JP6471954B2 - Air conditioning optimum control system and air conditioning optimum control method - Google Patents

Description

  The present invention relates to an air conditioning optimum control system and an air conditioning optimum control method.

  As an air conditioning system in which a plurality of air conditioners are combined, for example, there is a latent heat sensible heat separation air conditioner that separates latent heat treatment (dehumidification) and sensible heat treatment (cooling). In the latent heat sensible heat separation air conditioning, generally, the latent heat treatment is performed by a desiccant air conditioner, and the sensible heat treatment is performed by an air conditioner. The desiccant air conditioner performs control to keep indoor humidity at a set value. On the other hand, the air conditioner performs control to keep the indoor temperature constant.

  However, each of these air conditioners independently performs PID control that keeps the temperature and humidity constant in the room. For this reason, the desiccant air conditioner and the air conditioner are individually moved so as to keep the room temperature and humidity constant. For example, in a winter season, the desiccant air conditioner may heat and the cooling / heating device may be operated by cooling. Therefore, the operation status (operation mode) of the desiccant air conditioner and the air conditioner is confirmed, and when the conflicting operation is performed, it is avoided that the conflicting operation is performed by stopping the operation of one of the devices. A method has been proposed (Patent Document 1).

JP 2013-19642 A

However, in the control method described in Patent Document 1, when a plurality of air conditioners perform conflicting operations as described above, the control of the predetermined air conditioners is stopped. For this reason, by stopping the operation of any one of the air conditioners, it takes longer time to set temperature / humidity than usual. In addition, since it becomes impossible to make use of the characteristics of each air conditioner, waste of input energy occurs when the room is set to a target set temperature / humidity. As a result, there has been a problem that costs due to air conditioning such as CO ₂ emissions and air conditioning operating costs increase.

  The present invention has been made in view of such circumstances, and in an air conditioning system in which a plurality of air conditioners are combined, the energy consumed by the plurality of air conditioners when each of the indoor temperature and humidity is brought close to a set value. It is to provide an air conditioning optimum control system and an air conditioning optimum control method capable of optimizing the air conditioning.

Further, according to one embodiment of the present invention, in an air conditioning system in which a plurality of air conditioners including a radiation panel are combined, a measurement unit that measures a physical quantity including a room temperature that is a control target, and the physical quantity measured by the measurement unit , on the basis of the temperature and quantity of heat removed of the radiation panel of the indoor components, the estimated physical quantity to that chamber state estimating unit creates a state estimation model for calculating the representative of the future physical quantity, the estimated from the state estimation model and physical quantity, calculates the quantity of heat removed which the smallest error between the reference trajectory for follow the estimated physical quantity to the set value, based on the temperature difference between the flowing water in with the calculated the quantity of heat removed entrance of the radiation panel And an optimization calculation unit that calculates an operation amount of a control valve that controls the amount of cold water used in the radiation panel .

  One embodiment of the present invention is the above-described air conditioning optimal control system, wherein the indoor state estimation unit calculates a temperature of a constituent member in the room from the measured physical quantity and the measured disturbance.

Another embodiment of the present invention is the air conditioning method in which a plurality of air conditioners are combined, including a radiation panel, a step of measuring a physical quantity comprising the temperature in the room to be controlled, the physical quantity measured by said measuring step , based on the quantity of heat removed temperature and the radiation panel of the indoor components, and the estimated physical quantity from said state estimation model, the error of the reference trajectory for to follow the set value the estimated physical quantity is minimized the calculates the quantity of heat removed, the said quantity of heat removed with the calculated on the basis of the temperature difference between the flowing water at the inlet and outlet ports of the radiating panel, and calculates the operation amount of the control valve for controlling the amount of cold water used in the radiation panel And an optimization calculation step.

As described above, according to the present invention, in an air conditioning system in which a plurality of air conditioners are combined, a future physical quantity is predicted from the current time using a state estimation model from the physical quantity in the room. Further, the operation amount of each of the plurality of air conditioners that minimizes the input energy of the plurality of air conditioners necessary for the predicted physical quantity to become the set value is calculated. Thereby, the cost by the air conditioning such as the CO ₂ emission amount and the air conditioning operation cost can be reduced.

It is a structural example of the air-conditioning optimal control system in the 1st Embodiment of this invention. It is a block diagram of the control apparatus 4 in the 1st Embodiment of this invention. It is a block diagram of the control apparatus 4 shown in FIG. It is a structural example of the air-conditioning optimal control system in the 2nd Embodiment of this invention. It is a block diagram of 4 A of control apparatuses in the 2nd Embodiment of this invention. It is a figure explaining the reference track in the 2nd Embodiment of the present invention. It is an example of the panel valve opening degree and flow volume characteristic in the 2nd Embodiment of this invention. It is a block diagram of 4 A of control apparatuses in the 2nd Embodiment of this invention. It is the figure which showed the result of having compared the conventional PID control and the model prediction control of this invention. It is a figure explaining the conditions of the internal heat_generation | fever of the experimental result shown in FIG.

(First embodiment)
A first embodiment according to the present invention will be described with reference to the drawings. FIG. 1 is a schematic diagram showing an air conditioning optimum control system in the present embodiment. The air conditioning optimum control system according to the present embodiment includes a desiccant air conditioner 2, a radiation panel 3 (cooling / heating device), and a control device 4.

  As shown in FIG. 1, the desiccant air conditioner 2 includes an air supply passage 10, an exhaust passage 20, a total heat exchange rotor 11, a first cooling coil 12, a second cooling coil 13, a dehumidifying rotor 14, and a reheating coil 15. I have.

  The air supply channel 10 supplies outside air OA (Outdoor Air) flowing in from the outside to the room (control target space) 1, and the first blower fan 21 takes in the outside air OA from the opening 2 a, and the outside air OA. Is supplied to the room (control target space) 1 as supply air SA (Supply Air).

  The exhaust passage 20 is for exhausting the return air RA flowing in from the room to the outside. The 2nd ventilation fan 22 takes in the return air RA (Return Air) which is the air of the room | chamber interior (control object space) 1 from the opening part 2b, and exhausts this return air RA outside from the opening part 2c. The exhaust passage 20 is arranged in parallel with the air supply passage 10.

The total heat exchange rotor 11 is a rotary type, and is formed of, for example, a honeycomb rotor having a shape in which an aluminum sheet is processed into a corrugated board and wound into a cylindrical shape.
The total heat exchange rotor 11 exchanges heat and humidity between the outside air OA from the air supply passage 10 and the return air RA from the exhaust passage 20. For example, in the summer, the return air RA is air from the room (control target space) 1 and therefore has a lower temperature and lower humidity than the outside air OA. Therefore, the outside air OA undergoes total heat exchange with the return air RA by the total heat exchange rotor 11, so that the temperature decreases and the humidity decreases. The return air RA that has undergone total heat exchange by the total heat exchange rotor 11 is discharged to the outside by the second blower fan 22 as exhaust air EA (Exhaust Air).

  The first cooling coil 12 is supplied with cold water via the control valve 5. The second cooling coil 13 is supplied with cold water via the control valve 6. Each of the control valves 5 and 6 is connected to the control device 4. The control device 4 controls the opening degree of the control valves 5 and 6 to adjust the amount of cold water supplied to each of the first cooling coil 12 and the second cooling coil 13. As a result, the outside air OA that has been totally heat-exchanged by the total heat-exchange rotor 11 passes through the first cooling coil 12 and gradually decreases its temperature.

The dehumidifying rotor 14 dehumidifies the air that has passed through the rotor 14. The dehumidifying rotor 14 is of a rotary type, for example, a disk-shaped rotor member (not shown) formed of a bundle of honeycomb-shaped passages, and a silica gel-based adsorbent or a polymer that has soaked into the surface of the honeycomb-shaped passages of the rotor member. Sorbent.
The dehumidification rotor 14 adsorbs moisture contained in the air that has passed through the first cooling coil 12 with the adsorbent described above. Thereby, the humidity of the air which passed the dehumidification rotor falls. Further, since the temperature of the air that has passed through the dehumidifying rotor 14 is somewhat higher than before the passage, it is cooled to about room temperature by the second cooling coil 13. The cooled air is sent to the room (control target space) 1 by the first blower fan 21. However, the adsorbent provided in the dehumidifying rotor 14 has a limit in the amount of moisture that can be adsorbed. For this reason, when the rotor member of the dehumidifying rotor 14 rotates, the adsorbent that has adsorbed moisture is positioned in the exhaust flow path 20. Therefore, the outside air OA from the opening 2d heated by the reheating coil 15 passes through the adsorbent adsorbing moisture. Thus, moisture adsorbed on the adsorbent is desorbed by the outside air OA from the warmed opening 2d. The control valve 7 supplies hot water to the reheating coil 15. Since the outside air OA from the opening 2d has absorbed the moisture adsorbed on the adsorbent, the humidity becomes high. Therefore, the outside air OA from the opening 2d is discharged to the outside by the third blower fan 23.

  The radiant panel 3 includes a heat medium flow pipe 3a through which cold water as a heat medium flows and a panel body 3b. The radiation panel 3 is disposed on the ceiling of the room (control target space) 1, for example. The radiating panel 3 raises or lowers the temperature of the panel body 3b by flowing cold water through the control valve 8 through the heat medium flow pipe 3a. Thus, the radiating panel 3 performs air conditioning with the radiant heat from the panel body 3b. The control valve 8 is connected to the control device 4. The control device 4 controls the opening degree of the control valve 8 and adjusts the amount of cold water supplied to the heat medium flow pipe 3a.

FIG. 2 is a block diagram of the control device 4 in the embodiment of the present invention. The control device 4 includes an accumulation unit 41, a planning unit 42, a measurement unit 43, a calculation unit 44, and a control unit 45.
The accumulation unit 41 accumulates data on physical quantities of the past room (control target space) 1 such as temperature / humidity values, ceiling temperature, indoor heat generation, etc. in time series. Further, the structure data of the room 1 such as the size of the room 1 and the material of the wall of the room 1 is stored in advance.
The planning unit 42 stores in advance disturbance patterns such as data based on weather forecasts such as the outside air temperature, the amount of solar radiation, and the outside air humidity, the lighting state of the lighting in the room 1 (Lr), and the number of people in the room 1 (Pr). The disturbance pattern such as the lighting state (Lr) of the lighting in the room 1 and the number of persons (Pr) in the room 1 causes the person 1 to go out of the room 1 during the lunch break, for example, so that the human body heat generation in the room 1 is lowered and the temperature is lowered. Also, during lunch breaks, etc., the lighting is often turned off, so that the heat generated by the lighting is lowered and the temperature is lowered. Thus, the disturbance pattern is data indicating how much the lighting state (Lr) of the lighting in the room 1 and the number of persons (Pr) in the room 1 are reduced or increased.
The measurement unit 43 acquires physical quantity data of the room (control target space) 1 at the current time measured from a sensor (not shown) installed in the room 1. Further, the measurement unit 43 stores in advance a set value s (t) of a physical quantity in the room (control target space) 1. As an explanation in the present embodiment, a case is considered where the physical quantity x (t) of the room (control target space) 1 at time t is a temperature Tr (t) and a humidity Tr (t).

  The calculation unit 44 includes a state estimation unit 44_1, an optimization calculation unit 44_2, and a storage unit 44_3. In the storage unit 44_3, the operation amount is stored for each unit time by the calculation unit 44. The operation amount may be written and stored in the storage unit 44_3 in time series for each unit time. Further, the operation amount may be updated from the past operation amount every unit time and written and stored in the storage unit 44_3.

In the storage unit 44_3, the function α, the function β, and the function γ are written and stored in advance. The function α, the function β, and the function γ are obtained in advance by experiments. The function α is a function representing the influence of the environmental condition at the current time t on the future temperature and humidity of the room 1. For example, α _T is a function representing the influence of the environmental condition at the current time t on the future temperature of the room 1. Α _X is a function representing the influence of the environmental condition at the current time t on the future humidity of the room 1.
The function β is a function that represents the influence of the opening degree of each control valve on the future temperature and humidity of the room 1. β _pT is a function representing the influence of the opening degree of the control valve 8 on the future temperature of the room 1. β _d1T is a function representing the influence of the opening degree of the control valves 5 and 6 on the future temperature of the room 1. β _d2T is a function representing the influence of the opening degree of the control valve 7 on the future temperature of the room 1. β _d1X is a function representing the influence of the opening degree of the control valves 5 and 6 on the future humidity of the room 1. β _d2X is a function representing the influence of the opening degree of the control valve 7 on the future humidity of the room 1.
The function γ is a function representing the influence of disturbance on temperature and humidity. For example, γ _LT is a function that represents the influence of the lighting state of the illumination on the future temperature of the room 1. γ _PT is a function representing the influence of the number of people in the room 1 on the future temperature of the room 1. γ _PX is a function representing the influence of the number of people in the room 1 on the future humidity of the room 1.

  The state estimation unit 44_1 acquires the physical quantity data of the past room (control target space) 1 from the storage unit 41. In addition, the state estimation unit 44_1 acquires disturbance pattern data from the planning unit 42, for example, the lighting state Lr (t) of the lighting and the number of people Pr (t) in the room 1. In addition, the state estimation unit 44_1 acquires the set value s (t), the physical quantity x (^) (t), the temperature Tr (t), and the humidity Xr (t) from the measurement unit 43. In addition, the state estimation unit 44_1 acquires the operation amount u (t), the function α, the function β, and the function γ from the storage unit 44_3. The state estimation unit 44_1 uses the equation (1) of the state equation set in advance from the acquired data at the current time t, and uses the physical quantity at a time t + 1 in the future time from the current time, that is, a time t + 1 after the time t. (Hereinafter, referred to as “estimated physical quantity.”) An expression of x (^) (t + 1) is obtained. In the following equation (1), the physical quantity x (^) (t + 1) at time t + 1 is expressed with “^ (hat)” written above x. Further, the physical quantity x (^) (t) at time t is expressed with “^ (hat)” written above x.

  Note that x (^) (t), u (t), and w (t) are given by the determinant of Expression (2).

  In the following formula (3), A is a determinant including the function α. B is a determinant including the function β. D is a determinant including the function γ. C is a determinant for obtaining z (^) (t + 1) in equation (4).

  In addition, the state estimation unit 44_1 calculates z (^) (t + 1) from the equation (4) in order to predict the optimum operation amount of the cycle by the evaluation function (described later). In the following formula (1), z (^) (t + 1) is expressed with “^ (hat)” written at the top of z.

  The optimization calculation unit 44_2 calculates an optimum operation amount for each period so that the evaluation function shown in Expression (5) is minimized. N represents the number of time steps for which prediction is performed. Therefore, the optimization calculation unit 44_2 performs prediction for N control cycles of k + 1, k + 2,..., K + N as the current time k. That is, the optimization calculation unit 44_2 calculates the optimum operation amount for the time from the current time k to a future point k + N in order to reduce the difference between the estimated physical quantity and the set value. In addition, the optimization calculation unit 44_2 sends only the optimal operation amount to the control unit 45 in each cycle.

The first term of this evaluation function is a term that brings the estimated physical quantity close to the set value, and Q is a parameter matrix that follows the set values of temperature and humidity. The second term indicates the opening degree of the control valve for the time from the current time k to a future point k + N. The second term is also a cost evaluation term for calculating the cost from the opening degree of the control valve, and R ₁ is a parameter matrix for cost evaluation of the desiccant air conditioner 2 and the radiating panel 3. Details regarding the second term will be described later. The third term is a term that suppresses rapid fluctuations in the operation amount. Δu indicates the amount of change between the operation amount of the cycle and the operation amount at the current time, and R ₂ is a parameter matrix of the control stability of the control valves of the desiccant air conditioner 2 and the radiation panel 3.
For example, in the cost evaluation term of the second term, the following processing is performed. Assuming that the flow rate F (t) of each cold / hot water used in the desiccant air conditioner 2 and the radiant panel 3 is proportional to the opening degree V (t) of each control valve, the relationship shown in Expression (6) is established. Note that m is a proportionality constant obtained in advance through experiments or the like.

  In addition, the amount of heat consumed Q (t) consumed by each of the desiccant air conditioner 2 and the radiant panel 3 is expressed by the flow rate F (t) of cold / hot water and the inlet of the pipe through which the cold / hot water flows as shown in the equation (7). And the product of the temperature difference dT of the cold / hot water at the outlet.

As a result, the heat quantity Q (t), which is the energy consumption, can be obtained from the equation (7) as the cost index C (t) of each of the desiccant air conditioner 2 and the radiating panel 3. Further, as shown in the equation (8), the cost index C (t) can be obtained as related to the operating cost, CO _{2 and the} like in consideration of the basic unit c. The basic unit c indicates, for example, the cost required to produce energy per unit and the CO ₂ emission amount per unit energy.

Under the above-described procedure, the relationship between the element of the manipulated variable u (t) and the cost index is derived as in Expression (9), and the value of the matrix R ₁ is determined.

As in Equation (9), when representing a relationship between the elements and the cost index of the manipulated variable u (t), the coefficient r ₁₁ that associates the opening of the control valve is a cost and control variables is shown by the formula (10) can get.

  Thus, the cost can be set from the opening degree of the control valve.

  The control unit 45 sends a control command to the desiccant air conditioner 2 to instruct to perform control based on the optimum operation amount of each cycle of the control valves 5 to 8 that minimizes the evaluation function obtained by the optimization calculation unit 44_2. Transmit to the radiation panel 3.

Next, the control operation of the control device 4 shown in FIG. 1 will be described with reference to FIG. FIG. 3 is a block diagram of the control device 4 shown in FIG.
At the current time t, the physical quantity in the room (control target space) 1 is x (^) (t), the set value is s (t), and the manipulated variable is u (t). The physical quantity x (^) (t) is the temperature / humidity or ceiling temperature of the room (control target space) 1 at the current time t, the amount of heat generated in the room (control target space) 1, and the room (control target space). ) An acquired value of a sensor (not shown) installed at 1. The set value s (t) is a set value of the physical quantity at the current time t. The manipulated variable u (t) is the opening of the control valve at the current time t.

  The state estimation unit 44_1 acquires the physical quantity x (^) (t), the set value s (t), and the operation quantity u (t) in the room (control target space) 1. In addition, the state estimation unit 44_1 substitutes the physical quantity x (^) (t), the set value s (t), and the manipulated variable u (t) for the function of the state estimation model that has been set in advance. create. Thus, the state estimation unit 44_1 estimates the estimated physical quantity and outputs it to the optimization calculation unit. The optimization calculation unit 44_2 calculates an optimum operation amount so that the estimated physical quantity becomes a preset setting value s (t) with the minimum energy for each period, and the optimum operation amount of the calculated period is calculated by the control unit 45. Output to. The control part 45 transmits the control command which instruct | indicates performing control based on the optimal operation amount of a period to each of the control valves 5-8. Each of the control valves 5 to 8 performs control based on the control command, so that the physical quantity in the room 1 at time t + 1 including the temperature and humidity in the room 1 approaches the set value. The physical quantity in the room (control target space) 1 at time t + 1 becomes a value according to, for example, the following expression (11) due to the influence of the external information w (t) and the optimum operation amount of the cycle. Note that f (x (t), u (t)) is a function indicating the influence of the manipulated variable u (t) on the physical quantity x (t). The external information w (t) is data or a disturbance pattern based on a weather forecast such as the outside air temperature, the amount of solar radiation, and the outside air humidity at the current time t.

  The control device 4 repeats the above-described control operation every unit time.

As described above, according to the present embodiment, in the air conditioning system in which a plurality of air conditioners are combined, a future physical quantity is predicted from the current time using the state estimation model from the physical quantity in the room 1. In addition, the operation amount of each of the control valves 5 to 8 that minimizes the input energy of the desiccant air conditioner 2 and the radiating panel 3 necessary for the predicted physical amount to become the set value is calculated, and based on the calculated operation amount Control. Thereby, the operating cost of air-conditioning, such as the amount of primary energy, CO ₂ emission, and air-conditioning operation cost, can be reduced.
In addition, by including a term that suppresses the change in the operation amount in the evaluation function, it is possible to design a control policy that simultaneously considers factors such as the restriction on the change amount of the control valve.
In addition, since the state estimation model is created using the predicted disturbance pattern, stable control can be performed against the disturbance. Thereby, the rapid change of the temperature and humidity of the room 1 due to disturbance can be suppressed, and the comfort of the room 1 is improved.

In addition, this invention is not restricted to the above-mentioned embodiment, In the range which does not deviate from the meaning of this invention, what added the various change to the above-mentioned embodiment is included.
In the present embodiment, the indoor state estimation unit calculates the state estimation model from the physical quantity at the current time, the manipulated variable at the current time, and the disturbance pattern. However, the present invention is not limited thereto. By reflecting the past physical quantity data and structure data of the room 1 stored in advance in 41 on the state estimation model, it is possible to calculate the influence of the disturbance and the operation amount on the room 1 with sufficient accuracy.

  In the present embodiment, the indoor state estimation unit estimates the future temperature and humidity of the room 1 in the state estimation model. However, the present invention is not limited to this, and for example, the outside air temperature, the amount of solar radiation, and the outside air humidity In addition to external information on the interior of the room 1 based on the weather forecast data, information on the interior of the room 1 such as the ceiling temperature and the amount of heat generated in the room is reflected in the state estimation model, thereby estimating the required cooling amount and the required dehumidification / humidification amount in the room. be able to.

Further, in an air conditioning system that controls the temperature of a plurality of rooms to a set temperature set for each room, the present invention can be applied by using an appropriate state estimation model.

(Second Embodiment)
Hereinafter, a second embodiment of the present invention will be described with reference to the drawings. FIG. 4 is a schematic diagram showing an air conditioning optimum control system according to the second embodiment of the present invention. In addition, the same code | symbol is attached | subjected to the same structure as 1st Embodiment, and the description is abbreviate | omitted. The air conditioning optimum control system of the present embodiment is a specific implementation of the control method of the radiating panel 3 by the model predictive control of the first embodiment.
The air conditioning optimum control system of the present embodiment includes a desiccant air conditioner 2, a radiation panel 3 (cooling / heating device), and a control device 4A. The configuration except that the control device 4 is replaced with the control device 4A is the same as that in the first embodiment.

Each of the control valves 5 and 6 is connected to the control device 4A. The control device 4 </ b> A controls the opening degree of the control valves 5 and 6 to adjust the amount of cold water supplied to each of the first cooling coil 12 and the second cooling coil 13. As a result, the outside air OA that has been totally heat-exchanged by the total heat-exchange rotor 11 passes through the first cooling coil 12 and gradually decreases its temperature.
The control valve 8 is connected to the control device 4A. The control device 4A controls the opening degree of the control valve 8 and adjusts the amount of cold water supplied to the heat medium flow pipe 3a.

FIG. 5 is a block diagram of the control device 4A in the present embodiment. The control device 4 includes an accumulation unit 41, a planning unit 42A, a measurement unit 43A, a calculation unit 44A, and a control unit 45.
The planning unit 42A determines external conditions that can be set or estimated such as the set value of the temperature of the room 1 that is the target room, the planned value of the heat generation amount of the room 1, the temperature set value of the adjacent room, and the like as a pattern file. Accumulated. In addition, the planning unit 42A stores in advance, as a pattern file, disturbance patterns such as data based on weather forecasts such as the outside air temperature, the amount of solar radiation, and the outside air humidity, the lighting state of the lighting in the room 1, and the number of people in the room 1. For example, since a person leaves the room 1 during a lunch break or the like, the human body heat generation in the room 1 decreases, and the temperature inside the room 1 decreases. Further, since the lighting is often turned off during lunch breaks, the heat generated by the lighting is lowered, and the temperature inside the room 1 is lowered. In this way, the pattern file is predictable disturbance data such as how much the lighting state of the room 1 or the number of people in the room 1 decreases or increases.

  The measurement unit 43A acquires data on the physical quantity x (t) in the room (control target space) 1 at the current time measured from a sensor (not shown) installed in the room 1. Further, the measurement unit 43A acquires disturbance data such as the temperature of the outside air at the current time measured by a sensor (not shown). Note that. In the present embodiment, a case is considered where the physical quantity x (t) in the room 1 at time t is the internal temperature of the room 1 (hereinafter referred to as “room temperature”) y (t).

  The calculation unit 44A includes a state estimation unit 44A_1, an optimization calculation unit 44A_2, and a storage unit 44A_3. In the storage unit 44A_3, the operation amount is stored for each unit time by the state estimation unit 44A_1. The operation amount may be written and stored in the storage unit 44A_3 in time series for each unit time. Further, the operation amount may be updated from the past operation amount every unit time and written and stored in the storage unit 44A_3.

  The state estimation unit 44A_1 sets the reference trajectory so that the room temperature of the room 1 approaches the set value of the room temperature of the room 1. The reference trajectory is expressed by the following equation.

  In equation (12), k is the current step, s (k) is the room temperature setting temperature, y (k) room temperature, Ts is the sampling period, and Tref is the time constant. Here, k + i | k represents the predicted value of the future step k + i at the current step k. Further, the reference trajectory represents an ideal time change until the room temperature reaches the set temperature, as shown in FIG. In the present embodiment, for example, the set temperature is reached exponentially over 30 minutes from the current room temperature. However, this embodiment is not limited to the above-described 30 minutes until the set temperature is reached.

Next, the state estimation unit 44A_1 sets a room model that is a state estimation model of the room 1, and estimates a future temperature y (k + i | k) of the room 1. For this purpose, the state estimation unit 44A_1 acquires a pattern file from the planning unit 42A. The state estimation unit 44A_1 acquires the room temperature at the current time from the measurement unit 43A. Further, the state estimation unit 44A_1 calculates the temperature of the members constituting the room 1, such as the temperature inside the wall or floor, which cannot be measured or difficult to measure, by simulation. In this simulation, the temperature of the members constituting the walls, floors, etc. was calculated by calculation under the same conditions as in the room 1, and given as the initial value of the room model.
Specifically, for example, the target room is divided into a space in the room 1 and a ceiling space above the radiation panel 3, and a one-dimensional heat transfer model is created between the room 1 and the adjacent room or between the room 1 and the outside air. It is expressed by the equation of state expressed by the equation. Then, as the state variables of the room model, 22 temperature points are set for the structural members such as walls, windows, ceilings, floors, etc., and the amount of heat exchanged with the adjacent areas based on the thermal characteristics of each structural member. And the state of the room is estimated using this model.

Here, x (k + i | k) represents a state variable of the room model, u (k + i | k) represents the amount of heat removed from the radiant panel air conditioning, and u _A (k + i | k) represents a future disturbance estimated value. A, B, and B _A are matrices representing the influences of x (k + i | k), u (k + i | k), and u _A (k + i | k) on the room state at step k + i + 1, respectively.

  The state estimation unit 44A_1 calculates room temperature y (k + i | k) by substituting Equation (13) into the following equation. The state estimation unit 44A_1 calculates the future value of the state variable of the room model up to a predetermined number of intervals, and obtains room temperature y (k + i | k). For example, the prediction cycle and the control cycle for the radiant ceiling panel are set to once every 5 minutes, and the state estimation unit 44A_1 determines the future of the state variables of the room model from the start of calculation every 5 minutes until 30 minutes (6 steps). Calculate the value and determine the room temperature.

  C is a matrix representing an operation for extracting the room temperature from the state variable.

  The optimization calculation unit 44A_2 calculates the removed heat quantity u (k) at each step so that the error between the reference trajectory shown in Expression (12) and the room temperature shown in Expression (14) is minimized. That is, as shown in the following equation, the heat removal amount u (k) is calculated such that the sum of squares of errors between the reference trajectory and room temperature at 6 points after 5, 10, 15, 20, and 25 minutes after the current is minimized. To do.

  The optimization calculation unit 44A_2 obtains the flow rate L from the calculated removal heat amount u and the actually measured value of the inlet / outlet water temperature of the radiation panel 3 using the following equation.

  Here, Q represents the heat quantity of the removed heat quantity u. K represents specific heat, Tout represents the temperature of flowing water passing through the outlet of the radiating panel 3, and Tin represents the temperature of flowing water passing through the inlet of the radiating panel 3.

  FIG. 7 is an example of the opening and flow characteristics of the panel valve, that is, the control valve 8 in the present embodiment. The flow rate characteristic is a flow rate of water passing through the control valve 8 corresponding to the valve opening degree of the control valve 8 of the radiation panel 3. The valve opening degree θ of the control valve 8 can be expressed by, for example, the following approximate expression from the valve opening degree and flow rate characteristics of the control valve 8 of the radiation panel 3 shown in FIG.

  The optimization calculation unit 44A_2 calculates the valve opening θ of the control valve 8 by substituting the calculated flow rate L into the equation (17). The optimization calculation unit 44A_2 outputs the obtained valve opening θ of the control valve 8 to the control unit 45.

  The control unit 45 transmits a control command value for instructing the valve opening degree θ of the control valve 8 obtained by the optimization calculation unit 44A_2 to the radiation panel 3.

  Next, the control operation of the control device 4A will be described with reference to FIG. FIG. 8 is a block diagram of the control device 4A. The state estimation unit 44A_1 acquires a pattern file as a predicted value such as disturbance or internal heat generation from the planning unit 42A. The state estimation unit 44A_1 acquires an initial value every 5 minutes. As described above, the initial value is the room temperature acquired from the measurement unit 43A and the temperature of the constituent members such as the walls and the floor calculated by the simulation.

The state estimation unit 44A_1 estimates the future temperature y (k + i | k) in the room 1 from the pattern file and the initial value. Further, the state estimation unit 44A_1 calculates a reference trajectory from the measured room temperature.
The optimization calculation unit 44A_2 performs the removal heat amount u = (u1, u2, u3 so that the sum of squares of the error between the reference trajectory and the room temperature at 6 points after 5, 10, 15, 20, and 25 minutes is minimized. , U4, u5, u6). The optimization calculation unit 44A_2 calculates the valve opening θ of the control valve 8 from the heat removal u and outputs the calculated valve opening θ to the control unit 45. The control unit 45 transmits a control command value for instructing the valve opening degree θ of the control valve 8 obtained by the optimization calculation unit 44A_2 to the radiation panel 3.

  The measuring unit 43A acquires the room temperature at the current time measured from a sensor (not shown) installed in the room 1. The measurement unit 43 acquires a disturbance and internal heat generation that can be measured at a current time measured from a sensor (not shown).

  The state estimation unit 44A_1 uses the room temperature, the disturbance, and the internal heat generation acquired from the measurement unit 43A as initial values, and calculates the temperature of a constituent member such as the temperature of a wall or floor that cannot be measured or is difficult to measure by simulation. The state estimation unit 44A_1 estimates the future temperature y (k + i | k) of the room 1 using the temperature of the component obtained by the simulation as an initial value.

  By repeating the above operation, feedback control can be performed in consideration of external factors for the next 30 minutes every 5 minutes.

Next, experimental results using the model predictive control of the present embodiment will be described with reference to the drawings. FIG. 9 is a diagram showing a result of comparing the conventional PID control and the model predictive control of the present embodiment. As a condition, the set temperature was set to 25 ° C. Further, as shown in FIG. 10, the heat generation amount of the internal heat generation (simulated heating element) was suddenly changed from 1.7 kW to 0.7 kW 45 minutes after the start of the test. The change in the heat generation amount is incorporated in the control condition as a predicted value, that is, a pattern file in the flow shown in FIG. The average value of the outside temperature during the test period was 24.2 ° C., and the total solar radiation amount was 302.2 W / m ² . The blinds in the target room were fully opened during the test period, but there was almost no direct solar radiation from the building direction. The air conditioning room in the adjacent room was air-conditioned at a temperature close to the set room temperature of the target room. In addition, since the desiccant air conditioner has a large fluctuation in the supply air temperature, the desiccant air conditioner was stopped during the test period and cooled only by the radiation panel.

  As shown in FIG. 9, in the conventional PID control, the control valve is throttled 45 minutes before the internal heat generation load is changed, but the follow-up of the room temperature is delayed, and the temperature fluctuation is the lowest with a set temperature of 25 ° C. as a reference. But it becomes -0.3 ° C. In addition, the temperature has not yet recovered to the set temperature due to the cold water remaining in the radiant panel. On the other hand, in the model predictive control of this embodiment, the set temperature is stably controlled within 25 ° C. to ± 0.1 ° C. even when the internal heat generation load is changed. As a result, the model predictive control of the present embodiment can improve indoor comfort as compared with the conventional PID control.

As described above, according to the present embodiment, in the air conditioning system in which a plurality of air conditioners are combined, a future room temperature is predicted using a room model that is a state estimation model of the room 1. That is, the target room is divided into a space in the room 1 and a ceiling space on the upper side of the radiant ceiling panel, a one-dimensional heat transfer model is created between adjacent rooms and between the outside air, and is represented by a state equation of Expression (13). Then, as the state variables of the room model, 22 temperature points are set on walls, window surfaces, ceilings, floors, etc., and the relationship between the amount of heat exchanged with adjacent regions based on the thermal characteristics of each component member is set. Modeling is performed, and the state of the room is estimated using this model. In addition, since the predicted room temperature becomes a set value, a reference trajectory is set, and the amount of heat to be removed is determined so that the error between the reference trajectory and the predicted room temperature is minimized. Then, the operation amount of the control valve 8 is calculated from the removed heat amount, and the room temperature is adjusted to the set value according to the reference trajectory by controlling based on the calculated operation amount. Thereby, the operating cost of air-conditioning, such as the amount of primary energy, CO ₂ emission, and air-conditioning operation cost, can be reduced. Moreover, if fluctuation information is obtained in advance for internal load fluctuations, control without affecting the temperature of the room 1 is possible. As a result, the comfort in the room can be improved.

  The embodiment of the present invention has been described in detail with reference to the drawings. However, the specific configuration is not limited to the above-described embodiment, and includes design and the like within the scope not departing from the gist of the present invention.

DESCRIPTION OF SYMBOLS 1 Indoor 2 Desiccant air conditioner 3 Radiation panel 3a Heat-medium distribution pipe 3b Panel main body 4, 4A Control device 5-8 Control valve 10 Air supply flow path 20 Exhaust flow path 11 Total heat exchange rotor 12 First cooling coil 13 Second cooling coil 14 Dehumidification rotor 15 Reheating coil 21 1st ventilation fan 22 2nd ventilation fan 23 3rd ventilation fan 2a, 2b, 2c, 2d Opening part 41 Accumulation part 42, 42A Planning part 43, 43A Measurement part 44, 44A Calculation part 44_1 44A_1 State estimation unit 44_2, 44A_2 Optimization calculation unit 44_3, 44A_3 Storage unit 45 Control unit

Claims (3)

In an air conditioning system that combines multiple air conditioners including radiant panels,
A measurement unit that measures a physical quantity including the temperature of the room being controlled;
An indoor state estimation unit that creates a state estimation model that calculates an estimated physical quantity representing a future physical quantity based on the physical quantity measured by the measurement unit, the temperature of the indoor structural member, and the amount of heat removed from the radiation panel;
The removal heat quantity that minimizes an error between the estimated physical quantity and a reference trajectory for causing the estimated physical quantity to follow a set value from the state estimation model is calculated, and the calculated removal heat quantity and the entrance / exit of the radiation panel An optimization calculation unit that calculates an operation amount of a control valve that controls an amount of cold water used in the radiation panel based on a temperature difference of flowing water;
Air conditioning optimum control system. The air conditioning optimum control system according to claim 1 , wherein the indoor state estimation unit calculates the temperature of a constituent member in the room from the measured physical quantity and the measured disturbance. In an air conditioning method in which a plurality of air conditioners including a radiation panel are combined,
A measurement step for measuring a physical quantity including a room temperature to be controlled;
An indoor state estimation step for creating a state estimation model for calculating an estimated physical quantity representing a future physical quantity based on the physical quantity measured in the measurement step, the temperature of the indoor structural member, and the amount of heat removed from the radiation panel;
The removal heat quantity that minimizes an error between the estimated physical quantity and a reference trajectory for causing the estimated physical quantity to follow a set value from the state estimation model is calculated, and the calculated removal heat quantity and the entrance / exit of the radiation panel An optimization calculation step for calculating an operation amount of a control valve for controlling an amount of cold water used in the radiation panel based on a temperature difference of flowing water;
An air conditioning optimum control method.

Images