WO2021045258A1 - Appareil de traitement de linge à intelligence artificielle - Google Patents

Appareil de traitement de linge à intelligence artificielle Download PDF

Info

Publication number
WO2021045258A1
WO2021045258A1 PCT/KR2019/011430 KR2019011430W WO2021045258A1 WO 2021045258 A1 WO2021045258 A1 WO 2021045258A1 KR 2019011430 W KR2019011430 W KR 2019011430W WO 2021045258 A1 WO2021045258 A1 WO 2021045258A1
Authority
WO
WIPO (PCT)
Prior art keywords
information
cloth
laundry
unit
driving course
Prior art date
Application number
PCT/KR2019/011430
Other languages
English (en)
Korean (ko)
Inventor
박윤식
Original Assignee
엘지전자 주식회사
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 엘지전자 주식회사 filed Critical 엘지전자 주식회사
Priority to PCT/KR2019/011430 priority Critical patent/WO2021045258A1/fr
Publication of WO2021045258A1 publication Critical patent/WO2021045258A1/fr

Links

Images

Classifications

    • D—TEXTILES; PAPER
    • D06—TREATMENT OF TEXTILES OR THE LIKE; LAUNDERING; FLEXIBLE MATERIALS NOT OTHERWISE PROVIDED FOR
    • D06F—LAUNDERING, DRYING, IRONING, PRESSING OR FOLDING TEXTILE ARTICLES
    • D06F33/00—Control of operations performed in washing machines or washer-dryers 
    • D06F33/30—Control of washing machines characterised by the purpose or target of the control 
    • D—TEXTILES; PAPER
    • D06—TREATMENT OF TEXTILES OR THE LIKE; LAUNDERING; FLEXIBLE MATERIALS NOT OTHERWISE PROVIDED FOR
    • D06F—LAUNDERING, DRYING, IRONING, PRESSING OR FOLDING TEXTILE ARTICLES
    • D06F39/00—Details of washing machines not specific to a single type of machines covered by groups D06F9/00 - D06F27/00 
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/08—Learning methods

Definitions

  • the present invention relates to an artificial intelligent laundry treatment device, and more particularly, to an artificial intelligent laundry treatment device capable of providing information on the fabrication information of the calculated laundry cloth and optimal driving course information corresponding to the input laundry cloth information. will be.
  • laundry treatment equipment uses physical effects such as chemical decomposition between water and detergent and friction between water and laundry, such as clothing, bedding, etc.(hereinafter abbreviated as'laundry' or'washing cloth' or'cloth'. It is a collective term for a device that separates contaminants from the ).
  • Laundry treatment equipment is largely divided into agitation type, vortex type and drum type laundry treatment equipment.
  • the drum type laundry treatment device includes a storage tank containing water, and a washing tank rotatably provided in the storage tank to accommodate laundry. A plurality of through holes through which water passes are formed in the washing tub.
  • the washing operation is generally divided into a washing operation, a rinsing operation and a dehydration operation. The progress of such an administration can be confirmed through a display provided on the control panel.
  • the washing process removes contaminants from the laundry by the frictional force of the water stored in the water tank and the laundry stored in the drum, and the chemical action of the detergent stored in the water.
  • the rinsing cycle is to rinse the cloth by supplying water in which the detergent is not dissolved into the storage tank.
  • the detergent absorbed by the cloth is removed during the washing cycle.
  • a fabric softener may be supplied along with water.
  • the spin-drying cycle is to spin the cloth by rotating the washing tank at high speed after the rinsing cycle is completed. Normally, all operations of the laundry treatment machine are terminated by the completion of the spin-drying cycle. However, in the case of a laundry treatment machine for drying combined use, a drying cycle may be added after the spin-drying cycle.
  • the washing operation is set according to the amount of cloth injected into the washing tank (hereinafter, also referred to as'the amount of cloth').
  • the water supply level, washing intensity, draining time and spinning time are set in the amount of cloth.
  • the laundry performance varies not only with the amount of fabric but also with the type of laundry (hereinafter, also referred to as “cloth”).
  • the amount of fabric is considered when setting the washing operation, sufficient washing performance cannot be expected.
  • Korean Patent Publication No. 10-1841248 discloses a control method for detecting the amount of fabric using the speed of a motor as input data of an artificial neural network previously learned by machine learning.
  • the prior art is to detect only the amount of cloth, and there is a problem of the above-described washing performance and laundry damage.
  • Another object of the present invention is to provide a laundry treatment device capable of determining the fabric quality of a laundry cloth according to learning based on a deep neural network, and providing information on the fabric quality and optimal driving course information corresponding to the input laundry cloth information. Is in.
  • Another object of the present invention is a laundry treatment device capable of determining information on a cloth quality of a laundry cloth and contamination level of laundry according to learning based on a deep neural network, and providing optimal driving course information according to the cloth quality information and contamination level information It is in providing.
  • An artificial intelligent laundry treatment device for achieving the above object includes a casing, a washing tub in the casing, a motor rotating the washing tub, and an output current that drives the motor and detects an output current flowing through the motor.
  • a driving unit having a detection unit, a communication unit for communicating with an external mobile terminal or server, an input unit for generating an input signal, and a control unit for controlling the driving unit, the control unit, based on the output current from the output current detection unit, Performs deep neural network-based learning, determines the cloth information of the laundry cloth in the washing tank according to the learning, receives the cloth information from the input unit or communication unit, and drives the course based on the cloth information and the cloth cloth information Determine the information.
  • control unit performs learning based on a deep neural network based on the output current from the output current detection unit, determines the cloth information of the laundry cloth in the washing tub according to the learning, and based on the input signal from the input unit or the communication unit. , Determine the contamination level information of the laundry cloth, and determine driving course information based on the contamination level information and the cloth quality information of the laundry cloth.
  • control unit performs learning based on a deep neural network based on an input signal from an input unit or a communication unit, and determines contamination level information of the laundry cloth according to the learning.
  • the controller performs driving based on the determined driving course information and controls to output driving course information.
  • the laundry treatment apparatus further includes a memory for storing information on a plurality of driving courses, and the determined driving course information is additional driving course information that is not stored in the memory.
  • the memory stores additional driving course information.
  • control unit determines the contamination level information of the laundry cloth based on the sound information from the input unit or the communication unit.
  • control unit performs classification for setting a driving course based on the level of the cloth information of the laundry cloth in the washing tank and the level of the contamination level information, and determines driving course information based on the classification.
  • control unit accesses an external server, performs a search based on classification information according to classification, and determines driving course information based on the search result.
  • the controller accesses an external server, performs a search based on the level of the cloth information of the laundry cloth in the washing tub and the level of the pollution level information, and determines driving course information based on the search result.
  • control unit performs learning based on a deep neural network based on the output current from the output current detection unit, and according to the learning, determines the cloth information and the cloth information of the washing cloth in the washing tank, and an input signal from the input unit or the communication unit On the basis of, the pollution degree information of the laundry cloth is determined, and driving course information is determined based on the pollution degree information and the cloth quality information and the amount information of the laundry cloth.
  • control unit performs classification for setting a driving course based on the level of the cloth information of the laundry cloth, the level of the cloth amount information of the cloth, and the level of the pollution level information, and based on the classification, the driving course information Decide.
  • control unit accesses an external server, performs a search based on classification information according to classification, and determines driving course information based on the search result.
  • control unit connects to an external server to perform a search based on the level of the cloth information of the laundry cloth, the level of the cloth amount information of the cloth, and the level of the pollution level information, and based on the search result, driving course information Decide.
  • An artificial intelligent laundry treatment device includes a driving unit including a casing, a washing tub in the casing, a motor rotating the washing tub, and an output current detecting unit that drives the motor and detects an output current flowing through the motor.
  • a communication unit that communicates with an external mobile terminal or server, an input unit that generates an input signal, and a control unit that controls the driving unit, and the control unit performs deep neural network-based learning based on the output current from the output current detection unit. Then, according to the learning, it determines the fabric information of the laundry cloth in the washing tub, receives the laundry cloth information from the input unit or the communication unit, and determines driving course information based on the laundry cloth information and the cloth fabric information of the laundry cloth. Accordingly, it is possible to provide the calculated fabric information of the laundry cloth and optimal driving course information corresponding to the input laundry cloth information.
  • control unit performs learning based on a deep neural network based on the output current from the output current detection unit, determines the cloth information of the laundry cloth in the washing tub according to the learning, and based on the input signal from the input unit or the communication unit. , Determine the contamination level information of the laundry cloth, and determine driving course information based on the contamination level information and the cloth quality information of the laundry cloth. Accordingly, according to learning based on a deep neural network, it is possible to determine the cloth quality of the laundry cloth, and provide information on the cloth material and optimal driving course information corresponding to the input pollution degree information.
  • control unit performs learning based on a deep neural network based on an input signal from an input unit or a communication unit, and determines contamination level information of the laundry cloth according to the learning. Accordingly, according to learning based on a deep neural network, it is possible to provide information on an optimal driving course corresponding to the cloth quality and pollution level information of the laundry cloth.
  • the controller performs driving based on the determined driving course information and controls to output driving course information. Accordingly, the driving course information can be easily recognized by the user.
  • the laundry treatment apparatus further includes a memory for storing information on a plurality of driving courses, and the determined driving course information is additional driving course information that is not stored in the memory. Accordingly, it is possible to provide new driving course information.
  • the memory stores additional driving course information. Accordingly, it is possible to store new driving course information.
  • control unit determines the contamination level information of the laundry cloth based on the sound information from the input unit or the communication unit. Accordingly, it is possible to easily determine the pollution level information based on the user's voice.
  • control unit performs classification for setting a driving course based on the level of the cloth information of the laundry cloth in the washing tank and the level of the contamination level information, and determines driving course information based on the classification. Accordingly, it is possible to provide optimal driving course information based on the foam information and the pollution degree information.
  • control unit accesses an external server, performs a search based on classification information according to classification, and determines driving course information based on the search result. Accordingly, driving course information can be provided using an external server.
  • the controller accesses an external server, performs a search based on the level of the cloth information of the laundry cloth in the washing tub and the level of the pollution level information, and determines driving course information based on the search result. Accordingly, driving course information can be provided using an external server.
  • control unit performs learning based on a deep neural network based on the output current from the output current detection unit, and according to the learning, determines the cloth information and the cloth information of the washing cloth in the washing tank, and an input signal from the input unit or the communication unit
  • the pollution degree information of the laundry cloth is determined, and driving course information is determined based on the pollution degree information and the cloth quality information and the amount information of the laundry cloth. Accordingly, it is possible to provide information on an optimum driving course corresponding to the cloth information, the cloth amount information, and the pollution degree information.
  • control unit performs classification for setting a driving course based on the level of the cloth information of the laundry cloth, the level of the cloth amount information of the cloth, and the level of the pollution level information, and based on the classification, the driving course information Decide. Accordingly, it is possible to provide information on an optimum driving course corresponding to the cloth information, the cloth amount information, and the pollution degree information.
  • control unit accesses an external server, performs a search based on classification information according to classification, and determines driving course information based on the search result. Accordingly, driving course information can be provided using an external server.
  • control unit connects to an external server to perform a search based on the level of the cloth information of the laundry cloth, the level of the cloth amount information of the cloth, and the level of the pollution level information, and based on the search result, driving course information Decide. Accordingly, driving course information can be provided using an external server.
  • FIG. 1 is a perspective view showing a laundry treatment device according to an embodiment of the present invention.
  • FIG. 2 is an internal block diagram of the laundry treatment device of FIG. 1.
  • FIG. 3 is an internal circuit diagram of the motor driving unit of FIG. 2.
  • FIG. 4 is an internal block diagram of the inverter control unit of FIG. 3.
  • FIG. 8 shows a process of processing current output current values obtained by the current detection unit as input data of an artificial neural network.
  • FIG. 9 is a schematic diagram showing an example of an artificial neural network.
  • FIG. 10 is a schematic diagram showing a process of determining fabric quantity and fabric quality using a current output current value of the motor, divided into a learning process and a recognition process.
  • 11 is a graph (a) showing the current output current value detected by the current detection unit and a graph (b) showing average values obtained by processing a moving average filter.
  • FIG. 13 is a graph showing values processed to use the current values of the graph shown in FIG. 11 as input data of an artificial neural network.
  • FIG. 14 is a flow chart showing a control method of the laundry treatment device.
  • 15 is a graph in which current patterns for each load are superimposed.
  • 16 is a graph for classifying current patterns corresponding to a load of 0 to 6 kg in FIG. 15.
  • 17 is a graph for classifying current patterns corresponding to a load of 7 to 9 kg in FIG. 15.
  • FIG. 18 is a flowchart illustrating a control method of a laundry treatment device according to an embodiment of the present invention.
  • 19A to 19D are diagrams referenced for explaining the control method of FIG. 18.
  • module and “unit” for the constituent elements used in the following description are given in consideration of only the ease of writing in the present specification, and do not impart a particularly important meaning or role by themselves. Therefore, the “module” and “unit” may be used interchangeably with each other.
  • FIG. 1 is a perspective view showing a laundry treatment device according to an embodiment of the present invention.
  • a laundry treatment device 100 is a front load type laundry treatment device in which a carriage is inserted into a washing tub in a front direction.
  • the laundry treatment device 100 is a drum-type laundry treatment device, a casing 110 forming the exterior of the laundry treatment device 100, and disposed inside the casing 110, and the casing 110
  • the washing tub 120 supported by the washing tub 120, the drum 122, which is a washing tub that is disposed inside the washing tub 120 and in which the cloth is washed, a motor 130 driving the drum 122, and the cabinet body 111, are disposed outside the cabinet body 111
  • a washing water supply device (not shown) for supplying washing water into the casing 110, and a drainage device (not shown) formed under the washing tub 120 and discharging the washing water to the outside.
  • a plurality of through holes 122A are formed in the drum 122 so that the washing water passes, and after the laundry is lifted to a certain height when the drum 122 is rotated, a lifter on the inner side of the drum 12 so as to fall by gravity. 124 can be deployed.
  • the casing 110 includes a cabinet body 111 and a cabinet cover 112 disposed on and coupled to the front of the cabinet body 111, and a control disposed on the upper side of the cabinet cover 112 and coupled to the cabinet body 111 It includes a panel 115 and a top plate 116 disposed above the control panel 115 and coupled to the cabinet body 111.
  • the cabinet cover 112 includes a fabric entry hole 114 formed to allow entry and exit of the fabric, and a door 113 disposed to be rotatable left and right to allow the opening and closing of the fabric entry hole 114.
  • the control panel 115 includes an input unit 117 for manipulating the operating state of the laundry treatment device 100, and a display 118 disposed on one side of the input unit 117 and displaying the operating state of the laundry treatment device 100. Includes.
  • the input unit 117 and the display 118 in the control panel 115 are electrically connected to a control unit (not shown), and the control unit (not shown) electrically controls each component of the laundry treatment device 100.
  • the operation of the control unit (not shown) will be omitted with reference to the operation of the control unit 210 of FIG. 3.
  • an auto balance (not shown) may be provided on the drum 122.
  • the auto balance (not shown) is for reducing vibrations generated according to the amount of eccentricity of laundry accommodated in the drum 122, and may be implemented as a liquid balance, a ball balance, or the like.
  • the laundry treatment device 100 may further include a vibration sensor that measures the amount of vibration of the drum 122 or the amount of vibration of the casing 110.
  • FIG. 2 is an internal block diagram of the laundry treatment device of FIG. 1.
  • the laundry treatment device 100 includes a motor 230 for rotating the washing tub 120, a driving unit 220 for driving the motor 230, a communication unit 270, and an input unit 117. ), a display 118, a control unit 210 for controlling each unit in the laundry treatment device 100, and a speed detection unit 245 may be provided.
  • the input unit 117 may include an operation key (a power key, an operation key, etc.) for operating the laundry treatment device 100.
  • the input unit 117 may further include a microphone (not shown) for recognizing a user's voice.
  • the display 118 displays an operating state of the laundry treatment device 100.
  • the display 118 may display information on a driving course in which the laundry treatment device 100 is operating.
  • the memory 240 includes a programmed artificial neural network, current patterns for each fabric and/or fabric, a database (DB) built through machine learning-based learning based on the current pattern, a machine learning algorithm, and an output current detector (The current output current value detected by E), the averaged value of the current output current values, the averaged values processed according to a parsing rule, data transmitted and received through the communication unit 270, etc. Can be saved.
  • DB database
  • E output current detector
  • the memory 240 includes various control data for overall control of the operation of the laundry processing device, laundry setting data input by the user, laundry time calculated according to the laundry setting, data on the laundry course, etc., and errors in the laundry processing device. Data for determining whether or not it occurs may be stored.
  • the communication unit 270 may communicate with an external mobile terminal 600 or a server 500.
  • the communication unit 270 may include one or more communication modules such as an Internet module and a mobile communication module.
  • the communication unit 270 may receive various data such as learning data and algorithm updates from the server.
  • the controller 210 may update the memory 240 by processing various types of data received through the communication unit 270. For example, if the data input through the communication unit 270 is update data for a driving program previously stored in the memory 240, the data is updated to the memory 240 using this, and the input data is a new driving program. In this case, it may be additionally stored in the memory 240.
  • Deep Learning is an artificial intelligence technology that teaches computers how to think, based on an artificial neural network (ANN) for constructing artificial intelligence, and allows computers to learn like humans without humans teaching them.
  • the artificial neural network (ANN) may be implemented in the form of software or may be implemented in the form of hardware such as a chip.
  • the laundry treatment device 100 processes the current values detected by the output current detection unit E based on machine learning, and the characteristics of the laundry (cloth) injected into the washing tub 120 (hereinafter referred to as fabric characteristics). .) can be grasped.
  • fabric characteristics may be exemplified by the amount of fabric and the state of the fabric (hereinafter, also referred to as “foam”), and the controller 210 may determine the fabric quality for each fabric volume based on machine learning.
  • control unit 210 may obtain the amount of fabric and determine which of the categories pre-classified according to the fabric quality is included.
  • the condition of these fabrics is the material of the fabric, its softness (e.g., soft/hard fabric), the ability of the fabric to hold water (i.e. moisture content), and the volume difference between the dry and wet fabric. It can be defined based on several factors such as the composition of the fabric (that is, the mixing ratio of the soft fabric and the hard fabric).
  • the controller 210 uses the current output current value detected by the output current detection unit E until the target speed is reached, and input data of an artificial neural network previously learned by machine learning ( It can be used as input data) to detect the amount of fabric.
  • control unit 210 may control the motor driving unit 220.
  • the motor driving unit 220 drives the motor 230. Accordingly, the washing tub 120 rotates by the motor 230.
  • the control unit 210 operates by receiving an operation signal from the input unit 117. Accordingly, washing, rinsing, and spin-drying may be performed.
  • controller 210 may control the display 118 to display a washing course, a washing time, a spinning time, a rinsing time, or a current operation state.
  • control unit 210 controls the motor driving unit 220, and the motor driving unit 220 controls the motor 230 to operate.
  • the motor driving unit 220 is for driving the motor 230, an inverter (not shown), an inverter control unit (not shown), and an output current detection unit (E in FIG. 3) that detects an output current flowing through the motor 230. ) Can be provided.
  • the motor driving unit 220 may be a concept that further includes a converter or the like that supplies DC power input to an inverter (not shown).
  • the inverter control unit (430 in FIG. 3) in the motor driving unit 220 estimates the position of the rotor of the motor 230 based on the output current io. Then, based on the estimated rotor position, the motor 230 is controlled to rotate.
  • the inverter control unit (430 in Fig. 3) generates a pulse width modulation (PWM) switching control signal (Sic in Fig. 3) based on the output current (io) and outputs it to the inverter (not shown). Then, the inverter (not shown) performs a high-speed switching operation to supply AC power having a predetermined frequency to the motor 230. Then, the motor 230 is rotated by an AC power source having a predetermined frequency.
  • PWM pulse width modulation
  • the motor driving unit 220 will be described later with reference to FIG. 3.
  • control unit 210 may detect the amount of cloth based on the current i o detected by the current detection unit 220, or the like. For example, while the washing tub 120 is rotating, the amount of cloth may be detected based on the current value i o of the motor 230.
  • the controller 210 may detect an eccentric amount of the washing tub 120, that is, an unbalance (UB) of the washing tub 120.
  • the detection of the amount of eccentricity may be performed based on a ripple component of the current i o detected by the current detection unit 220 or a change in the rotational speed of the washing tub 120.
  • control unit 210 may perform learning and recognition by a treadmill, and to this end, may include a learning module 213 and a recognition module 216.
  • the speed detection unit 245 may include a Hall sensor inside or outside the motor 230 for detecting a rotor position of the motor. Accordingly, the speed sensing unit 245 may be referred to as a position sensing unit.
  • the position signal H output from the speed detection unit 245 is input to the control unit 210, and the control unit 210 detects the speed of the motor 230 based on the position signal H. You will be able to.
  • FIG. 3 is an internal circuit diagram of the motor driving unit of FIG. 2.
  • the motor driving unit 220 is for driving a sensorless motor, and includes a converter 410, an inverter 420, an inverter control unit 430, and a dc terminal. It may include a voltage detector (B), a smoothing capacitor (C), an output current detector (E), and an output voltage detector (F). In addition, the motor driving unit 220 may further include an input current detection unit A, a reactor L, and the like.
  • the reactor L is disposed between the commercial AC power source 405, v s and the converter 410, and performs power factor correction or boosting operation. In addition, the reactor L may perform a function of limiting harmonic current due to high-speed switching of the converter 410.
  • the input current detection unit A can detect an input current i s input from the commercial AC power supply 405. To this end, as the input current detection unit A, a current trnasformer (CT), a shunt resistor, or the like may be used.
  • CT current trnasformer
  • the detected input current i s is a pulsed discrete signal and may be input to the inverter controller 430.
  • the converter 410 converts the commercial AC power source 405 that has passed through the reactor L into a DC power source and outputs it.
  • the commercial AC power source 405 is shown as a single-phase AC power source, but may be a three-phase AC power source.
  • the internal structure of the converter 410 also varies according to the type of the commercial AC power source 405.
  • the converter 410 may be made of a diode or the like without a switching element and may perform a rectification operation without a separate switching operation.
  • diodes in the case of single-phase AC power, four diodes may be used in the form of a bridge, and in the case of three-phase AC power, six diodes may be used in the form of a bridge.
  • the converter 410 for example, a half-bridge type converter in which two switching elements and four diodes are connected may be used, and in the case of a three-phase AC power source, six switching elements and six diodes may be used. .
  • the converter 410 When the converter 410 includes a switching element, step-up operation, power factor improvement, and DC power conversion may be performed by a switching operation of the switching element.
  • the smoothing capacitor C smooths the input power and stores it.
  • one device is illustrated as the smoothing capacitor C, but a plurality of devices are provided to ensure device stability.
  • DC power may be directly input.
  • DC power from a solar cell is supplied to the smoothing capacitor C. It can be directly input or can be input by direct current/DC conversion.
  • the parts illustrated in the drawings will be mainly described.
  • DC power is stored at both ends of the smoothing capacitor C, it may be referred to as a dc terminal or a dc link terminal.
  • the dc terminal voltage detector B may detect the dc terminal voltage Vdc, which is both ends of the smoothing capacitor C. To this end, the dc terminal voltage detection unit B may include a resistance element, an amplifier, and the like. The detected dc voltage Vdc may be input to the inverter controller 430 as a discrete signal in the form of a pulse.
  • the inverter 420 includes a plurality of inverter switching elements, converts a DC power supply (Vdc) smoothed by an on/off operation of the switching element into a three-phase AC power supply (va, vb, vc) of a predetermined frequency, It can be output to the synchronous motor 230.
  • Vdc DC power supply
  • va, vb, vc three-phase AC power supply
  • a pair of upper arm switching elements (Sa, Sb, Sc) and lower arm switching elements (S'a, S'b, S'c) connected in series with each other, respectively, is a pair of three pairs of upper and lower arms.
  • the switching elements are connected to each other in parallel (Sa&S'a,Sb&S'b,Sc&S'c).
  • Diodes are connected in reverse parallel to each of the switching elements Sa, S'a, Sb, S'b, Sc, S'c.
  • the switching elements in the inverter 420 perform on/off operations of each switching element based on the inverter switching control signal Sic from the inverter controller 430. Accordingly, three-phase AC power having a predetermined frequency is output to the three-phase synchronous motor 230.
  • the inverter controller 430 may control a switching operation of the inverter 420 based on a sensorless method. To this end, the inverter control unit 430 may receive an output current i o detected by the output current detection unit E and an output voltage v o detected by the output voltage detection unit F.
  • the inverter controller 430 outputs an inverter switching control signal Sic to the inverter 420 in order to control the switching operation of the inverter 420.
  • Inverter switching control signal (Sic) is a pulse width modulation (PWM) switching control signal, the output current (i o ) detected by the output current detection unit (E) and the output voltage detected by the output voltage detection unit (F) ( It is generated and output based on v o)
  • PWM pulse width modulation
  • the output current detection unit E detects an output current i o flowing between the inverter 420 and the three-phase motor 230. That is, the current flowing through the motor 230 is detected.
  • the output current detection unit E may detect all of the output currents ia, ib, ic of each phase, or may detect the output currents of two phases using three-phase balance.
  • the output current detection unit E may be located between the inverter 420 and the motor 230, and for current detection, a current trnasformer (CT), a shunt resistor, or the like may be used.
  • CT current trnasformer
  • a shunt resistor When a shunt resistor is used, three shunt resistors are located between the inverter 420 and the synchronous motor 230, or the three lower arm switching elements (S'a, S'b, S'c) of the inverter 420 ) Can be connected to each end. On the other hand, using three-phase equilibrium, it is also possible to use two shunt resistors. Meanwhile, when one shunt resistor is used, a corresponding shunt resistor may be disposed between the capacitor C and the inverter 420 described above.
  • the detected output current i o may be applied to the inverter control unit 430 as a discrete signal in the form of a pulse, and the inverter switching control signal Sic based on the detected output current i o Is created.
  • the detected output current i o is a three-phase output current (ia, ib, ic).
  • the three-phase motor 230 includes a stator and a rotor, and AC power of each phase of a predetermined frequency is applied to the coils of the stator of each phase (a, b, c phase), so that the rotor rotates. Will do.
  • Such a motor 230 is, for example, a surface-mounted permanent magnet synchronous motor (Surface-Mounted Permanent-Magnet Synchronous Motor; SMPMSM), a built-in permanent magnet synchronous motor (Interior Permanent Magnet Synchronous Motor; IPMSM), and a synchronous relay. It may include a Synchronous Reluctance Motor (Synrm) or the like. Among them, SMPMSM and IPMSM are Permanent Magnet Synchronous Motor (PMSM), and Synrm is characterized by no permanent magnet.
  • the inverter controller 430 may control a switching operation of the switching element in the converter 410.
  • the inverter control unit 430 may receive an input current i s detected by the input current detection unit A.
  • the inverter controller 430 may output a converter switching control signal Scc to the converter 410 in order to control the switching operation of the converter 410.
  • the converter switching control signal Scc is a pulse width modulation (PWM) type switching control signal, and may be generated and output based on the input current i s detected from the input current detector A.
  • PWM pulse width modulation
  • FIG. 4 is an internal block diagram of the inverter control unit of FIG. 3.
  • the inverter control unit 430 includes an axis conversion unit 510, a speed calculation unit 520, a current command generation unit 530, a voltage command generation unit 540, an axis conversion unit 550, and A switching control signal output unit 560 may be included.
  • the axis conversion unit 510 receives the output currents (ia, ib, ic) detected by the output output current detection unit E, and receives the two-phase currents i ⁇ , i ⁇ of the stationary coordinate system, and the two-phase current of the rotational coordinate system. It can be converted to (id,iq). At this time, id represents the magnetic flux current of the motor 230, and iq may represent the torque current of the motor 230.
  • the axis conversion unit 510 the converted two-phase currents (i ⁇ , i ⁇ ) of the stationary coordinate system, the two-phase voltages (v ⁇ , v ⁇ ) of the stationary coordinate system, and the two-phase currents (id, iq) of the rotating coordinate system and
  • the two-phase voltage (vd,vq) of the rotary coordinate system can be output to the outside.
  • the speed calculating unit 520 receives the axis-converted, two-phase currents i ⁇ , i ⁇ of the stationary coordinate system and the two-phase voltages v ⁇ , v ⁇ of the stationary coordinate system from the axis conversion unit 510, and the motor 230 The rotor position ( ⁇ ) and the speed ( ⁇ ) can be calculated.
  • the current command generation unit 530 generates a current command value (i * q ) based on the calculation speed () and the speed command value ( ⁇ * r ). For example, the current command generation unit 530 performs PI control in the PI controller 535 based on the difference between the calculation speed () and the speed command value ( ⁇ * r ), and the current command value (i * q ) Can be created.
  • the q-axis current command value (i * q ) is illustrated as the current command value, but unlike the drawing, it is also possible to generate the d-axis current command value (i * d) together. Meanwhile, the value of the d-axis current command value (i * d ) may be set to 0.
  • the current command generation unit 530 may further include a limiter (not shown) that limits the level so that the current command value i * q does not exceed the allowable range.
  • the voltage command generation unit 540 the d-axis and q-axis currents (i d , i q ) axially transformed from the axis conversion unit into a two-phase rotational coordinate system, and a current command value ( Based on i * d ,i * q ), d-axis and q-axis voltage command values (v * d ,v * q ) can be generated.
  • the voltage command generation unit 540 performs PI control in the PI controller 544 based on the difference between the q-axis current (i q ) and the q-axis current command value (i * q ), and q It is possible to generate the axis voltage command value (v * q ).
  • the voltage command generation unit 540 performs PI control in the PI controller 548 based on the difference between the d-axis current (i d ) and the d-axis current command value (i * d ), and the d-axis voltage You can create a setpoint (v * d ). Meanwhile, the value of the d-axis voltage command value (v * d ) may be set to 0 corresponding to the case where the value of the d-axis current command value (i * d) is set to 0.
  • the voltage command generation unit 540 may further include a limiter (not shown) that limits the level of the d-axis and q-axis voltage command values (v * d ,v * q) to not exceed the allowable range. .
  • the generated d-axis and q-axis voltage command values (v * d and v * q ) may be input to the axis conversion unit 550.
  • the axis conversion unit 550 receives the operation position () and the d-axis and q-axis voltage command values (v * d , v * q ) from the speed calculation unit 520 and performs axis conversion.
  • the axis conversion unit 550 converts from a two-phase rotational coordinate system to a two-phase stationary coordinate system.
  • the calculation position () may be used in the speed calculation unit 520.
  • the axis conversion unit 550 converts from a two-phase stationary coordinate system to a three-phase stationary coordinate system. Through this conversion, the axis conversion unit 1050 may output a three-phase output voltage command value (v * a, v * b, v * c).
  • the switching control signal output unit 560 generates a switching control signal Sic for an inverter according to a pulse width modulation (PWM) method based on a three-phase output voltage command value (v * a, v * b, v * c). And print it out.
  • PWM pulse width modulation
  • the output inverter switching control signal Sic may be converted into a gate driving signal by a gate driver (not shown) and may be input to the gates of each switching element in the inverter 420. Accordingly, each of the switching elements Sa, S'a, Sb, S'b, Sc, S'c in the inverter 420 may perform a switching operation.
  • 5 shows a current pattern applied to the motor according to the fabric and the amount of load (package).
  • 6 shows a current pattern for each fabric.
  • 7 shows a current pattern for each load while controlling the speed of the motor in a preset manner.
  • Each graph shown in FIG. 5 represents the current output current measured while accelerating the washing tub 120 to a preset target rotational speed (for example, 80rpm), and these graphs show the composition of the fabric (i.e., smooth It was measured while varying the mixing ratio of soft and stiff fabric) and load.
  • a preset target rotational speed for example, 80rpm
  • FIG. 6 shows a pattern of the current output current for each fabric configuration (fabric).
  • C0.0 represents 100% of soft fabric
  • C0.25, C0.5, and C0.75 in turn represent 100% of soft fabric
  • the ratio of soft fabric: stiff fabric is 1:3, 1:1, 3:1
  • C1.0 represents the case of 100% stiff fabric, and in each case, the total amount of fabric (load) plus soft fabric and stiff fabric is constant.
  • the graphs show that the current output current pattern is different even though the load is the same when the fabric configuration is different. Therefore, it is possible to classify according to the fabric composition (or fabric) based on the machine learning of the current pattern.
  • the fabrication/foaming detection may be repeated a plurality of times, and in the embodiment, it was repeated three times, but the number of times is not limited thereto.
  • the fabric/foam detection may be repeated multiple times in the same step, or may be repeated multiple times in different steps.
  • the control unit 210 may set or change a washing algorithm according to a result of detecting the amount of fabric/fabric, and control the operation of the laundry treatment device according to the setting.
  • the graphs P1, P3, P5, P7, P9, and P15 shown in FIG. 7 represent when the amount of fabric is 1, 3, 5, 7, 9, and 15kg, respectively.
  • the graphs are in a form in which the current output current value rapidly rises to a certain level at the beginning of the acceleration of the washing tub 120, and then a constant value converges toward the second half.
  • the deviation of the current output current value according to the amount of cloth is remarkable at the beginning of acceleration of the washing tub 120.
  • the control unit 210 may include a learning module 213 and a recognition module 216.
  • the learning module 213 may perform machine learning using a current output current value detected by the output current detection unit E or a value processed by the current output current value. Through such machine learning, the learning module 213 may update a database stored in the memory 240.
  • the recognition module 216 may determine a level according to the amount of cloth based on the data learned by the learning module 213.
  • the determination of the amount of cloth may be an operation of classifying the cloth injected into the washing tub 120 into a plurality of preset levels according to the weight (load).
  • the fabric is classified into five stages (levels), and the load (kg) corresponding to each stage is shown in Table 1 below.
  • Table 1 statistically shows the number of households constituting the household when a household puts the amount of cloth into a laundry treatment machine.
  • Determination of the quality of the fabric is to classify the fabric injected into the washing tub 120 according to a preset standard, such a standard may be the material of the fabric, the degree of softness or stiffness, the moisture content, the volume difference between the dry cloth and the wet cloth. It may be a volume difference between the liver and the like.
  • the recognition module 216 Based on the current output current value obtained from the output current detector (E), the recognition module 216 corresponds to which of the plurality of washing steps the cloth injected into the washing tub 120 corresponds to, and which classification is the cloth at this time ( In other words, it is possible to determine the quality of each fabric).
  • fabrics are classified into 5 stages (levels), and the types corresponding to each stage are shown in Table 2 below, and referring to Table 2 below, the soft, weakly durable clothes series are 1 level, 1 level clothes series. Clothing with stronger durability is at Level 3, and at Level 3 is more durable than clothing at Level 3, and is at Level 5 for stiff clothes, Level 2 and Level 3 and Level 5 above when the 1st and 3rd levels are mixed. It can be judged as level 4 when the clothes series of are mixed.
  • the recognition module 216 may mount artificial neural networks (ANNs) previously learned by machine learning. This artificial neural network may be updated by the learning module 213.
  • ANNs artificial neural networks
  • the recognition module 216 may determine the amount and quality of fabric based on the artificial neural network. As in the embodiment, when the steps of the fabric are classified into five steps, the recognition module 216 uses the current output current value detected by the output current detection unit E as input data of the artificial neural network (ANN). You can determine the stage in which the quantity belongs, and furthermore, the stage in which the foam belongs.
  • ANN artificial neural network
  • the recognition module 216 may include an artificial neural network (ANN) that has been trained to classify fabrics and fabrics into stages according to predetermined criteria, respectively.
  • ANN artificial neural network
  • the capacity recognition module 216 is a deep neural network (DNN) such as a convolutional neural network (CNN), a recurrent neural network (RNN), and a deep belief network (DBN) learned by deep learning. ) Can be included.
  • DNN deep neural network
  • CNN convolutional neural network
  • RNN recurrent neural network
  • DBN deep belief network learned by deep learning.
  • RNN Recurrent Neural Network
  • RNN is widely used for natural language processing, etc., and is an effective structure for processing time-series data that changes with the passage of time, and can construct an artificial neural network structure by stacking layers every moment. .
  • DBN Deep Belief Network
  • RBM Restricted Boltzman Machine
  • DBN Deep Belief Network
  • CNN Convolutional Neural Network
  • learning of the artificial neural network can be accomplished by adjusting the weight of the connection line between nodes (if necessary, adjusting the bias value) so that a desired output is produced for a given input.
  • the artificial neural network can continuously update the weight value by learning. Methods such as back propagation may be used for learning of artificial neural networks.
  • the recognition module 216 takes the current output current value as input data, and based on weights between nodes included in the deep neural network (DNN), the output from the output layer is input to the washing tub 120. You can determine at least one of the amount and quality of the fabric.
  • DNN deep neural network
  • 9 is a schematic diagram showing an example of an artificial neural network.
  • 10 is a schematic diagram showing a process of determining fabric quantity and fabric quality using a current output current value of the motor, divided into a learning process and a recognition process.
  • Deep learning technology a kind of machine learning, is to learn by going down to the deep level in multiple stages based on data.
  • Deep learning may represent a set of machine learning algorithms that extract core data from a plurality of data while sequentially passing through hidden layers.
  • the deep learning structure may include an artificial neural network (ANN), for example, the deep learning structure consists of a deep neural network (DNN) such as a convolutional neural network (CNN), a recurrent neural network (RNN), and a deep belief network (DBN).
  • DNN deep neural network
  • CNN convolutional neural network
  • RNN recurrent neural network
  • DBN deep belief network
  • an artificial neural network may include an input layer, a hidden layer, and an output layer. Having multiple hidden layers is called a Deep Neural Network (DNN).
  • DNN Deep Neural Network
  • Each layer includes a plurality of nodes, and each layer is associated with the next layer. Nodes can be connected to each other with a weight.
  • An output from an arbitrary node belonging to the first hidden layer 1 becomes an input of at least one node belonging to the second hidden layer 2.
  • the input of each node may be a value to which a weight is applied to the output of the node of the previous layer.
  • Weight may mean the strength of the connection between nodes.
  • the deep learning process can also be viewed as a process of finding an appropriate weight.
  • the computer distinguishes light and dark pixels from the input image according to the brightness of the pixels, and distinguishes simple shapes such as borders and edges, and then a little more. Can distinguish between complex shapes and objects.
  • the computer can grasp the form that defines the human face. In this way, the specificity of the feature (regulation of the shape of the human face) is finally obtained from the output layer through the hidden layers of the middle layer.
  • the memory 240 may store input data for detecting a fabric amount and data for learning a deep neural network (DNN).
  • DNN deep neural network
  • motor speed data and/or speed data acquired by the speed detector 245 may be summed for each predetermined section or data processed by calculation may be stored.
  • the memory 240 may store weights and biases constituting a deep neural network (DNN) structure.
  • weights and biases constituting the deep neural network structure may be stored in an embedded memory of the recognition module 216.
  • the learning module 213 may perform learning by using the current output current value detected through the output current detection unit E as training data.
  • the learning module 213 adds the determination result to the database whenever it recognizes or determines the amount of fabric and/or fabric, and updates the structure of a deep neural network (DNN) such as weight or bias, or After a predetermined number of training data is secured, a deep neural network (DNN) structure such as weight may be updated by performing a learning process using the acquired training data.
  • DNN deep neural network
  • the laundry treatment device transmits the current output current data obtained from the output current detection unit E through the communication unit 270 to the server 500 connected to the communication network, and transmits the current output current data obtained from the server 500 to the machine. You can receive data related to running.
  • the laundry processing device may update the artificial neural network based on machine learning-related data received from the server.
  • the controller 210 performs learning using a deep neural network using the current output current data acquired by the output current detection unit E, and is related to machine learning. Data can be derived.
  • 11 is a graph (a) showing the current output current value detected by the current detection unit and a graph (b) showing average values obtained by processing a moving average filter.
  • 12 is a graph showing current values by the current detector.
  • 13 is a graph showing values processed to use the current values of the graph shown in FIG. 12 as input data of an artificial neural network.
  • 14 is a flow chart showing a control method of the laundry treatment device. Hereinafter, a method of determining the fabric amount and fabric quality will be described with reference to FIGS. 11 to 14.
  • the controller 210 controls the motor 230 to rotate at a preset target rotation speed (S1, S2, S3, S4, S5). While the motor 230 is rotating, the rotational speed of the washing tub 120 (or the motor 230) is detected by the speed detection unit 245 (S2).
  • the target rotational speed is the rotational speed of the washing tub 120 capable of maintaining a state attached to the carriage drum 122 when the washing tub 120 maintains the target rotational speed and continuously rotates one or more rotations in one direction. It can be determined as. That is, the target rotation speed may be determined as the rotation speed of the washing tub 120 capable of rotating integrally with the carriage drum 122.
  • a centrifugal force acting on the fabric by the rotation of the washing tub 120 may be greater than the gravity acting on the fabric.
  • the target rotation speed may be 60 to 80 rpm, preferably 80 rpm.
  • the carriage is flowed in the drum 122. That is, the carriage rises to a predetermined height by the rotation of the drum 122 and then falls.
  • the target rotation speed may be determined based on a state in which water is supplied into a storage tank (not shown) and the washing tank 120 is partially submerged in water. That is, when the washing tub 120 is partially submerged in water and rotated at the target rotational speed, the fabric may flow. In other words, while the washing tub 120 is rotating, the carriage may not always stick to the drum 122, but may rise to a predetermined height and then fall.
  • Current output current values used to determine the amount and quality of the fabric include those taken in a section in which the flow of the fabric occurs during the rotation of the washing tub 120. That is, the control unit 210 may take necessary current output current values based on the rotational speed of the washing tub 120 (or the rotational speed of the motor 230) detected by the speed detection unit 245.
  • control unit 210 instructs the motor driving unit 220 to accelerate the motor 230, and then when the rotational speed detected by the speed detection unit 245 reaches a preset first rotational speed V1,
  • the current output current value from that time may be stored in the memory 240 (S3 to S4).
  • the control unit 210 When the rotational speed (V) of the washing tub 120 reaches a preset second rotational speed (V2), the control unit 210 does not store the current output current value anymore and processes the current output current value. It can be done (S5 to S6).
  • the second rotational speed V2 is the aforementioned target rotational speed.
  • the acceleration slope in the section accelerating from the first rotation speed V1 to the second rotation speed V2 may be constant. In order to increase the reliability of detection of current pattern change, it is preferable that the acceleration slope is kept constant.
  • the acceleration slope should not be too high so that the change trend of the fabric flow in the washing tub 120 can be clearly seen.
  • the acceleration slope is preferably 1.5 to 2.5 rpm/s, more preferably 2.0 rpm/s, but is not necessarily limited thereto.
  • the acceleration slope may have a value as small as possible within a range that can be controlled by the controller 210.
  • the current output current values Iq obtained at preset time points are processed according to a predetermined algorithm, and the input layer of the artificial neural network is This is a process of generating input data (In1, In2, In3, In4,%) (S6).
  • This process may include obtaining an average of current output current values Iq, and generating input data of an artificial neural network by processing the obtained average values according to a preset parsing rule.
  • the number of input data processed by the parsing rule is less than the number of average values.
  • control unit 210 may extract 211 a current value at regular time intervals through the output current detection unit E.
  • a total of 545 current output current values were obtained at regular time intervals.
  • the control unit 210 may accumulate 212 and average the current output current values obtained in this way every predetermined time interval.
  • the controller 210 may use a moving average filter.
  • Moving average is to calculate the average by moving the interval so that the change of the trend can be known. For example, current output current values Iq1, Iq2, Iq2...
  • Iqn M1 is obtained by averaging Iq1 to Iql(l ⁇ n), and from Iqm(m>l) to Iqm+s-1 (s is the number of Iq used to calculate each moving average). Average and find M2. In this way, moving averages can be obtained while continuing to move through the section.
  • the control unit 210 obtains 50 moving averages from 545 current output current values Iq using a moving average filter.
  • the controller 210 may generate input data (In1, In2, In3, In4 ”) by processing the current output current A and moving averages according to a preset parsing rule.
  • the parsing rule may be configured to select a section in which the final input data is obtained so that the desired feature (packing amount/fog quality) is clearly revealed.
  • a total of 14 input data are generated, and the input data are 9 current output current values (16th to 24th current values) obtained at the beginning of acceleration of the motor 230.
  • the calculation can be processed more quickly than the sum of the current output current values in each section.
  • the input data (In1, In2, In3, In4, ... In14) thus obtained become input values of each node of the input layer.
  • the weights and biases given to nodes constituting the artificial neural network are determined through machine learning, and this machine learning is repeated based on a current pattern or current output current values.
  • the current pattern (or current output current value) reflects the characteristics of the cloth and/or cloth as described above, an accurate result (that is, the correct cloth and cloth put into the current washing tub 120) is obtained. It is possible to set improved or accurate weights and bias values by performing machine learning or neural network learning 213 on data previously stored until derived or added by the operation of the laundry treatment device.
  • the output of the output layer will reflect the packing-fog information, and based on the node that outputs the largest value among the nodes of the output layer, the control unit 210 The amount and/or quality of the fabric can be determined.
  • the control unit 210 may input the input data generated in step S6 to the artificial neural network, and obtain the fabric amount and/or the fabric as an output from the output layer (S7).
  • the controller 210 may set a washing algorithm based on the amount of cloth and/or cloth obtained in step S7, and control the operation of the laundry treatment device according to the setting.
  • the washing algorithm may include a water level, an execution time of washing, rinsing, spinning, drying, etc., a driving pattern of a motor in each stroke (eg, rotation speed, rotation time, acceleration, braking), and the like.
  • 15 is a graph in which current patterns for each load are superimposed.
  • 16 is a graph for classifying current patterns corresponding to a load of 0 to 6 kg in FIG. 15.
  • 17 is a graph for classifying current patterns corresponding to a load of 7 to 9 kg in FIG. 15.
  • P0 to P9 shown in these figures represent loads (packages) of 0 to 9 kg, respectively.
  • current patterns P0 to P6 at a load of 0 to 6 kg and current patterns P7 to P9 at 7 to 9 kg show a distinct difference. That is, in the case of a large amount of fabric (7 to 9 kg in the embodiment), it can be seen that the current output current value periodically increases or decreases (or vibrates) at the beginning of the acceleration of the washing tub 120. This is because the motion of some fabrics is constrained by the door 113, and when the washing tub 120 interferes with the constrained fabric, the load of the motor 230 increases, and when the interference weakens or disappears, the load of the motor 230 is again Because it decreases. That is, a change in the load of the motor 230 due to the restraint of the cloth is generated in response to the rotation period of the washing tub 120.
  • the load variation pattern may be learned through machine learning, and the learning result may be converted into a database and stored in the memory 240. Using these learning results, an artificial neural network can be constructed. Based on the artificial neural network configured in this way, the control unit 210 may determine the fabric constraint (or trapping) based on the output of the output layer.
  • FIG. 18 is a flow chart showing a control method of the laundry treatment device according to the first embodiment of the present invention.
  • a control method of a laundry treatment apparatus according to a first embodiment of the present invention will be described with reference to FIG. 18.
  • the controller 210 performs learning based on a deep neural network based on the output current from the output current detection unit E, and detects the fabric of the laundry cloth in the washing tub 120 according to the learning. (S1810).
  • steps S1 to S7 of FIG. 14 are performed, and accordingly, the cloth quality of the laundry cloth in the washing tub 120 may be detected.
  • control unit 210 may receive laundry cloth information from the input unit 117 or the communication unit 270 (S1820).
  • control unit 210 may include an input signal from an operation key of the input unit 117, a user voice signal from a microphone of the input unit 117, or an input signal by operation from the mobile terminal 600, or A user voice signal from a microphone of the mobile terminal 600 may be received as an input signal.
  • control unit 210 may extract laundry cloth information from the input signal.
  • the laundry cloth information may be a concept including information on the degree of contamination of the laundry cloth, irrespective of the cloth material information, the cloth amount information, and the like.
  • the pollution degree information may be a concept including information on a pollution source, information on a location of pollution, and information on an amount of pollution.
  • FIG. 19A shows a first user voice 1910 such as “there is a little spot on the bottom”, a second user voice 1920 such as “there is a lot of spots on the bottom", or “bottoms, tops,” A third user's voice 1930, such as “There is a lot of spots on”, is received, and laundry cloth information is wirelessly transmitted to the laundry treatment device 100 through voice recognition.
  • the first user voices 1910 to the third user voices 1930 are examples of pollution level information, including information on a location of pollution, information on a quantity of pollution, and the like.
  • controller 210 may perform classification based on the fabric information and the laundry fabric information (S1830).
  • FIG. 19B illustrates a classification table 1940 classified into 15 stages by dividing the foam into 1 to 5 stages, and by dividing the pollution degree into 3 stages by strong, medium, and weak.
  • the controller 210 may perform classification as shown in FIG. 19B by using the internal learning module 213 or the recognition module 216.
  • classification as shown in FIG. 19B may be performed using a deep neural network.
  • control unit 210 may perform classification as shown in FIG. 19B using an external server 500.
  • the controller 210 may determine a driving course based on the classification (S1840). Accordingly, it is possible to provide the calculated fabric information of the laundry cloth and optimal driving course information corresponding to the input laundry cloth information.
  • the controller 210 may perform driving according to the determined driving course (S1850). Then, driving course information may be output as a display 118 or sound (S1650).
  • 19C is a diagram illustrating that a driving course is determined according to classified information.
  • first driving course such as "Foam wear-resistant laundry”
  • second driving course such as “Forma wear + washing power strengthening”
  • third driving course such as “Laundry strengthening washing”
  • These first to third driving courses may be an added driving course that has not been stored in the memory 240, that is, a new driving course.
  • the controller 210 may control to perform the driving course according to the determined driving course. That is, according to the set driving course, while performing the washing, rinsing, and spinning, rotational speed, rotational time, applied current, and the like may be varied.
  • the determined driving course information may be output as shown in FIG. 19D.
  • 19D illustrates outputting various driving courses as sound through an audio output unit of the laundry treatment device 100.
  • FIG. 19D(a) outputs information on the first driving course of FIG. 19C as sound 1962
  • FIG. 19D(b) outputs information on the second driving course of FIG. 19C as sound 1964
  • (c) of FIG. 19D exemplifies outputting information on the third driving course of FIG. 19C as a sound 1966.
  • the user can intuitively recognize the set driving course information.
  • the control unit 210 performs learning based on a deep neural network, based on the output current from the output current detection unit E, and according to the learning, the washing cloth in the washing tub 120 is fabricated.
  • the information is determined, the laundry cloth information is received from the input unit 117 or the communication unit 270, and driving course information is determined based on the laundry cloth information and the cloth information of the laundry cloth. Accordingly, it is possible to provide the calculated fabric information of the laundry cloth and optimal driving course information corresponding to the input laundry cloth information.
  • the control unit 210 performs learning based on a deep neural network, based on the output current from the output current detection unit E, and according to the learning, the washing cloth in the washing tub 120 is fabricated.
  • the information is determined, based on the input signal from the input unit 117 or the communication unit 270, the contamination level information of the laundry cloth is determined, and the driving course information is determined based on the contamination level information and the cloth quality information of the laundry cloth. Accordingly, according to learning based on a deep neural network, it is possible to determine the cloth quality of the laundry cloth, and provide information on the cloth material and optimal driving course information corresponding to the input pollution degree information.
  • the controller 210 performs learning based on a deep neural network based on an input signal from the input unit 117 or the communication unit 270, and according to the learning, the pollution degree of the laundry cloth Judge the information. Accordingly, according to learning based on a deep neural network, it is possible to provide information on an optimal driving course corresponding to the cloth quality and pollution level information of the laundry cloth.
  • the controller 210 performs a driving operation based on the determined driving course information and controls to output driving course information. Accordingly, the driving course information can be easily recognized by the user.
  • the controller 210 may control the determined driving course information to be stored in the memory 240. Accordingly, it is possible to store new driving course information.
  • control unit 210 determines the contamination level information of the laundry cloth based on sound information from the input unit 117 or the communication unit 270. Accordingly, it is possible to easily determine the pollution level information based on the user's voice.
  • the controller 210 performs classification for setting a driving course based on the level of the cloth quality information and the level of the pollution level information in the washing tub 120, and classifies On the basis of, the driving course information is determined. Accordingly, it is possible to provide optimal driving course information based on the foam information and the pollution degree information.
  • control unit 210 accesses the external server 500, performs a search based on classification information according to the classification, and based on the search result, Determine driving course information. Accordingly, driving course information can be provided using the external server 500.
  • the controller 210 accesses the external server 500 and performs a search based on the level of the cloth quality information of the laundry cloth in the washing tub 120 and the level of the pollution level information. , Based on the search results, determine the driving course information. Accordingly, driving course information can be provided using the external server 500.
  • the controller 210 performs learning based on a deep neural network based on the output current from the output current detection unit E, and according to the learning, the washing cloth in the washing tub 120
  • determine the cloth quality information and the cloth information based on the input signal from the input unit 117 or the communication unit 270, determine the pollution level information of the laundry cloth, and based on the pollution level information and the cloth quality information and the cloth amount information , Determine driving course information. Accordingly, it is possible to provide information on an optimum driving course corresponding to the cloth information, the cloth amount information, and the pollution degree information.
  • control unit 210 performs classification for setting a driving course based on the level of the cloth information of the laundry cloth, the level of the cloth amount information of the cloth, and the level of the pollution degree information. And, based on the classification, driving course information is determined. Accordingly, it is possible to provide information on an optimum driving course corresponding to the cloth information, the cloth amount information, and the pollution degree information.
  • the controller 210 accesses the external server 500, performs a search based on classification information according to the classification, and determines driving course information based on the search result. . Accordingly, driving course information can be provided using the external server 500.
  • control unit 210 is connected to the external server 500, based on the level of the cloth information of the laundry cloth, the level of the cloth amount information, and the level of the pollution degree information. A search is performed and, based on the search result, driving course information is determined. Accordingly, driving course information can be provided using the external server 500.
  • the laundry treatment apparatus is not limited to the configuration and method of the embodiments described as described above, but the embodiments are all or part of each embodiment so that various modifications can be made. May be configured by selectively combining.
  • the method of operating a laundry treatment device may be implemented as a code readable by a processor on a recording medium readable by a processor provided in the laundry treatment device.
  • the processor-readable recording medium includes all types of recording devices that store data that can be read by the processor.

Landscapes

  • Engineering & Computer Science (AREA)
  • Textile Engineering (AREA)
  • Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Evolutionary Computation (AREA)
  • General Health & Medical Sciences (AREA)
  • Biomedical Technology (AREA)
  • Biophysics (AREA)
  • Computational Linguistics (AREA)
  • Data Mining & Analysis (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Artificial Intelligence (AREA)
  • Molecular Biology (AREA)
  • Computing Systems (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Mathematical Physics (AREA)
  • Software Systems (AREA)
  • Health & Medical Sciences (AREA)
  • Control Of Washing Machine And Dryer (AREA)

Abstract

La présente invention concerne une machine de traitement de linge à intelligence artificielle. Une machine de traitement de linge à intelligence artificielle, selon un mode de réalisation de la présente invention, comprend : un bâti ; une cuve de lavage à l'intérieur du bâti ; un moteur pour faire tourner la cuve de lavage ; une unité d'entraînement qui entraîne le moteur et qui comprend une unité de détection de courant de sortie pour détecter un courant de sortie circulant à travers le moteur ; une unité de communication pour communiquer avec un terminal mobile externe ou un serveur ; une unité d'entrée pour générer un signal d'entrée ; et une unité de commande pour commander l'unité d'entraînement. L'unité de commande : effectue, sur la base du courant de sortie de l'unité de détection de courant de sortie, un apprentissage basé sur un réseau neuronal profond ; détermine, selon l'apprentissage, des informations sur la matière du linge dans la cuve de lavage ; reçoit des informations sur le linge de l'unité d'entrée ou de l'unité de communication ; et détermine, sur la base des informations de linge et des informations de matière de linge, des informations de cycle d'entraînement. En conséquence, il est possible de fournir des informations de cycle d'entrainement optimales correspondant aux informations de matière de linge calculées et aux informations de linge entrées.
PCT/KR2019/011430 2019-09-04 2019-09-04 Appareil de traitement de linge à intelligence artificielle WO2021045258A1 (fr)

Priority Applications (1)

Application Number Priority Date Filing Date Title
PCT/KR2019/011430 WO2021045258A1 (fr) 2019-09-04 2019-09-04 Appareil de traitement de linge à intelligence artificielle

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/KR2019/011430 WO2021045258A1 (fr) 2019-09-04 2019-09-04 Appareil de traitement de linge à intelligence artificielle

Publications (1)

Publication Number Publication Date
WO2021045258A1 true WO2021045258A1 (fr) 2021-03-11

Family

ID=74852629

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/KR2019/011430 WO2021045258A1 (fr) 2019-09-04 2019-09-04 Appareil de traitement de linge à intelligence artificielle

Country Status (1)

Country Link
WO (1) WO2021045258A1 (fr)

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPH07171289A (ja) * 1993-12-21 1995-07-11 Matsushita Electric Ind Co Ltd 洗濯機
KR19990062394A (ko) * 1997-12-10 1999-07-26 구자홍 전자동 세탁기의 포질 감지방법
KR20120038271A (ko) * 2010-10-13 2012-04-23 삼성전자주식회사 세탁기의 제어 방법
KR20190081856A (ko) * 2017-12-29 2019-07-09 엘지전자 주식회사 세탁기 및 세탁기의 동작방법
KR20190092334A (ko) * 2019-07-19 2019-08-07 엘지전자 주식회사 세탁물 처리 장치 및 그의 구동 방법

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPH07171289A (ja) * 1993-12-21 1995-07-11 Matsushita Electric Ind Co Ltd 洗濯機
KR19990062394A (ko) * 1997-12-10 1999-07-26 구자홍 전자동 세탁기의 포질 감지방법
KR20120038271A (ko) * 2010-10-13 2012-04-23 삼성전자주식회사 세탁기의 제어 방법
KR20190081856A (ko) * 2017-12-29 2019-07-09 엘지전자 주식회사 세탁기 및 세탁기의 동작방법
KR20190092334A (ko) * 2019-07-19 2019-08-07 엘지전자 주식회사 세탁물 처리 장치 및 그의 구동 방법

Similar Documents

Publication Publication Date Title
WO2021045259A1 (fr) Séchoir à intelligence artificielle
WO2017183883A1 (fr) Appareil d'entraînement de pompe de vidange, et appareil de traitement du linge le comprenant
WO2020159119A1 (fr) Machine à laver et son procédé de commande
WO2012005510A2 (fr) Machine à laver
WO2012005511A2 (fr) Machine à laver et son procédé de commande
WO2020046080A2 (fr) Machine à laver à intelligence artificielle et procédé pour commander une machine à laver à intelligence artificielle
WO2018199581A1 (fr) Dispositif de traitement de linge
WO2020060269A1 (fr) Procédé de commande d'appareil pour le linge
WO2020159145A1 (fr) Machine à laver et procédé de commande de machine à laver
WO2018062827A1 (fr) Lave-linge et procédé de commande dudit lave-linge
WO2019107871A1 (fr) Système d'entraînement de moteur et procédé de commande de ce dernier
WO2020046078A2 (fr) Machine à laver à intelligence artificielle et procédé pour commander une machine à laver à intelligence artificielle
WO2020050691A1 (fr) Appareil d'entraînement de pompe de vidange et machine de traitement du linge le comprenant
WO2021045258A1 (fr) Appareil de traitement de linge à intelligence artificielle
WO2020009533A1 (fr) Appareil de traitement de linge et procédé de commande associé
WO2020009536A1 (fr) Machine de traitement de linge et son procédé de commande
WO2020046079A2 (fr) Machine à laver à intelligence artificielle et procédé de commande d'une machine à laver à intelligence artificielle
WO2021025196A1 (fr) Appareil de traitement de vêtements doté d'une caméra et son procédé de commande
WO2019088619A1 (fr) Appareil de traitement de vêtements, et son procédé de commande
WO2020009535A1 (fr) Appareil de manipulation de linge
WO2022169122A1 (fr) Lave-linge et son procédé de commande
WO2020009526A1 (fr) Machine à laver le linge
WO2020009521A1 (fr) Dispositif d'entraînement de pompe de circulation et appareil de traitement de linge
WO2019107872A1 (fr) Appareil d'entraînement de moteur et appareil électroménager le comportant
WO2020009530A1 (fr) Appareil de manipulation de linge

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 19944571

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 19944571

Country of ref document: EP

Kind code of ref document: A1