JP4782664B2 - Method for detecting the kneading behavior of food dough - Google Patents

Description

  The present invention is used for bread dough, hamburger, marine products, etc. (viscoelastic products such as food dough) and other food fields where proper kneading is required, and the kneader is kneaded by rotating a rotor having a rotating shaft. Observe the behavior of the food dough made of viscoelastic material and objectively grasp the kneading state, and without this, it is necessary to visually observe the kneading state of the food dough without requiring the judgment of a highly skilled expert. The present invention relates to a method for detecting the behavior of a kneaded state of a food dough that can be accurately grasped.

  Conventionally, in bread dough, hamburger, marine products, etc., kneading of the dough (in the present invention, dough refers to a food material or food made of a viscous elastic material adjusted to be processed in the next step) This is the most important step in deciding. For example, when using a mixer equipped with stirring blades with fish meat surimi (mixing ingredients as necessary), the rotation speed of the stirring blades is about 500 to 2000 rpm. Mix for about 1 to 5 minutes. Furthermore, the mixing is preferably performed at 1000 rpm to 2000 rpm for 1 to 3 minutes. In such a case, too much kneading results in a soft surimi that reduces the texture. The kneading operation is extremely important.

  Also in bread dough, it is important to accurately determine the end point of the mixing operation. When hydrolyzed wheat flour is kneaded, the swollen gluten acts as a binder to join together the starch particles, resulting in a viscoelastic bread dough. In this case, when kneading is insufficient, it becomes a hard bread with less elasticity, and conversely, when it is excessively kneaded, it becomes a soft bread like cake. “How much to knead” and “when to stop kneading operation” are the most important points in bread making.

  Such bread dough is made by adding water to wheat flour and further adding fat and oil (such as butter) as a shortening and seasonings (sugar, salt, eggs, flavors, etc.), and yeast if necessary. The vertical kneader and spiral kneader used at the bakery store level are in a state where the bowl (container that kneads the dough) is visible. The operation of the kneader can be stopped while watching.

  On the other hand, large horizontal kneaders installed at bread factories (with rotating shafts arranged horizontally) perform stirring with the container closed due to its structure, so that the dough looks like a vertical kneader. The kneading machine cannot be stopped by judging the stirring state of the dough while observing. Therefore, when kneading bread dough with a large horizontal kneader, the operation is stopped by craftsman experience and cans, but it is extremely difficult to always stop the operation with an optimal finish.

For this reason, in the case of a horizontal kneader for business use, the container is opened once in the middle of kneading, the state of the dough is extended by hand, and the finish of the dough is observed. In this case, if kneading is insufficient, the container is closed again and restarted. Conversely, if you stop driving and overmixing, the fabric becomes a loss fabric and is disposed of.
Therefore, in order to avoid disposal, it is necessary to repeat the process of stopping the operation in the middle of kneading, stretching the dough by hand, and restarting the operation. Therefore, it is inefficient and wasteful in the manufacturing process. Furthermore, for the dough, if the dough is hurt by kneading and then stopped, then the condition settles down and it is hurt again by re-running, etc. Repeating the process over and over again may lead to quality degradation. It was.

In view of the above, there has been a demand for a device that allows the horizontal kneader to observe the kneading state of the dough from the outside.
For example, Patent Document 1 (Japanese Patent Application Laid-Open No. 61-219333) discloses a method for monitoring changes in the state of bread dough and the like by analyzing fluctuations in a kneader load (load power detection value of an electric motor) during kneading bread dough and the like. Is disclosed.
In this method, the kneader load fluctuation is displayed as a tendency fluctuation curve obtained by removing the pulsation fluctuation component and smoothing appropriately and a pulsation amplitude curve indicating a short period pulsation fluctuation of the kneader load. Changes in the state of bread dough and the like in the kneader are observed and determined in a pattern. That is, the fluctuation of the tendency fluctuation curve is considered as a change in viscous resistance due to a change in viscoelasticity of bread dough or the like, and the peak time is regarded as the optimum kneading time. The pulsation amplitude curve represents the magnitude of the load fluctuation applied to the kneader arm during one rotation of the bread dough, etc., and has a constant change pattern as the properties of the bread dough etc. during the mixing change. However, a specific determination method focusing on the pulsation amplitude curve is not disclosed.

  In addition, the present inventors have previously proposed a method of determining the progress of the kneading operation from the rotor torque waveform of the kneader in Patent Document 2 (Japanese Patent Laid-Open No. 5-56744). In this method, water and auxiliary materials are added to wheat flour, and the dough is produced by stirring and kneading with a kneader. In this method, the value of power for driving the rotary stirring rod of the kneader is detected, and the rotary stirring rod rotates. The change in the driving force within the minute time is detected by detecting the change in the driving force that occurs due to the change in the reaction force received from the fabric or the cloth material, and the peak value of the driving force change within the minute time is determined every minute time. Is a technique for determining the progress of the kneading operation based on the above, but in particular, the value of power for driving the rotating stirrer bar of the kneader is detected, and the minimum value of power change is that the rotating stirrer bounces off the dough It corresponds to the idling state, and the peak value of power change is regarded as equivalent to the state where the rotating stirrer is hit against the dough lump, and the peak value within the same minute time is obtained every minute time. ,same To track changes in the over click value, a method for determining and which is the completion of kneading when turned to decrease from increase.

Japanese Patent Laid-Open No. 61-219333 JP-A-5-56744

  Although the technique disclosed in Patent Document 1 focuses on the pulsation fluctuation of the kneader load, the fluctuation trend is observed mainly by focusing on a trend fluctuation curve that is considered to have been the moving average process of the kneader load. 3 and 4 of Patent Document 1 showing a trend variation curve, the tendency does not necessarily appear clearly. For the pulsation amplitude curve, the peak value of the pulsation amplitude curve shown in FIG. This frequency simply represents the rotation timing (frequency) of the stirring bar of the mixer and is not considered to be caused by the viscoelastic behavior of the bread dough itself. It has not reached the point where monitoring is possible.

  In addition, the method disclosed in Patent Document 2 determines the peak value of the power change every minute time, tracks the change of the peak value, and determines that the kneading is completed when the tendency from the increasing tendency to the decreasing tendency is reached. Similar to Patent Document 1, the average fluctuation tendency of the kneader load is still observed, and therefore, it has not been possible to accurately determine and monitor the properties of bread dough and the like.

  In view of the problems of the prior art, the present invention, when kneading viscoelastic food dough containing a polymer substance such as protein and at least partially pregelatinized bread dough, etc. without requiring the judgment of a skilled technician, In addition, in a horizontal kneader where the food dough cannot be visually observed, it is possible to determine the best kneading state, that is, the time when the kneading is completed or the time when the auxiliary material is charged. A method that can monitor the physical property behavior of the material and grasp the kneading state in real time, for example, grasp the proper timing of adding additional materials such as secondary materials, or grasp the optimal end point of the kneading. The purpose is to provide.

  The present invention achieves such an object, and in a method for detecting the behavior timing of a kneaded state of a food dough made of a viscoelastic material that is kneaded in a kneader by rotating a rotor having a rotating shaft, The detected torque detection value is frequency-analyzed at each time interval, and the DC component of zero frequency and the frequency peak intensity in a specific frequency band are extracted from the analyzed frequency spectrum time-series data, and the time-dependent change of the DC component In addition, the time-dependent change in peak intensity in the specific frequency band is displayed in a two-dimensional or three-dimensional manner by incorporating a time axis, and the behavior of the kneaded state of the viscoelastic material is detected based on the display of the two time-dependent changes. It is characterized by doing.

In the present invention, in the torque detection value of the rotor in the kneader, the frequency component is focused on the DC component having a frequency of zero, the frequency analysis is performed, and the viscoelastic behavior of the viscoelastic food dough can be observed by the analysis. Obtaining knowledge, the present invention has been achieved.
In general, when measuring the dynamic viscoelasticity of a viscoelastic material, there is a method of measuring the phase difference between distortion and stress waves generated in the viscoelastic material by applying vibration to the viscoelastic material to be measured (vibration method). ). The elasticity increases in proportion to the magnitude of deformation (Hooke's law), and the viscosity increases in proportion to the speed of deformation. There is a rule in the vibration motion of an object that the speed is minimum when the deformation amount is maximum and the speed is maximum when the deformation amount is minimum. That is, the deformation speed curve is formed by shifting the deformation amount curve by a quarter (90 degrees) phase of one cycle.

When a vibration stimulus is applied to a complete elastic body, the stress increases in proportion to the amount of deformation, so there is no phase difference between the wave of deformation and the wave of stress. When a vibration stimulus is applied to a complete viscous body, the stress increases in proportion to the deformation speed, so that the deformation amount wave and the stress wave have a phase difference of 90 degrees. A viscoelastic body having an intermediate property between viscosity and elasticity has a phase difference of 45 degrees.
Therefore, the food dough made of viscoelastic material has an influence on the behavior of the rotor in the kneader (rotor shape / speed) due to the property ratio of elasticity and viscosity, which affects the behavior of the entire viscoelastic material and becomes the amount of torque applied to the rotor.
By utilizing the fact that viscosity and elasticity are reflected in vibration motion in this way, in the present invention, the viscosity and elasticity of the food dough are determined from the DC component of the food dough made of the target viscoelastic material and the frequency analysis result. The state of kneading of the food dough is monitored by determining the transition.

  In the present invention, when the detected torque value applied to the rotor for kneading the food made of viscoelastic material installed in the kneader is analyzed, the torque average value corresponds to the DC component, and is received when the rotor rotates from the DC component. It can be seen that the resistance value is the sum of the weight load of the fabric, the elastic resistance value of the fabric, and the viscous resistance value of the fabric. We have obtained the knowledge that the change in vibration caused by the behavior of the whole food dough, that is, the change in dough physical properties (viscoelasticity) (mixing progress) can be found by the frequency component of the torque. When frequency analysis is performed in the following slow frequency range, a specific peak (near 0.25 Hz) appears, and it is found that the change in the peak position and intensity and the change in the DC component represent the change in the properties of the food dough. It is a thing.

That is, in the present invention, since the change with time of the DC component corresponds to the change of the average torque value, the change with time of the resistance value received when the rotor rotates can be understood, and the food that moves according to the rotation period of the rotor. Since the vibration behavior of the dough is involved in the expression of the peak strength, the fluctuations based on the time-dependent change (mixing progress) of the dough physical properties (viscoelasticity, especially the packing / extensibility) in the food dough kneader. It is considered that the change with time of the dynamic motion appears as the change with time of the frequency peak intensity in the specific frequency band, and the physical property (viscoelasticity) change of the food dough is determined or monitored by the change with time of the frequency peak intensity. .
As described above, since the frequency at which the peak intensity appears due to the shaking of the food dough appears differently, it is desirable to set the specific frequency band to a frequency band corresponding to the shaking of the kneaded food dough.

For example, the collective state of the entire fabric can be determined by the peak intensity appearing or disappearing in the time-dependent change of the frequency peak intensity in the specific frequency band. In addition, when it is time to complete the mixing of the food dough made of viscoelastic material, after the point of time when the DC component gradually changes and once again tends to rise again, in addition to a certain frequency region, It can be considered that the peak intensity having the intensity is in an optimum kneading state at the time when the frequency range changes (moves), and is the completion timing of kneading.
The behavior and the timing thereof are not limited to the completion of kneading of the food dough, which is a viscoelastic material, and can be used for the addition timing of the auxiliary material if the dough hardness and the mixing progress are known.

  Fast Fourier transform (FFT) is often used as a frequency analysis method, but since this method assumes that the analysis target is stationary, identification of a system whose characteristics change over time and signal Since it cannot be used for analysis, the frequency analysis for each time interval is analyzed using one of the short-time Fourier transform (STFT), Wavelet transform (WT), and Wigna analysis (WD). It is good to become.

  Specifically, the torque detection value is subjected to a short-time Fourier transform process by applying a window function, and a DC component having a frequency of zero and a frequency peak intensity in a specific frequency band are extracted from the processed frequency spectrum, and the DC Along with changes in the components, the changes over time in the peak intensity of the specific frequency band are displayed two-dimensionally or three-dimensionally by incorporating a time axis, and the kneaded state of the viscoelastic material is displayed based on the two changes over time. It is better to detect the behavior of

  According to the present invention, the torque applied to the rotor for kneading the food dough made of viscoelastic material installed in the kneading machine is detected, the torque detection value is subjected to frequency analysis, and the food dough is kneaded from the analyzed frequency distribution. By monitoring the state by, for example, a method such as displaying it in a distribution graph, it is possible to knead the food dough without requiring the judgment of an expert skilled in the art and without visually checking the kneading state of the food dough. The state can be accurately grasped in real time.

  Therefore, in general foods, it becomes extremely easy to grasp the kneading state of the food dough and grasp, for example, the proper timing of adding the additional material, or grasp the finished end point. For this reason, when mixing bread dough etc. using a horizontal kneader, the optimum kneading state can be grasped without temporarily stopping the operation of the apparatus as in the past, so that the operation efficiency of the apparatus is improved at each stage. improves.

  Preferably, the detected torque value of the rotor is subjected to Fourier transform processing for each time band, and thereafter, the frequency peak intensity of the specific frequency band of the rotor torque waveform and the change over time of the DC component are displayed, and the frequency peak of the specific frequency band is displayed. By monitoring the kneading state of the food dough based on changes over time in strength and DC component, the behavior of the kneading state of the food dough can be accurately and in real time.

  In addition, in the case of bread dough, after the time point when the DC component changes gradually with time and once again tends to rise again, the frequency distribution of the rotor torque waveform changes over time in a certain frequency range, for example, 0.2 to Since the peak intensity around a frequency of 0.5 Hz near the frequency of 0.5 Hz can be regarded as being in an optimal kneading state at the time when the frequency range changes (moves), it is optimally objectively and clearly The kneading state can be grasped.

In the present invention, preferably, the torque detection value is sampled at a set time interval, and the Fourier transform process and the subsequent process are repeated at the same time interval, and the time interval is set appropriately. For example, by setting the constant value within 1 to 10 seconds, the analysis accuracy is not lowered and the peak band can be picked up without omission.
In addition to kneading bread dough, the present invention is a viscoelastic food dough containing a high molecular weight substance such as protein or at least partially gelatinized starch, such as hamburger, marine product, whipped cream, etc. It is widely applicable to the whole food dough that needs to be monitored.

  For example, only fish meat such as guchi, walleye pollock, hokke, wallazuka, flounder, eso, and prickly fish are taken out, water-soluble proteins are removed by water exposure, dehydrated, and additives such as sucrose, sorbitol, and polyphosphate are mixed. Then, while thawing frozen frozen surimi, add a suitable amount of water, salt, other egg whites, starch, sugar, glutamine soda, fats and other seasonings or additives, so-called salting, and further When the same confirmation experiment was conducted when producing a paste-like fish meat product product with high viscosity, the DC component represents the firmness of the surimi during kneading, and the frequency spectral intensity distribution over time It was confirmed that the dynamic viscoelastic state of surimi was expressed, and the kneading state of surimi could be determined.

Hereinafter, the present invention will be described in detail with reference to the embodiments shown in the drawings. However, the dimensions, materials, shapes, relative arrangements, and the like of the component parts described in this example are not intended to limit the scope of the present invention only to specific examples unless otherwise specified. Only.
FIG. 1 is a process diagram showing a specific analysis procedure (real-time frequency analysis by short-time Fourier transform) of the present invention, FIG. 2 is a time-dependent change (a) in the rotor torque waveform of the horizontal kneader according to FIG. It is a (c) monitor screen showing the change over time (b) of the DC component (0 Hz) and the peak intensity of the high frequency component after 0.08 Hz.

  FIGS. 7A and 7B are schematic views showing an example of a horizontal kneader applied to the present invention. In FIG. 7, the casing of this horizontal kneader is a hermetically sealed type, and a rotary agitator 1 b is connected to the rotary shaft 1 a of the kneader 1, and the rotary shaft 1 a is connected via a pulley 1 c and a V belt 2. The motor 3 is rotationally driven. The motor 3 is connected to a power source via a relay 4, and a torque detector 5 that detects a torque value is attached to the output shaft of the motor 3. Further, T is a temperature sensor for detecting the temperature of the dough in the kneading machine 1, 1d is an opening / closing lid for the kneading machine 1, and 1e is a handle for opening / closing the opening / closing lid 1d.

The torque detection signal a detected by the torque detector 5 is input to the measurement / analysis PC 6 to analyze the change with time of the power and display the analysis information b on the monitor 8.
The measurement / analysis PC 6 is a system capable of performing time / frequency analysis in real time while collecting torque changes. The measurement / analysis PC includes a computer (CPU: Pentium) having a processing speed capable of withstanding real-time frequency analysis processing. (Registered Trademark) III (about 800 MHz) is used, and the torque data detector 5 is a shaft output detection torque detector that can convert electric power to torque while taking into account internal loss of the motor. The detected torque value obtained by the output torque detector 5 is taken in via the A / D conversion board 9 having a maximum sampling frequency of about 200 kHz attached to the analysis PC 6.

  Further, the structure of the rotor of the horizontal kneader is such that 2 to 4 horizontal rotating stirring rods rotate. For example, the operation speed is a low-to-high two-stage switching type.・ Create a network. The low-speed operation time and high-speed operation time can be set with a timer, but even if it is operated according to the conventional setting time, the dough will hardly be produced with the same time schedule, and the expert can stop the kneader according to the situation at that time Is adjusted flexibly. In addition, the casing of the kneader is generally a hermetic type.

Time / frequency analysis was performed on the torque detection value data obtained from the rotor of the horizontal kneader by short-time frequency analysis.
The short-time Fourier transform was selected because of the simplicity and high speed of the algorithm. FIG. 1 shows this analysis procedure.

  The present invention was obtained from the torque detector of FIG. 1 (step 1) so that the frequency peak of the vibration component could change as the kneading progressed, so that the frequency peak of the vibration component could be cut. Time / frequency analysis (TFR) was performed on the detected torque value. There is Fast Fourier Transform (FFT) as a frequency analysis method, but since this method assumes that the analysis target is stationary, it should be used for identification of systems whose characteristics change over time and analysis of signals. It was understood that it was not possible. Although FFT was tried about the torque detection value actually obtained from the torque detector installed in the horizontal mixer, it did not reach a significant feature.

Typical techniques for TFR include short-time Fourier transform (STFT), WAVELET transform (WT), and Wigna distribution (WD). In this embodiment, STFT was used because of the simplicity and high speed of the algorithm.
For the experiment, the horizontal mixer equipped with a torque detector was operated until the dough was let down (excess kneading) while changing the dough production conditions in various ways, such as the amount of moisture and the presence or absence of sugar. VTR images were taken so that the analysis results could be compared with the actual fabric at a later date.

  In the analysis method, as shown in FIG. 1, the temporal change of the torque change of the 128 point data is extracted from the measurement start point from the obtained torque change data (step 1), and a window function (Hamming) is preprocessed as the data. (Step 2), FFT conversion is performed (step 3), and the frequency spectrum (see FIG. 1 (a)) in the series of analysis processes is shifted by one sample (= 1 second) until the measurement end time. The frequency spectrum analysis was performed, and then the DC component (frequency 0) (FIG. 1B) of the frequency spectrum analysis change with time change and the peak intensity of the high frequency component after 0.08 Hz (FIG. 1C). )) Are extracted and displayed on the monitor 8 as changes over time.

FIG. 2 shows the change over time of the torque change obtained from the torque detector and the change over time of the DC component (frequency 0) and the change over time of the peak intensity of the high frequency component after 0.08 Hz, respectively (a). , (B), and (c) are displayed on the monitor. Note that a range indicated by a in FIG. 2B indicates a time zone in which the fabric is considered to be in an optimum state as judged from the video from the VTR.
As shown in FIG. 2 (b), the DC component (0 Hz) temporarily decreases immediately after switching to high-speed operation, but after that, when it approaches the optimum state (range indicated by a), it starts to increase, and is further kneaded. Then, the spectrum of the DC component tended to decrease again when proceeding to the letdown state.

  In addition, when looking at FIG. 2 (c), it has been found that the change in the peak intensity of the high frequency component after 0.08 Hz gradually appears in the spectrum around 0.25 Hz and tends to increase with time. It was. Paying particular attention to the spectrum around 0.25 Hz, the peak frequency shifts to the high frequency side before a certain period of time (up to 1 minute) when the fabric is in the optimum state, and returns to the low frequency side after the optimum state, letting down. It turned out that the peak disappeared when approaching. It was also found that the same tendency was shown regardless of the amount of water and the amount of sugar.

Therefore, it has been found from the results of this experiment that there is a possibility of being a mixer stop index if the spectral change of the DC component and the spectral change in the vicinity of 0.2 Hz are followed in combination. Detailed determination criteria by performing real-time waveform analysis while observing actual conditions such as which frequency to focus on in the vicinity of 0.2 Hz and how long it takes to reach the peak To consider.
Furthermore, if the display of the real-time analysis status is devised, for example, when the DC component display has different colors when the DC component is rising and falling, it becomes a visual and intuitive mixer stop determination index different from the conventional one. Another effect that can be expected.

  From the above, a window function (Hamming window) is applied as preprocessing (FIG. 1 (step 2)), and then FFT transformation is performed (FIG. 1 (step 3)). It has been found that if the change and the change with time of the spectrum in the vicinity of 0.2 Hz are chased in a composite manner, it may become a kneader stop index. Each criterion can be set by performing a rotor torque waveform analysis in real time for each bread dough.

  In addition, so that the correlation between the torque change and the temperature inside the kneading machine can be examined, the output of the temperature sensor mounted in the kneading machine can also be taken in, and the time-dependent change in the DC component of the frequency spectrum of the two kneading machines If the spectrum can be measured simultaneously with the change with time in the vicinity of 0.2 Hz, the determination accuracy is further improved.

Next, the confirmation result of the present invention will be described. These confirmation results were obtained using the mixer and raw materials shown in FIG.
Example 1 shows the results of using pastry dough and mixing conditions shown in FIG. 4 and the following mixing conditions. Along with the time-dependent change in torque (power consumption) of the target kneader, the analysis results after performing the FFT conversion are displayed as a change over time in the DC component and in the spectral intensity distribution in the vicinity of 0.2 to 0.5 Hz. Is.

[Mixing conditions]
(1) After adding flour, sugar, salt, dough improving material and yeast to a horizontal mixer, add whole egg and water and start stirring the mixer.
(2) The stirring conditions were low speed 3 minutes 55 seconds and high speed 12 minutes 00 seconds, mixing was performed, the mixing load was measured, and the dough state was monitored.
In addition, -8 ° C. chiller water was allowed to flow into the mixer jacket during mixing to suppress the rise in the dough temperature. The temperature at the time of kneading was 16.5 ° C.

  Example 1 is a pastry dough, blended with soft flour, and used as a raw material a dough without adding fats and oils. The DC component gradually increases in amplitude (elasticity) in the low-speed stirring region, and rapidly increases in the high-speed stirring region, and thereafter changes around 600. Then, the amplitude decreases slightly, but finally increases, and at the same time, the frequency region also enters the high-speed agitation region, and the peak of 0.26 Hz appearing from the initial stage increases, and further to the high frequency side of 0.27 Hz. It shows a state where the fabric is shifted and connected.

Example 2 shows the results obtained by using the dough for anpan and mixing with the mixing composition shown in FIG. 5 and the following mixing conditions. FIG. 5 shows the analysis results after performing the FFT conversion together with the torque (power consumption) change with time of the target kneader. Is displayed on the monitor.
[Mixing conditions]
(1) After adding wheat flour, sugar, nonfat dry milk, shortening, dough improving material and yeast to the horizontal mixer, add whole egg, fermented flavor liquid and water, and start stirring the mixer.
(2) Stirring conditions are low speed 3 minutes and high speed 4 minutes. After mixing, salt is added.
Again, mixing was performed at a low speed of 1 minute and at a high speed of 10 minutes to obtain an anpan dough.
The mixing load was measured and the dough condition was monitored.
In addition, -10 degreeC chiller water was continuously poured into the mixer jacket during mixing, and the rise in the dough temperature was suppressed. The temperature of the dough at the time of kneading was 16.8 ° C.

Example 2 is a dough mainly composed of strong powder, with a large amount of sugar and shortening added, and a dough that is easy to collapse. The composition is also a hard dough, and the DC component gradually increases in amplitude (elasticity) in the initial low-speed stirring region and exceeds 800 in the next high-speed stirring region, but the amplitude decreases sharply by adding salt, After that, it enters the high-speed stirring zone, and the amplitude increases rapidly and exceeds 600 and remains unchanged.
On the other hand, in the frequency range, after addition of salt, a peak appears in the vicinity of 0.26 Hz from the initial stage, and mixing proceeds while repeating the peak intensity between frequencies of 0.25 to 0.26 Hz. Eventually, the DC component rises, and at the same time, the peak in the frequency region also forms a large peak in the vicinity of 0.26 Hz, and further shifts to the high frequency side of 0.27 Hz, indicating a state where the fabric is connected.

Example 3 shows the results of using a cream bread dough with the mixing composition shown in FIG. 6 and the following mixing conditions. FIG. 6 shows the analysis results after performing the FFT conversion together with the torque (power consumption) change with time of the target kneader. Is displayed on the monitor.
[Mixing conditions]
(1) Add wheat flour, sugar, nonfat dry milk, butter, dough improving material and yeast into a horizontal mixer, then add whole egg, fermented flavor liquid and water, and start stirring the mixer.
(2) Stirring conditions are low speed 3 minutes and high speed 4 minutes. After mixing, salt is added.
Again, mixing was performed at a low speed of 1 minute and a high speed of 28 minutes to obtain a dough for cream bread.
The mixing load was measured and the dough condition was monitored.
In addition, -10 degreeC chiller water was continuously poured into the mixer jacket during mixing, and the rise in the dough temperature was suppressed. The temperature of the dough at the time of kneading was 16.8 ° C.

  In Example 3, a dough mainly composed of strong powder is used, and reduced maltose (liquid sugar) is added in addition to addition of sugar and butter. The DC component tends to increase in amplitude (elasticity) in the initial low-speed stirring region, but exceeds 600 in the next high-speed stirring region. The amplitude drops sharply after the addition of salt, but in the subsequent high-speed agitation zone, the amplitude is less than 600 and is mixed at around 500 due to the softly blended dough. As the mixing progresses, the dough is connected and the DC component rises to nearly 600. At the same time, a peak in the frequency range also appears in the vicinity of 0.25 Hz, and finally a large peak is formed in the vicinity of 0.26 Hz (increase in viscosity), indicating a state where the dough is connected.

  As described above, according to these examples, during the kneading operation of bread dough, the analysis results after performing the FFT conversion as shown in FIGS. If you display the change over time in the spectral intensity distribution near 0.5 Hz on the monitor screen, you can operate while looking at the monitor screen, so you can easily understand the progress of mixing and the mixing failure, and the time and frequency distribution of the rotor torque waveform Can grasp the kneading end point objectively, clearly and optimally by combining the time point when a peak appears in the vicinity of 0.2 Hz and the change with time of the DC component.

  According to the present invention, not only kneading of bread dough, but also widely applicable to all foods that require monitoring of kneading state such as hamburger, fish paste product, whipped cream, and objectively without the need to visually check the kneading state This makes it possible to determine the completion of the kneading operation accurately and easily without having to stop the apparatus in the conventional manner and visually observe the inside of the casing, which is beneficial for improving the quality of the product after kneading. It is possible to detect the behavior of the kneading state and determine the behavior.

It is process drawing which shows the analysis procedure (real-time frequency analysis by short-time Fourier transform) of this invention. It is a monitor screen which shows the time-dependent change of the rotor torque waveform of the horizontal type kneader according to the invention, the time-dependent change of the DC component (0 Hz), and the time-dependent change of the spectral intensity distribution in the vicinity of 0.08 to 0.5 Hz. It is a chart which shows the mixer and raw material which were used in the Example of this invention. It is a monitor screen which shows a time-dependent change of the rotor torque waveform of the kneading | mixing state of bread dough in Example 1 of this invention, a time-dependent change of DC component (0 Hz), and a time-dependent change of the spectral intensity distribution in 0.08-0.5 Hz vicinity. . It is a monitor screen which shows the time-dependent change of the rotor torque waveform of the kneading | mixing state of bread dough in Example 2 of this invention, the time-dependent change of DC component (0 Hz), and the time-dependent change of the spectral intensity distribution in 0.08-0.5 Hz vicinity. . It is a monitor screen which shows a time-dependent change of the rotor torque waveform of the kneading | mixing state of bread dough in Example 3 of this invention, a time-dependent change of DC component (0 Hz), and a time-dependent change of the spectrum intensity distribution in 0.08-0.5 Hz vicinity. . It is the schematic which shows an example of the horizontal type kneader applied to this invention, (a) is an elevation, (b) is a side view.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Kneading machine 1a Rotating shaft 1b Stirring bar 2 V belt 3 Drive motor 5 Torque detector 6 Measurement / analysis PC
8 Monitor

Claims (7)

In a method for detecting the behavior of the kneading state of the food dough made of a viscoelastic material that is kneaded in a kneader by rotating a rotor having a rotating shaft,
The torque detection value detected from the rotating shaft is subjected to frequency analysis for each time interval, and a DC component having a frequency of zero and a frequency peak intensity in a specific frequency band are extracted from the analyzed frequency spectrum time series data, and the DC Along with changes in the components, the changes over time in the peak intensity of the specific frequency band are displayed two-dimensionally or three-dimensionally by incorporating a time axis, and the kneaded state of the viscoelastic material is displayed based on the two changes over time. A method for detecting the behavior of a kneaded state of a food dough, characterized by detecting the behavior of the food.   The method for detecting a behavior of a kneading state of a food dough according to claim 1, wherein the change with time of the torque detection value is displayed together with the display of the two changes with time.   2. The frequency analysis for each time interval is performed using one analysis method among short-time Fourier transform (STFT), Wavelet transform (WT), and Wigna analysis (WD). A method for detecting the behavior of the kneaded state of the food dough as described.   The torque detection value is subjected to a short-time Fourier transform process by applying a window function, and a DC component having a frequency of zero and a frequency peak intensity in a specific frequency band are extracted from the frequency spectrum after the processing, along with the change with time of the DC component The time-dependent change of the peak intensity in the specific frequency band is displayed two-dimensionally or three-dimensionally by incorporating a time axis, and the behavior of the kneaded state of the viscoelastic material is detected based on the display of the two time-dependent changes. The method for detecting the behavior of a kneaded state of a food dough according to claim 1.   2. The method for detecting the behavior of a kneaded state of a food dough according to claim 1, wherein the behavior is a kneading completion timing of the food dough, which is a viscoelastic material, or an addition timing of the auxiliary material.   When the food dough is bread dough, a DC component having a frequency of zero and a frequency component in the vicinity of 0.08 Hz to 0.5 Hz are extracted from the frequency spectrum and the change with time is displayed. 3. A method for detecting the behavior of a kneaded state of a food dough according to 3.   5. The method for detecting the behavior of a kneading state of a food dough according to claim 1 or 4, wherein in the DC component display, the color is varied depending on whether the DC component is rising or falling.

Images