JP3090571B2 - 1 / f fluctuation controller and 1 / f fluctuation air conditioning system - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a 1 / f fluctuation controller and a 1 / f fluctuation controller using a chaotic oscillator for creating a more comfortable environment.
f. It relates to a fluctuation air conditioning system.

[0002]

2. Description of the Related Art Fluctuations have been found in various places. In the wind, sound (music), smell, etc., which are indispensable elements in human comfortable space, and in the heartbeat that has been thought to pulse regularly and evenly at rest,
Fluctuations are detected.

[0003] Fluctuations are classified into three types: 1 / f fluctuations, 1 / f ² fluctuations, and white fluctuations (White Noise). It is said that the 1 / f fluctuations therein provide a comfortable feeling to a person. Conversely, the 1 / f ² fluctuation and the white fluctuation give discomfort. However,
Here, “f” is a variable frequency. Since the fluctuation frequency indicates how quickly the oscillation frequency fluctuates, care must be taken so as not to be confused with the actual spread of the oscillation frequency spectrum.

[0004] The 1 / f fluctuation is included in various natural phenomena such as a breeze on a plateau and the sound of waves on the seashore. Spectral analysis is performed to investigate the nature of the fluctuation.
^The 1 / f fluctuation has the same slope as ^-1 . Humans constantly expect various things in their daily lives. A space with 1 / f fluctuation moderately answers the expectation or betrays the expectation. There seems to be a betrayal that humans feel comfortable in the 1 / f fluctuation.

[0005] Comfort is a state that is well suited to the mind and body and is comfortable, and comfort can be said to be a unique feature inherent in comfort. In other words, the human perceives the state of comfort as one characteristic of comfort, and feels it with the sensitivity of comfort.
In other words, it can be said that people accept comfort by the work of sensitivity.

[0006] Here, the word "sensitivity" has come out, which is a general term for the ability of human beings to sense external stimuli, and has a variety of meanings. Sensibility is sometimes used in the sense of "sensitivity", or perceived as "sense", or "acceptance". Human sensibility is deeply and widely developed. The sensibility can be said to be an overall feeling of feeling and feeling, a feeling unique to the person.

The four human senses (visual, auditory, olfactory, and tactile sensations) are pillars for obtaining a sense of comfort, and the sensibility (psychology) reacts to this to be able to feel the comfort.
Even if all four senses are not satisfied, if there is an element that feels "comfortable" in any one, it is thought that the other senses will be tuned in a direction that feels comfortable due to the sensitivity. .

[0008] It is very difficult to evaluate comfort. Although comfort is naturally evaluated based on comfort, there is an individual difference in the comfort, and it is difficult to create a standard. Although evaluation should be objective in the first place, this is also why such an evaluation method has not been established.

Here, the most general subjective evaluation method is introduced, and an objective evaluation method which is being established is also described. First of all, the subjective evaluation method (questionnaire method) is to ask the subject to experience something (for example, music or smell) that actually contains 1 / f fluctuation, It is to have you answer. The contents of the questionnaire are about 20 items, such as "calm, refreshing, annoying, boring", and the answer methods are ± 3 (comfortable, uncomfortable), ± 2 (very comfortable, quite uncomfortable), ± 1 (somewhat comfortable, Slightly uncomfortable), and 0 (normal)
One of the seven types is selected. The objective evaluation method (method of detecting physiological data) also causes the subject to experience 1 / f fluctuations, during which the heart rate and brain waves (α waves) are measured and analyzed, and the analysis result is 1 If the / f trajectory is shown, it is comfortable, and if the white fluctuation is shown, it is uncomfortable. However, this method only requires "analysis" at this stage, so more analyzes and experiments are needed to be established.

Since the above method cannot be evaluated unless there is an object to be compared, not only the 1 / f fluctuation but also the 1 / f ² fluctuation and the white fluctuation can be experienced at the same time. It does not become.

[0011] The "comfort" required by humans, of course, may vary somewhat from individual to individual, but since it is considered to be in the same direction, the extracted answers are expected to be highly reliable. .

When engineering 1 / f fluctuations, 1
It is necessary to create time-series data having a trajectory of / f fluctuation. For that purpose, there is a method using the concept of chaos technology.

Chaos = deterministic chaos (periodic motion ← → prechaos (quasi-periodic motion) ← → chaos) Chaos (Cha)
os) refers to the phenomenon that looks very messy but is actually governed by simple rules.

[0014]

(Equation 1)

Equation (1) is a broken line model of a discrete dynamical system that causes a typical chaotic state as shown in FIG. How this track is changed to vary according to changes in a (initial value is x _0). However, in this example, the operation of at least x _4, point # 8 below is truncated. That is, a rounding error occurs. It should be noted that no matter how much the accuracy of the computer is increased, there is a problem that a true chaos trajectory is not tracked in a numerical calculation of a computer or the like.

The apparently complex movements can be divided into two parts: the synthesis of a large number of periodic movements and the complex time intervals resulting from different mechanisms in principle. In the conventional signal processing technology, information on a source of a complicated motion is extracted by analyzing a frequency component.

Chaos can be found in various places on the earth. However, when chaos is applied in engineering, the chaos phenomenon cannot be used as it is. This is because chaos (complex behavior) has not been fully elucidated, and it is difficult to identify system equations.

When data of various types (natural style, music, etc.) are actually taken, it seems that at first glance it is very random and the regularity is not found anywhere. However, this involves intermittent chaos. It can be seen from the analysis that a clean 1 / f fluctuation is shown. As described above, in order to obtain a 1 / f fluctuation trajectory, it is a shortcut to focus on chaos.

Then, how should the 1 / f trajectory be determined? Currently, two methodologies are being considered. One is a more engineering method using white fluctuation, and the second is a method of directly sampling data and analyzing it. However, when the latter method is used, 1 / f fluctuation cannot be engineered unless the function is approximated by following a scenario leading to chaos.

For the white fluctuation, a random number incorporated in a computer or the like can be used. In order to make this a 1 / f orbit, the white fluctuation may be integrated by ぎ times. Thus, equation (2) is obtained by considering n times differentiation of the data Xm at the discretized time interval δ and formally converting it into an equation corresponding to integration. However, when r = 0, the coefficient of Xm is 1.

[0021]

(Equation 2) If n = 1/2 in this equation, equation (2) becomes

[0022]

(Equation 3) Becomes Also, if you expand this further,

[0023]

(Equation 4)

[0024]

(Equation 5)

[0025]

(Equation 6) Becomes

[0026]

As described above, 1
There are two methods for creating the / f fluctuation, but the above two methods have the following disadvantages. [When white fluctuation is used] (1) A new idea of incorporating the concept of chaos cannot be used.

(2) It is expected that a slightly larger microcomputer system will be required. [When Data is Measured and Analyzed] (1) The time and place for that are required, and it is difficult to secure it.

(2) Function approximation must be performed. The present invention has been made in view of the above circumstances, and artificially (engineering) a chaotic trajectory to create a 1 / f trajectory based on the chaotic trajectory. An object of the present invention is to provide a 1 / f fluctuation controller and a 1 / f fluctuation air-conditioning system which can be acquired and implemented in hardware, eliminate the trouble of measuring data, and thereby eliminate the need for function approximation.

[0029]

In order to achieve the above object, a 1 / f fluctuation controller according to the present invention comprises a control section for generating a timing signal, and a timing signal added from the control section to generate a chaotic orbit. An initial setting section for setting values of a starting point and a vertex; a chaos oscillator section to which a timing signal is added from the control section to cause a chaotic trajectory based on the set value of the initial setting section; 1 / is obtained by integrating a plurality of data obtained as discrete values by half.
A 1 / f fluctuation generator that generates time-series data of f orbitals, and D (digital) / A that converts digital data sent from the 1 / f fluctuation generator into an analog signal.
(Analog) conversion unit.

The 1 / f fluctuation air conditioning system of the present invention is characterized in that the above 1 / f fluctuation controller is used in an air conditioner. Therefore, by creating a chaotic trajectory artificially (engineering) and creating a 1 / f trajectory based on it, a new attempt to connect the chaotic concept with the 1 / f fluctuation can be made. Omission, thereby eliminating the need for function approximation.

[0031]

Embodiments of the present invention will be described below in detail with reference to the drawings. FIG. 1 is a circuit diagram showing a 1 / f fluctuation controller according to an embodiment of the present invention, and FIG. 2 is a timing chart showing signal timings of respective parts in FIG.

That is, the 1 / f fluctuation controller is divided into five parts: an initial setting unit 1, a control unit 2, a chaotic oscillator unit 3, a 1 / f fluctuation generation unit 4, and a D / A conversion unit 5. . The initial setting unit 1 has a role of setting values of the starting point of the chaotic trajectory and the vertices of the triangle. The chaotic trajectory starts from the input initial value, but since the initial value is used only once, this part is automatically closed when the initial input ends.

The control unit 2 generates a timing signal.
Control is performed so that data can smoothly flow through the circuit based on a predetermined time interval. The chaotic vibrator 3 has a role of causing a chaotic orbit and sending data obtained therefrom to the 1 / f fluctuation generator 4.

The 1 / f fluctuation generator 4 generates time-series data of the 1 / f trajectory according to the following flow. Data sent from the chaotic vibrator portion 3, can store up to four (in the current data and x _m, past three data becomes _{x m-1, x m-} 2, x m-3) . A solution obtained by integrating the four data as discrete values １ ／ times becomes the required data. However, since the data in the chaotic orbit will be incremented, the oldest data is discarded when the fifth data has been stored in the x _m.

The D / A converter 5 includes a 1 / f fluctuation generator 4
This is a part for converting the numerical value (digital data) obtained in the above into an analog signal. Next, FIGS. 1 and 2 will be described in detail.

The control unit 2 includes a clock pulse generation circuit 1 having a power supply voltage application terminal T2, a switch S1, and a resistor R3.
The clock pulse generating circuit 1 generates a clock pulse CK, and the timing signal generating circuit 2 generates a timing signal Q. The timing signal generating circuit 2 includes a demultiplexer 3, gate circuits G3 and G4, and inverters I4 and I5.
A, QB, QC, QD, QE, and QF are generated. ARS
Is an automatic return signal. The timing signal δ is generated by the switch S1, and the time interval of the timing signal δ is one cycle. The timing signal δ indicates a time until the timing signal QF is inverted.

The initial setting section 1 is composed of a power supply voltage application terminal T1, resistors R1 and R2, a capacitor C1, a diode D, inverters I1, I2 and I3, and a flip circuit of gate circuits G1 and G2. Signal QF is applied.

The operation of the initial setting unit 1 is determined by the node B and the timing signal QF. When the board rises, both the node B and the timing signal QF are at L (low level), so that as shown in FIG.
1) = L, R (gate circuit G2) = H (high level), so that the signal sent to the chaotic vibrator unit 3 becomes H. Therefore, the gate circuits G5, G in the chaotic vibrator section 3
6, the initial value x ₀ is to be selected which is determined by the setting of the switch 4 of the chaotic oscillator section 3 at from consisting selection circuit SEL1 G7.

However, after the second time, the chaotic vibrator unit 3
Initial value x ₀ which is determined by the setting of the switch 4 is not selected, so that the signal from xm is selected. The reasons are as follows.

That is, as shown in FIG. 4, the node B is automatically inverted from L to H after a short period of time, and keeps its state. On the other hand, the timing signal QF repeats inversion every cycle. In the second time, since S = H and R = L, the signal sent to the chaotic vibrator unit 3 becomes L. However, since the state of S = H continues since there is no inversion at the node B, the output state becomes L no matter how many times R is inverted. As described above, after the second time, the initial value x ₀ determined by the setting of the switch 4 of the chaotic vibrator unit 3 is not selected.

The chaos vibrator section 3 has three dip switches S2, S3, S4. Dip switch S4
Is important only for the first cycle,
The other two dip switches S2 and S3 always play an important role.

The curve ROM 5 in the chaotic vibrator section 3 has 25
Six types of function patterns can be stored, and the dip switch S3 is a switch for selecting one of the patterns. As shown in FIG. 5, the data xn flowing through the chaos oscillator unit 3 through the selection circuit SEL1 and the latch circuit 4 including the D-type flip-flop is substituted into the selected function, and thereafter, the data xn is multiplied. (MU
LT; algebraic product) 7. In the multi 7, the data xn sent from the curve ROM 5 and the arbitrarily selected data aj are algebraically multiplied (2 * aj * xn), as shown in FIG. S
It is the role of 2. However, the data aj is divided into 256 data from 0 to 255, which is made to correspond to data from 0 to 1.

The data value obtained by the algebraic product becomes data xm and is sent to the 1 / f fluctuation generating unit 4, while data xn + 1
Is returned to the chaotic vibrator unit 3 via the selection circuit SEL2 and the latch circuit 6 including a D-type flip-flop.

The chaotic vibrator section 3 is provided with power supply terminals T3, T4, T5 and resistors R4, R5, R6. Now, why the value a of the vertex of the triangle of the chaotic orbit is changed will be described here. This is based on the fact that the state of the broken line model of the discrete dynamical system shown in FIG. 10 changes variously depending on the change of the value a of the vertex of the triangle. 0 <a <0.
5, the data xn always converges under the conditions of a = 0.5 and a = 1, but under the condition of 0.5 <a <1, the data xn becomes infinitely many periodic Draw a trajectory. This “infinitely many infinite periodic orbits” evolve from prechaos (quasi-periodic motion) to chaos. In this case, a = 0.99 is the best condition for obtaining the 1 / f trajectory. In the control unit 2, the time interval (second) of the timing signal δ can be selected as follows according to the state of the DIP switch S1.

[0045]

(Equation 7)

The timing signal Qa in the control section 2 becomes L when the timing signals CK = 1, QA = 1, QB = 1, and QC = 0. When the timing signal Qa becomes L, it enters the demultiplexer 3 in the control unit 2 along the flow, where it is crossed with the timing signals QD and QE. When the timing signals QD = 0 and QE = 0 (A in FIG. 2), this leads to the timing signal QDE0. Timing signal QD
Since E0 is connected to the latch circuit 10 composed of a D-type flip-flop in the 1 / f fluctuation generator 4, when the timing signal QDE0 becomes L, the value of the data xm-2 is moved and held as the data xm-3, The value of the data xm-3 is automatically discarded. Timing signal QD = 1, QE = 0
At this time (a in FIG. 2), this leads to the timing signal QDE1. Since the timing signal QDE1 is connected to the latch circuit 9 composed of a D-type flip-flop in the 1 / f fluctuation generator 4, when the timing signal QDE1 becomes L, the value of the data xm-1 is moved and held as the data xm-2. You. When the timing signals QD = 0 and QE = 1 (C in FIG. 2), this leads to the timing signal QDE2. The timing signal QDE2 is connected to a latch circuit 8 composed of a D-type flip-flop in the 1 / f fluctuation generator 4 and a latch circuit 6 composed of a D-type flip-flop in the chaotic vibrator 3. When the timing signal QDE2 becomes L, the value of the data xm is moved to and held in the data xm-1 in the latch circuit 8, and the data xn + 1 ( x
m) flows into the chaotic vibrator portion 3. When the timing signals QD = 1 and QE = 1 (d in FIG. 2), this leads to the timing signal QDE3. Timing signal QDE3
Is connected to the latch circuit 4 in the chaotic vibrator section 3, so that when the timing signal QDE3 becomes L, the data xn +
1 (xm) is moved and held to become data xn, and then passes through the curve ROM 5 and the multi 7 to obtain data xn + 1 (xm).
become. With this, the data value xm required to generate 1 / f fluctuation
, Xm-1, xm-2, and xm-3, are calculated by the corresponding arithmetic circuits 14, 13, 12, and 11, respectively.
The calculation formula is as follows.

[0047]

(Equation 8) Since the required data is four, N = 3, so

[0048]

(Equation 9) Becomes Here, n represents a function using the value a of the vertex of the triangle of the chaotic orbit as a variable.

Based on this equation, the data xm, xm-1 and x
m−2 and xm−3 are added by adders (ADD) 15 and 16 in the 1 / f fluctuation generator 4 respectively, and the answer is further added by an adder (ADD) 17. The result is data ym. This data ym is sent to the D / A converter 5.

The D / A conversion section 5 includes a D / A conversion circuit 19, operational amplifiers 18, 21, 22 and power supply voltage application terminals T6, T
7, T8, T9, resistors R7, R8, R9, R10, R1
1, R12, R13, R14, R15, R16 and output terminals T10, T11.

The D / A converter 5 converts the digital data into an analog signal.
/ A SET works. The timing signal D / A SET includes timing signals CK = 1, QA = 1, QB = 0, QC =
It is inverted only when 1, QD = 1 and QE = 1 (e in FIG. 2). This is because the timing signal D / AS is performed after all the operations from the timing signals QDE0 to QDE3 are completed.
This indicates that the ET is activated (see FIG. 2).

The D / A converter 5 converts the data ym represented by decimal numbers 0 to 1,023 into analog signals 0 to 5
Works to assign to V. By the above operation, one 1 / is set within the time interval (= 1 cycle) of the timing signal δ.
An f-fluctuation output is obtained.

Next, a simulation will be described.
The simulation shows whether or not a 1 / f trajectory can be created when a chaotic trajectory is generated and integrated １ ／ times.

As shown in FIG. 11, 1 /
The twice integration is made up of three parts.
1 generates a chaotic orbit. Since the vertices of the triangle may converge at 100%,
It is set to 99%. In addition, the initial setting (Starti
ng Point) is arbitrarily determined. Next, Figu
re-2 represents a numerical value obtained by plotting the chaotic trajectory obtained in FIG. 1 in accordance with the set time interval, and then integrating it １ ／ times. The horizontal axis indicates time, and the vertical axis indicates a change range after integrating the chaotic trajectory by ２ times. Finally, Figure-3 shows the results of Figure-2 as orbitals. The horizontal axis indicates the fluctuation frequency (Hz), and the vertical axis indicates the power level. And it turns out that the result has become almost 1 / f orbit.
This has revealed that a 1 / f orbit can be obtained from a chaotic orbit under certain conditions. The number of samplings in the simulation is 2,048 points per second, where one analysis is one second, so that the time interval δ
Is very short, 1 / 2,048 seconds. Although the frequency is considerably high because the time interval is short, the time width is variable (that is, one analysis may be one hour or one day).

In this embodiment, development of a board is considered when generating 1 / f fluctuation. The most important reason is that once a board is developed,
/ F fluctuation application device. Although it is assumed that a commercially available IC is used, there is an advantage that the cost can be reduced.

Next, another embodiment of the present invention will be described. The application of 1 / f fluctuation can be considered in a wide range of fields from home appliances such as electric fans and massage machines to medical devices such as pain relief, but in the present embodiment, it is desired to make use of this in the air conditioning field. Today, with the changes in the work content and the improvement of the work environment, a review of the comfortable space has been called out from various directions. In such a situation, 1 / f fluctuation is applied to air conditioning which is indispensable for a comfortable space.

There are two types of air conditioning: a method of changing the air supply temperature without changing the air supply amount and a method of changing the air supply amount without changing the air supply temperature. Adopt the latter.

FIG. 7 shows an example of a 1 / f fluctuation air conditioning system, which comprises a personal air conditioning system and a general air conditioning system. That is, the personal air-conditioning system sets the initial value and the air volume, and sets the chaotic vibrator unit 3, the 1 / f fluctuation generating unit 4 and the D
Using a 1 / f fluctuation controller composed of a / A conversion unit 5 and the like,
The personal air conditioner 72 including an air conditioner and a variable air valve VAV is inverter-controlled. In addition, the general air conditioning system adds output data of the 1 / f fluctuation generator 4 of the 1 / f fluctuation controller to the integrated processing device 71 and branches the data into a plurality of data, and converts each data into a plurality of D / A converters 5. And an air conditioner 73 including an air conditioner, a variable air valve VAV, and the like.

FIG. 8 shows a specific example of a 1 / f fluctuation air conditioning system, in which the output of an air conditioner (AHU) 82 is applied to a blower 83 to blow air to a variable air valve VAV. The variable air valve VAV receives a 1 / f fluctuation signal from the air conditioner 82 and controls the 1 / f nozzle air blowing speed to the controlled room 81 whose room temperature is controlled to be constant. RA is the return air.

If FIG. 11 in FIG. 11 is directly adapted to the air conditioning control, the vertices of the triangle can be applied to the air supply amount. Since the air supply temperature is not changed, the air supply amount may be changed as shown in FIG. 9 to change the room temperature. Therefore, the increase or decrease of the air supply amount can be determined by fuzzy inference.

[0061]

As described above, according to the present invention, a chaotic trajectory is artificially (engineered) and the 1 / f
By creating a trajectory, 1 / f
Acquires fluctuations and can be implemented in hardware, eliminating the need to measure data, thereby eliminating the need for function approximation

[Brief description of the drawings]

FIG. 1 is a circuit diagram showing a 1 / f fluctuation controller according to an embodiment of the present invention.

FIG. 2 is a timing chart showing timings of signals of respective units in FIG. 1;

FIG. 3 is an explanatory diagram illustrating an example of a function of an initial setting unit according to the present invention.

FIG. 4 is a characteristic diagram showing a response characteristic of the auto reset circuit according to the present invention.

FIG. 5 is a characteristic diagram for explaining an operation in the curve ROM according to the present invention.

FIG. 6 is a characteristic diagram for explaining an operation of a multi (MULT) according to the present invention.

FIG. 7 is a configuration explanatory view showing an example of a 1 / f fluctuation air conditioning system according to the present invention.

FIG. 8 is a configuration explanatory view showing a specific example of a 1 / f fluctuation air conditioning system according to the present invention.

FIG. 9 is a characteristic diagram for explaining a range of an air supply amount of the 1 / f fluctuation air conditioning system according to the present invention.

FIG. 10 is a characteristic diagram illustrating a broken line model of a discrete dynamical system that causes a chaotic state.

FIG. 11 is a measurement diagram showing a simulation of 回 integration by a chaotic orbit.

[Explanation of symbols]

DESCRIPTION OF SYMBOLS 1 ... Initial setting part, 2 ... Control part, 3 ... Chaos vibrator part, 4
... 1 / f fluctuation generator, 5... D / A converter.

Claims (2)

(57) [Claims] A control unit for generating a timing signal; an initial setting unit for adding a timing signal from the control unit to set a starting point and a vertex value of a chaotic trajectory; and a timing signal for adding a timing signal from the control unit. A chaotic oscillator that generates a chaotic trajectory based on the set value of the initial setting unit, and time-series data of a 1 / f trajectory obtained by integrating a plurality of data sent from the chaotic oscillator as discrete values by １ ／ times 1 / f fluctuation control section, comprising: a 1 / f fluctuation generating section for generating a digital signal; and a D / A conversion section for converting digital data sent from the 1 / f fluctuation generating section into an analog signal. vessel. 2. A 1 / f fluctuation air conditioning system, wherein the 1 / f fluctuation controller according to claim 1 is used for an air conditioner.