JP2753136B2 - rice cooker - Google Patents

Description

Description: TECHNICAL FIELD The present invention relates to a rice cooker capable of adjusting the degree of hardness of rice at the end of rice cooking.

2. Description of the Related Art A conventional so-called microcomputer rice cooker has a configuration including a microcomputer having a rice cooking process determined by a manufacturer to be optimal, and cooks rice according to the rice cooking process.

Problems to be Solved by the Invention However, there is a demand for a user to cook rice with a desired hardness, and in order to satisfy this demand, when using the rice cooker having the conventional configuration, the amount of rice is increased. The user responded by changing the ratio of the amount of water to water. However, this method has a problem that it is difficult to cook delicious rice because the amount of water is difficult to determine and the amount of water is not always appropriate for the rice cooking process performed by the rice cooker.

Means for Solving the Problems In order to solve the above problems, a rice cooker according to the present invention has a rice cooking control means for determining a rice cooking process relating to a temperature or a time during a pre-cooking process according to a set rice hardness. Wherein the rice cooking control means determines a rice cooking process using a neuro-fuzzy inferencer.

Further, as another means, the cooked rice is cooked according to the set hardness of the rice, the amount of heating electric power at the time of raising the temperature of the cooking process, and the gradient of the temperature when the object to be heated reaches a predetermined temperature. Rice cooking control means for determining a rice cooking process relating to the amount of heating power at the time of boiling in the process;
A rice cooker that determines the rice cooking process using a fuzzy inference device.

As still another means, rice cooking control means for determining a rice cooking process relating to the amount of heating electric power at the time of heating or boiling during the cooking step, according to the set hardness of rice and the total amount of electric power at the time of the previous cooking step. Wherein the rice cooking control means determines a rice cooking process using a neuro-fuzzy inference device.

Operation According to the present invention, the hardness of cooked rice can be set to the hardness desired by the user.

In addition, by determining the rice cooking process by fuzzy inference, it becomes possible to perform the rice cooking operation in accordance with the know-how of rice cooking that humans know empirically, and the hardness of the cooked rice can be further reduced by the user's favorite hardness. At that time, the parameters of the fuzzy inference such as the shape of the membership function of the antecedent part of the fuzzy inference, the shape of the membership function of the consequent part, the number of inference rules, etc., are used for the neural network such as the descent method. By optimizing according to the learning rule, the rice cooking process can be easily designed, and a highly accurate design can be realized. The descent method is a learning rule used for neural networks.It is a method that automatically learns by minimizing the error function, and uses fuzzy inference that represents the rice cooking process that you want to realize by using a neuro-fuzzy inference device. It can be learned automatically.

Embodiment A rice cooker according to a first embodiment of the present invention will be described with reference to FIGS. First Example showing the electric circuit of the present embodiment
In the figure and FIG. 2 showing the overall configuration, 1 is a main body,
A heater 3 such as a heater and a triac or a heater and a current control circuit is provided at the bottom, a lid 5 is provided at the top, and an inner pot 2 is provided inside. A thermistor 4 which is a temperature detecting element for detecting the temperature of the object to be cooked is provided at the center bottom of the inner pot 2. Further, on the surface of the lid 5, an input means 6 for inputting the degree of hardness of the finished rice cooked by the user is provided. Further, the main body 1 determines a rice cooking process from the input means 6 and the information of the temperature detecting element 4.
A fuzzy inference unit 8, a control unit 9 for controlling the heating means 3 under the rice cooking conditions determined by the neuro-fuzzy inference unit 8, and a timing unit 10 for timing pre-cooking, cooking up, and re-cooking of the rice cooking process. I have. The neuro-fuzzy inference unit 8, the control unit 9, and the timing unit 10 form the rice cooking control unit 7, and are configured by a microcomputer in this embodiment.

Hereinafter, the operation and operation of the present embodiment will be described. FIG. 3 is a diagram showing a change in the temperature of the object to be cooked when rice cooking is performed. The rice cooking process can be broadly divided into pre-cooking, cooking, and re-cooking. The pre-cooking process is a process for absorbing water in rice, and maintains a temperature of about 50-60 ° C.
The cooking process consists of a heating process and a boiling maintenance process. In the heating process, the food is heated rapidly to raise the temperature,
Maintain that temperature in the boiling maintenance process. In this cooking step, in addition to the water absorption in the pre-cooking step, water is further absorbed into rice, and rice starch is gelatinized. Thereafter, unnecessary water is evaporated during the additional cooking process. Therefore, depending on the temperature and time of the rice cooking process such as the pre-cooking step and the cooking step, the amount of moisture contained in the cooked food changes, and the degree of hardness of the rice at the end of cooking, that is, rice cooking is determined. Things. In this embodiment, the rice cooking process is determined as follows. The information on the hardness of the rice at the end of rice cooking set by the user in the input means 6 is a neuro-
It is transmitted to the fuzzy inference unit 8. The neuro-fuzzy inference unit 8 determines a rice cooking process with an inference configuration as shown in FIGS. 4, 5, and 6. That is, the inference rules are, for example, "If soft, the pre-cooking temperature is low, the pre-cooking time is short, the heating power is small, and the boiling maintenance power is large." The three rules shown in FIG. Consists of A qualitative concept such as “softness” in hardness and “small” heating power is quantitatively expressed by a membership function shown in FIG. At this time, the inference rule and the membership function are optimally set in advance by a learning rule such as the steepest descent method (one of the learning rules used for the neural network and a method for minimizing an error function). . Next, a method of the inference operation will be described with reference to FIG. First, the hardness matching degree calculating means 12 obtains a matching degree as an antecedent with respect to an input, that is, an input value of the input means 6, according to a rule stored in the rice cooking process inference rule storing means 20. The determination of the degree of matching is determined by taking the maximum of the input value and the membership function stored in the hardness membership function storage means 15. Next, the consequent part minimum operation means 13 includes a pre-cooking temperature membership function storage means 16, a pre-cooking time membership function storage means 17, a heating power amount membership function storage means 18, and a boiling maintenance power amount membership function storage. The MIN of the membership function stored in the means 19 and the antecedent part fitness described above is taken as the conclusion of the rule. Further, after obtaining the respective conclusions for all the rules stored in the rice cooking process inference rule storage means 20, the center-of-gravity calculating means 14 obtains the MAX of all the conclusions and obtains the center of gravity. Thus, an optimal rice cooking process can be obtained as a final conclusion.

According to such fuzzy inference, it is possible to determine an optimal rice cooking process for cooking rice having an arbitrary degree of hardness input by the input unit, and to perform optimal rice cooking. In the present embodiment, the consequent variable of the fuzzy inference is a general triangular type. However, a method of realizing the trapezoidal type, a real value, or a function may be considered.

Next, a second embodiment of the present invention will be described with reference to FIGS. The present embodiment provides a specific example of the neuro-fuzzy inference device of the first embodiment. 23 is a pre-cooking temperature determining means for determining a pre-cooking temperature, 24
Is a pre-cooking time determining means for determining a pre-cooking time, 25 is a first heating power amount determining means for determining a heating power amount, and 26 is heating at the time of boiling of a cooking step which is a part of a rice cooking step. This is a first boiling maintaining power amount determining means for determining the power amount.
The neuro-fuzzy inference unit 22 of the present embodiment has the above-described units. Reference numeral 27 denotes a control unit that controls the heating unit 3 based on the determination result of each unit described above. Reference numeral 11 denotes a time measuring means similar to that of the first embodiment.

Hereinafter, the operation of this embodiment will be described. User input means 6
The information indicating the hardness of the rice at the end of the cooking, which is set in the above, is input from the input means 6 to the pre-cooking temperature determining means 23 and the pre-cooking time determining means.
The power is transmitted to the first heating power determining means 25 and the first boiling maintenance power determining means 26. Neuro-fuzzy reasoner
22 performs fuzzy inference using the input value of the input means 6 as an input variable to determine a rice cooking process. The inference rule is, for example, "If soft, lower the pre-cooking temperature.""If soft, increase the pre-cooking time.""If soft, reduce the heating power.""If soft, reduce the boiling power." It consists of 12 rules shown in Fig. 8. A qualitative concept such as “soft” in hardness and “small” in heating power is quantitatively expressed by a membership function shown in FIG. At this time, the inference rule and the membership function are optimally set in advance by a learning rule such as the steepest descent method (one of the learning rules used for the neural network and a method for minimizing an error function). .

Next, a method of an inference operation executed by the neuro-fuzzy inference device 22 of the present embodiment will be described with reference to FIGS. 10, 11, and 12. FIG. FIG. 10 shows a specific configuration of the pre-cooking temperature determining means 23. First, the hardness matching degree calculating means 28 is stored in the hardness membership function storing means 30 with respect to the input, that is, the input value of the input means 6, in accordance with the rule stored in the pre-cooking temperature inference rule storing means 29. The degree of fitness as the antecedent is obtained by taking the membership function and MAX. Next, in the consequent part minimum calculating means 31, the MI of the membership function stored in the pre-cooking temperature membership function storage means 32 and the antecedent part fitness is stored.
Take N as the conclusion for that rule. Further, respective conclusions are obtained for all the rules stored in the pre-cooking temperature inference rule storage means 29. Next, the center-of-gravity calculating means 33 obtains the MAX of all the conclusions and obtains the center of gravity. Thus, the final result is the pre-cooking temperature. FIG. 11 shows a specific configuration of the pre-cooking time determination means 24. First, the hardness matching degree calculation means 34 is stored in the hardness membership function storage means 36 for the input, that is, the input value of the input means 6, according to the rule stored in the pre-cooking time inference rule storage means 35. The degree of fitness as the antecedent is obtained by taking the membership function and MAX. Next, the consequent part minimum operation means 37 calculates the MIN between the membership function stored in the pre-cooking time membership function storage means 38 and the antecedent part fitness degree, and makes a conclusion based on the rule. Further, for all rules stored in the pre-cooking time inference rule storage means 35, respective conclusions are obtained. The respective conclusions are obtained for all the rules stored in the inference rule storage means 60. The center-of-gravity calculating means 66 takes the MAX of all the conclusions and finds the center of gravity. In this way, the final conclusion is the amount of boiling maintenance power.

According to such fuzzy inference, it is possible to determine an optimal rice cooking process for performing rice cooking of an arbitrary degree of hardness input by the input unit according to a load, and to perform optimal rice cooking. . In the present embodiment, the consequent variable of the fuzzy inference is a general triangular type. However, a method of realizing the trapezoidal type, a real value, or a function may be considered.

Next, a fourth embodiment will be described with reference to FIGS. 19 to 22. This embodiment uses a neuro-fuzzy inference device.
As 69, the pre-cooking total power amount calculation means 71 for calculating the total heating power amount in the pre-cooking process, and the boiling maintenance power amount from the input information of the input means 6 and the information of the pre-cooking total power amount calculation means 71 There is provided third boiling maintenance power amount determination means 72 for determination. The other units are the same as those in the second and third embodiments, and the description is omitted.

Hereinafter, the operation of this embodiment will be described. In the pre-cooking process,
The control unit 70 controls the heating power amount so as to maintain the temperature determined by the pre-cooking temperature determining means 54. In other words, if the amount of food to be cooked, that is, the amount of cooked rice is large, the total amount of electric power in the pre-cooking process will be large. The pre-cooking total power amount calculating means 71 calculates the total power amount in the pre-cooking process as an integral value of the power amount at each instant. The third boiling maintaining power amount determining means 72 receives the input value of the input means 6 and the calculated value of the pre-cooking total power amount calculating means 17 as inputs, and fuzzy infers the subsequent power amount, that is, the total power amount in the cooking process. Determined by First, input means 6
The optimum temperature curve in the boiling maintenance process is determined from the information on the hardness, which is the input value of. Similarly, the pre-cooking temperature can be determined from the input value of the input means 6. Next, the load amount is estimated from the value calculated by the pre-cooking total power amount calculation means 71 and the pre-cooking temperature. In accordance with the estimated load amount, the boiling position power amount necessary to realize the optimum temperature curve is determined. The inference rules are, for example, "If the total amount of pre-cooking power is large and the hardness is normal, the power amount is slightly increased." The nine rules shown in FIG. 20 are used. Qualitative concepts such as hardness being "normal" and electric energy being "slightly larger"
It is quantitatively expressed by the membership function shown in Fig. 21. At this time, the inference rule and the membership function are determined in advance by the steepest descent method (one of the learning rules used for the neural network, which is a method for minimizing the error function).
It is set optimally by learning rules such as. Next, a method of the inference operation will be described. FIG. 22 shows a specific configuration of the third boiling maintenance power amount determining means 72. First, according to the rules stored in the boiling maintenance power amount inference rule storage means 77, the pre-cooking total power amount conformity calculating means 73 calculates the input, that is, the total By taking the MAX with the membership function stored in the pre-cooking total power amount membership function storage means 74, the degree of conformity as the antecedent is obtained. The hardness fitting degree calculating means 75 calculates the membership function stored in the hardness membership function storing means 76 with respect to the input, that is, the input value of the input unit 6, and the MAX.
To obtain the degree of conformity as the antecedent part. Next, the antecedent part minimum calculation means 78 obtains the smaller one of the two degrees of conformity based on the antecedent part of the boiling maintenance power amount inference rule storage first stage 77, and sets it as the degree of conformity according to the rule. Next, in the consequent part minimum operation means 79, the MIN between the membership function stored in the boiling maintenance power amount membership function storage means 81 and the antecedent part conformity which is the conclusion in the antecedent part minimum operation means 78 is obtained. And take the conclusion in that rule.
Further, for each rule stored in the boiling maintenance power amount inference rule storage means 77, respective conclusions are obtained. The center-of-gravity calculating means 80 takes the MAX of all the conclusions and finds the center of gravity. In this way, the final conclusion is the amount of boiling maintenance power.

According to such fuzzy inference, it is possible to determine an optimal rice cooking process for performing rice cooking of an arbitrary degree of hardness input by the input unit according to the load, and to perform optimal rice cooking. In the present embodiment, the consequent variable of the fuzzy inference is a general triangular type. However, a method of realizing the trapezoidal type, a real value, or a function may be considered.

Next, a fifth embodiment will be described with reference to FIGS. 23 to 29. In the present embodiment, the neuro-fuzzy inference unit 83 includes a second heating power amount determining means 86 and a fourth boiling maintenance power amount determining means 87. The second temperature-raising power amount determining means 86 receives the information indicating the hardness of the rice at the end of the cooking input to the input means 6 and the value calculated by the pre-cooking total power amount calculating means 89 described in the fourth embodiment. Is used as input to determine the heating power amount at the beginning of the cooking process. The fourth boiling maintenance power amount determining means 87 obtains the boiling maintenance power amount at the time of boiling in the cooking process from the information indicating the hardness of the rice at the end of the cooking input to the input means 6 and the detection value of the temperature detecting element 4. To determine. Other configurations are the same as those of the above-described embodiments, and the description is omitted.

Hereinafter, the operation of this embodiment will be described. In the present embodiment, the second heating power amount determining means 86 receives the information indicating the hardness of the rice at the end of the cooking input to the input means 6 and the pre-cooking total power amount calculating means 89 as input values, The subsequent heating power, that is, the heating power in the boiling maintenance process is determined by fuzzy inference.
First, the optimum temperature curve in the heating process at the beginning of the cooking process is determined from the hardness of the rice at the end of cooking, which is an input value of the input means 6. Similarly, the pre-cooking temperature can be determined from the input value of the input means 6. Next, the load amount is estimated from the value calculated by the pre-cooking total power amount calculation means 17 and the pre-cooking temperature.
In accordance with the estimated load amount, a heating power amount required to realize an optimal temperature curve is determined. The inference rule is, for example, "If the total amount of pre-cooked electric power is large and the hardness is normal, slightly increase the amount of electric power for heating."
Consists of rules. A qualitative concept of hardness being “normal” or heating power amount being “slightly larger”
It is quantitatively expressed by the membership function shown in the figure. Further, the fourth boiling maintenance power amount determining means 87 receives the input value of the input means 6, the temperature gradient θ of the food to be cooked, and the calculated value of the pre-cooking total power amount calculating means 89, and receives the subsequent heating power amount, that is, boiling. The heating power in the maintenance process is determined by fuzzy inference. First, an optimal temperature curve in the boiling maintenance process is determined from the hardness as the input value of the input means 6. Similarly, from the input value of the input means 6, the pre-cooking temperature and the amount of electric power for temperature rise can be known. Next, a first estimation of the load amount is performed using the calculated value of the pre-cooking total power amount calculation means 89 and the pre-cooking temperature, and the temperature gradient θ
And a second estimation of the load amount from the heating power amount. From the first and second two types of estimated load values, more accurate load estimation is performed. In accordance with the finally estimated load amount, the amount of boiling maintenance power necessary to realize an optimal temperature curve is determined. An inference rule is, for example, "If the temperature gradient θ is small, the total amount of pre-cooked electric power is normal, and the hardness is normal, the electric power is slightly increased."
Consists of rules. The qualitative concept of hardness being “normal” or electric energy “slightly larger” is shown in Fig. 26
It is quantitatively expressed by the membership function shown in Figure 27. At this time, the inference rule and the membership function are determined in advance by the steepest descent method (one of the learning rules used for the neural network, which is a method for minimizing the error function).
It is set optimally by learning rules such as.

Next, a method of the inference operation will be described. FIG. 28 shows a specific configuration of the second heating power amount determining means 86. First, according to the rules stored in the heating power amount inference rule storage means 91, the pre-cooking total power amount conformity calculating means 90 is an input, i.e., the value calculated by the pre-cooking total power amount calculation means 89 at the time of pre-cooking. For the total electric energy, the membership function stored in the pre-cooking total electric energy membership function storage means 92 and MAX are used to determine the degree of suitability as the antecedent part. The hardness suitability calculating means 93 calculates the membership function stored in the hardness membership function storage means 94 and the MAX value for the input, that is, the input value of the input means 6.
To obtain the degree of conformity as the antecedent part. Next, the antecedent part minimum calculating means 95 obtains the smaller one of the two degrees of conformity based on the antecedent part of the heating power amount inference rule storage means 91, and sets the smaller one as the degree of conformity according to the rule. Next, in the consequent part minimum operation means 96, the MIN between the membership function stored in the heating power amount membership function storage means 97 and the antecedent part conformity which is the conclusion of the antecedent part minimum operation means 95 is obtained. And take the conclusion in that rule. Further, respective conclusions are obtained for all the rules stored in the heating power amount inference rule storage means 91. Next, the center-of-gravity calculating means 98 takes the MAX of all the conclusions and finds the center of gravity. In this way, the final result is the heating power amount.

Next, a specific configuration of the fourth boiling maintenance power amount determining means 87 will be described with reference to FIG. First, the temperature gradient suitability calculating means 99 obtains the antecedent by taking the membership function stored in the temperature gradient membership function storage means 101 and MAX for the input, that is, the change in the detection value of the temperature detecting element 4. Find the degree of conformity as a part. In the pre-cooking total power amount adaptability calculating means 102, the pre-cooking total power amount membership function storage means 10 is used for the input, that is, the total power amount at the time of pre-cooking which is the value calculated by the pre-cooking total power amount calculating means 89.
The fitness as the antecedent is obtained by taking the membership function stored in 3 and MAX. The hardness suitability calculating means 104 calculates the fitness as the antecedent part by taking the input, that is, the input value of the input means 6 and the membership function stored in the hardness membership function storage means 105 and MAX. Ask. Next, the antecedent minimum calculation means 10
In step 6, based on the antecedent part of the boiling maintenance power amount inference rule storage means 100, the smallest one of the three conformances is determined, and is determined as the conformity according to the rule. Next, in the consequent part minimum operation means 107, the membership function stored in the boiling maintenance electric energy membership function storage means 108 and the antecedent part conformity MIN which is the conclusion of the antecedent part minimum operation means 106 are calculated. So let's conclude with that rule. further,
With respect to all the rules stored in the boiling maintenance power amount inference rule storage means 100, respective conclusions are obtained. Next, the center-of-gravity calculating means 109 obtains the MAX of all the conclusions and obtains the center of gravity. In this way, the final conclusion is the amount of boiling maintenance power.

According to such fuzzy inference, it is possible to determine an optimal rice cooking process for performing rice cooking of an arbitrary degree of hardness input by the input unit according to the load, and to perform optimal rice cooking. In the present embodiment, the consequent variable of the fuzzy inference is a general triangular type. However, a method of realizing the trapezoidal type, a real value, or a function may be considered.

In the above embodiment, only the temperature detecting element is used as the sensor, but the weight sensor, the volume sensor,
A method of using a detection value of a sensor such as a moisture content sensor as an input variable of a neuro-fuzzy inference device is also conceivable.

Effect of the Invention As described above, according to the present invention, the hardness of cooked rice can be set to the hardness desired by the user.

In addition, by determining the rice cooking process by fuzzy inference, it becomes possible to perform the rice cooking operation in accordance with the know-how of rice cooking that humans know empirically, and the hardness of the cooked rice can be further reduced by the user's favorite hardness. At that time, the parameters of the fuzzy inference such as the shape of the membership function of the antecedent part of the fuzzy inference, the shape of the membership function of the consequent part, the number of inference rules, etc., are used for the neural network such as the descent method. By optimizing according to the learning rule, the rice cooking process can be easily designed, and a highly accurate design can be realized. The descent method is a learning rule used for neural networks.It is a method that automatically learns by minimizing the error function, and uses fuzzy inference that represents the rice cooking process that you want to realize by using a neuro-fuzzy inference device. It can be learned automatically.

[Brief description of the drawings]

FIG. 1 is a block diagram showing a first embodiment of the present invention, and FIG.
Fig. 3 is a cross-sectional view showing the overall configuration, Fig. 3 is a characteristic diagram showing a temperature change in the rice cooking process, Fig. 4 is a diagram showing the fuzzy inference rule, and Fig. 5 is a membership function of the fuzzy inference. FIG. 6, FIG. 6 is a diagram showing the configuration of the fuzzy inference device, FIG. 7 is a block diagram showing the second embodiment, FIG. 8 is a diagram showing the fuzzy inference rules, and FIG. Diagram showing membership functions of inference, Figs. 10 and 11
FIG. 12 and FIG. 13 are diagrams showing the configuration of the fuzzy inference device.
FIG. 14 is a block diagram showing the third embodiment, FIG. 15 is a characteristic diagram showing a temperature change in the rice cooking process, FIG. 16 is a diagram showing the fuzzy inference rule, and FIG. 17 is a diagram of the fuzzy inference FIG. 18 shows a membership function, FIG. 18 shows a configuration of the fuzzy inference device, FIG. 19 is a block diagram showing the fourth embodiment, FIG. 20 shows a fuzzy inference rule, FIG. Fig. 22 shows the membership function of the fuzzy inference, Fig. 22 shows the configuration of the fuzzy inference device, Fig. 23 is a block diagram showing the fifth embodiment, Figs. 24 and 25 are FIGS. 26 and 27 are diagrams showing the fuzzy inference rules, FIGS. 26 and 27 are diagrams showing membership functions of the fuzzy inference, and FIGS. 28 and 29 are diagrams showing the configuration of the fuzzy inference device. 3 ... heating means, 4 ... temperature detection element, 6 ... input means, 7, 21, 52, 68, 82 ... rice cooking control means, 8, 22, 53
・ 69 ・ 83… Neuro fuzzy inference device, 9 ・ 27 ・ 58 ・
70 · 88 ···· Control unit, 23 · 54 ··· Pre-cooking temperature determination means, 24
· 55 · 84 ··· Pre-cooking time determining means, 25 · 56 ··· First heating power amount determining means, 26 ··· First boiling maintaining power amount determining means, 57 ··· Second boiling maintaining power amount Determining means, 71: pre-cooking total electric energy calculating means, 72, ... third boiling maintaining electric energy determining means, 86, ... second heating power electric energy determining means, 87 ... fourth boiling maintaining electric energy Decision means.

Continuation of front page (72) Inventor Haruo Terai 1006 Kazuma Kadoma, Kadoma, Osaka Prefecture Inside Matsushita Electric Industrial Co., Ltd. (56) References JP-A-63-197414 (JP, A) JP-A-1-227720 (JP) , A) JP-A-2-44126 (JP, A)

Claims (3)

(57) [Claims] 1. A rice cooking control means for determining a rice cooking process relating to a temperature or a time during a pre-cooking process according to a set hardness of rice.
A rice cooker that determines the rice cooking process using a fuzzy inference device. 2. The cooking process according to the set hardness of the rice, the amount of heating electric power at the time of raising the temperature of the cooking process, and the gradient of the temperature when the object to be heated reaches a predetermined temperature. A rice cooker comprising rice cooking control means for determining a rice cooking process relating to the amount of heating power at the time of boiling, wherein the rice cooking control means determines a rice cooking process by a neuro-fuzzy inference device. 3. A rice-cooking control means for determining a rice-cooking process relating to the amount of heating power at the time of heating or boiling during the cooking process according to the set rice hardness and the total power during the pre-cooking process. A rice cooker, wherein the rice cooker control means determines a rice cook process using a neuro-fuzzy inference device.