JP6280733B2 - Air conditioning control system and air conditioning control method - Google Patents

Description

  The present invention relates to a technique of an air conditioning control system and an air conditioning control method for adjusting an operating state of an air conditioning unit so that a comfort level of a subject is improved.

  Conventionally, the technology of the air-conditioning control system and the air-conditioning control method which adjusts the driving | running state of an air-conditioning means so that a subject's comfort level improves is known. For example, as described in Patent Document 1.

  Patent Document 1 describes an air conditioning control system including ambient air conditioning means (ambient air conditioning equipment) that adjusts the living room environment as a whole and task air conditioning means (task air conditioning equipment) that locally adjusts the living room environment. Has been. The task air conditioning means enables local air conditioning by blowing out the adjusted air from a plurality of air outlets provided on the floor.

  In the air conditioning control system configured in this way, in addition to air conditioning by the ambient air conditioning means, local air conditioning that blows out air from an arbitrary air outlet of the task air conditioning means suppresses uneven temperature distribution in the room. The air conditioning according to the comfort of the target person (person) in the room can be performed.

  Further, Patent Document 1 can be configured to automatically control the operating state of the air conditioning means (specifically, opening and closing of the air outlet of the task air conditioning means) using a control means such as a personal computer. The effect is also described.

  However, Patent Document 1 does not specifically describe the control content of the operation state of the air conditioning means, and it is difficult to control the improvement of the comfort level of the subject flexibly and efficiently.

JP 2011-2104 A

  The present invention has been made in view of the above circumstances, and a problem to be solved is to provide an air conditioning control system and an air conditioning control method capable of improving comfort level efficiently and flexibly. It is.

  The problem to be solved by the present invention is as described above. Next, means for solving the problem will be described.

That is, in claim 1, the air conditioning means for performing air conditioning in the room, the comfort level detecting means for detecting the comfort level that is the degree of comfort of the target person with respect to the air conditioning by the air conditioning means, and the comfort level of the target person. An air conditioning control system comprising: control means for adjusting the operating state of the air conditioning means so as to improve, wherein the control means is the air conditioning that affects the comfort level of the subject and the comfort level. And the desired value of the operating state of the air conditioning means capable of improving the comfort level of the subject person from the relationship between the learned comfort level of the subject person and the factor relating to the air conditioning. Calculate and add the noise calculated based on the biological fluctuation theory to the desired value to calculate the target value of the operating state of the air conditioning unit, and control the operating state of the air conditioning unit to be the target value A series of It is configured to repeat the air conditioning control, when the target person is more than one, and calculates the magnitude degree of which impact the effect of factors relating to the air conditioning has on the comfort of the subject, the impact The average value of the comfort level of each subject weighted by is calculated, noise is added to the desired value that maximizes the average value of the comfort level, and the target value of the operating state of the air conditioning means is calculated. is there.

In Claim 2 , The said air-conditioning means contains the ambient air-conditioning means which air-conditions the whole room, and the task air-conditioning means which performs the local air-conditioning around the said subject, The said control means is the said ambient air-conditioning. And the operation of the task air conditioning means are individually controlled.

According to a third aspect of the present invention, the operation state of the task air conditioning unit includes an air volume of air released by the task air conditioning unit during air conditioning.

According to a fourth aspect of the present invention, the control means uses a nonlinear multiple regression analysis when learning the relationship between the comfort level of the subject and the factors related to the air conditioning that affect the comfort level.

According to a fifth aspect of the present invention, the control means is set so that the noise variance decreases as the comfort level of the subject increases.

The air conditioning control method according to claim 6 , wherein the air conditioning control method adjusts the operating state of the air conditioning means so that the comfort level, which is the degree of comfort of the target person with respect to the air conditioning by the air conditioning means, is improved. And the factors related to the air conditioning that affect the comfort level, and the comfort level of the target person can be improved from the relationship between the learned comfort levels of the target person and the factors related to the air conditioning. Calculating a desired value of the operating state of the air conditioning unit, adding a noise calculated based on the biological fluctuation theory to the desired value to calculate a target value of the operating state of the air conditioning unit, It is configured to repeat a series of air-conditioning control for controlling the operation state to be the target value . When there are a plurality of target persons, the influence of the air-conditioning factors on the comfort level of each target person is large. The The degree of influence is calculated, the average value of the comfort level of each subject weighted by the degree of influence is calculated, and noise is added to the desired value that maximizes the average value of the degree of comfort. The target value of the operation state of the means is calculated.

  As effects of the present invention, the following effects can be obtained.

In claim 1, by simultaneously learning the driving state that can improve the comfort level of the subject by learning the relationship between the comfort level of the subject and the factors related to air conditioning that affect the comfort level. By using the biological fluctuation theory, it is possible to flexibly search for a driving state capable of improving the comfort level of the subject from the noise range. Moreover, the comfort levels of a plurality of subjects can be improved at the same time.

In claim 2 , even when the ambient air conditioning means and the task air conditioning means are used in combination, it is possible to control the driving state so as to improve the comfort level of the subject.

In Claim 3 , the comfort level of the subject can be improved by controlling the air volume of the task air conditioning means.

In Claim 4 , the relationship between a subject's comfort level and the factor regarding the air conditioning which affects the said comfort level can be learned appropriately.

According to the fifth aspect , the comfort level of the subject can be improved more efficiently and flexibly.

In claim 6 , by simultaneously learning the driving state that can improve the comfort level of the subject by learning the relationship between the comfort level of the subject and the factors relating to air conditioning that affect the comfort level. By using the biological fluctuation theory, it is possible to flexibly search for a driving state capable of improving the comfort level of the subject from the noise range. Moreover, the comfort levels of a plurality of subjects can be improved at the same time.

The schematic diagram which showed the whole structure of the air-conditioning control system which concerns on one Embodiment of this invention. (A) The flowchart which showed the process of the air conditioning control which concerns on this embodiment. (B) The flowchart which showed the detail of the process of learning. (A) The figure which showed an example of the content of the term regarding a subject's sensible temperature among comfort levels. (B) Similarly, the figure which showed an example of the content of the term regarding whether the air volume of an electric fan is the air volume which can be accept | permitted.

  First, the structure of the air-conditioning control system 4 which concerns on one Embodiment of this invention is demonstrated using FIG.

  The air conditioning control system 4 performs indoor air conditioning (air conditioning) so as to improve the comfort level of a person (target person 2) who is present inside the room 1 (in the room). The air conditioning control system 4 mainly includes air conditioners 10 and 10, a fan 20, a temperature sensor 30, a comfort level reporting dial 40, and a personal computer 50.

  The air conditioners 10 and 10 are an embodiment of an air conditioning unit and an ambient air conditioning unit according to the present invention. The air conditioners 10 and 10 perform air conditioning of the entire room. The air conditioners 10 and 10 are arranged at appropriate positions in the room 1. The air conditioner 10 can blow out air (cold air) cooled to a low temperature. The air conditioner 10 can adjust the temperature of the entire room to an arbitrary temperature by switching presence / absence of cold air blowing.

  The electric fan 20 is an embodiment of the air conditioning means and the task air conditioning means according to the present invention. The electric fan 20 performs local air-conditioning around the subject 2 in the room. The electric fan 20 is disposed in the vicinity of the subject 2. The electric fan 20 can change the air volume into four stages of “weak” (the weakest wind), “medium”, “strong” (the strongest wind), or “off” (the state where the blowing is stopped). it can. Further, the electric fan 20 can arbitrarily change (swing) the blowing direction within a predetermined range. By appropriately controlling the operation of the electric fan 20 (that is, the four-stage air volume and the air blowing direction), local air conditioning around the subject 2 can be arbitrarily performed.

  The temperature sensor 30 detects the temperature in the room 1 (indoor). The temperature sensor 30 is disposed in the room 1 at a position where it is not affected by the air conditioning by the fan 20 (not receiving the wind from the fan 20).

  The comfort level reporting dial 40 is an embodiment of the comfort level detecting means according to the present invention. The comfort level reporting dial 40 detects the comfort level of the subject 2 with respect to factors relating to air conditioning.

  Here, the “comfort level” is a factor related to the surrounding air conditioning by the subject 2 (in this embodiment, (1) the temperature of the subject 2 to be described later and (2) the amount of air that can be allowed by the fan 20. It indicates the degree of comfort (pleasure / discomfort) felt from The comfort level is higher as the subject 2 feels more comfortable, and the comfort level is lower as the subject 2 feels uncomfortable.

  The comfort level reporting dial 40 is disposed on the desk 3 on which the subject 2 is seated. The target person 2 can detect the current comfort level of the target person 2 by indicating the comfort level he / she feels with respect to factors related to the surrounding air conditioning with the comfort level reporting dial 40.

  The personal computer 50 is an embodiment of the control means according to the present invention. The personal computer 50 controls the operation of the air conditioner 10 and the electric fan 20 so that the comfort level of the subject 2 is improved (improved). The personal computer 50 includes a storage unit such as a RAM and a ROM, an arithmetic processing unit such as a CPU, and the like, and can perform predetermined arithmetic processing and storage based on various information.

The personal computer 50 is connected to the air conditioner 10 and can control the operating state of the air conditioner 10.
The personal computer 50 is connected to the electric fan 20 and can control the operating state of the electric fan 20.
The personal computer 50 is connected to the temperature sensor 30 and can always acquire information on the indoor temperature detected by the temperature sensor 30.
The personal computer 50 is connected to the comfort level reporting dial 40, and can always acquire information regarding the comfort level of the subject 2 detected by the comfort level reporting dial 40.

Here, the “operating state” of the air conditioner 10 controlled by the personal computer 50 refers to a target value (set value) of the temperature in the room 1 adjusted by the air conditioner 10. That is, when the operation state (target value) of the air conditioner 10 is set, the personal computer 50 controls the operation of the air conditioner 10 so that the temperature in the room 1 detected by the temperature sensor 30 becomes the target value.
The “operating state” of the electric fan 20 controlled by the personal computer 50 includes the air volume (“weak”, “medium”, “strong” or “off”) of the electric fan 20 and the direction of air blowing (swinging speed). , Frequency, angle, etc.). That is, when the operating state of the electric fan 20 (air volume and direction of blowing) is set, the personal computer 50 controls the operation of the electric fan 20 in accordance with the set operating state.

In FIG. 1, for convenience of explanation, a state in which only one subject 2 is present in the room is illustrated, but the present invention is not limited to this, and two or more subjects 2 are shown. May be in the room. The air conditioning control system 4 can control the operating states of the air conditioner 10 and the electric fan 20 so that the comfort levels of all the subjects 2 in the room 1 are improved. In this case, a comfort level reporting dial 40 is provided for each subject 2.
Further, in FIG. 1, for convenience of explanation, a state in which only one electric fan 20 is provided in the room is illustrated, but the present invention is not limited to this. For example, one electric fan 20 may be provided for each of the plurality of subjects 2. Moreover, you may air-condition the several subject 2 by changing the direction which the one fan 20 blows.

  Next, the state (air conditioning control method) of the air conditioning in the room 1 (control of the air conditioner 10 and the electric fan 20) by the air conditioning control system 4 configured as described above will be described with reference to FIG.

  The air conditioning control system 4 constantly detects the comfort level of each target person 2 so that the comfort levels of all the target persons 2 in the room are improved, and based on the comfort level, the air conditioner (air conditioner) 10 and the fan 20) are automatically controlled. At this time, in order to efficiently and flexibly search for an air conditioning pattern (combination of control values of the operating state of the air conditioner 10 and the fan 20) that improves the comfort level of each subject person 2, "learning" and "biological body" "Fluctuation theory" is used. Hereinafter, the control by the air conditioning control system 4 will be described in order.

  As a premise of control by the personal computer 50, in the initial state, the personal computer 50 sets the air conditioner 10 and the electric fan 20 to a predetermined operation state (initial value).

In step S <b> 110 shown in FIG. 2A, the personal computer 50 acquires the comfort level of each subject person 2 in the room detected using the comfort level reporting dial 40.
After performing the process of step S110, the personal computer 50 proceeds to step S120.

In step S120, the personal computer 50 determines each of the air conditioners (air conditioner 10 and fan 20) from the history of combinations of the control values (air conditioning pattern) of the operating state and the comfort level of each subject 2 obtained in step S110. While learning the relationship between the comfort level of the subject 2 and the factors relating to air conditioning that affect the comfort level, each air conditioning device (the air conditioner 10 and The control value candidates for the operating state of the electric fan 20) are calculated.
Details of the processing content of step S120 will be described later.
After performing the process of step S120, the personal computer 50 proceeds to step S130.

In step S130, the personal computer 50 determines an actual control value (target value) of the operating state of each air conditioner (air conditioner 10 and fan 20) based on the learning result in step S120.
Specifically, a value obtained by adding noise based on the biological fluctuation theory to the control value of the operating state of each air conditioner calculated based on the learning result in step S120 is determined as the actual control value.
Details of the processing content of step S130 will be described later.
After performing the process of step S130, the personal computer 50 proceeds to step S140.

  In step S140, the personal computer 50 controls the operation of each air conditioner so that the operation state of each air conditioner becomes the control value (target value) determined in step S130.

In the air conditioning control system 4, by repeating a series of processing from step S <b> 110 to step S <b> 140 (the series of processing is simply referred to as “air conditioning control”), a plurality of subjects 2 who are present in the room are selected. The air conditioner can be automatically controlled so that the comfort level is improved.
In step S120 of this air conditioning control, it is possible to improve the comfort level of each target person 2 by learning the relationship between the comfort level of each target person 2 and the factors relating to air conditioning that affect the comfort level. It becomes possible to search efficiently the control value of the operating state of each air conditioner. Further, by repeating the learning with the repetition of the air conditioning control, an improvement in the accuracy (certainty) of the learning is expected.
Further, by using the biological fluctuation theory in step S130 of the air conditioning control, it is possible to search for a flexible control value of the operating state in consideration of the learning uncertainty in step S120.

  Next, the processing content of step S120 (learning) will be described in detail with reference to FIG.

  In step S <b> 121, the personal computer 50 determines the comfort level of each target person 2 obtained in step S <b> 110 and the control value of the operating state of each air conditioner when the comfort level is obtained. The relationship between the comfort level and factors related to air conditioning that affect the comfort level is learned (calculated) by nonlinear multiple regression analysis.

  Here, in the present embodiment, the “factor” relating to air conditioning that affects the comfort level of each subject 2 is defined as (1) the sensible temperature of the subject 2 and (2) the amount of air that the air flow of the fan 20 can allow. There are two categories:

  Regarding the factor (1), the temperature of the subject 2 is the temperature in the room 1 minus the decrease in which the temperature of the subject 2 decreases when the subject 2 receives the wind of the electric fan 20. The temperature in the room 1 is detected by the temperature sensor 30. As the decrease value of the sensible temperature by the electric fan 20, a value corresponding to the air volume of the electric fan 20 is stored in the personal computer 50 in advance. For example, when the air volume is “weak”, 1.5 (° C.), when it is “medium”, 2.3 (° C.), when it is “strong”, 2.8 (° C.), “ When it is “OFF”, it is stored as 0 (° C.) or the like.

  Furthermore, since the electric fan 20 can change (swing) the direction in which the air is blown, there is a case where there is a time when the wind does not hit the subject 2 depending on the direction. In such a case, a weighted average value between the time when the wind is hitting the subject 2 and the time when the wind is not hit is calculated and the weighted average value is used as the decrease value. For example, when the air volume of the electric fan 20 is “strong” and the ratio of the time when the subject 2 is hit by the wind is 25 (%), the decrease in the sensible temperature is 2.8 × 25/100 = 0.7. (° C).

  In this way, the sensible temperature of the subject 2 is calculated by subtracting the decrease in the sensible temperature when the subject 2 receives the wind of the electric fan 20 from the temperature in the room 1. For example, if the temperature in the room 1 is 25 (° C.) and the reduction value of the sensible temperature by the electric fan 20 is 0.7 (° C.), the sensible temperature is 25−0.7 = 24.3 (° C.). Further, as described above, even when the wind from the electric fan 20 does not always strike the subject 2 (the electric fan 20 is swung), the sensible temperature can be calculated.

  Regarding the above factor (2), not only the sensory temperature but also the air volume itself of the electric fan 20 is assumed to be unacceptable for the subject 2. The case where the air volume of the fan 20 is unacceptable for the subject 2 is, for example, when the air volume of the fan 20 is too strong and the eyes are dry, or when the fan 20 is too strong and the document is likely to be skipped, making it difficult to perform the work. The case where it becomes is assumed.

  The personal computer 50 learns (calculates) the relationship between the comfort level of each subject person 2 and factors related to air conditioning that affect the comfort level, as shown in the following equation (1).

  As shown in the above equation 1, the regression function of the comfort level of the subject 2 includes the above-described factor (1) the term related to the temperature of the subject 2 (the first term on the right side) and the factor (2) of the electric fan 20. And a term (second term on the right side) relating to whether the air volume is an allowable air volume.

  Of the above formula 1, in the factor (1) the term relating to the sensible temperature of the subject 2 (the first term on the right side), the center and variance are determined according to the Gaussian function. FIG. 3A schematically shows the first term on the right side.

  Here, as shown to Fig.3 (a), the center of the said Gaussian function is the body temperature (feeling body temperature) when the subject 2 feels most comfortable.

  Further, as shown in FIG. 3A, the degree of influence of the sensible temperature of the subject 2 (see the above equation 1) is relatively small (the variance of the Gaussian function is large), and the comfort level is relatively high at a wide range of sensible temperatures. Is obtained. On the other hand, as the influence level of the sensible temperature of the subject 2 is relatively large (the variance of the Gaussian function is small), the range of sensible temperature at which a high comfort level is obtained becomes relatively narrow.

  As described above, the influence level of the sensible temperature of the target person 2 represents the degree of the magnitude of the influence of the sensible temperature on the comfort level of the target person 2. That is, when the degree of influence of the sensible temperature of the subject 2 is large, the comfort level also changes abruptly when the sensible temperature changes. On the other hand, when the degree of influence of the sensible temperature of the subject 2 is small, the comfort level changes relatively slowly as the sensible temperature changes.

  Among the above formula 1, the factor (2) whether the airflow of the fan 20 is an allowable airflow (second term on the right side) is that the airflow of the fan 20 for the subject 2 exceeds the allowable airflow of the subject 2 This means that a constant is subtracted from the comfort level only if the For example, as shown in FIG. 3B, when it is determined by learning that the allowable air volume of a target person 2 is “medium”, the air volume of the electric fan 20 for the target person 2 is “strong” (allowable). In the case of exceeding the air volume), the value of the second term on the right side of Equation 1 is −M. That is, the constant (M) is subtracted from the comfort level of the first term on the right side. On the other hand, when the air volume of the electric fan 20 for the subject 2 is “medium” or less (“medium”, “weak” or “cut”) (below the allowable air volume), the second term on the right side of the above equation 1 The value of is 0.

  After performing the process of step S121, the personal computer 50 proceeds to step S122.

In step S122, the personal computer 50 performs post-processing of nonlinear multiple regression analysis.
Specifically, candidates for control values of the operating state of each air conditioner (air conditioner 10 and fan 20) are calculated using the following equation (2).

Here, there are a plurality of patterns of control value candidates (combinations) for the operating state of each air conditioner that satisfy the condition shown in Equation 2 above (that is, the predicted comfort level of all subjects 2 exceeds a certain value). Can do.
For example, (a) the operation state (target value) of the air conditioner 10 is 25 (° C.), the operation state (air volume) of the electric fan 20 is “weak”, and the wind always hits the specific subject 2; ) When the operation state of the air conditioner 10 is 26 (° C.), the operation state of the electric fan 20 is “medium”, and the wind always hits the specific subject 2, the comfort level of the subject 2 is substantially the same. Conceivable.
Therefore, in the present embodiment, a set of control value candidates for the operating state of each air conditioner in which the predicted comfort levels of all the subjects 2 exceed a certain value is calculated using the above formula 2.

  After performing the process of step S122, the personal computer 50 ends the process of step S120, and proceeds to step S130 (see FIG. 2A).

  As described above, in the process of step S120 (learning), the relationship between the comfort level and the factors affecting the comfort level is learned from the control value of the operating state of each air conditioner and the acquired comfort level of the target person 2. To do. As described above, it is possible to improve the comfort level efficiently by learning and using the relationship between the comfort level and the factors that affect the comfort level (see Equation 1) from limited data.

  Next, the processing contents of step S130 (control value determination) will be described in detail.

In step S130, the personal computer 50 calculates (determines) the actual control value (target value) of each air conditioner based on the learning result in step S120.
Specifically, the personal computer 50 determines the target value of the actual operating state of each air conditioner using the following Equation 3.

  Here, the first term on the right-hand side of the first equation of Equation 3 is the control value of the operating state of each air conditioner considered to improve the comfort level of each subject 2 (hereinafter simply referred to as “desired value”). It is. Specifically, the first term on the right side is normalized so that the total degree of influence of the sensible temperature of each subject 2 is 1, and the average value of the predicted comfort level weighted by the degree of influence is the maximum. This means that a candidate for a control value is determined as a “desired value”. Thus, the desired value is a value such that the air volume of each fan 20 is equal to or less than the allowable air volume of each subject 2 and the sensible temperature of each subject 2 is close to the comfortable sensible temperature of each subject 2. .

  In addition, the second term on the right side of the first expression of Equation 3 is noise based on biological fluctuation theory (fluctuation).

Here, “biological fluctuation theory (fluctuation)” is a technique for changing the magnitude of noise according to the goodness of the situation and searching for a control value from the range of the noise.
That is, when the situation is good, the control value is carefully searched from the vicinity (narrow range) using small noise. On the other hand, when the situation is bad, the control value is searched from a wide range using a large noise.

  In the present embodiment, the good situation means “comfort level of each subject 2”. That is, a small noise is given when the total comfort level of each subject 2 is large, and a large noise is given when the total comfort level of each subject 2 is small. (Refer to the second and third formulas in the above equation 3).

  In this way, the personal computer 50 adds the second term (noise) to the first term (desired value) on the right side of the first equation of Equation 3 above, and the actual control value (target value) of the operating state of each air conditioner. ) Is calculated (determined). That is, the personal computer 50 searches for the next control value from the range of noise centered on the desired value.

  As described above, in step S130 (control value determination), noise based on the comfort level of each subject 2 is set, and the next control value is searched for within the range of the noise. In this way, by using the biological fluctuation theory in addition to the learning in step S120, it is possible to search for a flexible control value in consideration of the uncertainties of the learning without being limited only by the learning results.

In addition, when an appropriate control value is searched for by repeatedly performing the above-described air conditioning control, the control value of the air conditioner 10 that consumes a large amount of power does not become lower than necessary. Therefore, an energy saving effect can also be expected. .
Further, by repeatedly performing the air conditioning control described above, it is possible to follow changes in the situation (for example, changes in the preference of the subject 2, movement of the subject 2, etc.).

As described above, the air conditioning control system 4 according to the present embodiment is
Air conditioning means (air conditioner 10 and fan 20) for air conditioning the room;
A comfort level reporting dial 40 (comfort level detecting means) for detecting a comfort level that is a degree of comfort of the subject 2 with respect to air conditioning by the air conditioning means;
A personal computer 50 (control means) for adjusting the operating state of the air conditioning means so that the comfort level of the subject 2 is improved;
An air conditioning control system 4 comprising:
PC 50
Learning the relationship between the comfort level of the subject 2 and the factors related to the air conditioning that affect the comfort level (the above formula 1),
Calculate a desired value of the operating state of the air conditioning means capable of improving the comfort level of the subject person 2 from the relationship between the learned comfort level of the subject person 2 and the factors related to the air conditioning (the above formula 2),
Add the noise calculated based on the biological fluctuation theory to the desired value to calculate the target value of the operating state of the air conditioning means (the above formula 2),
A series of air-conditioning control for controlling the air-conditioning means so that the operation state of the air-conditioning means becomes the target value is repeated.
By configuring in this way, by learning the relationship between the comfort level of the subject person 2 and the factors relating to air conditioning that affect the comfort level, the driving state that can improve the comfort level of the subject person 2 is improved. At the same time as searching well, by using the biological fluctuation theory, it is possible to flexibly search the driving state that can improve the comfort level of the subject 2 from the range of noise. That is, it is possible to achieve both efficiency and high accuracy of control for improving the comfort level.

In addition, the personal computer 50
If there are multiple Target Persons 2,
Calculating the degree of influence, which is the degree of the influence of the factors related to air conditioning on the comfort level of each subject 2 (the above formula 1),
An average value of the comfort level of each subject 2 weighted by the influence level is calculated (the above formula 2),
The target value of the operating state of the air conditioning means is calculated by adding noise to the desired value at which the average value of the comfort level is maximized (the above formula 2).
By comprising in this way, the comfort level of the some subject 2 can be improved simultaneously.

The air-conditioning means includes
An air conditioner 10 (ambient air conditioning means) for air conditioning the entire room,
A fan 20 (task air-conditioning means) that performs local air-conditioning around the subject 2;
Including
PC 50
The operation of the air conditioner 10 and the electric fan 20 is individually controlled.
With this configuration, even when the air conditioner 10 and the electric fan 20 are used in combination, it is possible to control the driving state so that the comfort level of the subject 2 is improved.

The operating state of the fan 20 is
The electric fan 20 includes the amount of air discharged during air conditioning.
By configuring in this way, the comfort level of the subject 2 can be improved by controlling the air volume of the electric fan 20.

In addition, the personal computer 50
Nonlinear multiple regression analysis is used when learning the relationship between the comfort level of the subject 2 and the factors relating to the air conditioning that affect the comfort level.
By comprising in this way, the relationship between the comfort level of the subject person 2 and the factor regarding the air conditioning which affects the said comfort level can be learned appropriately.

In addition, the personal computer 50
The variance of the noise is set to be smaller as the comfort level of the subject 2 is higher.
By comprising in this way, the comfort level of the subject 2 can be improved more efficiently and flexibly.

In addition, the air conditioning control method according to this embodiment
An air conditioning control method for adjusting the operating state of the air conditioning means so that the comfort level, which is the degree of comfort of the subject 2 with respect to air conditioning by the air conditioning means (air conditioner 10 and fan 20), is improved.
Learn the relationship between the comfort level of the subject 2 and the factors related to the air conditioning that affect the comfort level,
Calculating a desired value of the operating state of the air conditioning means capable of improving the comfort level of the subject 2 from the relationship between the learned comfort level of the subject 2 and the factors relating to the air conditioning;
Add the noise calculated based on the biological fluctuation theory to the desired value to calculate the target value of the operating state of the air conditioning means,
A series of air-conditioning control for controlling the air-conditioning means so that the operation state of the air-conditioning means becomes the target value is repeated.
By configuring in this way, by learning the relationship between the comfort level of the subject person 2 and the factors relating to air conditioning that affect the comfort level, the driving state that can improve the comfort level of the subject person 2 is improved. At the same time as searching well, by using the biological fluctuation theory, it is possible to flexibly search the driving state that can improve the comfort level of the subject 2 from the range of noise.

  In the present embodiment, the configuration of the air conditioning control system 4 has been described using FIG. 1, but the configuration of the present invention is not limited to the configuration of the air conditioning control system 4. That is, the present invention does not limit the number of air conditioners 10 and fans 20, installation locations, and the like.

  In the present embodiment, the air volume of the electric fan 20 is adjusted to any one of “weak”, “medium”, “strong”, and “off” (four steps), but the present invention is not limited to this. is not. That is, the adjustable stage of the fan 20 can be set arbitrarily.

  Moreover, in this embodiment, although the air conditioner 10 was illustrated as an ambient air conditioning means, this invention is not restricted to this, What is necessary is just to be able to air-condition the whole room. Moreover, although the air conditioner 10 which concerns on this embodiment shall adjust the temperature of the whole room by switching the presence or absence of the blowing of cold air, this invention is not limited to this, For example, the air conditioner 10 is cold air or warm air The temperature of the entire room may be adjusted by blowing out the air.

  Moreover, in this embodiment, although the electric fan 20 was illustrated as a task air conditioning means, this invention is not restricted to this, What is necessary is just what can perform the local air conditioning around the subject 2. .

  In the present embodiment, the comfort level reporting dial 40 is used to detect the comfort level of the subject 2, but the present invention is not limited to this. For example, the configuration may be such that the facial expression or behavior of the target person 2 is detected and the comfort level is detected based on the facial expression or behavior.

  Moreover, in this embodiment, although it was set as the structure by which the air conditioning control is performed by the personal computer 50, this invention is not limited to this, The control means which performs air conditioning control can be comprised arbitrarily. It is also possible to connect a display device capable of displaying various types of information to the control means and display information related to air conditioning control.

  In this embodiment, learning is performed using nonlinear multiple regression analysis in step S120, but the present invention is not limited to this. For example, it is possible to perform learning using another technique such as a neural network.

4 Air conditioning control system 10 Air conditioner (air conditioning means, ambient air conditioning means)
20 Fan (air conditioning means, task air conditioning means)
30 Temperature sensor 40 Comfort level declaration dial (Comfort level detection means)
50 PC (control means)

Claims (6)

Air conditioning means for air conditioning in the room;
Comfort level detection means for detecting a comfort level that is a degree of comfort of the subject with respect to air conditioning by the air conditioning means;
Control means for adjusting the operating state of the air conditioning means so as to improve the comfort level of the subject;
An air conditioning control system comprising:
The control means includes
Learn the relationship between the comfort level of the subject and the factors related to the air conditioning that affect the comfort level,
Calculating a desired value of the operating state of the air conditioning means capable of improving the comfort level of the subject person from the relationship between the learned comfort level of the subject person and the factors relating to the air conditioning;
Add the noise calculated based on the biological fluctuation theory to the desired value to calculate the target value of the operating state of the air conditioning means,
It is configured to repeat a series of air-conditioning control for controlling the operation state of the air-conditioning means to be the target value ,
If there are multiple subjects,
Calculating the degree of influence, which is the magnitude of the influence of the air conditioning factor on the comfort level of each subject person,
Calculate the average value of the comfort level of each subject weighted by the degree of influence,
The target value of the operating state of the air conditioning means is calculated by adding noise to the desired value at which the average value of the comfort level is maximized ,
Air conditioning control system. The air conditioning means includes:
Ambient air conditioning means for air conditioning the entire room,
Task air conditioning means for performing local air conditioning around the subject,
Including
The control means includes
The operation of the ambient air-conditioning means and the task air-conditioning means are individually controlled,
The air conditioning control system according to claim 1. The operation state of the task air conditioning means is:
The task air-conditioning means includes the air volume of air released during air-conditioning,
The air conditioning control system according to claim 2 . The control means includes
Nonlinear multiple regression analysis is used when learning the relationship between the comfort level of the subject and the factors relating to the air conditioning that affect the comfort level,
The air conditioning control system according to any one of claims 1 to 3 . The control means includes
The degree of comfort of the subject is higher and the variance of the noise is set to be smaller.
The air conditioning control system according to any one of claims 1 to 4 . An air-conditioning control method for adjusting the operating state of the air-conditioning means so as to improve the comfort level, which is the degree of comfort of the subject with respect to air-conditioning by the air-conditioning means,
Learn the relationship between the comfort level of the subject and the factors related to the air conditioning that affect the comfort level,
Calculating a desired value of the operating state of the air conditioning means capable of improving the comfort level of the subject person from the relationship between the learned comfort level of the subject person and the factors relating to the air conditioning;
Add the noise calculated based on the biological fluctuation theory to the desired value to calculate the target value of the operating state of the air conditioning means,
It is configured to repeat a series of air-conditioning control for controlling the operation state of the air-conditioning means to be the target value,
If there are multiple subjects,
Calculating the degree of influence, which is the magnitude of the influence of the air conditioning factor on the comfort level of each subject person,
Calculate the average value of the comfort level of each subject weighted by the degree of influence,
The target value of the operating state of the air conditioning means is calculated by adding noise to the desired value at which the average value of the comfort level is maximized,
Air conditioning control method.

Images