turned on and operating parameters of components of the heat pump system 6 gradually reach stability. To make the collected drain characteristic parameter more intuitively reflect a progress of a formal drying stage, the drain characteristic parameter is not collected in the preheat stage, or the drain characteristic parameter collected in the preheat stage is not used for subsequent fabric moisture level analysis.
If it is determined in step S111 that the preheat phase does not end, such determining continues to be performed in step S111. If it is determined in step S111 that the preheat stage ends, and the formal drying stage starts, in the following step S112, the water pressure information of the determined position in the drain pipe of the fabric processing appliance is obtained, and the drain characteristic parameter of the fabric processing appliance is obtained based on the water pressure information.
In the embodiment shown in FIG. 1, when the pressure sensor 221 is arranged at a bottom portion of the section 92 (which is briefly referred to as a vertical section 92 for ease of description below) of the drain pipe 9 extending in the vertical direction, the measured water pressure information may be converted into flow information of the liquid water according to the following formula: $h = \frac{p}{ρg}$ $q = \frac{{hπr}^{2}}{t} = \frac{{pπr}^{2}}{ρgt}$ q is a drain flow (for example, the volume flow) of liquid water extracted from a fabric in the vertical section of the drain pipe, p is a water pressure at the bottom portion of the vertical section (for example, a difference between a reading of the pressure sensor and an atmospheric pressure), h is a liquid level height in the vertical section, ρ is a density of the liquid water guided by the drain pipe, r is a cross-sectional radius of the vertical section, t is a duration form a moment at which the drain pump 10 is turned off each time, and g is a gravitational acceleration.
In step S113, information about a water volume accumulated in the vertical section 92 of the drain pipe 9 is obtained. For example, the liquid level height in the vertical section 92 may be detected by using a liquid level sensor, and the liquid level height is used as the water volume information. For another example, the water pressure information obtained in the previous step may alternatively be determined as the water volume information.
In step S114, whether the water volume accumulated in the vertical section 92 of the drain pipe 9 reaches a preset limit is detected. The preset limit may also have a plurality of representation forms according to a specific sensor type for collecting the water volume information. For example, it may be determined that the water volume reaches the preset limit, when the liquid level in the vertical section 92 of the drain pipe 9 reaches a preset height threshold, and/or, when the water pressure at the bottom portion of the vertical section 92 reaches a preset pressure threshold.
If it is determined in step S114 that the water volume does not reach the preset limit, the drain pump 10 is kept being turned off and the drain characteristic parameter continues to be determined by using the reading of the pressure sensor 221 in step S112.
If it is determined in step S114 that the water volume reaches the preset limit, in step S115, the drain pump 10 is controlled to be turned on, to pump the liquid water accumulated in the vertical section to the external environment. At the same time, in step S116, the pressure sensor 221 is further enabled to temporarily stop recording the water pressure information during a turning-on period of the drain pump 10, and a timer is reset, to prevent a detection result obtained during the turning-on period of the drain pump 10 from interfering with the fabric moisture level determining.
Next, in step S117, whether the water in the vertical section 92 is completely drained is detected. In a case that it is assumed that a drain rate of the drain pump 10 is constant, in this step, whether turning-on time of the drain pump 10 reaches a preset time threshold is detected. In addition, such determining may alternatively be performed based on detection results of the liquid level sensor and the pressure sensor 221.
If it is found that the water is not completely drained, the drain pump 10 is kept being turned on, and such detection continues to be performed in step S117. Once it is found that the water is completely drained, the drain pump 10 is turned off in step S118, detection of the water pressure information is resumed in step S112, and timing is restarted from this moment. In this way, a complete water pressure recording period is completed. A time period between each turning-off moment and a next turning-on moment of the drain pump 10 may be defined as the complete water pressure recording period.
In this manner, it can be ensured that the water volume in the drain pipe 9 is sufficient during collection of the drain characteristic parameter, to improve measurement accuracy, and avoid energy consumption and noise problems caused by continuous turning-on of the drain pump 10.
In step S2, the moisture level of the fabric in the fabric processing appliance 1 is determined based on at least the air temperature parameter and the drain characteristic parameter. The trained first machine learning model 231 is used to determine the moisture level of the fabric based on at least the air temperature parameter and the drain characteristic parameter. It is considered that input variables of the first machine learning model 231, that is, the air temperature parameter and the drain characteristic parameter, are parameters collected in real time in the process of drying the fabric, the first machine learning model 231 may determine the real-time moisture level of the fabric based on the parameters collected in real time.
FIG. 7 is a schematic principle diagram of determining a moisture level of a fabric in a fabric processing appliance by using the first machine learning model 231. In the first machine learning model 231, for example, the moisture level of the fabric may be determined by using two links 2311 and 2312. In the first link 2311, the initial weight of the fabric in the fabric processing appliance 1 may be determined according to the time change characteristic parameter of the air temperature. Optionally, a look-up table of the time change characteristic parameter of the air temperature (the first duration Δt₁ and the second duration Δt₂ are exemplarily shown) and the initial weight w of the fabric may be prestored in the first link 2311. Therefore, the initial weight w of the fabric in the fabric processing appliance 1 may be determined based on the time change characteristic parameter of the air temperature. Then, in the second link 2312, based on an air temperature parameter T_a and a drain characteristic parameter q that are collected in real time, and with reference to the time change characteristic parameter of the air temperature, or the initial weight w of the fabric that is determined in the first link 2311, a real-time moisture level y of the fabric can be determined more accurately. In this process, the initial weight w of the fabric functions as an implicit factor that affects the working process of the entire first machine learning model 231. The implicit factor does not need to be measured and inputted by the outside, and is not directly outputted as a result, but is used as an intermediate result to construct a mapping relationship between a model input and a model output.
For example, the first machine learning model 231 may be configured as a distributed gradient boosting library model and/or a support vector machine model, and can balance computing precision and computing resources. The first machine learning model 231 may be trained by using a flowchart of a method for measuring the moisture level of a fabric in a fabric processing appliance according to another exemplary embodiment of this application shown in FIG. 8. Only differences from the embodiment shown in FIG. 3 are described below, and for brevity, same steps are not described again.
As shown in FIG. 8, the method may further include steps S201 to S203. In step S201, first sample data during the drying operation performed by the fabric processing appliance 1 on the fabric in a historical time period is collected, where the first sample data includes the air temperature parameter and the drain characteristic parameter. The collected first sample data is to have a certain amount, and in particular is to be able to cover different fabric moisture levels, different fabric initial weights and/or different drying modes, to ensure that the collected first sample data can simulate diversity of actual drying scenarios, so that the first machine learning model has a good generalization capability. In addition, during sample collection, for example, fabrics of the same initial weight may be numbered or distinguished by using identifiers or identification codes, to prevent similar individuals from repeatedly appearing for a plurality of times.
In step S202, the first sample data is marked in terms of the moisture level of the fabric. For example, during a test, the drying operation may be periodically suspended and the fabric is taken out, and then a moisture sensor is placed in the fabric for measuring the moisture level of the fabric, or a water content of the fabric is calculated based on a measured weight of the fabric and converted into the moisture level of the fabric. Next, the corresponding first sample data is manually marked based on the moisture level of the fabric.
In step S203, a parameter of the first machine learning model 231 is adjusted based on marked first sample data until a performance evaluation indicator of the first machine learning model 231 satisfies a preset condition and/or until a preset quantity of training steps is reached. The first sample data may be divided into a training set and a verification set, to train and optimize a constructed model. Specifically, the air temperature parameter and the drain characteristic parameter may be used as an input of the first machine learning model 231, and a corresponding marked moisture level of the fabric may be used as an output of the first machine learning model 231. Then, the first sample data in the training set is continuously used to adjust the parameter of the first machine learning model 231, thereby training the constructed model. When performance of the first machine learning model 231 is evaluated by using the first sample data in the verification set, for example, a mean square error (MSE) may be used as an evaluation indicator to measure an error value between a model predicted value and a true value, and the error value is defined as: $MSE = \frac{1}{n} \sum_{i = 1}^{n} {(yi - \hat{y} i)}^{2}$ where n represents a quantity of samples, yi represents a measured moisture level of a fabric of an i^th sample, and ŷi represents a predicted moisture level of a fabric of the i^th sample. A smaller MSE indicates a smaller prediction error of the model and better model performance. When the MSE is less than a preset threshold, or when a quantity of training steps of the first machine learning model 231 reaches the preset quantity of training steps, the first machine learning model 231 completes a training process. It should be noted that, another performance evaluation indicator, such as a mean absolute error (MAE), a root mean square error (RMSE), or a determining coefficient (R^2), may also be selected to verify the first machine learning model 231.
FIG. 9 is a flowchart of a method for measuring the moisture level of a fabric in a fabric processing appliance according to another exemplary embodiment of this application. Only differences from the embodiment shown in FIG. 3 are described below, and for brevity, same steps are not described again.
As shown in FIG. 9, the method may further include step S3. In step S3, based on at least the determined moisture level of the fabric and/or the determined time change characteristic parameter of the air temperature, timing information for enabling the fabric in the fabric processing appliance 1 to reach a predetermined drying degree is obtained. The "predetermined drying degree" may be represented by a moisture level level of the fabric. For example, when the moisture level of the fabric is less than a preset moisture level threshold, and in particular, reaches a zero moisture level point, it may be considered that the fabric reaches the predetermined drying degree.
For example, in a preliminary experiment or calibration process, the initial weight of the fabric in the dry state can be measured, and then the fabric in the drum can be taken out periodically during the drying operation for weighing and checking the dry state. When it is found that the weight obtained by weighing is consistent with the initial weight of the fabric in the dry state, it is considered that the zero moisture level point of the fabric is reached, or the predetermined drying degree is reached.
In another embodiment, the "predetermined drying degree" may alternatively be set personalized according to a user preference. For example, setting of a moisture level threshold may enable the user to specify an expected drying degree. This may be, for example, presented to the user in a form of a moisture level or a grading level (for example, full-dry, basic-dry, or basic-wet) of the fabric, for the user to make a selection.
The "timing information" may include, for example, time required by the fabric processing appliance to reach the predetermined drying degree. In addition, the "timing information" may alternatively not represent a period of time, but represent a moment or a condition corresponding to the moment. When the moment is reached or the condition corresponding to the moment is satisfied, it means that the drying process is to end.
In addition, it is considered that the initial weight w of the fabric in the fabric processing appliance 1 may be determined based on the time change characteristic parameter of the air temperature, and in a case of the fabrics with different initial weights, a time change rate of the air temperature is different, and a final air temperature when the fabric reaches a drying state (that is, the moisture level is zero) is also different, the initial weight of the fabric is also one of important influencing factors in the fabric drying process.
In another embodiment, by using a second machine learning model 232, based on at least the determined moisture level of the fabric and/or the determined time change characteristic parameter of the air temperature, the timing information for enabling the fabric in the fabric processing appliance 1 to reach the predetermined drying degree may be obtained. The second machine learning model 232 may also be configured as, for example, a distributed gradient boosting library model and/or a support vector machine model. FIG. 10 is a schematic principle diagram of obtaining, by using the second machine learning model 232, timing information for enabling a fabric in a fabric processing appliance to reach a predetermined drying degree. In the second machine learning model 232, for example, the moisture level of the fabric may also be determined by using two links 2321 and 2322. In the first link 2321, the initial weight w of the fabric in the fabric processing appliance 1 may be determined according to the time change characteristic parameter of the air temperature. Optionally, a look-up table of the time change characteristic parameter of the air temperature (the first duration Δt₁ or the second duration Δt₂ are exemplarily shown) and the initial weight w of the fabric may be prestored in the first link 2321. Therefore, the initial weight w of the fabric in the fabric processing appliance 1 may be determined based on the time change characteristic parameter of the air temperature. Then, in the second link 2322, based on the determined real-time moisture level y of the fabric and with reference to the time change characteristic parameter of the air temperature, or the initial weight w of the fabric that is determined in the first link 2321, timing information z for enabling the fabric in the fabric processing appliance 1 to reach the predetermined drying degree may be obtained. In this process, the initial weight w of the fabric functions as an implicit factor that affects the working process of the entire second machine learning model 232. The implicit factor does not need to be measured and inputted by the outside, and is not directly outputted as a result, but is used as an intermediate result to construct a mapping relationship between a model input and a model output.
A training process of the second machine learning model 232 is described in detail below with reference to a flowchart of a method for measuring the moisture level of a fabric in a fabric processing appliance according to another exemplary embodiment of this application shown in FIG. 11. Only differences from the embodiment shown in FIG. 9 are described below, and for brevity, same steps are not described again.
As shown in FIG. 11, the method may further include steps S301 to S303. In step S301, second sample data during the drying operation performed by the fabric processing appliance 1 on the fabric in a historical time period is collected, where the second sample data may include the moisture level of the fabric and/or the time change characteristic parameter of the air temperature.
The collected second sample data is to have a certain amount, and in particular is to be able to cover different fabric moisture levels, different fabric initial weights and different drying modes, to ensure that the collected first sample data can simulate diversity of actual drying scenarios, so that the first machine learning model has a good generalization capability. In addition, during sample collection, for example, fabrics of the same initial weight may be numbered or distinguished by using identifiers or identification codes, to prevent similar individuals from repeatedly appearing for a plurality of times.
For example, during a test, the drying operation may be periodically suspended and the fabric is taken out, and then a moisture sensor is placed in the fabric for measuring the moisture level of the fabric. In addition, during the test, based on the air temperature of the at least one position from the outlet of the drum 2 to the inlet of the dehumidifier 7 that is collected by the temperature sensor, the duration Δt₁ or Δt₂ through which the air temperature reaches the temperature threshold T₁ or T₂ is recorded.
In step S302, the second sample data is marked in terms of the timing information for enabling the fabric to reach the predetermined drying degree. For example, during the test, the drying operation may be periodically suspended and the fabric is taken out, then a current drying level of the fabric is determined through manual observation or a touch operation, and when it is determined according to experience that the fabric reaches an expected dry state or smoothness, a duration of the drying process until now is manually recorded and is marked as the timing information.
In step S303, a parameter of the second machine learning model 232 is adjusted based on marked second sample data until a performance evaluation indicator of the second machine learning model 232 satisfies a preset condition and/or until a preset quantity of training steps is reached.
The second sample data may also be divided into a training set and a verification set, to train and optimize a constructed model. Specifically, the real-time moisture level of the fabric and/or the time change characteristic parameter of the air temperature may be used as an input of the second machine learning model 232, and corresponding marked timing information may be used as an output of the second machine learning model 232. Then, the second sample data in the training set is continuously used to adjust the parameter of the second machine learning model 232, thereby training the constructed model. When performance of the second machine learning model 232 is evaluated by using the second sample data in the verification set, for example, a mean square error (MSE) may be used as an evaluation indicator to measure an error value between a model predicted value and a true value, and the error value is defined as: $MSE = \frac{1}{n} \sum_{i = 1}^{n} {(zi - \hat{z} i)}^{2}$ n represents a quantity of samples, zi represents actual timing information of an i^th sample, and ŷi represents predicted timing information of the i^th sample. A smaller MSE indicates a smaller prediction error of the model and better model performance. When the MSE is less than a preset threshold, or when a quantity of training steps of the second machine learning model 232 reaches the preset quantity of training steps, the second machine learning model 232 completes a training process. It should be noted that, another performance evaluation indicator, such as a mean absolute error (MAE), a root mean square error (RMSE), or a determining coefficient (R^2), may also be selected to verify the second machine learning model 232.
In addition, in an embodiment that is not shown, on the premise that the second machine learning model 232 is preliminarily trained based on steps S301 to S303, feedback behavior information of the timing information obtained by the user from the trained second machine learning model 232 continues to be obtained in a process of using the model, and the second machine learning model 232 is trained based on at least the feedback behavior information.
Specifically, an event of forced correction performed by the user on the timing information determined by using the second machine learning model 232 may be recorded, or a behavior of the user actively ending the drying process or actively turning off the fabric processing appliance 1 when the timing information determined by the second machine learning model 232 is not reached may be recorded, and these events are recorded as negative feedback behaviors. Generally, an active interference behavior performed by the user on the timing information determined based on the second machine learning model 232 may be classified as a negative feedback behavior, and a behavior that is not interfered with may be classified as a positive feedback behavior. By knowing a quantity of times and a degree of active intervention of the user, satisfaction of the user with the current second machine learning model 232 may be reflected, so that these feedback behaviors may be used as supplementary training data again to update an internal parameter of the second machine learning model 232.
FIG. 12 is a flowchart of a method for measuring the moisture level of a fabric in a fabric processing appliance according to another exemplary embodiment of this application. Only differences from the embodiment shown in FIG. 9 are described below, and for brevity, same steps are not described again.
As shown in FIG. 12, the method may further include step S4. In step S4, running of the fabric processing appliance 1 is controlled based on the obtained timing information. The fabric processing appliance 1 may be controlled to terminate the drying operation when the timing information is satisfied. For example, the motor 4 is stopped by using an instruction sent by the control module 23. According to test data, a root mean square error (RMSE) between energy consumed in the entire drying process and energy consumed in an ideal drying process when the drying operation is terminated based on the timing information obtained according to the method of this application may be limited to a range of 0.034 kWh.
Optionally, the fabric processing appliance 1 may further be controlled, so that the display 11 not only displays the determined moisture level of the fabric and/or the collected air temperature parameter, but also displays estimated drying time and/or remaining drying time of the fabric processing appliance 1, and the like, so that the user can learn a real-time progress of the fabric drying process in time.
In addition, it should be noted that sequence numbers of steps described herein do not necessarily represent a sequence, but are merely reference signs. The sequence may be changed according to specific situations, as long as the technical objectives of this application can be achieved.
It should be understood that, the expressions "first", "second", "third", and the like are merely used for descriptive purposes, and should not be understood as indicating or implying relative importance, nor should
be understood as implicitly specifying a quantity of indicated technical features.
According to another embodiment of this application, a machine-readable storage medium such as a CD-ROM is provided. The machine-readable storage medium includes a computer program. When the computer program is executed, a computer or a processor performs the method according to embodiments of the present disclosure. The machine-readable storage medium is, for example, an optical storage medium or a solid-state medium supplied with or as part of other hardware.
If an embodiment includes an "and/or" association between a first feature and a second feature, this should be read as follows: According to an implementation, this embodiment not only has a first feature but also has a second feature. According to another implementation, this embodiment has only a first feature or only a second feature.
Although specific embodiments have been described above, these embodiments are not intended to limit the scope of the disclosure of this application, even if only a single embodiment is described with respect to specific features. The feature examples provided in the disclosure of this application are intended to be illustrative, not limiting, unless otherwise stated. During specific implementation, a plurality of features can be combined with each other according to the actual needs when technically feasible. Without departing from the spirit and scope of this application, various replacements, changes, and modifications may be conceived.

Claims

A method for measuring a moisture level of a fabric in a fabric processing appliance (1), the fabric processing appliance (1) being configured to perform a drying operation on a fabric, and the method comprising the following steps:
collecting an air temperature parameter and a drain characteristic parameter of the fabric processing appliance (1) during the drying operation performed by the fabric processing appliance (1), wherein the air temperature parameter is related to air circulating in the fabric processing appliance (1) during the drying operation, and the drain characteristic parameter is related to liquid water extracted from the fabric in the fabric processing appliance (1) through the drying operation; and

determining a moisture level of the fabric in the fabric processing appliance (1) based on at least the air temperature parameter and the drain characteristic parameter.
The method according to claim 1, characterized in that the air temperature parameter comprises an air temperature of at least one position from an outlet of a drum (2) to an inlet of a dehumidifier (7) in an air circulation loop of the fabric processing appliance (1) during the drying operation and/or a time change characteristic parameter of the air temperature, and/or in that the drain characteristic parameter comprises a mass flow, a volume flow, and a flow rate of the liquid water extracted from the fabric in the fabric processing appliance (1) in a drain pipe (9) of the fabric processing appliance (1) and/or time change characteristic parameters of the mass flow, the volume flow, and the flow rate, and/or in that water pressure information of a determined position in the drain pipe (9) of the fabric processing appliance (1) is measured by using a pressure sensor (221), and the drain characteristic parameter is obtained based on the water pressure information, and/or in that the determined position comprises at least a section (92) of the drain pipe (9) of the fabric processing appliance (1) that extends along a vertical direction.
The method according to claim 1, characterized in that the moisture level of the fabric in the fabric processing appliance (1) is determined based on at least the air temperature parameter and the drain characteristic parameter by using a trained first machine learning model (231), and in that the first machine learning model (231) is trained in the following manner:
collecting first sample data during the drying operation performed by the fabric processing appliance (1) on the fabric in a historical time period, wherein the first sample data comprises the air temperature parameter and the drain characteristic parameter;

marking the first sample data in terms of the moisture level of the fabric; and

adjusting a parameter of the first machine learning model (231) based on marked first sample data until a performance evaluation indicator of the first machine learning model (231) satisfies a preset condition and/or until a preset quantity of training steps is reached.
The method according to claim 2, characterized in that the time change characteristic parameter of the air temperature of the at least one position from the outlet of the drum to the inlet of the dehumidifier during the drying operation is collected and determined, and the moisture level of the fabric in the fabric processing appliance (1) is determined with reference to the time change characteristic parameter of the air temperature.
The method according to claim 4, characterized in that a step of determining the time change characteristic parameter of the air temperature comprises:
determining whether the air temperature of the at least one position from the outlet of the drum (2) to the inlet of the dehumidifier (7) reaches a first temperature threshold T₁;

determining, if the air temperature reaches the first temperature threshold T₁, whether first duration Δt₁ through which the air temperature reaches the first temperature threshold T₁ exceeds a preset time threshold t_w;

determining, if the first duration Δt₁ does not exceed the preset time threshold t_w, the first duration Δt₁ as the time change characteristic parameter of the air temperature; and

determining, if the first duration Δt₁ exceeds the preset time threshold t_w, second duration Δt₂ through which the air temperature reaches a second temperature threshold T₂ as the time change characteristic parameter of the air temperature, wherein the second temperature threshold T₂ is greater than the first temperature threshold T₁.
The method according to claim 4, characterized in that the method further comprises:
obtaining, based on at least the determined moisture level of the fabric and/or the determined time change characteristic parameter of the air temperature, timing information for enabling the fabric in the fabric processing appliance (1) to reach a predetermined drying degree.
The method according to claim 6, characterized in that the timing information for enabling the fabric in the fabric processing appliance (1) to reach the predetermined drying degree is obtained based on at least the determined moisture level of the fabric and/or the determined time change characteristic parameter of the air temperature by using a trained second machine learning model (232).
The method according to claim 7, characterized in that the second machine learning model (232) is trained in the following manner:
collecting second sample data during the drying operation performed by the fabric processing appliance (1) on the fabric in a historical time period, wherein the second sample data comprises the moisture level of the fabric and/or the time change characteristic parameter of the air temperature;

marking the second sample data in terms of the timing information for enabling the fabric to reach the predetermined drying degree; and

adjusting a parameter of the second machine learning model (232) based on marked second sample data until a performance evaluation indicator of the second machine learning model (232) satisfies a preset condition and/or until a preset quantity of training steps is reached.
The method according to claim 8, characterized in that the second machine learning model (232) is retrained based on at least feedback behavior information of the timing information obtained by a user from the trained second machine learning model (232).
The method according to claim 3 or 7, characterized in that the first machine learning model (231) and/or the second machine learning model (232) comprises a distributed gradient boosting library model and/or a support vector machine model.
The method according to claim 6, characterized in that the method further comprises:
controlling running of the fabric processing appliance (1) based on the obtained timing information, wherein the fabric processing appliance (1) is controlled to terminate the drying operation when the timing information is satisfied, and/or the fabric processing appliance (1) is controlled to display the determined moisture level of the fabric, the collected air temperature parameter, and estimated drying time and/or remaining drying time needed by the fabric to reach the predetermined drying degree.
A fabric processing appliance (1), the fabric processing appliance (1) comprising:
an air circulation loop, air in the fabric processing appliance (1) circulating along the air circulation loop during a drying operation;

a drain pipe (9), liquid water extracted from a fabric in the fabric processing appliance (1) through the drying operation being led to an external environment along the drain pipe (9); and

comprising an apparatus (20) for measuring a moisture level of a fabric in a fabric processing appliance (1), the apparatus (20) comprising the following components:
air temperature collecting modules (211 and 212), configured to collect an air temperature parameter of the fabric processing appliance (1) during a drying operation performed by the fabric processing appliance (1), wherein the air temperature parameter is related to air circulating in the fabric processing appliance (1) during the drying operation;

a drain characteristic collecting

module (22), configured to collect a drain characteristic parameter of the fabric processing appliance (1) during the drying operation performed by the fabric processing appliance (1), wherein the drain characteristic parameter is related to liquid water extracted from a fabric in the fabric processing appliance (1) through the drying operation; and

a control module (23), configured to perform the method according to any one of claims 1 to 11.
The fabric processing appliance (1) according to claim 12, characterized in that the air temperature collecting modules (211 and 212) comprise a first temperature sensor (211) arranged at an outlet of a drum (2) in the air circulation loop of the fabric processing appliance (1) and/or a second temperature sensor (212) arranged at an inlet of a dehumidifier (7) in the air circulation loop of the fabric processing appliance (1).
The fabric processing appliance (1) according to claim 12 or 13, characterized in that the drain characteristic collecting module (22) comprises a pressure sensor (221), and the pressure sensor (221) is arranged in the drain pipe (9) and is configured to measure water pressure information of a determined position in the drain pipe (9), wherein the drain characteristic parameter is obtained based on the water pressure information.
A computer program product, such as a computer-readable program carrier, comprising or storing computer program instructions, the computer program instructions, when executed by a processor, at least assistantly implementing steps of the method according to any one of claims 1 to 11.