Parking lot ventilation system, parking lot ventilation method, learning device, and inference device

Abstract

A parking lot ventilation system (1) for ventilating the interior of a parking lot (100) by discharging, from an exhaust port (102), air supplied into the parking lot (100) from an air supply port (101). The parking lot ventilation system (1) is provided with: an air supply blower (2) for supplying external air from the air supply port (101); an exhaust blower (3) for discharging internal air from the exhaust port (102); a plurality of conveying blowers (5a to 5d) that are sequentially installed on a ventilation air path (4) leading to the exhaust port (3) from the air supply port (2) to convey air; an entry / exit number detection device (6) for detecting the entry / exit number of vehicles entering / exiting the parking lot (100); and an air blowing control device (7a, 7c) that uses information on the entry / exit number detected by the entry / exit number detection device (6) to control an air blowing amount of one or more of the conveying blowers (5a to 5d).

Description

Parking lot ventilation system, parking lot ventilation method, learning device, and inference device

[0001] The present disclosure relates to a parking lot ventilation system, a parking lot ventilation method, a learning device, and an inference device for ventilating parking lots.

[0002] Patent Document 1 discloses a parking lot ventilation system that includes an intake fan installed near an air intake port, an exhaust fan installed near an exhaust port, multiple transport fans installed between the intake fan and the exhaust fan, a CO sensor installed near the exhaust fan to detect the CO gas concentration in the air around it, and a fan control device that controls the operation of at least one of the intake fan, exhaust fan, and transport fan based on the output signal of the CO sensor.

[0003] Japanese Patent Application Publication No. 2020-106222

[0004] However, in parking lots where the air intake and exhaust vents are close to each other, a CO sensor installed near the exhaust fan, as in the parking lot ventilation system described in Patent Document 1, can cause fresh air from the air intake to flow around the CO sensor. In this case, the CO sensor measures a CO gas concentration that is lower than the actual CO gas concentration in the parking lot, preventing effective airflow control based on sensor detection of CO gas concentration, and resulting in insufficient ventilation of the entire parking lot.

[0005] The present disclosure has been made to solve the above-mentioned problems, and aims to provide a parking lot ventilation system, a parking lot ventilation method, a learning device, and an inference device that can provide sufficient ventilation even in parking lots where the air intake and exhaust vents are close to each other.

[0006] The parking lot ventilation system according to the present disclosure is a system for ventilating a parking lot by supplying air into the parking lot through an air intake port and exhausting the air through an exhaust port. The parking lot ventilation system according to the present disclosure includes an intake fan that supplies outside air through the intake port, an exhaust fan that exhausts inside air through the exhaust port, multiple transport fans that are sequentially installed on a ventilation air path from the intake port to the exhaust port and transport air, an entry / exit number detection device that detects the number of vehicles entering and leaving the parking lot, and an air blowing control device that controls the airflow rate of one or more of the transport fans based on the number of entries and exits detected by the entry / exit number detection device.

[0007] The parking lot ventilation system, parking lot ventilation method, learning device, and inference device disclosed herein have the effect of enabling sufficient ventilation even in parking lots where the air intake and exhaust vents are close to each other.

[0008] 1 is a plan view showing a parking lot in which a parking lot ventilation system according to a first embodiment is installed. 2 is a diagram schematically showing the air flow in the parking lot caused by the parking lot ventilation system. 3 is an enlarged view of the area around a first corner. 4 is a diagram schematically showing a first specific example of placement restrictions. 5 is a diagram schematically showing a second specific example of placement restrictions. 6 is a diagram schematically showing the signal communication connections between the devices in the parking lot ventilation system. 7 (a) and (b) are diagrams showing the spread of the wind speed distribution of air blown out from the conveying blower. 8 is a cross-sectional view taken along line A-A in FIG. 2. 9 is a flowchart showing the operation of the parking lot ventilation system. 10 is a diagram schematically showing the ventilation state when the parking lot ventilation system is operated in weak operation mode. 11 is a diagram schematically showing the ventilation state when the parking lot ventilation system is operated in strong operation mode. 12 is a diagram schematically showing the ventilation state when the parking lot ventilation system is operated in entrance strong operation mode. 13 is a diagram schematically showing the ventilation state when the parking lot ventilation system is operated in innermost strong operation mode. 14 is a diagram schematically showing the ventilation state when the parking lot ventilation system is operated in corner stirring mode A. 1 is a diagram schematically showing the ventilation state when the parking lot ventilation system is operated in corner stirring mode B. FIG. 2 is a diagram schematically showing the ventilation state when the parking lot ventilation system is operated in center stirring mode A. FIG. 3 is a diagram schematically showing the ventilation state when the parking lot ventilation system is operated in center stirring mode B. FIG. 4 is a flowchart showing an example of operation in combined mode. (a) to (c) are diagrams showing examples of combined mode. FIG. 5 is a block diagram showing the configuration of a learning device according to embodiment 2. FIG. 6 is a diagram schematically showing the installation status of environmental sensors. FIG. 7 is a flowchart showing the flow of learning processing by the learning device. FIG. 8 is a block diagram showing the configuration of an inference device according to embodiment 3. FIG. 9 is a flowchart showing the flow of inference processing by the inference device. FIG. 10 is a diagram showing an example of the hardware configuration of a counting unit, an air blowing control device, a model generation unit, and an inference unit.

[0009] The parking lot ventilation system according to the embodiment of the present disclosure will be described in detail below with reference to the drawings. In each drawing, the same or corresponding parts are designated by the same reference numerals. Duplicate explanations of these parts will be appropriately simplified or omitted.

[0010] Embodiment 1. *** Description of Configuration *** Figure 1 is a plan view showing a parking lot 100 in which a parking lot ventilation system 1 according to embodiment 1 is installed. As shown in Figure 1, parking lot ventilation system 1 includes an intake air blower 2 that supplies outside air into parking lot 100 through intake vent 101 of parking lot 100, an exhaust air blower 3 that exhausts internal air to the outside of parking lot 100 through exhaust vent 102 of parking lot 100, and multiple transport air blowers 5a-5d that are sequentially installed on ventilation air duct 4 leading from intake vent 101 to exhaust vent 102 and transport air. In the following description, the multiple transport air blowers 5a-5d may be collectively referred to as transport air blower 5.

[0011] The parking lot ventilation system 1 also includes a vehicle entry / exit number detection device 6 that detects the number of vehicles, such as four-wheeled and two-wheeled motor vehicles, entering and leaving the parking lot 100. The vehicle entry / exit number detection device 6 includes an infrared emitting unit 6a that emits infrared rays, infrared sensors 6b and 6c that detect the infrared rays emitted by the infrared emitting unit 6a, and a counting unit 6d that counts the number of vehicles based on the output signals of the infrared sensors 6b and 6c. The infrared emitting unit 6a and the infrared sensors 6b and 6c are also referred to as the "vehicle detection unit" that detects vehicles entering and leaving the parking lot.

[0012] Furthermore, the parking lot ventilation system 1 includes air blowing control devices 7a, 7c that are provided on at least one of the multiple transport blowers 5a-5d and that control the air volume of one or more of the transport blowers 5. In this example, the air blowing control device 7a is provided on the transport blower 5a, and controls the air volume of the transport blower 5a based on the entry / exit number information detected by the entry / exit number detection device 6. Similarly, the air blowing control device 7c is provided on the transport blower 5c, and controls the air volume of the transport blower 5c based on the entry / exit number information detected by the entry / exit number detection device 6.

[0013] Parking lot 100 is a highly sealed parking lot, such as an underground parking lot or an indoor parking lot, and therefore ventilation is required if the air quality inside parking lot 100 deteriorates. Parking lot ventilation system 1 ventilates parking lot 100 using a ductless ventilation system that does not require duct space, is easy to install, and has advantages in terms of initial and running costs.

[0014] Parking lot 100 has a rectangular shape in plan view, with multiple parking spaces arranged in the center and periphery of parking lot 100. Parking lot 100 has four corners, arranged clockwise (right-handed) in plan view: first corner 103a, second corner 103b, third corner 103c, and fourth corner 103d. Conveying fan 5a is installed near first corner 103a, and conveying fan 5b is installed near second corner 103b. Conveying fan 5c is installed near third corner 103c, and conveying fan 5d is installed near fourth corner 103d. Conveying fan 5a has the shortest total distance from air inlet 101 and air outlet 102 compared to the other conveying fans 5b to 5d. Furthermore, the conveying blower 5c has the longest total distance from the air inlet 101 and the air outlet 102 compared to the other conveying blowers 5a, 5b, and 5d.

[0015] A vehicle entrance / exit is provided near first corner 103a, and vehicle entrance / exit path 104 is connected to this entrance / exit. Parking lot 100 is provided with a circular passage 105 that circles the interior of parking lot 100, and one end of circular passage 105 is connected to entrance path 104a of vehicle entrance / exit path 104. The other end of circular passage 105 is connected to exit path 104b of vehicle entrance / exit path 104. Ventilation air duct 4 is provided along circular passage 105. Therefore, air in circular passage 105, which is likely to generate CO gas due to vehicle movement, is transported to exhaust port 102 by multiple transport fans 5a to 5d, thereby enabling efficient ventilation within parking lot 100.

[0016] An entrance / exit gate 106 is provided at the vehicle entrance / exit road 104, and an infrared emitting unit 6a is installed between the entrance road 104a and the exit road 104b at the entrance / exit gate 106. The infrared emitting unit 6a emits infrared rays in two directions, the entrance road 104a and the exit road 104b. An infrared sensor 6b is installed at the end of the entrance / exit gate 106 on the entrance road 104a side. The infrared sensor 6b detects the infrared rays emitted by the infrared emitting unit 6a in the direction of the entrance road 104a. An infrared sensor 6c is installed at the end of the entrance / exit gate 106 on the exit road 104b side. The infrared sensor 6c detects the infrared rays emitted by the infrared emitting unit 6a in the direction of the exit road 104b.

[0017] The infrared sensors 6b and 6c transmit detection signals to the counting unit 6d, and the counting unit 6d counts the number of vehicles obtained from the detection signals. The infrared sensors 6b and 6c are connected to each other by signal lines.

[0018] Here, the infrared rays emitted from the infrared emitting unit 6a toward the infrared sensor 6b are blocked by a vehicle entering the parking lot through the entrance road 104a, so the infrared sensor 6b can detect the entering vehicle. Similarly, the infrared rays emitted from the infrared emitting unit 6a toward the infrared sensor 6c are blocked by a vehicle leaving the parking lot through the exit road 104b, so the infrared sensor 6c can detect the leaving vehicle.

[0019] Figure 2 is a diagram schematically showing the air flow within a parking lot 100 using the parking lot ventilation system 1. As shown in Figure 2, a conveying blower 5a is installed at a position where the outside air blown out from the air inlet 101 by the air supply blower 2 reaches. A conveying blower 5b is installed at a position where the air blown out from the conveying blower 5a reaches. A conveying blower 5c is installed at a position where the air blown out from the conveying blower 5b reaches. A conveying blower 5d is installed at a position where the air blown out from the conveying blower 5c reaches. An exhaust port 102 is located at a position where the air blown out from the conveying blower 5d reaches.

[0020] Therefore, outside air blown out from the air intake port 101 by the air intake blower 2 is supplied to the conveying blower 5a. The supplied air is blown out from the conveying blower 5a and transported toward the conveying blower 5b. The air that reaches the conveying blower 5b is blown out from the conveying blower 5b and transported toward the conveying blower 5c. The air that reaches the conveying blower 5c is blown out from the conveying blower 5c and transported toward the conveying blower 5d. The air that reaches the conveying blower 5d is blown out from the conveying blower 5d and transported toward the exhaust port 102. The air that reaches the exhaust port 102 is exhausted to the outside by the exhaust blower 3.

[0021] In this way, air supplied from outside is transported in sequence via multiple transport blowers 5a to 5d, thereby forming an air flow that circulates within parking lot 100 along circular passage 105. As a result, air supplied from outside through air intake 101 reaches every corner of parking lot 100, allowing effective ventilation within parking lot 100.

[0022] Next, the positional relationship between the air inlet 101, the exhaust outlet 102, and the conveying blower 5a will be described using Figure 3. Figure 3 is an enlarged view of the periphery of the first corner 103a. In parking lots like parking lot 100, where the air inlet 101 and the exhaust outlet 102 are close to each other, it is important to prevent ventilation short circuits. Note that ventilation short circuits refer to a phenomenon in which air taken in through the air inlet 101 is directly exhausted through the exhaust outlet 102 without ventilating the parking lot.

[0023] As shown in Figure 3, in the parking lot ventilation system 1, the conveying blower 5a is installed closer to the air inlet 101 than the exhaust port 102. As a result, much of the air supplied from the air inlet 101 flows toward and reaches the conveying blower 5a. The air that reaches the conveying blower 5a is then blown out by the conveying blower 5a so as to circulate within the parking lot 100. In this way, the air supplied from the air inlet 101 rarely flows directly toward the exhaust port 102, so the parking lot ventilation system 1 can effectively prevent short-circuit ventilation.

[0024] Here, there are limitations on the placement of the air inlet 101, the exhaust port 102, and the conveying blower 5a. First, a first specific example of the placement limitation will be described with reference to FIG. 4. FIG. 4 is a diagram schematically illustrating the first specific example of the placement limitation. As shown in FIG. 4, if the distance between the air inlet 101 and the conveying blower 5a is L1, the distance between the air inlet 101 and the exhaust port 102 is L2, and the distance between the exhaust port 102 and the conveying blower 5a is L3, the following equations (1) and (2) hold between the distances L1, L2, and L3.

[0025] L1<L2…(1) L1<L3…(2)

[0026] If the positions of the air inlet 101 and the air outlet 102 are fixed, the position of the conveying blower 5a that satisfies formula (1) will be inside circle B. Similarly, if the positions of the air inlet 101 and the air outlet 102 are fixed, the position of the conveying blower 5a that satisfies formula (2) will be above line C in FIG. 4 (on the air inlet 101 side). As a result, the conveying blower 5a will be located within area D, which is inside circle B and above line C. In this way, if the conveying blower 5a is located within area D, the air inlet 101 will be closer to the conveying blower 5a than the air outlet 102, and most of the air supplied from the air inlet 101 will flow toward the conveying blower 5a. Therefore, the parking lot ventilation system 1 can effectively suppress short-circuit ventilation.

[0027] Next, a second specific example of the restriction on placement will be described with reference to Fig. 5. Fig. 5 is a diagram schematically illustrating the second specific example of the restriction on placement. As shown in Fig. 5, if the distance between the air supply port 101 and the conveying blower 5a is L1 and the distance between the air supply port 101 and the exhaust port 102 is L2, the following equation (3) holds between the distance L1 and the distance L2.

[0028] L1 < L2 / 2 ... (3)

[0029] When the positions of the air inlet 101 and the air outlet 102 are fixed, the position of the conveying blower 5a that satisfies formula (3) is inside area F, which is inside circle E. In this way, if the conveying blower 5a is placed inside area F, the air inlet 101 will be closer to the conveying blower 5a than the air outlet 102, so most of the air supplied from the air inlet 101 will flow toward the conveying blower 5a. For this reason, the parking lot ventilation system 1 can effectively suppress short-circuiting of ventilation.

[0030] Next, the signal communication connections between the devices in the parking lot ventilation system 1 will be explained using FIG. 6. FIG. 6 is a diagram that schematically illustrates the signal communication connections between the devices in the parking lot ventilation system 1. As shown in FIG. 6, the supply air blower 2, the exhaust air blower 3, the transport air blowers 5a-5d, the entry / exit number detection device 6, and the air blower control devices 7a and 7c are connected via a network 110. This allows signals to be sent and received between the devices. Note that, because the devices are located within a distance of approximately several tens of meters, it is desirable for them to be connected via wireless communication. Connecting via wireless communication simplifies the wiring installation work.

[0031] Next, the results of verifying the relationship between the distance between the conveying blower 5 and the ceiling surface of the parking lot 100 and the conveyance distance of the air blown out from the conveying blower 5 will be described with reference to Figures 7(a) and 7(b). Figure 7(a) is a diagram showing the spread of the wind speed distribution of the air blown out from the conveying blower 5 when the distance between the conveying blower 5 and the ceiling surface of the parking lot 100 is 0.1 m. Figure 7(b) is a diagram showing the spread of the wind speed distribution of the air blown out from the conveying blower 5 when the distance between the conveying blower 5 and the ceiling surface of the parking lot 100 is 2.0 m. In Figures 7(a) and 7(b), the horizontal direction is the direction in which air is blown out from the conveying blower 5, and the vertical direction is the height direction of the parking lot 100.

[0032] 7(a) and 7(b), the conveying distance when the separation distance shown in Fig. 7(a) is 0.1 m is 1.2 to 1.6 times longer than the conveying distance when the separation distance shown in Fig. 7(b) is 2.0 m. This shows that if the separation distance between the conveying blower 5 and the ceiling surface of the parking lot 100 is short, the Coanda effect works effectively, and the air blown out from the conveying blower 5 is conveyed farther even if the operating conditions of the conveying blower 5 (wind direction, air volume, wind speed) are the same.

[0033] Therefore, all of the conveying blowers 5a to 5d are installed at a distance of 0.5 m or less from the ceiling surface of the parking lot 100. This will be explained using FIG. 8. FIG. 8 is a cross-sectional view taken along line A-A in FIG. 2. As shown in FIG. 8, the conveying blowers 5a and 5b are installed at a distance of 0.5 m from the ceiling surface of the parking lot 100. The ceiling surface of the parking lot 100 is a wall surface parallel to the wind direction of the air blown out from the conveying blower 5a. Therefore, the air blown out from the conveying blower 5a is transported along the ceiling surface of the parking lot 100 to the conveying blower 5b due to the Coanda effect.

[0034] In this way, when the distance between the conveying blower 5 and the ceiling surface of the parking lot 100 is 0.5 m or less, the Coanda effect works effectively, so even if the conveying blower 5a is operated at a low airflow rate, the air blown out from the conveying blower 5 can be transported farther, achieving efficient ductless ventilation. Furthermore, because CO gas, which is lighter than air, tends to accumulate in the upper part of the parking lot 100, a conveying blower 5 installed near the ceiling surface of the parking lot 100 can efficiently exhaust air with a high CO gas concentration that has accumulated in the upper part of the parking lot 100.

[0035] ***Explanation of Operation*** Next, the operation of the parking lot ventilation system 1 according to the first embodiment will be explained using Figure 9. Figure 9 is a flowchart showing the operation of the parking lot ventilation system 1. When the parking lot ventilation system 1 starts operation, the four conveying fans 5a to 5d start in the initial setting operation mode (step S10).

[0036] When a vehicle travels along the entrance road 104a and passes through the entrance / exit gate 106, the optical axis of the infrared rays emitted from the infrared emitting unit 6a toward the infrared sensor 6b is blocked by the vehicle. As a result, the amount of infrared rays detected by the infrared sensor 6b is significantly reduced. The counting unit 6d receives an infrared detection signal from the infrared sensor 6b and detects a vehicle entering the parking lot 100 based on the received detection signal. In other words, when the amount of infrared rays included in the received detection signal is significantly reduced, the counting unit 6d determines that a vehicle has entered the parking lot 100. Then, the counting unit 6d counts the number of vehicles entering the parking lot 100 (step S11).

[0037] Similarly, when a vehicle travels along the circular passage 105 and passes through the entrance / exit gate 106, the optical axis of the infrared light emitted from the infrared light emitter 6a toward the infrared sensor 6c is blocked by the vehicle. As a result, the amount of infrared light detected by the infrared sensor 6c is significantly reduced. The counting unit 6d receives an infrared detection signal from the infrared sensor 6c and detects vehicles exiting the parking lot 100 based on the received detection signal. That is, when the amount of infrared light included in the received detection signal is significantly reduced, the counting unit 6d determines that a vehicle has exited the parking lot 100. Then, the counting unit 6d counts the number of vehicles exiting the parking lot 100 (step S11).

[0038] The counting unit 6d continues counting the number of vehicles entering and leaving the parking lot until a predetermined first time period has elapsed (step S12). When the first time period has elapsed, the counting unit 6d compares the cumulative number of vehicles entering and leaving the parking lot with a predetermined threshold value (step S13). If the comparison in step S13 indicates that the cumulative number of vehicles entering and leaving the parking lot is equal to or greater than the threshold value, the counting unit 6d transmits a notification signal to the two air blowing control devices 7a and 7c to notify them of a deterioration in the air quality within the parking lot 100 (step S14). That is, the counting unit 6d determines that an increase in the number of vehicles moving within the parking lot 100 is causing a large amount of CO gas to be generated within the parking lot 100, and transmits a notification signal to the two air blowing control devices 7a and 7c.

[0039] When the air blowing control device 7a receives the notification signal from the counting unit 6d, the air blowing control device 7a sends a control signal to the conveying blower 5a to increase the air blowing volume (step S15). The conveying blower 5a, which has received the control signal, controls itself to increase the air blowing volume (step S16). This control increases the amount of outside air supplied from the air inlet 101, accelerating the introduction of fresh air into the parking lot 100. Furthermore, increasing the air blowing volume of the air blowing control device 7a reliably prevents the above-mentioned short-circuiting of ventilation.

[0040] Similarly, when the air blowing control device 7c receives the notification signal from the counting unit 6d, the air blowing control device 7c transmits a control signal to the conveying blower 5c to increase the air blowing volume (step S15). Upon receiving the control signal, the conveying blower 5c controls itself to increase the air blowing volume (step S16). This control causes the air with a high CO gas concentration accumulating in the area surrounding the third corner 103c to be transported to the exhaust port 102, where it is quickly exhausted from the exhaust port 102. In particular, the area surrounding the third corner 103c is the farthest from the air inlet 101 and the exhaust port 102, making it prone to CO gas accumulation. Therefore, by increasing the air blowing volume of the conveying blower 5c installed near the third corner 103c, the air with a high CO gas concentration can be efficiently exhausted.

[0041] Then, after a predetermined second time has elapsed (step S17), the air blowing control devices 7a, 7c determine that the parking lot 100 has been sufficiently ventilated, and control the conveyance fans 5a, 5c to transition to their initial setting operation mode (step S18). This control reduces the airflow volume of the conveyance fans 5a, 5c. Then, the process returns to step S11. Furthermore, if the comparison in step S13 shows that the cumulative number of vehicles entering and leaving the parking lot is less than the threshold, the counting unit 6d resets the count of the cumulative number, and the process returns to step S11 (step S19).

[0042] As described above, when the cumulative number of vehicles entering and leaving the parking lot 100 exceeds a threshold value, the parking lot ventilation system 1 according to the first embodiment determines that the CO gas concentration within the parking lot 100 is rising and increases the airflow rate of one of the conveying fans 5a to 5d. As a result, the amount of outside air supplied from the air inlet 101 increases, accelerating the introduction of fresh air into the parking lot 100. Furthermore, the air with a high CO gas concentration is conveyed to the exhaust outlet 102, allowing the air with a high CO gas concentration to be quickly exhausted from the exhaust outlet 102.

[0043] Ventilation with an increased airflow rate in this manner quickly improves the air quality within the parking lot 100. After the air quality has improved, the operation mode of the conveying blower 5 is immediately returned to the initial setting, thereby reducing the airflow rate of the conveying blower 5. This prevents excessive ventilation from continuing, reducing running costs.

[0044] In the first embodiment, two airflow control devices 7a and 7c were used to adjust the airflow rates of the conveying blowers 5a and 5c. However, the conveying blowers 5 that adjust the airflow rates are not limited to the conveying blowers 5a and 5c. For example, if the ceiling surface of the second corner 103b is high, CO gas tends to accumulate around the second corner 103b. Furthermore, if a pedestrian access corridor is located near the second corner 103b, the amount of parking near the access corridor is large, which tends to cause CO gas to accumulate around the second corner 103b. In this way, if CO gas tends to accumulate around the second corner 103b, an airflow control device may be provided for the conveying blower 5b to adjust the airflow rate of the conveying blower 5b. Similarly, if CO gas tends to accumulate around the fourth corner 103d, an air blowing control device may be provided on the conveying blower 5d so that the amount of air blown by the conveying blower 5d can be adjusted.

[0045] Furthermore, the air flow rate of only one of the four conveying blowers 5a to 5d may be adjusted. In other words, if there is one corner where CO gas is likely to accumulate, the air flow rate of the conveying blower 5 installed near that corner may be adjusted.

[0046] 9, a single threshold value is used for comparison with the cumulative number of vehicles that have entered and left the facility, but multiple threshold values ​​that increase in stages may be used for comparison with the cumulative number of vehicles. In this case, in step S15, upon receiving a notification signal from the counting unit 6d (step S14), the air blowing control devices 7a and 7c transmit control signals to the conveying fans 5a and 5c so that the air blowing volume increases according to the level of the threshold value used in step S13. Then, in step S16, the conveying fans 5a and 5c that have received this control signal increase the air blowing volume so that the air blowing volume varies depending on the threshold value.

[0047] ***Explanation of Control Modes*** Next, details of the control modes for controlling the multiple conveying fans 5a to 5d will be explained using Figures 10 to 17. In Figures 10 to 17, the air flows 8a to 8d generated by the conveying fans 5a to 5d are indicated by arrows. The air volume of each air flow 8a to 8d is represented by the number of arrows. That is, a single arrow indicates a weak air flow, and two or three arrows indicates a strong air flow.

[0048] The parking lot ventilation system 1 shown in Figures 10 to 17 is equipped with four air blowing control devices 7a to 7d, one for each of the four transport fans 5a to 5d. Therefore, each transport fan 5a to 5d operates in one of three modes: high-power operation (high airflow), low-power operation (low airflow), or off. A combination of operating modes is called a control mode. Control modes include the low-power operation mode shown in Figure 10, the high-power operation mode shown in Figure 11, the high-power entrance operation mode shown in Figure 12, the high-power innermost operation mode shown in Figure 13, corner agitation modes A and B shown in Figures 14 and 15, and center agitation modes A and B shown in Figures 16 and 17.

[0049] ***Explanation of Low Operation Mode*** Figure 10 is a diagram showing the ventilation state when the parking lot ventilation system 1 is operated in low operation mode. As shown in Figure 10, the four conveyance fans 5a-5d are operated in low operation with a low air volume under the control of the air blowing control devices 7a-7d. In step S10 of the flowchart in Figure 9, the four conveyance fans 5a-5d are started in the default operation mode, which is the low operation mode. While the number of vehicles entering the parking lot 100 is low, the parking lot ventilation system 1 operates in low operation mode as a standby state.

[0050] ***Explanation of High Operation Mode*** Figure 11 is a diagram showing the ventilation state when the parking lot ventilation system 1 is operated in high operation mode. As shown in Figure 11, the four conveyance fans 5a-5d are operated in high operation mode with a high airflow volume under the control of the airflow control devices 7a-7d. In the flowchart of Figure 9, if the cumulative number of vehicles entering and leaving the parking lot is equal to or greater than the threshold (step S13), each of the airflow control devices 7a-7d receives a notification signal from the counting unit 6d (step S14) and sends a control signal to each of the conveyance fans 5a-5d to increase the airflow volume (step S15). Upon receiving this control signal, each conveyance fan 5a-5d increases its airflow volume and operates in high operation mode with a high airflow volume (step S16). In this way, in the strong operation mode in which each conveying fan 5a to 5d is operated at high power, the strong air currents 8a to 8d allow outside air to flow into every corner of the parking lot 100, thereby enabling ventilation throughout the parking lot 100.

[0051] ***Explanation of Inlet Strong Operation Mode*** Figure 12 is a diagram showing the ventilation state when the parking lot ventilation system 1 is operated in inlet strong operation mode. As shown in Figure 12, the conveying blower 5a operates in strong operation with a large air volume under the control of the air blower control device 7a. Furthermore, the conveying blowers 5b to 5d each operate in weak operation with a small air volume under the control of the air blower control devices 7b to 7d.

[0052] In the flowchart of Figure 9, if the cumulative number of vehicles entering and leaving the warehouse is equal to or greater than the threshold value (step S13), the airflow control device 7a receives a notification signal from the counting unit 6d (step S14) and sends a control signal to the conveyance blower 5a to increase the airflow volume (step S15). The conveyance blower 5a, upon receiving this control signal, increases the airflow volume and operates at high airflow volume (step S16). Furthermore, each of the conveyance blowers 5b, 5c, and 5d maintains the low airflow volume set by the airflow control devices 7b, 7c, and 7d in step S10.

[0053] In this manner, in the inlet strong operation mode in which the conveying blower 5a operates at a strong power setting, the airflow 8a generated by the conveying blower 5a spreads outward compared to the other airflows 8b to 8d. That is, the airflow 8a generated by the conveying blower 5a does not only flow straight toward the conveying blower 5b, but also spreads outward. As a result, the circulation passage 105 between the conveying blowers 5a and 5b can be sufficiently ventilated.

[0054] In particular, CO gas tends to accumulate around the entrance / exit gate 106 due to vehicles passing through the entrance / exit gate 106. Therefore, by operating the parking lot ventilation system 1 in the strong entrance operation mode, the parking lot 100 can be efficiently ventilated.

[0055] ***Explanation of Deepest Area Strong Operation Mode*** Figure 13 is a diagram showing the ventilation state when the parking lot ventilation system 1 is operated in deepest area strong operation mode. As shown in Figure 13, under the control of the air blowing control device 7c, the conveying blower 5c operates in strong operation with a large air volume. Also, under the control of the air blowing control devices 7a, 7b, and 7d, the conveying blowers 5a, 5b, and 5d each operate in weak operation with a small air volume.

[0056] In the flowchart of Figure 9, if the cumulative number of vehicles entering and leaving the warehouse is equal to or greater than the threshold value (step S13), the airflow control device 7c receives a notification signal from the counting unit 6d (step S14) and sends a control signal to the conveyance blower 5c to increase the airflow volume (step S15). Upon receiving this control signal, the conveyance blower 5c increases the airflow volume and operates at high airflow volume (step S16). Furthermore, each of the conveyance blowers 5a, 5b, and 5d maintains the low airflow volume set by the airflow control devices 7a, 7b, and 7d in step S10.

[0057] In this way, in the deepest strong operation mode in which the conveying blower 5c operates at a strong power setting, the airflow 8c generated by the conveying blower 5c spreads outward compared to the other airflows 8a, 8b, and 8d. That is, the airflow 8c generated by the conveying blower 5c does not only flow straight toward the conveying blower 5d, but also spreads outward. As a result, the circulation passage 105 between the conveying blower 5c and the conveying blower 5d can be sufficiently ventilated.

[0058] In particular, CO gas tends to accumulate more easily around the third corner 103c, which is the farthest from the air intake 101 and the exhaust 102, than around the first corner 103a, the second corner 103b, and the fourth corner 103d. Therefore, by operating the parking lot ventilation system 1 in the innermost strong operation mode, the parking lot 100 can be ventilated efficiently.

[0059] ***Explanation of Corner Stirring Mode A*** Figure 14 is a diagram showing the ventilation state when the parking lot ventilation system 1 is operated in corner stirring mode A. As shown in Figure 14, under the control of the air blowing control devices 7a and 7b, each of the conveying fans 5a and 5b operates at high speed with a large air volume. Also, under the control of the air blowing control devices 7c and 7d, each of the conveying fans 5c and 5d operates at low speed with a small air volume.

[0060] In the flowchart of FIG. 9 , if the cumulative number of vehicles entering and leaving the warehouse is equal to or greater than the threshold (step S13), the airflow control device 7a sends a control signal to the conveyance blower 5a to increase the airflow rate (step S15). Upon receiving this control signal, the conveyance blower 5a increases its airflow rate and operates at high airflow rate (step S16). Similarly, upon receiving a notification signal from the counting unit 6d (step S14), the airflow control device 7b sends a control signal to the conveyance blower 5b to increase its airflow rate (step S15). Upon receiving this control signal, the conveyance blower 5b increases its airflow rate and operates at high airflow rate (step S16). Furthermore, the conveyance blowers 5c and 5d each maintain the low airflow rate set by the airflow control devices 7c and 7d in step S10.

[0061] Thus, in corner agitation mode A, in which the conveying fans 5a and 5b operate at high power, the air flow 8a generated by the conveying fan 5a and the air flow 8b generated by the conveying fan 5b are air flows that spread out more widely than the other air flows 8c and 8d. That is, the air flow 8a generated by the conveying fan 5a not only travels straight toward the conveying fan 5b but also spreads out around the periphery. As a result, the circulation passage 105 between the conveying fans 5a and 5b can be sufficiently ventilated. Similarly, the air flow 8b generated by the conveying fan 5b not only travels straight toward the conveying fan 5c but also spreads out around the periphery. As a result, the circulation passage 105 between the conveying fans 5b and 5c can be sufficiently ventilated.

[0062] In particular, in corner stirring mode A, fresh air from air intake 101 is transported to the periphery of second corner 103b by the strong wind of conveying blower 5a, and air with a high CO gas concentration stagnating around second corner 103b is transported toward conveying blower 5c by the strong wind of conveying blower 5b. Such corner stirring mode A allows for concentrated ventilation around second corner 103b. Therefore, even if the CO gas concentration around second corner 103b increases due to structural reasons of parking lot 100, such as a high ceiling, the air with a high CO gas concentration around second corner 103b can be quickly exhausted, thereby achieving efficient ventilation within parking lot 100.

[0063] ***Explanation of Corner Agitation Mode B*** Figure 15 is a diagram showing the ventilation state when the parking lot ventilation system 1 is operated in corner agitation mode B. As shown in Figure 15, the air blowing control devices 7a and 7d control each of the conveying fans 5a and 5d to operate at high speed with a large air volume. Also, the air blowing control devices 7b and 7c control each of the conveying fans 5b and 5c to operate at low speed with a small air volume.

[0064] In the flowchart of FIG. 9 , if the cumulative number of vehicles entering and leaving the warehouse is equal to or greater than the threshold (step S13), the airflow control device 7a sends a control signal to the conveyance blower 5a to increase the airflow rate (step S15). Upon receiving this control signal, the conveyance blower 5a increases its airflow rate and operates at high airflow rate (step S16). Similarly, upon receiving a notification signal from the counting unit 6d (step S14), the airflow control device 7d sends a control signal to the conveyance blower 5d to increase its airflow rate (step S15). Upon receiving this control signal, the conveyance blower 5d increases its airflow rate and operates at high airflow rate (step S16). Furthermore, the conveyance blowers 5b and 5c each maintain the low airflow rate set by the airflow control devices 7b and 7c in step S10.

[0065] Thus, in corner agitation mode B, in which the conveying fans 5a and 5d are operated at high power, the air flow 8a generated by the conveying fan 5a and the air flow 8d generated by the conveying fan 5d are air flows that spread out more widely than the other air flows 8b and 8c. That is, the air flow 8a generated by the conveying fan 5a not only travels straight toward the conveying fan 5b but also spreads out around the periphery. As a result, the circulation passage 105 between the conveying fans 5a and 5b can be sufficiently ventilated. Similarly, the air flow 8d generated by the conveying fan 5d not only travels straight toward the conveying fan 5a but also spreads out around the periphery. As a result, the circulation passage 105 between the conveying fans 5a and 5d can be sufficiently ventilated.

[0066] In particular, in corner mixing mode B, fresh air from air intake 101 is transported to the periphery of second corner 103b by the strong wind of conveying blower 5a, and air with a high CO gas concentration stagnating around fourth corner 103d is transported to exhaust outlet 102 by the strong wind of conveying blower 5d. Such corner mixing mode B allows for concentrated ventilation around first corner 103a. Therefore, even if, for example, vehicles are congested near entrance / exit gate 106 during the morning and evening rush hours and the CO gas concentration around first corner 103a increases, the air with a high CO gas concentration around first corner 103a can be quickly exhausted, thereby achieving efficient ventilation within parking lot 100.

[0067] ***Description of Central Stirring Mode A*** Figure 16 is a diagram showing the ventilation state when the parking lot ventilation system 1 is operated in central stirring mode A. As shown in Figure 16, the air blowing control devices 7a and 7c control each of the conveying fans 5a and 5c to operate at high speed with a large air volume. Also, the air blowing control devices 7b and 7d control each of the conveying fans 5b and 5d to operate at low speed with a small air volume.

[0068] In the flowchart of FIG. 9 , if the cumulative number of vehicles entering and leaving the warehouse is equal to or greater than the threshold (step S13), the airflow control device 7a sends a control signal to the conveyance blower 5a to increase the airflow rate (step S15). Upon receiving this control signal, the conveyance blower 5a increases its airflow rate and operates at high airflow rate (step S16). Similarly, upon receiving a notification signal from the counting unit 6d (step S14), the airflow control device 7c sends a control signal to the conveyance blower 5c to increase its airflow rate (step S15). Upon receiving this control signal, the conveyance blower 5c increases its airflow rate and operates at high airflow rate (step S16). Furthermore, the conveyance blowers 5b and 5d each maintain the low airflow rate set by the airflow control devices 7b and 7d in step S10.

[0069] Thus, in central agitation mode A, in which conveying fans 5a and 5c operate at high power, air flow 8a generated by conveying fan 5a and air flow 8c generated by conveying fan 5c become air flows that spread more outward than the other air flows 8b and 8d. A portion of air flow 8a that spreads toward the center of parking lot 100 becomes air flow 9a that circulates clockwise in a plan view around the center of parking lot 100. Similarly, a portion of air flow 8c that spreads toward the center of parking lot 100 becomes air flow 9b that circulates clockwise in a plan view around the center of parking lot 100. As a result, air flows 9a and 9b circulating around the center of parking lot 100 agitate the air with a high CO concentration stagnating around the center of parking lot 100, promoting ventilation around the center of parking lot 100.

[0070] ***Explanation of Central Stirring Mode B*** Figure 17 is a diagram showing the ventilation state when the parking lot ventilation system 1 is operated in central stirring mode B. As shown in Figure 17, the conveying fans 5b and 5d are operated in high-speed operation with a large air volume under the control of the air blowing control devices 7b and 7d, respectively. Also, the conveying fans 5a and 5c are operated in low-speed operation with a small air volume under the control of the air blowing control devices 7a and 7c, respectively.

[0071] In the flowchart of FIG. 9 , if the cumulative number of vehicles entering and leaving the warehouse is equal to or greater than the threshold (step S13), the airflow control device 7b sends a control signal to the conveyance blower 5b to increase the airflow rate (step S15). Upon receiving this control signal, the conveyance blower 5b increases its airflow rate and operates at high airflow rate (step S16). Similarly, upon receiving a notification signal from the counting unit 6d (step S14), the airflow control device 7d sends a control signal to the conveyance blower 5d to increase its airflow rate (step S15). Upon receiving this control signal, the conveyance blower 5d increases its airflow rate and operates at high airflow rate (step S16). Furthermore, the conveyance blowers 5a and 5c each maintain the low airflow rate set by the airflow control devices 7a and 7c in step S10.

[0072] Thus, in central agitation mode B, in which conveying fans 5b and 5d operate at high power, airflow 8b generated by conveying fan 5b and airflow 8d generated by conveying fan 5d become airflows that spread more outward than the other airflows 8a and 8c. A portion of airflow 8b that spreads toward the center of parking lot 100 becomes airflow 9c that circulates clockwise in a plan view around the center of parking lot 100. Similarly, a portion of airflow 8d that spreads toward the center of parking lot 100 becomes airflow 9d that circulates clockwise in a plan view around the center of parking lot 100. As a result, airflows 9c and 9d circulating around the center of parking lot 100 agitate the air stagnating around the center of parking lot 100, promoting ventilation around the center of parking lot 100.

[0073] As described above, the parking lot ventilation system 1 according to the first embodiment is equipped with many control modes, allowing ventilation within the parking lot 100 in the optimal control mode according to the structure of the parking lot 100, etc. Furthermore, the parking lot ventilation system 1 according to the first embodiment can efficiently exhaust CO gas by stirring not only the ventilation air duct 4 extending from the air inlet 101 to the air outlet 102, but also the entire parking lot 100, including the corners and center of the parking lot 100. Note that, although four air blowing control devices 7a to 7d each control four conveying fans 5a to 5d in FIGS. 10 to 17, a single air blowing control device may be used to centrally control the four conveying fans 5a to 5d.

[0074] 9, a single threshold value is used for comparison with the cumulative number of vehicles that have entered and left the facility, but multiple threshold values ​​that increase in stages may be used for comparison with the cumulative number of vehicles. In this case, in step S15, the air blowing control devices 7a and 7c send control signals to the conveying fans 5a and 5c so that the conveying fans 5a and 5c operate in control modes that correspond to the threshold values ​​used in step S13. Then, in step S16, the conveying fans 5a and 5c that have received these control signals operate in different control modes depending on the threshold values.

[0075] For example, if there are three thresholds to be compared in step S13, the thresholds are designated S, M, and L in ascending order. If the threshold used in step S13 is S, the conveying blower 5 operates in the entrance strong operation mode or the innermost strong operation mode. If the threshold used in step S13 is S, the conveying blower 5 operates in the corner stirring mode A or the corner stirring mode B. If the threshold used in step S13 is L, the conveying blower 5 operates in the center stirring mode A or the center stirring mode B. In this way, by operating the conveying blower 5 in the optimal control mode depending on the cumulative number of vehicles entering and leaving the parking lot 100, the parking lot 100 can be effectively ventilated. The control mode selection here is merely an example, and other control modes may also be selected.

[0076] ***Explanation of Combined Mode*** Here, when the combined mode, which periodically switches between multiple control modes, is used, the airflow periodically changes, generating turbulence, further promoting agitation of the air within the parking lot 100. For example, when operating from the strong operation mode, passing through multiple other control modes, and then returning to the strong operation mode, the position of the turbulence generated within the parking lot 100 changes periodically, allowing for more effective agitation of the air stagnating within the parking lot 100. An example of the operation of the combined mode will be described using FIG. 18. In this example, the processing of steps S15 to S17 in the flowchart of FIG. 9 has been modified. FIG. 18 shows the process flow after the modification.

[0077] As shown in Figure 18, when a notification signal is sent from the counting unit 6d to each of the airflow control devices 7a-7d in step S14, each of the airflow control devices 7a-7d that received the notification signal controls each of the conveying fans 5a-5d. Through this control, each of the conveying fans 5a-5d operates in high-power operation mode for a predetermined third time period (step S20). After operation in the high-power operation mode ends, each of the airflow control devices 7a-7d cooperate to determine the next control mode (step S21). The control modes that can be determined here are modes other than the low-power operation mode and the high-power operation mode.

[0078] After determining the control mode, each of the air blowing control devices 7a to 7d controls each of the conveying fans 5a to 5d. Through this control, each of the conveying fans 5a to 5d operates in the determined control mode for a third time period (step S22). Next, it is determined whether a predetermined second time period has elapsed (step S23). If the second time period has not elapsed, the process returns to step S21. If it is determined in step S23 that the second time period has elapsed, each of the conveying fans 5a to 5d operates in the high operation mode for a third time period (step S24).

[0079] The above process realizes a combined mode that switches from the high operation mode to multiple other control modes and then back to the high operation mode. The control mode determined by the air flow control devices 7a to 7d in cooperation with each other may be determined according to a predetermined control mode pattern, or may be determined randomly each time step S21 is performed.

[0080] 19(a) to 19(c) are diagrams showing examples of combined modes. The combined mode shown in FIG. 19(a) is a mode that, after a strong operation mode, alternates between an entrance strong operation mode and an innermost strong operation mode, and then returns to the strong operation mode. By operating in this combined mode, it is possible to sufficiently agitate the air around the entrance of the parking lot 100 and around the innermost part of the parking lot 100, thereby promoting ventilation within the parking lot 100. The combined mode shown in FIG. 19(b) is a mode that, after a strong operation mode, alternates between a corner agitation mode A and a corner agitation mode B, and then returns to the strong operation mode. By operating in this combined mode, it is possible to sufficiently agitate the air around each corner of the parking lot 100, thereby promoting ventilation within the parking lot 100.

[0081] The combined mode shown in FIG. 19(c) is a mode in which the strong operation mode is followed by repeated cycles of center agitation mode A and center agitation mode B, and then returns to the strong operation mode. Operating in this combined mode allows for concentrated agitation of the air around the center of the parking lot 100, thereby promoting ventilation within the parking lot 100. The combined mode examples shown in FIGS. 19(a) to 19(c) are merely examples, and other combinations are also possible, such as combining corner agitation mode A with center agitation mode A. Furthermore, three or more control modes may be combined in addition to the strong operation mode. Furthermore, control may be performed in which different control modes are combined for each cycle, with the operation from the strong operation mode through multiple other control modes and then returning to the strong operation mode being considered as one cycle. Furthermore, in the first embodiment, CO gas was described as an example of a gas exhausted from the parking lot 100, but the exhaust gas is not limited to CO gas and may be other exhaust gases such as NOx gas.

[0082] Embodiment 2. ***Description of the Configuration of the Learning Device*** Figure 20 is a block diagram showing the configuration of a learning device 120 according to embodiment 2. The learning device 120 is connected to the parking lot ventilation system 1 via a communication line (not shown). The learning device 120 learns airflow control information that controls the airflow rate of the conveying fan 5 shown in Figure 1, and generates a trained model to be used in the parking lot ventilation system 1 shown in embodiment 1. As in embodiment 1, the ventilation area ventilated by the conveying fan 5 is the parking lot 100.

[0083] As shown in FIG. 20 , the learning device 120 includes a data acquisition unit 121 and a model generation unit 122. The data acquisition unit 121 acquires the air flow rate of the conveyance blower 5 and the number of vehicles entering and leaving the parking lot 100 as learning data. The learning data acquired by the data acquisition unit 121 is output to the model generation unit 122. The model generation unit 122 learns air flow rate control information for the conveyance blower 5 based on the learning data input from the data acquisition unit 121, including the air flow rate of the conveyance blower 5 and the number of vehicles entering and leaving the parking lot 100. In other words, it generates a trained model that infers optimal air flow rate control information for the conveyance blower 5 from the number of vehicles entering and leaving the parking lot 100 obtained from the train-entry / exit number detection device 6. The generated trained model is then stored in the trained model storage unit 130.

[0084] The air volume may include not only the air volume of the conveying blower 5 but also the air volume of the supply blower 2 and the exhaust blower 3. In this case, the air volume control information is information for controlling the air volumes of the supply blower 2, the exhaust blower 3, and the conveying blower 5.

[0085] The learning algorithm used by the model generation unit 122 may be a known algorithm such as supervised learning, unsupervised learning, or reinforcement learning. As an example, a case where reinforcement learning is applied will be described. In reinforcement learning, an agent (subject of action) in a certain environment observes the current state (environmental parameters) and decides on an action to be taken. The environment changes dynamically depending on the agent's actions, and the agent is given a reward according to the change in the environment. The agent repeats this process and learns a course of action that will obtain the most reward through a series of actions. Q-learning and TD-learning are known as representative reinforcement learning methods. For example, in the case of Q-learning, a general update formula for the action value function Q(s, a) is expressed by Equation 1:

[0086]

[0087] In equation 1, st represents the state of the environment at time t, and at represents the action at time t. The state changes to st+1 due to the action at. rt+1 represents the reward obtained due to the change in state, γ represents the discount rate, and α represents the learning coefficient. Note that γ is in the range of 0<γ≦1, and α is in the range of 0<α≦1. The air volume of the conveying blower 5 represents the action at, the number of entries and exits from the parking lot 100 represents the state st, and the best action at in the state st at time t is learned.

[0088] The update formula expressed by Equation 1 increases the action value Q if the action value Q of the action a with the highest Q value at time t+1 is greater than the action value Q of the action a executed at time t, and decreases the action value Q in the opposite case. In other words, the action value function Q(s, a) is updated so that the action value Q of the action a at time t approaches the best action value at time t+1. As a result, the best action value in a certain environment is propagated sequentially to the action value in the previous environment.

[0089] As described above, when generating a trained model by reinforcement learning, the model generation unit 122 includes a reward calculation unit 122a and a function update unit 122b.

[0090] The reward calculation unit 122a calculates a reward based on the airflow volume of the conveyance blower 5 and the number of vehicles entering and leaving the parking lot 100. The reward calculation unit 122a calculates a reward r based on an exhaust gas concentration standard within the parking lot 100. For example, if the exhaust gas concentration within the parking lot 100 decreases as a result of operating the conveyance blower 5 at this airflow volume, the reward r is increased (for example, a reward of "1" is given). On the other hand, if the exhaust gas concentration within the parking lot 100 increases as a result of operating the conveyance blower 5 at this airflow volume, the reward r is decreased (for example, a reward of "-1" is given).

[0091] The function update unit 122b updates the function for determining the airflow control information for the conveying blower 5 in accordance with the reward calculated by the reward calculation unit 122a, and outputs the updated function to the learned model storage unit 130. For example, in the case of Q-learning, the action value function Q(st, at) expressed by Equation 1 is used as a function for calculating the airflow control information for the conveying blower 5.

[0092] The learning process described above is repeated. The learned model storage unit 130 stores the action-value function Q(st,at) updated by the function update unit 122b, i.e., the learned model.

[0093] ***Description of Environmental Sensors*** The exhaust gas concentration in the parking lot 100, which is used as a basis when the reward calculation unit 122a calculates rewards, is measured as follows. As shown in FIG. 21 , environmental sensors 140 are installed on multiple pillars and other components within the parking lot 100, and these environmental sensors 140 measure the exhaust gas concentration. The environmental sensors 140 are installed at a height of approximately 1 to 2 meters above the floor and measure the exhaust gas concentration in the air. The multiple environmental sensors 140 are connected to a network (not shown), and exhaust gas concentration data measured by each environmental sensor 140 is transmitted to the data acquisition unit 121. The environmental sensors 140 are installed in multiple locations, including the center and periphery of the parking lot 100, to grasp the air quality of the entire parking lot 100. The environmental sensors 140 are installed at a height of approximately 1 to 2 meters above the floor because the concentration near a person's mouth and nose when standing is the most problematic. Exhaust gases to be emitted from the parking lot 100 include CO gas and NOx gas.

[0094] Environmental sensors 140 are expensive, and it is not practical to permanently install environmental sensors 140 at multiple locations within parking lot 100. Therefore, in this embodiment, multiple environmental sensors 140 are temporarily installed in parking lot 100, and when the learning period by learning device 120 ends, multiple environmental sensors 140 are collected from parking lot 100. In this way, environmental sensors 140 can be used in rotation at multiple parking lots, thereby reducing the cost of learning by learning device 120.

[0095] Furthermore, parking lots come in a variety of shapes, and the optimal control differs for each parking lot. However, in this embodiment, it is possible to learn airflow control that is tailored to each parking lot, so that appropriate airflow control can be performed regardless of the shape of the parking lot.

[0096] ***Explanation of Operation of the Learning Device*** Next, the operation of the learning device 120 will be described using Figure 22. Figure 22 is a flowchart showing the flow of the learning process by the learning device 120. The data acquisition unit 121 acquires airflow volume data of the conveying blower 5 and data on the number of entries and exits of the parking lot 100 as learning data (step S30). The model generation unit 122 calculates a remuneration based on the airflow volume data of the conveying blower 5 and the data on the number of entries and exits of the parking lot 100 (step S31). Specifically, the remuneration calculation unit 122a acquires the airflow volume data of the conveying blower 5 and the data on the number of entries and exits of the parking lot 100, and determines whether to increase or decrease the remuneration based on a predetermined exhaust gas concentration standard within the parking lot 100 (step S32).

[0097] If the conveying blower 5 operates at this air volume and the amount of exhaust gas in the parking lot 100 decreases, the remuneration calculation unit 122a determines that the remuneration should be increased, and increases the remuneration (step S32). On the other hand, if the conveying blower 5 operates at this air volume and the amount of exhaust gas in the parking lot 100 increases, the remuneration calculation unit 122a determines that the remuneration should be decreased, and decreases the remuneration (step S33).

[0098] The function update unit 122b updates the action value function Q(st, at) represented by Equation 1 and stored in the trained model storage unit 130 based on the reward calculated by the reward calculation unit 122a (step S34). The learning device 120 repeatedly executes the above steps S30 to S34 and stores the generated action value function Q(st, at) as a trained model in the trained model storage unit 130.

[0099] As described above, the learning device 120 according to the second embodiment learns during a learning period how the exhaust gas concentration in the parking lot 100 changes due to changes in the airflow rate of the conveyance blower 5 operated to ventilate the parking lot 100 when a vehicle actually enters or leaves the parking lot 100, and generates an optimal trained model. Therefore, the learning device 120 according to the second embodiment can generate a trained model necessary for efficiently ventilating the parking lot 100.

[0100] In addition, the learning device 120 in this embodiment is configured to store the learned model in a learned model storage unit 130 provided outside the learning device 120, but the learned model storage unit 130 may also be provided inside the learning device 120.

[0101] In addition, although the present embodiment has been described with reference to a case where supervised learning is applied to the learning algorithm used by the model generation unit 122, the present invention is not limited to this. As for the learning algorithm, reinforcement learning, unsupervised learning, semi-supervised learning, or the like can also be applied in addition to supervised learning.

[0102] Furthermore, the model generation unit 122 may learn the airflow rate control information according to learning data created for multiple parking lot ventilation systems. Furthermore, the learning device 120 that has learned the airflow rate control information for one parking lot ventilation system may be applied to another parking lot ventilation system, and the learning device 120 may re-learn and update the airflow rate control information for the other parking lot ventilation system.

[0103] Furthermore, the learning algorithm used in the model generation unit 122 may be deep learning, which learns to extract the features themselves, or machine learning may be performed according to other known methods, such as genetic programming, functional logic programming, or support vector machines.

[0104] Embodiment 3. ***Description of the Configuration of the Inference Device*** Figure 23 is a block diagram showing the configuration of an inference device 150 according to embodiment 3. The inference device 150 is connected to the parking lot ventilation system 1 via a communication line (not shown). As shown in Figure 23, the inference device 150 includes a data acquisition unit 151 and an inference unit 152. The data acquisition unit 151 acquires data on the number of entries and exits from the parking lot 100. The data acquisition unit 151 outputs the data on the number of entries and exits to the inference unit 152. The inference unit 152 reads out the trained model stored in the trained model storage unit 130 and infers airflow control information obtained using this trained model. That is, by inputting the data on the number of entries and exits acquired by the data acquisition unit 151 into this trained model, it is possible to output airflow control information for operating the conveyance fan 5 at an airflow rate appropriate for the number of entries and exits.

[0105] The airflow rate control information output from the inference unit 152 is input to the airflow control devices 7a to 7d, and the airflow control devices 7a to 7d control the conveying fans 5a to 5d based on this airflow rate control information. Note that, in the present embodiment, the airflow rate control information is output using a trained model generated by the model generation unit 122 of the learning device 120, but it is also possible to acquire a trained model from an external source and output the airflow rate control information based on this trained model.

[0106] ***Explanation of Operation of Inference Device*** Next, the operation of the inference device 150 will be described using FIG. 24. FIG. 24 is a flowchart showing the flow of inference processing by the inference device 150. The data acquisition unit 151 acquires data on the number of entries and exits from the parking lot 100 (step S40). The inference unit 152 reads out the trained model stored in the trained model storage unit 130 and inputs the data on the number of entries and exits acquired by the data acquisition unit 151 into this trained model to obtain airflow control information (step S41). The inference unit 152 outputs the obtained airflow control information to the airflow control devices 7a to 7d of the parking lot ventilation system 1 (step S42). The airflow control devices 7a to 7d control the conveying fans 5a to 5d based on this airflow control information (step S43).

[0107] As described above, the inference device 150 according to the third embodiment infers airflow rate control information based on the trained model learned by the learning device 120. This allows the conveyance blower 5, which is controlled based on this airflow rate control information, to operate optimally. As a result, the parking lot ventilation system 1 can provide efficient ventilation in accordance with the number of vehicles entering and leaving the parking lot 100, thereby improving the air quality within the parking lot 100. The airflow rate control information output from the inference device 150 includes information regarding the airflow rate of the conveyance blower 5 and information regarding the control mode for controlling the conveyance blower 5. Based on the information regarding the airflow rate of the conveyance blower 5, the timing for increasing the airflow rate of the conveyance blower 5 can be optimized. Furthermore, the information regarding the control mode for controlling the conveyance blower 5 allows a control mode effective for efficiently ventilating the parking lot 100 to be selected.

[0108] The learning device 120 according to the second embodiment and the inference device 150 according to the third embodiment may be connected to the parking lot ventilation system 1 via a network (not shown). The learning device 120 and the inference device 150 may reside on a cloud server (not shown).

[0109] ***Description of Hardware Configuration*** Fig. 25 is a diagram showing an example of the hardware configuration of counting unit 6d, air blowing control devices 7a to 7d, model generation unit 122, and inference unit 152. Counting unit 6d, air blowing control devices 7a to 7d, model generation unit 122, and inference unit 152 each have a processor 160, a memory 161, and an interface 162.

[0110] The processor 160 is a CPU (Central Processing Unit). The processor 160 may be a processing device, an arithmetic unit, a microprocessor, a microcomputer, or a DSP (Digital Signal Processor). The functions of the counting unit 6d, the airflow control devices 7a to 7d, the model generating unit 122, and the inference unit 152 are realized by the processor 160, software, firmware, or a combination of software and firmware. The software or firmware is written as a program and stored in the built-in memory 161. The memory 161 is a non-volatile or volatile semiconductor memory such as a RAM (Random Access Memory), a ROM (Read Only Memory), a flash memory, an EPROM (Erasable Programmable Read Only Memory), or an EEPROM (Electrically Erasable Programmable Read Only Memory).

[0111] The configurations shown in the above embodiments are merely examples, and may be combined with other known technologies. Furthermore, embodiments may be combined with each other, and some of the configurations may be omitted or modified without departing from the spirit of the invention.

[0112] 1 Parking lot ventilation system, 2 Air supply fan, 3 Exhaust fan, 4 Ventilation air duct, 5a to 5d Transport fan, 6 Entry / exit number detection device, 6a Infrared emitting unit (vehicle detection unit), 6b, 6c Infrared sensor (vehicle detection unit), 6d Counting unit, 7a to 7d Air blowing control device, 100 Parking lot, 101 Air supply port, 102 Exhaust port, 103a First corner, 103b Second corner, 103c Third corner, 103d Fourth corner, 104 Vehicle entry / exit, 105 Circular passage, 106 Entry / exit gate, 120 Learning device, 121 Data acquisition unit, 122 Model generation unit, 130 Learned model storage unit, 150 Inference device, 151 Data acquisition unit, 152 Inference unit.

Claims

1. A parking lot ventilation system that ventilates a parking lot by supplying air into the parking lot from an air intake port and exhausting the air from an exhaust port, an intake fan that draws in outside air through the intake port; an exhaust fan that exhausts the internal air through the exhaust port; a plurality of conveying fans that are sequentially installed on a ventilation air path extending from the air inlet to the exhaust port and convey air; a vehicle entry / exit number detection device for detecting the number of vehicles entering and leaving the parking lot; an air blowing control device that controls the air blowing volume of one or more of the conveying fans based on the number of incoming and outgoing goods detected by the incoming and outgoing goods number detection device, The parking lot ventilation system is configured so that, when the number of vehicles leaving the parking lot per unit time is large, the air blowing control device controls the air volume of the multiple transport fans so that one or more of the transport fans operate at high speed with a high air volume, and the remaining one or more transport fans operate at low speed with a lower air volume than the high speed operation.

2. A parking lot ventilation system as described in Claim 1, wherein the plurality of conveying fans are respectively arranged at multiple corners of the parking lot.

3. The parking lot ventilation system of claim 2, wherein the air blowing control device operates the conveying fans located at two adjacent corners at high speed, and operates the conveying fans located at the remaining corners at low speed.

4. The parking lot ventilation system of claim 2, wherein the air blowing control device operates the conveying fans located at two opposing corners at high speed, and operates the conveying fans located at the remaining corners at low speed.

5. A parking lot ventilation system that ventilates a parking lot by supplying air into the parking lot from an air intake port and exhausting the air from an exhaust port, an intake fan that draws in outside air through the intake port; an exhaust fan that exhausts the internal air through the exhaust port; a plurality of conveying fans that are sequentially installed on a ventilation air path extending from the air inlet to the exhaust port and convey air; a vehicle entry / exit number detection device for detecting the number of vehicles entering and leaving the parking lot; an air blowing control device that controls the air blowing volume of one or more of the conveying fans based on the number of incoming and outgoing goods detected by the incoming and outgoing goods number detection device, The air blowing control device is a parking lot ventilation system that controls the air flow volume of the multiple conveying fans by alternating between high-power operation with a high air flow volume and low-power operation with a lower air flow volume than the high-power operation when the number of pre-entry and post-entry vehicles per unit time in the parking lot is high.

6. A parking lot ventilation system as described in Claim 5, wherein the plurality of conveying fans are respectively arranged at multiple corners of the parking lot.

7. 6. A parking lot ventilation system according to claim 1, wherein the conveying blower is installed at a distance of 0.5 m or less from the ceiling surface of the parking lot.

8. 6. The parking lot ventilation system according to claim 1, wherein at least one of the conveying fans is equipped with the air blowing control device.

9. 9. The parking lot ventilation system according to claim 8, wherein the conveying fan having the shortest total distance from the air inlet and the air outlet is equipped with the air blowing control device.

10. 9. The parking lot ventilation system according to claim 8, wherein the conveying fan having the longest total distance from the air supply port and the air exhaust port is provided with the air blowing control device.

11. The incoming / outgoing number detection device is a vehicle detection unit that is installed at a vehicle entrance / exit path of the parking lot and detects vehicles entering and exiting the parking lot; 6. The parking lot ventilation system according to claim 1, further comprising a counting unit that counts the number of vehicles based on the output signal of the vehicle detection unit.

12. A parking lot ventilation method for ventilating a parking lot using a plurality of conveying fans arranged in the parking lot, comprising: This parking lot ventilation method operates one or more of the conveying fans at high speed with a large air volume when there are a large number of entries and exits per unit time in the parking lot, and operates the remaining one or more of the conveying fans at low speed with a smaller air volume than the high speed operation to ventilate the parking lot.

13. A parking lot ventilation method as described in Claim 12, wherein the plurality of conveying fans are respectively arranged at a plurality of corners of the parking lot.

14. A parking lot ventilation method as described in claim 13, wherein when the number of entries and withdrawals per unit time in the parking lot is large, the conveying fans located at two adjacent corners are operated at the high speed, and the conveying fans located at the remaining corners are operated at the low speed to ventilate the parking lot.

15. A parking lot ventilation method as described in claim 13, wherein when the number of entries and withdrawals per unit time in the parking lot is high, the conveying fans located at two opposing corners are operated at the high speed, and the conveying fans located at the remaining corners are operated at the low speed to ventilate the parking lot.

16. A parking lot ventilation method for ventilating a parking lot using a plurality of conveying fans arranged in the parking lot, comprising: This parking lot ventilation method ventilates the parking lot by alternately operating the multiple conveying fans between high-power operation with a large air volume and low-power operation with a smaller air volume than the high-power operation when there are a large number of entries and exits per unit time in the parking lot.

17. A parking lot ventilation method as described in Claim 16, wherein the plurality of conveying fans are respectively arranged at a plurality of corners of the parking lot.

18. A learning device connected to the parking lot ventilation system according to claim 1 or claim 5 via a communication line, a data acquisition unit that generates learning data including the number of incoming and outgoing goods detected by the incoming and outgoing goods number detection device and the air flow rate of the conveying fan for the number of incoming and outgoing goods; A learning device comprising: a model generation unit that uses the learning data to generate a trained model for inferring air volume control information for the conveying blower from the number of incoming and outgoing shipments.

19. An inference device connected to the parking lot ventilation system according to claim 1 or claim 5 by a communication line, a data acquisition unit that acquires the incoming and outgoing numbers detected by the incoming and outgoing number detection device; An inference device comprising an inference unit that outputs airflow control information for the conveying fan from the number of incoming and outgoing shipments acquired by the data acquisition unit, using a trained model for inferring airflow control information for the conveying fan from the number of incoming and outgoing shipments.