JP2977968B2 - Washing machine - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a washing machine for controlling a washing operation based on a plurality of data indicating the state of laundry or the like.

[0002]

2. Description of the Related Art A general automatic washing machine has a washing time,
A plurality of washing courses such as a standard course, a stubble stain course, a soaking course, and a hand washing course, each having a different washing water flow, are stored in the storage means in advance, and a desired washing course is selected from the stored courses using the course switching key. It is intended to be executed.

However, in such a general automatic washing machine, the condition of the laundry, that is, the amount of cloth as a cloth condition, the amount of dirt as a cloth condition and the dirt condition, and the dirt quality vary widely depending on household circumstances. Therefore, it is impossible to cope with these in a predetermined washing course, and it is not possible to obtain an optimum washing effect.

For this reason, recently, a plurality of data such as the state of the laundry, that is, the amount of cloth, the amount of cloth and the amount of dirt, and the amount of dirt are detected by a sensor, and a plurality of input data from the sensor is used as control means. It is considered that the optimum washing water flow and washing time are determined based on a plurality of input data and executed by the microcomputer.

[0005] In this case, specifically, first, cloth volume data is detected by a capacity sensor under a detected water level lower than the low water level and a detected water flow weaker than the weak water flow. After water is supplied to the water level, the middle water level or the high water level, the fabric data is detected by the rotation speed sensor by stirring with the standard water flow, and the weak water flow, the standard water flow, or the strong water flow is selectively executed according to the fabric data, Thereafter, after the generation of the water flow is stopped, the amount of dirt and the dirt quality data are detected by the transmittance sensor.

[0006]

In a configuration in which any one of the weak water flow, the standard water flow, and the strong water flow is selected and executed according to the above-described cloth data, the amount of dirt and the dirt quality are detected by the transmittance sensor. It seems that even with the agitated water flow in the previous stage for making the water flow, a reasonable water flow such as a weak water flow, a standard water flow, or a strong water flow according to the fabric can be executed to contribute to the washing effect at first glance. The degree of dirt removal from the laundry varies depending on the strength.

[0007] Therefore, when subsequently detecting the amount of dirt and the data of dirt quality by the transmittance sensor, it is necessary to change the basic data of the criterion and the processing program according to the strength of the water flow. The number of basic data stored in the memory increases, causing a problem that the program becomes complicated.

The present invention has been made in view of the above circumstances. An object of the present invention is to obtain a plurality of data corresponding to the state of laundry or the like from a sensor, thereby reducing the basic data of the control means and reducing the program. It is to provide a washing machine that is easy.

[0009]

SUMMARY OF THE INVENTION A washing machine according to the present invention is capable of controlling the state of a laundry or the like including a cloth amount, a cloth condition of a cloth and a dirt condition of a dirt amount in response to intermittent generation of a water flow. Provided a required number of sensors for sequentially detecting a plurality of data shown, and a control means for controlling the washing operation based on a plurality of input data from these sensors, wherein the control means, cloth data When obtaining the dirt amount data from the sensor, it is characterized in that a constant water flow is generated .

[0010]

According to the washing machine of the present invention, the control means
Since the water flow is controlled to be constant when the cloth data and the dirt amount data are obtained from the sensor, there is no difference in the degree of dirt removal of the laundry due to the difference in the water flow. We need less.

[0011]

An embodiment of the present invention will be described below with reference to the drawings.

First, an overall configuration of a fully automatic washing machine as a washing machine will be described with reference to FIG. Outer tub 2 inside outer box 1
Are arranged. The outer tub 2 is provided with a drive mechanism 4 mainly composed of a motor 3 and a drain valve 5 and a drain hose 6 located at a lower part of the outer tub. Further, the outer tank 2 has a driving mechanism 4 located inside thereof.
The inner tub 7 which is a washing tub that is rotated at the time of spin-drying is provided. Incidentally, a large number of dehydration holes 8 are formed in the peripheral wall of the inner tank 7. Further, in the inner tank 7, there is provided a stirring body 9 which is driven to rotate at the time of washing and rinsing by the drive mechanism 4.

On the other hand, a top cover 10 is mounted on the outer box 1, and inside the rear portion of the top cover 10,
A water supply valve 11 for supplying water to the inner tank 7 (in the outer tank 2) is provided, and a water level sensor 12 for detecting a water level in the outer tank 2 (in the inner tank 7) is provided. Further, inside the front portion of the top cover 10, a microcomputer 13 is disposed as control means that operates as described later.

Further, at the front end of the top cover 10, an operation panel 14 as shown in FIG. The operation panel 14 is provided with a washing course display section 15 in which “standard”, “stubble”, “place”, “separate washing”, “speed”, “big” and “large” Characters indicating the respective courses of “hand washing” are displayed, and light emitting diodes 15a and 15
b, 15c, 15d, 15e, 15f and 15g are additionally provided.

On the operation panel 14, a course switching key 16 is mounted, a start key (also used as a pause key) 17 is mounted, and a segment type display 18 is mounted. A light emitting diode 19 for sensor display is mounted.

A transmittance sensor 20 is attached to the inner bottom of the outer tank 2 described above. This transmittance sensor 20
As shown in FIG. 5, a light emitting element 22 composed of, for example, a light emitting diode and a light receiving element 23 composed of, for example, a phototransistor are disposed so as to be opposed to each other in a case 21. The portions 21a and 21b facing the light receiving element 23 are formed of a translucent material. In this case, the light receiving signal of the light receiving element 23 is A / D converted and output as a transmittance signal S20 from the transmittance sensor 20.

A temperature sensor 24 is provided on the outer bottom of the outer tank 2.
Are arranged. As shown in FIG. 6, the temperature sensor 24 has a configuration in which a thermistor 26 is disposed in an annular ridge 25 formed in the outer tank 2 and embedded and fixed with a resin 25a. And the detection signal of the thermistor 26 is
A / D conversion is performed and output as a water temperature detection signal S24.

Now, the electrical configuration will be described with reference to FIG. The input port of the microcomputer 13 is connected to an output terminal of a key operation input section 27 for outputting signals based on various key operations of the operation panel 14, and the other input ports are a water level sensor 12 and a transmittance sensor. 20 and the output terminal of the temperature sensor 24 are connected. Further, one input port of the microcomputer 13 is connected to an output terminal of a rotation speed sensor 28 for detecting the rotation speed of the motor 3, and the other input port is connected to an output terminal of a capacitance sensor 29.

The output port of the microcomputer 13 is connected to the motor 3 and the water supply valve 11 via drive circuits 30 and 31.
And the other output ports are connected to the light emitting diodes 15a to 15g and 19 via the drive circuit 32. One output port of the microcomputer 13 is connected to the display 18 via a drive circuit 33, and the other output port is connected to an electromagnet 35 via a drive circuit 34. In this case, the electromagnet 35
It is for the clutch of the drive mechanism 4 and for the drain valve 5.

In the above case, when the course switch key 16 shown in FIG. 4 is pressed, the microcomputer 13 sends a signal to the light emitting diodes 15a to 15g via the drive circuit 32, and the course switch key 16 is pressed. The light emission of the light-emitting diodes 15a to 15g is shifted every time or during the pressing operation, and when the pressing operation of the course switching key 16 is released, the shift of the light emission is stopped. Then, it is determined that the washing course corresponding to the light emitting diode emitting light at that time is selected.

Now, the configuration of the capacitance sensor 29 will be described with reference to FIG. The drive circuit 30 of the motor 3 is composed of bidirectional three-terminal thyristors 36 and 37, and the power supply voltage Vac of the AC power supply 38 is applied to the motor 3 via these thyristors 36 and 37. Motor 3
Is detected by the current transformer 39, and the detected current is converted into a voltage and supplied to the non-inverting input terminal (+) of the comparator 40. The inverting input terminal (-) of the comparator 40 is set to the ground potential. The power supply voltage Vac is divided and supplied to the non-inverting input terminal (+) of the comparator 41, and the inverting input terminal (-) of the comparator 41 is set to the ground potential. The output signals of the comparators 40 and 41 are supplied to the input terminal of an exclusive OR circuit 42, the output signal of the exclusive OR circuit 42 is supplied to an integration circuit 43, and the output signal of the integration circuit 43 is A / D conversion circuit 44.

Thus, the phase difference θm of the current Im flowing through the motor 3 with respect to the power supply voltage Vac decreases as the output of the motor 3 increases.
Is proportional to the laundry amount (cloth amount). Therefore, the amount of cloth can be detected by detecting the phase difference θm.

That is, as shown in FIG. 8, comparators 40 and 41 output pulses Vv according to current Im and power supply voltage Vac.
And Vi, and the exclusive OR circuit 42 outputs a phase difference pulse Vθ in response to this. Therefore, if the integration circuit 43 detects the average voltage Vθa of the phase difference pulse Vθ, the A / D conversion circuit 44 outputs the digitalized amount signal S29.

On the other hand, the microcomputer 13
The neural network 45 for neural control as shown in FIG. The neural network 45 is actually simulated by software using the microcomputer 13, but is shown here as being constituted by hardware for convenience of explanation.

That is, the neural network 45 includes an input layer I composed of seven units I1 to I7, an intermediate layer J composed of five units J1 to J5, and two units K
1 and K2, and an input layer I.
The units I1 to I7 of the intermediate layer J and the units J1 to J5 of the intermediate layer J are connected to each other by a link, and the units J1 to J5 of the intermediate layer J and the output layers K1 and K2 are connected to each other by a link.

Next, the operation of this embodiment will be described with reference to FIGS.
This will be described with reference to FIGS.

First, the laundry is put into the inner tub 7 together with the detergent, and the light emitting diode 15a is turned on by pressing the course switching key 16 shown in FIG. 4, so that the "standard" course is set as the washing course. select.

In the following, referring to FIG.
7 is pressed (time T0), the microcomputer 13 sends the “neuro sensor” via the drive circuit 32.
Is supplied with electricity to emit light. When the start key 17 is pressed, a "washing" step is first performed, and the microcomputer 13
Electricity is supplied to the water supply valve 11 via the drive circuit 31, and water is supplied into the inner tank 7 (in the outer tank 2). Thereafter, when the water level in the inner tub 7 becomes “low” among the “high”, “medium”, “low” and “small” water levels, the microcomputer 13 switches the water supply valve 11 based on the signal of the water level sensor 12. Is turned off (T1). At this time, the microcomputer 13 transmits the transmittance signal S20 from the transmittance sensor 20.
Is read and stored in the RAM as detection data A1. The microcomputer 13 supplies power to the motor 3 via the drive circuit 30.

In this case, the motor 3 is repeatedly energized for 0.6 seconds (forward rotation), cut off for 1 second, energized for 0.6 seconds (reverse rotation) and cut off for 1 second to rotate the agitator 9. As a result, a detected water flow is generated. Such detected water flow is about 1
The microcomputer 13 generates the amount signal S2 of the capacitance sensor 29 during the period of 5 to 20 seconds.
9 is read and stored in the RAM as the cloth amount data D1.
Based on this, the cloth amount is determined. After that, the microcomputer 13 cuts off the electric power of the motor 3 and again supplies the water supply valve 1.
1 (time T2).

The determination of the cloth amount by the microcomputer 13 is to determine three levels of "low", "medium" and "high", and the water level is determined to be "low", "low", "high" or "low", respectively.
Set "medium" and "high" water levels. Thereafter, when the water level in the inner tub 7 reaches the set “low”, “medium” or “high” water level, the microcomputer 13 turns off the water supply valve 11 and energizes the motor 3 (time T3). In this case, the motor 3 is repeatedly turned on for 1 second (forward rotation), cut off for 0.9 seconds, turned on for 1 second (reverse rotation), and cut off for 0.9 seconds.
As a result, a standard water flow as a constant water flow is generated. When this operation is performed for a predetermined time, the motor 13 is turned off by the microcomputer 13 (time T4), and thereafter, is stopped for a relatively short time (for example, 15 seconds). During the suspension period (between times T4 and T5), the microcomputer 13 controls the light emitting diode 19 of the “neuro sensor”.
Is changed from continuous lighting to blinking lighting so that the user does not misunderstand that the stop is due to a failure.

When the motor 3, that is, the stirring body 9, is stopped, the movement of the washing liquid stops, bubbles and the like rise, and the vibration of the outer tub 2 also stops, so that the permeability of the washing liquid is stabilized. I do. Then, the microcomputer 13 reads the transmittance signal S29 of the transmittance sensor 29 and stores it in the RAM as detection data A2. Then, as shown in FIG. 11, the microcomputer 13 compares the above-described detection data A1 with the current detection data A2, and if the difference is small, determines that the supplied detergent is a liquid detergent,
If the difference is large, it is determined that the detergent is a powder detergent, and the result is stored in the RAM as detergent quality data D6.

That is, the liquid detergent does not make the washing liquid so turbid, but the powder detergent becomes more turbid as it dissolves in the washing liquid. Thereafter, the motor 3, that is, the stirring body 9 is rotated by the standard water flow (time T5).

When the stirrer 9 is rotated in the normal or reverse direction, the number of revolutions of the motor 3 changes in accordance with the type of laundry, that is, the cloth, every time the stirring body 9 is started. The microcomputer 13 controls the motor 3
The change amount of the rotation speed of the rotation speed sensor 28
8 and the quality of the cloth is determined based on the magnitude of the change, and stored in the RAM as the cloth data D2 such as, for example, "stiff", "standard", and "smooth" (time T
6). At this time, the microcomputer 13 reads the temperature signal S24 of the temperature sensor 24, and stores the water temperature of the washing liquid in the RAM as water temperature data D7 based on the temperature signal S24.

After that, the microcomputer 13 sets the standard water flow, which is a constant water flow, again, drives the motor 3 to perform stirring. When such stirring is performed for a predetermined time, the microcomputer 13 stops the motor 3 again, and the microcomputer 3 is stopped (time T7). Also in this case, the light emitting diode 19 emits light by blinking. In this stopped state, the microcomputer 13 reads the transmittance signal S29 of the transmittance sensor 29 and stores it in the RAM as detection data A3. Thereby, the microcomputer 13 determines that the detergent amount is small if the difference between the detection data A2 and A3 is small, as shown in FIG.
If the difference is large, it is determined that the detergent amount is also large, and this is stored in the RAM as detergent amount data D5.

Thereafter, the microcomputer 13 operates the motor 3 with the standard water flow, which is a constant water flow set as described above.
That is, the stirrer 9 is rotated (time T8), and is stopped again after a predetermined time has elapsed (time T9). Again,
The light emitting diode 19 flashes. Then, the microcomputer 13 reads the transmittance signal S29 of the transmittance sensor 29 and stores it in the RAM as detection data A4 (time T10).

As shown in FIG. 11, the microcomputer 13 calculates A3-A2 = ΔV from the detected data A2, A3 and A4 at times T5, T8 and T10.
1, ΔV1 ′ and A4-A3 = ΔV2, ΔV2 ′ are calculated, and the difference between ΔV1 and ΔV2 is large. In this case, the quality of the dirt is determined to be, for example, a mud system, and ΔV1 ′ and ΔV2 are determined.
Since the difference from 'is small, the quality of the stain is determined to be, for example, oil based in this case. That is, in the case of mud-based dirt,
Since the degree of removal of the dirt is fast, the rate of change between the detection data A2 and A3 is large and the rate of change after that between the detection data A3 and A4 is large. On the other hand, in the case of oil-based dirt, the degree of dirt becomes little by little, so that the rate of change between the detection data A2-A3 and the rate of change between the detection data A3-A4 are very small. There is no difference. In this way, the microcomputer 13 calculates the dirt quality data D4 and stores it in the RAM.

The microcomputer 13 is arranged as shown in FIG.
As shown by, the absolute value of the detection data A2, A3, and A4 at times T5, T8, and T10 is used to calculate the contamination amount data D.
3 is calculated and stored in the RAM. This is based on the fact that the transmittance decreases as the amount of contamination increases.

As described above, the RAM of the microcomputer 13 stores the cloth amount data D1 and the cloth data D2 as the cloth conditions, the stain amount data D3 and the soil data D4 as the soil conditions, and further, the detergent. As conditions, detergent amount data D5 and detergent quality data D6 are stored, and in addition, water temperature data D7 is stored. Then, these data D1, D2, D3, D4, D5, D
6 and D7, as shown in FIG.
5 units I1, I2, I3, I4, I of the input layer I
5, I6 and I7 are respectively input to perform neuro control.

Now, the outline of the theory of the neural network used for the neural control will be described with reference to FIGS.

The neural network simulates a biological neural network, and is a circuit composed of units and links, as shown in FIG. Unit j has input / output characteristics Fj (Uj) as shown in FIG. Here, Fj is a sigmoid function, for example,

[0041]

(Equation 1)

Is represented by the following equation. The output Vj of the unit j is represented by the following equation.

[0043]

(Equation 2)

Here, Vi is the output of another unit i, W
ji is a weight coefficient indicating the degree of influence of the output of the unit i on the unit j, and θj is a threshold value.

Neural nets are classified into a complete connection type, a hierarchical type, and an intermediate type depending on the way of linking links. Here, an example of a hierarchical type is shown.

FIG. 16 shows a three-layered neural network in which units are arranged in three layers called an input layer, an intermediate layer and an output layer. They are distinguished by an index of k. In this neural network, signals are transmitted in one direction from an input layer to an output layer through an intermediate layer. In this case, a weight coefficient Wji is set for the link from the input layer to the hidden layer,
A weight coefficient Wkj is set for the link from the hidden layer to the output layer. As described above, the neural network is composed of a number of units as simple arithmetic elements, and a unit outputs a large output when the sum of inputs from other units exceeds a threshold value.

The features of the neural network include learning ability, high speed, and noise resistance.

Learning of the neural network is performed by adjusting link weighting factors so that a desired output pattern (teacher pattern) is obtained for an input pattern (case). In this case, the weight coefficient is initially set at random. In the neural network after learning, a plurality of pairs of learned input patterns and output patterns are associated with each other, and a desired output pattern can be obtained by inferring input patterns other than learned ones.

In this case, there is a back propagation method as an example of a neural network learning method. In the back propagation method, the weight coefficient is adjusted using an error function between a desired output pattern (teacher pattern) and an actual output pattern. In the neural network of FIG. 16, the error function E is defined as in the following equation.

[0050]

(Equation 3)

Here, Tk and Vk are the teacher data (desired output data) and the actual output data of the unit k of the output layer, respectively. The output data Vk is represented by the following equation.

[0052]

(Equation 4)

Then, in the back propagation method,
The correction amount of the weighting coefficient is calculated, and the correction is repeated until this value becomes a certain value or less. That is, the correction amount ΔWkj of the weight coefficient
And ΔWji are

[0054]

(Equation 5)

Is obtained. However,

[0056]

(Equation 6)

Is as follows. Here, η is a constant determined from the balance between the correction speed of the weighting coefficient and the stability of calculation. Then, from the correction amounts ΔWkj and ΔWji,

[0058]

(Equation 7)

The new weighting factors Wkj 'and Wji' are calculated as described above, and the learning is transferred to the next teacher pattern.

FIG. 17 shows the learning procedure. That is,
When learning a pair of N sets of input patterns and teacher patterns, first, the correction amounts ΔWkj, Δ
Calculate Wji and modify the weighting factor. Next, the correction amounts ΔWkj and ΔWji are calculated in the input pattern 2 to correct the weight coefficients. Hereinafter, similarly, the correction of the weight coefficient is performed up to the input pattern N. If the correction amounts .DELTA.Wkj and .DELTA.Wji are not below a certain value, the correction is repeated by repeating the input pattern 1, and if the correction amounts are below a certain value, the learning is terminated.

Incidentally, the above-described neural network can be constituted by hardware using a neural chip or the like called a neural processor, or can be simulated by software using a microcomputer. Yes, in any case, it fulfills the function of a neural network.

The operation of the neural network 45 according to the present embodiment based on the above-described theory will be described.

As shown in FIG. 9, cloth data D1, cloth data D2, stain data D3, stain data D4, detergent data D5, detergent data D6 and water temperature data D7.
Are units of I1, I2, I3, I4, I5, I5, I5
6 and I7, and takes one of 16 divisions from "1" to "16". The output data from the units K1 and K2 of the output layer K of the neural network 45 are wash water flow data O1 and wash time data O2.
Are output, and each of them is also output as a 4-bit signal. In this case, the washing time data O
As shown in FIG. 12, 1 is a 16-level classification from "1" (0.75 second energization and 1.35 second interruption) to "16" (1.1 second energization and 0.6 second interruption). It takes any value, and the washing time data O2 is "1" (7.5 minutes).
To "16" (15 minutes) in any of 16 sections every 0.5 minutes.

The operation of the neural network 45 will now be described.

Now, input data D1, D as input patterns
.., UI7, the units I1, I2,..., I7 of the input layer I output the data as output VI1, VI2,. That is,
VI1 = UI1, VI2 = UI2,... VI7 = UI7.

As for the units J1, J2,... J5 of the intermediate layer J, taking the unit J1 as an example, the input UJ1 is

[0067]

(Equation 8)

Is obtained. In this case, WJ1I1 is the unit I1
Weighting factor from unit to unit J1, WJ1I2 is unit I2
WJ1I7 is the weight of unit I from unit to unit J1
7 is a weighting factor for the unit J1, and θJ1 is a threshold value. Then, the unit J1 calculates the sigmoid function F using the above UJ1 as an input, and outputs the result to the output VJ1.
And That is,

[0069]

(Equation 9)

Is as follows. The same applies to the units J2 to J5.

As for the units K1 and K2 of the output layer K, taking the unit K1 as an example, its input UK1
Is

[0072]

(Equation 10)

Is obtained. In this case, WK1J1 is unit J1
Weighting factor from unit to unit K1, WK1J2 is unit J2
Weighting factor from unit to unit K1, WK1J5 is unit J
5 is a weighting factor for the unit K1, and θK1 is a threshold value. The unit K1 receives the above-mentioned UK1 as an input, calculates a Sigmond function F, and outputs the result to the output VK1.
And That is,

[0074]

[Equation 11]

Is as follows. The same applies to the unit K2.

Each of the weight coefficient W and the threshold value θ is represented by, for example, 4 bits, and can be any of positive, zero, and negative. For example, “0” is represented by “0000”, “1” by “0001”, and “−1” by “1111”.
That is, the most significant bit is a negative sign bit. The multiplication result (WV) of the weight coefficient W and the output V takes the upper 5 bits, and the most significant bit is a negative bit. Further, input U
Is represented by 8 bits, and the most significant bit is a negative bit. The output V is represented by 4 bits and takes a positive (or 0) value.

As for the threshold value θ, as shown in FIG. 13, a unit whose output is always “1” is set in the input layer I and the intermediate layer J, and the weight coefficient of the link from the unit is set to θK. , ΘJ, the actual weighting factors WKJ, WK
It can be handled in exactly the same way as JI. Here, the output "1" is the maximum value that the output of the unit can take,
That is, it represents the maximum value of the output of the sigmoid function F, and is "16" in the above example. The threshold value θ can be any of positive, zero, and negative similarly to the weight coefficient W, but the number of bits may be different. In the above case, as the unit that takes the output “1”, either one of the input layer I and the intermediate layer J of the neural network 45 may be set, or a new unit may be provided.

Learning of the neural network 45 is mainly performed at the stage of product development. In this case, it is not necessary to learn all the possible patterns of the input pattern. That is, FIG.
In which N is 20. If the weight coefficient W and the threshold value θ are determined as a result of the learning, the weight coefficient W and the threshold value θ may be the same for the same model for mass production.

In this manner, the washing water flow data O1 and the washing time data O2, which are output data, are calculated by the neural network 45, and based on the results, the microcomputer 13 determines whether the washing water flow data O1 is "1" to "16". Of the washing time data O2 and one of the values "1" to "16" of the washing time data O2, and the stirring after the time T10 in FIG. 1, that is, the "washing" step is executed. Let it.

When the microcomputer 13 completes the aforementioned “washing” process (time T11),
As shown in FIG. 10, the process shifts to the “rinse” process. In this “rinsing” step, “draining” is performed in which the drain valve 5 is opened by energizing the electromagnet 35 via the drive circuit 34, and the motor 3 is energized while the drain valve 5 is opened to rotate the inner tank 7. "Dehydration", "water supply" to open the water supply valve 11,
The “first rinsing” that executes the same “rinsing” as the agitation, and the “second rinsing” that executes the “water drainage”, “dehydration”, “water supply” and “rinsing” as described above. In this case, “water supply” is set to the water level set during the “washing” process, and “rinse” is set to the washing water flow set during the “washing” process.

Next, when the “rinse” step is completed, the micro-computer 13 performs the “dehydration” as shown in FIG.
Move to the process. In this “dehydration” step, “drainage” and “dehydration” are performed in the same manner as in the “rinse” step.

According to the present embodiment, the following effects can be obtained.

That is, after the microcomputer 13 obtains the cloth amount data D1 under the detected water flow by the capacity sensor 29, the microcomputer 13 changes the water flow into a standard water flow, which is a constant water flow, and intermittently interrupts the flow to obtain a transmittance sensor. 20, temperature sensor 24
And the rotation speed sensor 28 sequentially obtains the detergent quality data D6, the fabric quality data D2, the water temperature data D7, the detergent amount data D5, the soil quality data D4, and the soil amount data D3, so that at least data other than the fabric amount data D1. The detection of D2 to D7 is performed under a constant condition of a standard water flow, which is a constant water flow. Therefore, the microcomputer 1
3 does not need to store a large number of basic data for judgment corresponding to a change in water flow, and requires only a small amount of basic data capable of processing one type of pattern as shown in FIG. 11 and simplifies the program.

Further, as the condition of the laundry or the like, the cloth amount data D1 and the cloth data D2 as the cloth conditions, the dirt amount data D3 and the dirt data D4 as the dirt conditions, the detergent amount data D5 and the detergent quality data D6 as the detergent conditions. Then, the water temperature data D7 is stored in the neural network 4 of the microcomputer 13.
5 and the neural control controls the washing operation, that is, the washing water flow and the washing time in the "washing" step, so that the optimum washing effect can be obtained corresponding to various conditions,
The cloth can be prevented from being damaged due to excessive washing, and there is no trouble such as re-washing due to insufficient washing.

In the above embodiment, as input data of the state of the laundry or the like, cloth data D1, cloth data D2, stain data D3, stain data D4, detergent data D5, and detergent data D5.
Although the detergent quality data D6 and the water temperature data D7 are input to the neural network 45, it is needless to say that other conditions may be input as data, and at least the cloth amount data D1, which is a cloth condition. By inputting the cloth data D2 and the stain amount data D3 as a stain condition, a sufficient result can be obtained for neuro control.

In the above embodiment, the data D1 to D7
Is used to control the washing operation by neuro control. Alternatively, the washing operation may be controlled by, for example, fuzzy control or other general control.

Further, in the above embodiment, the cloth amount data D1 is detected under a detected water flow lower than the standard water flow, which is a constant water flow. However, if possible, it may be detected under the standard water flow. .

[0088]

As described above, the washing machine of the present invention has the following features.
When the cloth data and the dirt amount data are obtained from the sensor, the water flow is controlled to be constant, so that the basic data for the judgment of the control means can be reduced and the program can be simplified. .

[Brief description of the drawings]

FIG. 1 is a time chart for explaining an operation of a washing process showing one embodiment of the present invention.

FIG. 2 is an overall vertical sectional view.

FIG. 3 is a block diagram showing an electrical configuration.

FIG. 4 is a top view of the operation panel.

FIG. 5 is a longitudinal sectional view of a transmittance sensor.

FIG. 6 is a longitudinal sectional view of a temperature sensor.

FIG. 7 is an electrical connection diagram of a capacitance sensor.

FIG. 8 is a waveform chart for explaining the operation of the capacitance sensor.

FIG. 9 is a diagram for explaining an outline of neuro control.

FIG. 10 is a diagram showing a flow of input / output data and a process;

FIG. 11 is a graph showing transmittance characteristics.

FIG. 12 is a diagram showing an output result of a washing water flow.

FIG. 13 is an explanatory diagram of an operation for determining a threshold value.

FIG. 14 is a basic diagram of a neural network for explaining the principle of neuro control.

FIG. 15 is a diagram showing the sigmoid function.

FIG. 16 is a diagram showing a three-layer neural network.

FIG. 17 is a diagram showing a procedure of the learning.

[Explanation of symbols]

In the drawing, 2 is an inner tank, 3 is a motor, 5 is a drain valve, 7 is an outer tank, 9 is a stirring body, 11 is a water supply valve, 13 is a microcomputer (control means), 14 is an operation panel, and 20 is permeability. Sensor, 24 is a temperature sensor, 28 is a rotation speed sensor, 29
Indicates a capacitance sensor, and 45 indicates a neural network.

Claims (1)

(57) [Claims] 1. In response to a water stream being generated intermittently
A required number of sensors for sequentially detecting a plurality of data indicating the state of the laundry including the cloth amount, the cloth condition of the cloth, and the dirt condition of the dirt amount, and a plurality of input data from these sensors Control means for controlling the washing operation on the basis of the control information, wherein the control means is configured to generate a constant water flow when obtaining the cloth data and the stain amount data from the sensor. A washing machine, characterized in that: