Wind speed calculation device, wind speed calculation method, and program

Abstract

The wind speed calculation device (100) according to the present disclosure comprises: a boundary value calculation unit (2) that calculates velocity potential (uΓi) at a boundary on the basis of the shape of an area inside a structure, the position of and normal wind speed (q) at a ventilation facility (VF) installed at the boundary with the outside of the area, and the position of and normal wind speed (q) at a ventilation port (VP) installed at the boundary; and an interior point calculation unit (4) that calculates the wind speed (q') at a prescribed position in the area on the basis of the velocity potential (uΓi). Either the boundary value calculation unit (2) performs low-rank approximation to calculate the velocity potential (uΓi), or the interior point calculation unit (4) performs low-rank approximation to calculate the wind speed (q') at the prescribed position.

Description

[Technical field]

[0001] The present disclosure relates to a wind speed calculation device, a wind speed calculation method, and a program. [Background technology]

[0002] Conventionally, in order to ensure the safety of workers who maintain tunnels, which are underground culverts, it is necessary to properly ventilate the inside of the tunnel. For this reason, it is known to install a ventilation device in the tunnel and monitor the oxygen concentration at points where air tends to stagnate using an oxygen concentration sensor while the ventilation device is operating.

[0003] For example, Non-Patent Document 1 describes a method for simulating the wind speed distribution inside a tunnel by generating a simplified model of the entire tunnel using links and nodes based on a ventilation network analysis method and using an initial value of the wind speed.

[0004] However, tunnels, particularly those installed underground in urban areas, have complex shapes due to expansions, etc. In addition, multiple ventilation devices may be installed inside a tunnel. With the method described in Non-Patent Document 1, it was difficult to calculate the wind speed with high accuracy depending on complex conditions such as the shape of the tunnel and the locations where the ventilation devices are installed.

[0005] Therefore, Non-Patent Document 2 describes a method that uses nodes at the boundary between the tunnel area and the outside based on the boundary element method to calculate wind speed with high accuracy according to complex conditions such as the shape of the tunnel and the location where the ventilation equipment is installed. [Prior art documents] [Non-patent literature]

[0006] [Non-Patent Document 1] Nishima, Takayasu, and 3 others, "Aerodynamic Model for Designing Ventilation Systems in Complex Tunnels (Fluid Engineering, Fluid Machinery)", Transactions of the Japan Society of Mechanical Engineers, Series B, Vol. 68, No. 675, 2002, pp. 2990-2996 [Non-Patent Document 2] Tsukasa Nakayama, "Analysis of flows with free surfaces by the boundary element method (research into basic algorithms for numerical calculation)", Research Institute for Mathematical Sciences Kokyuroku (1985), February 1985, 553, 145-158 Summary of the Invention [Problem to be solved by the invention]

[0007] However, the method described in Non-Patent Document 2 has a problem in that it takes a long time to calculate the wind speed.

[0008] In consideration of the above circumstances, the object of the present disclosure is to provide a wind speed calculation device, a wind speed calculation method, and a program that can calculate wind speed with high accuracy and in a short period of time according to the complex conditions of an area within a structure. [Means for solving the problem]

[0009] In order to solve the above problems, the wind speed calculation device of the present disclosure comprises a boundary value calculation unit that calculates a velocity potential at the boundary based on the shape of an area within a structure, the position and normal wind speed of ventilation equipment installed on the boundary with the outside of the area, and the position and normal wind speed of a ventilation opening installed on the boundary, and an interior point calculation unit that calculates the wind speed at a specified position within the area based on the velocity potential, wherein the boundary value calculation unit calculates the velocity potential by performing a low-rank approximation, or the interior point calculation unit calculates the wind speed at the specified position by performing a low-rank approximation.

[0010] In addition, in order to solve the above problems, the wind speed calculation method of the present disclosure includes a boundary value calculation step of calculating a velocity potential at the boundary by performing a low-rank approximation based on the shape of an area within a structure, the position and normal wind speed of a ventilation equipment installed at the boundary with the outside of the area, and the position and normal wind speed of a ventilation opening installed at the boundary, and an interior point calculation step of calculating the wind speed at a specified position within the area based on the velocity potential, wherein the boundary value calculation step includes a step of calculating the velocity potential by performing a low-rank approximation, or the interior point calculation step includes a step of calculating the wind speed at the specified position by performing a low-rank approximation.

[0011] In order to solve the above problem, a program according to the present disclosure causes a computer to function as the wind speed calculation device described above. Effect of the Invention

[0012] According to the wind speed calculation device, wind speed calculation method, and program disclosed herein, it is possible to calculate wind speed with high accuracy and in a short time according to the complex conditions of an area within a structure. [Brief description of the drawings]

[0013] [Figure 1A] 1 is a schematic diagram of a wind speed calculation device according to an embodiment of the present invention. [Figure 1B] 2 is a schematic diagram of a condition storage unit shown in FIG. 1. [Figure 1C] 2 is a schematic diagram of a boundary value calculation unit shown in FIG. 1. [Figure 1D] FIG. 2 is a schematic diagram of a calculation result storage unit shown in FIG. [Figure 1E] FIG. 2 is a schematic diagram of an interior point calculation unit shown in FIG. [Diagram 2] 2 is a diagram showing an example of an area including wind speed calculation points where the wind speed calculation device shown in FIG. 1 calculates wind speeds; FIG. [Figure 3A] FIG. 13 is a diagram showing an example of nodes determined using the boundary element method. [Figure 3B] FIG. 13 is a diagram showing an example of nodes determined using a finite difference method. [Figure 4] 1D is a diagram for explaining low-rank approximation by the interpolation matrix calculation unit shown in FIG. 1C. FIG. [Diagram 5] FIG. 1D is a diagram showing an example of clusters generated by the tree structure generating unit shown in FIG. 1C. [Figure 6A] 4 is a part of a flowchart showing an example of an operation of the wind speed calculation device according to the present embodiment. [Figure 6B] 11 is a remaining part of the flowchart showing an example of the operation of the wind speed calculation device according to the present embodiment. [Figure 7A] FIG. 13 is a diagram showing the calculation time of the velocity potential at the boundary versus the number of nodes. [Figure 7B] FIG. 13 is a diagram showing the calculation time of the velocity potential at the wind speed calculation point in the domain relative to the number of nodes. [Figure 8] FIG. 13 is a diagram showing wind speeds at the positions of the respective wind speed calculation points. [Figure 9] FIG. 2 is a hardware block diagram of the wind speed calculation guide generating device. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0014] Hereinafter, the present embodiment will be described with reference to the drawings.

[0015] <<Configuration of the wind speed calculation device>> As shown in FIG. 1A, the wind speed calculation device 100 according to this embodiment includes a condition storage unit 1, a boundary value calculation unit 2, a calculation result storage unit 3, and an interior point calculation unit 4. The condition storage unit 1 and the calculation result storage unit 3 are configured with memories such as a hard disk drive (HDD), a solid state drive (SSD), an electrically erasable programmable read-only memory (EEPROM), a read-only memory (ROM), and a random access memory (RAM). The boundary value calculation unit 2 and the interior point calculation unit 4 form a control unit (controller). The control unit may be configured with dedicated hardware such as an application specific integrated circuit (ASIC) or a field programmable gate array (FPGA), may be configured with a processor, or may be configured with both.

[0016] <Configuration of condition storage unit> As shown in FIG. 1B, the condition storage unit 1 includes a shape storage unit 11, a ventilation equipment storage unit 12, and a ventilation opening storage unit 13.

[0017] The shape storage unit 11 stores information indicating the shape of an area Ω within a structure, including a predetermined position (wind speed calculation point SCP) where the wind speed calculation device 100 of this embodiment calculates the wind speed q'. The structure may be, for example, a tunnel as shown in FIG. 2. The information indicating the shape of the area Ω includes, for example, the position of the boundary Γ (e.g., wall, floor, ceiling) between the area Ω and the outside. In the example shown in FIG. 2, the area Ω includes three wind speed calculation points SCP1 to 3, which are a point where the cross-sectional area changes, a point where tunnels join, and a point where the area comes to a dead end, respectively.

[0018] The ventilation equipment storage unit 12 stores information about the ventilation equipment VF installed at the boundary Γ with the outside in the area Ω in which the shape storage unit 11 stores information indicating the shape. The information about the ventilation equipment VF includes the position where the ventilation equipment VF is installed and the wind speed in the normal direction (hereinafter referred to as "normal wind speed") q at the position of the ventilation equipment VF, which is determined by the specifications of the ventilation equipment VF. In the example shown in FIG. 2, the normal wind speed q is -1.0 (m / s). In this example, the normal wind speed q of the gas flowing out from the area Ω to the outside is shown as a positive value, and the normal wind speed q of the gas flowing into the area Ω from the outside is shown as a negative value. In this example, the ventilation equipment VF is an intake equipment that flows gas into the area Ω from the outside, but is not limited to this, and may be an exhaust equipment that flows gas out from the area Ω to the outside.

[0019] The ventilation opening memory unit 13 stores information about the ventilation opening VP installed at the boundary Γ with the outside in the area Ω in which the shape memory unit 11 stores information indicating the shape. The information about the ventilation opening VP includes the position where the ventilation opening VP is installed and the normal wind speed q at the position of the ventilation opening VP. The normal wind speed q at the position of the ventilation opening VP is obtained by actual measurement. In the example shown in FIG. 2, two ventilation openings VP are installed, and the normal wind speed q at the position of one ventilation opening VP is 0.4 (m / s), and the normal wind speed q at the position of the other ventilation opening VP is 0.3 (m / s). In this example, the ventilation opening VP functions as an exhaust port through which gas flows out from the area Ω to the outside, but is not limited to this, and may function as an intake port through which gas flows in from the outside into the area Ω.

[0020] <Configuration of boundary value calculation unit> The boundary value calculation unit 2 calculates a velocity potential u at the boundary Γ based on the shape of the area Ω inside the structure, the position and normal wind speed q of the ventilation equipment VF installed at the boundary Γ with the outside of the area Ω, and the position and normal wind speed q of the ventilation opening VP installed at the boundary Γ. Γi The boundary value calculation unit 2 performs low-rank approximation to calculate the velocity potential u Γi Here, the velocity potential u Γiis the velocity potential u at each of the i nodes on the boundary Γ, which has been discretized. As shown in Fig. 1C, the boundary value calculation unit 2 includes a shape reading unit 21, a model generation unit 22, a ventilation information reading unit 23, a boundary condition generation unit 24, a tree structure generation unit 25, an interpolation matrix calculation unit 26, a high-speed matrix-vector product execution unit 27, and a velocity potential calculation unit 28.

[0021] The shape reading unit 21 reads information indicating the shape of the region Ω, which is stored in the shape storage unit 11 .

[0022] The model generation unit 22 generates a model indicating the nodes on the boundary Γ with the outside of the region Ω. Specifically, the model generation unit 22 generates a model indicating the nodes on the boundary Γ with the outside of the region Ω based on the shape of the region Ω indicated by the information read by the shape reading unit 21. For example, the model generation unit 22 may determine the nodes on the boundary Γ by a collocation method using a piecewise constant basis, and generate a model indicating the nodes on the boundary Γ.

[0023] Thus, in this embodiment, the model generation unit 22 uses the finite element method to generate a model showing nodes (black circles in FIG. 3A) at the boundary Γ with the outside of the region Ω, as shown in Fig. 3A. This allows the model generation unit 22 to generate a model more easily than when using, for example, a finite difference method to generate a model showing nodes (black circles in FIG. 3B) throughout the region Ω, including the boundary and the inside, as shown in Fig. 3B.

[0024] The ventilation information reading unit 23 reads information about the ventilation equipment VF stored in the ventilation equipment memory unit 12. In addition, the ventilation information reading unit 23 reads information about the ventilation openings VP stored in the ventilation opening memory unit 13.

[0025] The boundary condition generating unit 24 generates boundary conditions (see equations (2) and (3)) for the governing equation (see equation (1)). In this example, the governing equation is Ω is the Laplace equation satisfied by

[0026] Δu Ω =0 in Ω (1)

[0027] Specifically, the boundary condition generating unit 24 generates boundary conditions as shown in formulas (2) and (3) based on the shape of the area Ω within the structure, the position and normal wind speed q of the ventilation equipment VF, and the position and normal wind speed q of the ventilation opening VP. In the examples of formulas (2) and (3) below, the boundary conditions are Neumann boundary conditions. Here, the position of the boundary Γ is determined based on the shape of the area Ω indicated by the information read by the shape reading unit 21. It is also used that the normal wind speed at the node in the boundary Γ at a position other than the ventilation equipment VF and the ventilation opening VP is q=0. In addition, in formula (3), N is the unit normal vector of the boundary Γ from the area Ω toward the outside. q - is the known normal wind speed.

[0028]

number

[0029]

number

[0030] The velocity potential u(x) that satisfies equation (1) at position x in the x-axis direction in the domain Ω is shown in equation (4). In equation (4), G(xy) is the fundamental solution of Laplace's equation. Also, dSy indicates integration with y along the boundary shape. In this example, y is the position in the y-axis direction perpendicular to the above-mentioned x-axis direction. In addition, in two dimensions, the fundamental solution of Laplace's equation can be expressed as G = -(1 / 2π)logr, where r is the distance between two points. In equations (3) and (4), the integral symbol with Γ at the bottom right indicates integration at the boundary Γ in the domain Ω.

[0031]

number

[0032] The velocity potential u shown in equation (4) Ω By taking the limit of (x) on the boundary Γ, the boundary integral equation shown in equation (5) is established on the boundary. In equation (5), u(x) is the velocity potential u on the boundary Γ.

[0033]

number

[0034] The tree structure generating unit 25 divides the boundary Γ of the region Ω with the outside into clusters of a binary tree to generate a cluster structure on the boundary Γ. Specifically, the tree structure generating unit 25 divides the boundary Γ so that nodes located spatially close to each other on the boundary Γ are included in the same cluster.

[0035] As an example, the tree structure generation unit 25 assigns a number that uniquely identifies the element of the node determined by the model generation unit 22. Then, the tree structure generation unit 25 divides the cluster CL1 including the element assigned a number equal to or less than 1 / 2 of the maximum number, and the cluster CL2 including the element assigned a number greater than 1 / 2 of the maximum number (first division). Further, the tree structure generation unit 25 similarly divides the cluster CL1 into clusters CL11 and CL12. At this time, the tree structure generation unit 25 similarly divides the cluster CL2 into clusters CL21 and CL22 (second division). Further, the tree structure generation unit 25 divides the cluster CL11 into clusters CL111 and CL112, and divides the cluster CL12 into clusters CL121 and CL122. At this time, tree structure generating unit 25 similarly divides cluster CL21 into cluster CL211 and cluster CL212, and divides cluster CL22 into cluster CL221 and cluster CL222 (third division).

[0036] In this way, the tree structure generating unit 25 repeats the division l times (l is an arbitrary integer) to divide the boundary Γ by 2l Divide the data into clusters.

[0037] The velocity potential u at each node shown in Fig. 2 is u Γi , the normal wind speed q is q i When discretized as , equation (5) is expressed as a linear combination of elements (velocity potential u, normal wind speed q) as shown in equation (6). i is an integer from 1 to n, and n is the total number of nodes. h in equation (6) ij and g ij are expressed by equations (7) and (8), respectively. Note that Γ is written in the lower right corner of equations (7) and (8). j The integral symbol with indicates the integral within one divided cluster.

[0038]

number

[0039]

number

[0040]

number

[0041] Furthermore, equation (6) can be expressed in matrix form as shown in equation (9).

[0042]

number

[0043] Furthermore, by summarizing the left side of equation (9), it is expressed as shown in equation (10).

[0044]

number

[0045] If the matrices shown in the determinant of equation (10) are matrix A and matrix B as shown in equations (11) and (12), respectively, equation (5) can be expressed as equation (13).

[0046]

number

[0047]

number

[0048] Au=Bq (13)

[0049] The interpolation matrix calculation unit 26 calculates the velocity potential u at the boundary Γ by a velocity potential calculation unit 28, which will be described in detail later. Γi Specifically, the interpolation matrix calculation unit 26 calculates an interpolation matrix by decomposing a matrix included in a determinant obtained by discretizing a boundary integral equation based on a boundary condition.

[0050] For example, the interpolation matrix calculation unit 26 decomposes the matrix A, which is multiplied with the velocity potential u, into the interpolation matrices U and V, which are shown in the equation (13) obtained by discretizing the boundary integral equation. A As shown in FIG. 4, the interpolation matrix calculation unit 26 calculates the n×n matrix A shown in equation (11) into an n×k interpolation matrix U and a k×k matrix R. A , and the k×n interpolation matrix V A where k is an integer smaller than n.

[0051] At this time, in order to perform low-rank approximation of the matrix A, the interpolation matrix calculation unit 26 calculates the interpolation matrix U and the interpolation matrix V based on the cluster structure at the boundary Γ generated by the tree structure generation unit 25. A Specifically, the interpolation matrix calculation unit 26 calculates the velocity potential u Γi In the calculation of the interpolation matrix U and the interpolation matrix V, all other nodes in the same cluster as the node are used. AAt this time, the interpolation matrix calculation unit 26 calculates the interpolation matrix U and the interpolation matrix V using some of the nodes included in a cluster different from the one node. A That is, the interpolation matrix calculation unit 26 calculates the interpolation matrix U and the interpolation matrix V without using the remaining nodes among the nodes included in a cluster different from the one node. A Calculate.

[0052] 5, the nodes are indicated by any one of dotted circles, black circles, and hatched circles. Here, for example, the interpolation matrix calculation unit 26 calculates the velocity potential u Γi In calculating the interpolation matrix U and the interpolation matrix V, all other nodes in the same cluster CL1 as the node in question (nodes indicated by black circles in CL1 in the example of FIG. 5) are used. A At this time, the interpolation matrix calculation unit 26 calculates the interpolation matrix U and the interpolation matrix V using some nodes included in a cluster CL2 different from the cluster CL1 (nodes included in CL2 indicated by black circles in the example of FIG. 5). A That is, the interpolation matrix calculation unit 26 calculates the interpolation matrix U and the interpolation matrix V without using some of the nodes included in a cluster CL2 different from the cluster CL1 (nodes included in CL2 indicated by hatched circles in the example of FIG. 5). A Calculate.

[0053] This creates a velocity potential u Γi In calculating the velocity potential u, all nodes that are close to the node (nodes indicated by black circles in CL1 in the example of FIG. 5) and some of the nodes that are far from the node (nodes indicated by black circles in CL2 in the example of FIG. 5) are used. Γi The remaining nodes (nodes indicated by hatched circles included in CL2 in the example of FIG. 5) that are far from the one node for which the velocity potential u is calculated are not used. Γi is the velocity potential u at one node ΓiIt is calculated by interpolating the integration results of the nodes used in the calculation of

[0054] In addition, the interpolation matrix calculation unit 26 decomposes the matrix B, which is multiplied with the normal wind speed q and is shown in the equation (13) obtained by discretizing the boundary integral equation, into the interpolation matrices U and V B The interpolation matrix calculation unit 26 calculates the n×n matrix B shown in equation (12) in the same manner as the above-mentioned matrix A, by dividing the n×k interpolation matrix U and the k×k matrix R. B , and the k×n interpolation matrix V B In addition, the interpolation matrix U may be the same as the interpolation matrix U obtained by the interpolation matrix calculation unit 26 by decomposing the matrix A. As a result, the equation (13) is expressed by the equation (14).

[0055] UR A V A u=UR B V B q (14)

[0056] Furthermore, the interpolation matrix calculation unit 26 calculates a k×k matrix R A Similarly, we obtain the k×k1 interpolation matrix U1 and the k1×k1 matrix R A1 , and the k1×k interpolation matrix V A1 Alternatively, the interpolation matrix calculation unit 26 may perform low-rank approximation by decomposing the matrix into k m ×k m Matrix R Am (m is an integer) to k m ×k m+1 Interpolation matrix U m+1 , k m+1 ×k m+1 Matrix R A(m+1) , and k m+1 ×k m Interpolation matrix V A(m+1) A low-rank approximation can be achieved by decomposing

[0057] Similarly, the interpolation matrix calculation unit 26 decomposes the k×k matrix R obtained by decomposing the matrix B in the low-rank approximation into a k×k1 interpolation matrix U1, a k1×k1 matrix R B1 , and the k1×k interpolation matrix V B1 By recursively repeating this process, the interpolation matrix calculation unit 26 can obtain k m ×k m Matrix R Bm k m ×k m+1 Interpolation matrix U m+1 , k m+1 ×k m+1 Matrix R B(m+1) , and k m+1 ×k m Interpolation matrix V B(m+1) A low-rank approximation can be achieved by decomposing

[0058] The high-speed matrix vector product execution unit 27 calculates the interpolation matrix U and the interpolation matrix V A Based on the matrix R A Similarly, the high-speed matrix-vector product execution unit 27 calculates the interpolation matrix U and the interpolation matrix V calculated by the interpolation matrix calculation unit 26. B Based on the matrix R B Calculate.

[0059] The high-speed matrix-vector product execution unit 27 calculates the right-hand side of the equation (14) expressed by the matrix that has been low-rank approximated by the interpolation matrix calculation unit 26, that is, the interpolation matrix U and the matrix R B , and the interpolation matrix V B Calculate the matrix-vector product of the product of and the matrix q.

[0060] The velocity potential calculation unit 28 calculates the velocity potential u Γi Calculate.

[0061] Specifically, the velocity potential calculation unit 28 calculates the interpolation matrix U and the matrix R, which are the left side of the equation (14) expressed by the matrix that is low-rank approximated by the interpolation matrix calculation unit 26. A , and the interpolation matrix V AThen, the velocity potential calculation unit 28 calculates the velocity potential u at each node of the boundary Γ using the calculation of the right side of the equation (14) calculated by the high-speed matrix-vector product execution unit 27. Γi (u1, u2, . . . , u n ) is calculated. The velocity potential calculation unit 28 also calculates the velocity potential u Γi (u1, u2, . . . , u n ) to be memorized.

[0062] <Configuration of calculation result storage unit> As shown in FIG. 1D, the calculation result storage unit 3 includes a model storage unit 31, a boundary condition storage unit 32, a tree structure storage unit 33, an interpolation matrix storage unit , and a velocity potential storage unit .

[0063] The model storage unit 31 stores the model generated by the model generation unit 22.

[0064] The boundary condition storage unit 32 stores the boundary conditions generated by the boundary condition generating unit 24 .

[0065] The tree structure storage unit 33 stores the cluster structure on the boundary Γ generated by the tree structure generating unit 25 .

[0066] The interpolation matrix storage unit 34 stores the interpolation matrix calculated by the interpolation matrix calculation unit 26 .

[0067] The velocity potential storage unit 35 stores the velocity potential u at each node of the boundary Γ calculated by the velocity potential calculation unit 28. Γi Remember.

[0068] <Configuration of interior point calculation part> The interior point calculation unit 4 calculates the velocity potential u ΓiBased on this, the interior point calculation unit 4 calculates the wind speed q' at the wind speed calculation point SCP (predetermined position) in the domain Ω. The interior point calculation unit 4 may calculate the wind speed q' by performing low-rank approximation. As shown in FIG. 1E, the interior point calculation unit 4 includes a model reading unit 41, a boundary condition reading unit 42, a tree structure reading unit 43, an interpolation matrix reading unit 44, a velocity potential reading unit 45, an interior point calculation location extraction unit 46, a wind speed calculation unit 47, and a determination unit 48.

[0069] The model reading unit 41 reads the model stored in the model storage unit 31 .

[0070] The boundary condition reading unit 42 reads the boundary conditions stored in the boundary condition storage unit 32 .

[0071] The tree structure reading unit 43 reads the cluster structure stored in the tree structure storage unit 33 .

[0072] The interpolation matrix reading unit 44 reads the interpolation matrix stored in the interpolation matrix storage unit 34 .

[0073] The velocity potential reading unit 45 reads the velocity potential u at each node of the boundary Γ stored in the velocity potential storage unit 35. Γi Load.

[0074] The interior point calculation point extraction unit 46 extracts wind speed calculation points SCP in the region Ω. For example, the interior point calculation point extraction unit 46 may extract at least one of a point where the cross-sectional area changes, a point where tunnels join, and a dead-end point in the region Ω as the wind speed calculation points SCP. The interior point calculation point extraction unit 46 may extract wind speed calculation points SCP input by an operator. Furthermore, the interior point calculation point extraction unit 46 may determine at least one of a point where the cross-sectional area changes, a point where tunnels join, and a dead-end point by any method based on the shape of the region Ω indicated by the information stored in the shape storage unit 11, and extract the determined point as the wind speed calculation point SCP.

[0075] The wind speed calculation unit 47 calculates the speed potential u Γi Based on this, the wind speed q' at the wind speed calculation point SCP within the region Ω is calculated.

[0076] Specifically, the wind speed calculation unit 47 calculates the velocity potential u Γi Based on this, the velocity potential u at each wind speed calculation point SCP Ω (u SCP1 , u SCP2 , , u SCPM ) is calculated. M is the number of wind speed calculation points SCP, and in the example shown in FIG. 2, M=3. Specifically, the wind speed calculation unit 47 calculates the velocity potential u Γi By substituting and calculating the velocity potential u Ω Calculate.

[0077] For example, the wind speed calculation unit 47 calculates the velocity potential u at the wind speed calculation point SCP in the domain Ω obtained from Equation (4) in a similar manner to the boundary value calculation unit 2 described above using the model, boundary conditions, cluster structure, and interpolation matrix read by the model read unit 41, boundary condition read unit 42, tree structure read unit 43, and interpolation matrix read unit 44, respectively. Ω The wind speed calculation unit 47 then calculates the equation represented by the low-rank approximation matrix to calculate the velocity potential u Ω At this time, the wind speed calculation unit 47 calculates the velocity potential u calculated by the velocity potential calculation unit 28 using the boundary conditions generated by the boundary condition generation unit 24. Γi Based on this, low-rank approximation is performed using the interpolation matrix calculated by the interpolation matrix calculation unit 26 to obtain the velocity potential u Ω may be calculated.

[0078] In addition, the wind speed calculation unit 47 calculates the velocity potential u Ω (u SCP1, u SCP2 , , u SCPM ) is directionally differentiated to obtain the wind speed q'(q' SCP1 , q' SCP2 , , q' SCPM ) is calculated.

[0079] The determination unit 48 determines whether the wind speed q' is equal to or lower than a reference wind speed. The reference wind speed is a predetermined wind speed, for example, a wind speed equal to or lower than the reference wind speed that is expected to require ventilation from a safety standpoint.

[0080] Furthermore, the determination unit 48 may output information indicating whether or not the wind speed q' is equal to or less than the reference wind speed. Furthermore, when the determination unit 48 determines that the wind speed q' at the wind speed calculation point SCP is equal to or less than the reference wind speed, the determination unit 48 may output information recommending the installation of an oxygen concentration sensor. In such a configuration, when the determination unit 48 determines that the wind speed q' at the wind speed calculation point SCP is greater than the reference wind speed, the determination unit 48 does not output information recommending the installation of an oxygen concentration sensor.

[0081] <Operation of the wind speed calculation device> Here, the operation of the wind speed calculation device 100 according to this embodiment will be described with reference to Fig. 6A and Fig. 6B. Fig. 6A and Fig. 6B are flowcharts showing an example of the operation of the wind speed calculation device 100 according to this embodiment. The operation of the wind speed calculation device 100 described with reference to Fig. 6A and Fig. 6B corresponds to an example of a wind speed calculation method of the wind speed calculation device 100 according to this embodiment.

[0082] As shown in Fig. 6A, first, in steps S11 to S18, the boundary value calculation unit 2 performs low-rank approximation based on the shape of the area Ω, the position and normal wind speed q of the ventilation equipment VF installed on the boundary Γ with the outside of the area Ω, and the position and normal wind speed q of the ventilation opening VP installed on the boundary Γ to calculate the velocity potential on the boundary Γ. The process of steps S11 to S18 will be described in detail below.

[0083] In step S11, the shape reading unit 21 reads information that is stored in the shape storage unit 11 and indicates the shape of the region Ω.

[0084] In step S12, the model generation unit 22 generates a model indicating nodes on the boundary Γ between the area Ω and the outside, based on the shape of the area Ω indicated by the information read by the shape reading unit 21. The model generation unit 22 also causes the model storage unit 31 to store the model.

[0085] In step S13, the ventilation information reading unit 23 reads information about the ventilation equipment VF stored in the ventilation equipment memory unit 12 and information about the ventilation opening VP stored in the ventilation opening memory unit 13.

[0086] In step S14, the boundary condition generating unit 24 generates boundary conditions for the governing equation. At this time, the boundary condition generating unit 24 generates the boundary conditions based on the position of the ventilation equipment VF included in the information on the ventilation equipment VF and the normal wind speed q at the position of the ventilation equipment VF, the position of the ventilation opening VP included in the information on the ventilation opening VP and the normal wind speed q at the position of the ventilation opening VP, and the position of the boundary Γ. In addition, the boundary condition generating unit 24 stores the boundary conditions in the boundary condition storage unit 32.

[0087] In step S15, the tree structure generating unit 25 divides the boundary Γ between the region Ω and the outside using a binary tree to generate a cluster structure in the boundary Γ. The tree structure generating unit 25 also causes the tree structure storage unit 33 to store the cluster structure.

[0088] In step S16, the interpolation matrix calculation unit 26 calculates the velocity potential u Γi At this time, the interpolation matrix calculation unit 26 performs low-rank approximation of the matrix A, which is multiplied with the matrix U and the matrix V, based on the cluster structure at the boundary Γ generated in step S15. AThe interpolation matrix calculation unit 26 calculates the interpolation matrix U and the interpolation matrix V A Memorize the following.

[0089] In step S17, the interpolation matrix calculation unit 26 performs low-rank approximation on the matrix B, which is multiplied with the normal wind speed q in the equation (13) obtained by discretizing the boundary integral equation. At this time, the interpolation matrix calculation unit 26 performs low-rank approximation on the matrix B based on the cluster structure on the boundary Γ generated in step S15 by using the interpolation matrix U and the interpolation matrix V B The interpolation matrix calculation unit 26 calculates the interpolation matrix U and the interpolation matrix V B Memorize the following.

[0090] In step S18, the velocity potential calculation unit 28 calculates the velocity potential u Γi Specifically, first, the high-speed matrix-vector product execution unit 27 calculates the interpolation matrix U and the interpolation matrix V A Based on the matrix R A In addition, the high-speed matrix vector product execution unit 27 calculates the interpolation matrix U and the interpolation matrix V B Based on the matrix R B Then, the high-speed matrix-vector product execution unit 27 calculates the interpolation matrix U and the matrix R on the right side of the equation (14) represented by the matrix that has been low-rank approximated by the interpolation matrix calculation unit 26. B , the interpolation matrix V B The high-speed matrix-vector product execution unit 27 calculates the matrix-vector product of the product of the matrix q and the interpolation matrix U and matrix R in the equation (14) represented by the matrix that has been low-rank approximated by the interpolation matrix calculation unit 26. A , the interpolation matrix V A Then, the velocity potential calculation unit 28 calculates the velocity potential u at each node of the boundary Γ by using the calculation of the right side of the above-mentioned equation (14). Γi (u1, u2, . . . , u n ) is calculated. The velocity potential calculation unit 28 also calculates the velocity potential uΓi (u1, u2, . . . , u n ) to be memorized.

[0091] As shown in FIG. 6B, in steps S19 to S26, the interior point calculation unit 4 calculates the velocity potential u Γi The wind speed q' at the wind speed calculation point SCP in the domain Ω is calculated based on the above equation (1). The processes from step S19 to step S26 will be described in detail below.

[0092] In step S19 , the model reading unit 41 reads the model stored in the model storage unit 31 .

[0093] In step S20, the boundary condition reading unit 42 reads the boundary conditions stored in the boundary condition storage unit 32.

[0094] In step S21, the tree structure reading unit 43 reads the cluster structure stored in the tree structure storage unit 33.

[0095] In step S22, the interpolation matrix reading unit 44 reads the interpolation matrix stored in the interpolation matrix storage unit .

[0096] In step S23, the velocity potential reading unit 45 reads the velocity potential u at the boundary Γ stored in the velocity potential storage unit 35. Γi (u1, u2, . . . , u n ) to load the

[0097] In step S24, the interior point calculation point extraction unit 46 extracts wind speed calculation points SCP within the domain Ω.

[0098] In step S25, the wind speed calculation unit 47 calculates the velocity potential u Ω At this time, the wind speed calculation unit 47 calculates the velocity potential u Ω By performing a low-rank approximation of the formula for calculating the velocity potential uΓi Based on the velocity potential u Ω Calculate.

[0099] In step S26, the wind speed calculation unit 47 calculates the velocity potential u Ω (u SCP1 , u SCP2 , , u SCPM ) based on the wind speed q' (q' SCP1 , q' SCP2 , , q' SCPM ) to calculate

[0100] Furthermore, in step S27, the determination unit 48 determines whether or not the wind speed q' at the wind speed calculation point SCP is equal to or lower than the reference wind speed.

[0101] <Calculation results> Next, the calculation time by the wind speed calculation device 100 according to this embodiment will be compared with that by the conventional technology to explain the calculation time. Also, the accuracy of the values ​​calculated by the wind speed calculation device 100 will be compared with the values ​​calculated by the conventional technology to explain the calculation time. Note that in the calculation by the conventional technology, the above-mentioned processing by the tree structure generation unit 25 and the interpolation matrix calculation unit 26 is not performed, and the velocity potential u is calculated using a matrix-vector product that is not approximated at low rank. Γi and the wind speed q' is calculated.

[0102] (calculation speed) Figure 7A shows the element (velocity potential u Γi ) for the number of nodes where the velocity potential u on the boundary Γ is unknown Γi 7A is a diagram showing the calculation time of the wind speed calculation device 100 according to the present embodiment and the conventional technology. As shown in FIG. 7A, the calculation time becomes longer as the number of nodes with unknown elements increases, both in the wind speed calculation device 100 according to the present embodiment and in the conventional technology. However, the calculation time by the wind speed calculation device 100 according to the present embodiment is shorter by about a power of 10 compared to the calculation time by the conventional technology. Furthermore, the difference between the calculation time by the wind speed calculation device 100 and the calculation time by the conventional technology becomes more noticeable as the number of nodes with unknown elements increases.

[0103] Figure 7B shows the velocity potential u at the wind speed calculation point SPC in the domain Ω versus the number of nodes with unknown elements. Ω 7B is a diagram showing the calculation time for the wind speed calculation points SPC. In the example shown in Fig. 7B, there are three wind speed calculation points SPC. As shown in Fig. 7B, the calculation time by the wind speed calculation device 100 of this embodiment is shorter than the calculation time by the conventional technology.

[0104] (calculation accuracy) The velocity potential u of each node on the boundary Γ calculated by the wind speed calculation device 100 of this embodiment Γi are approximately the same as the velocity potentials of the same nodes on the boundary Γ calculated by the prior art.

[0105] 2. Specifically, FIG. 8 shows the wind speed q' calculated by the wind speed calculation device 100 of this embodiment and the wind speed calculated by the conventional technology. As shown in FIG. 8, the wind speed q' calculated by the wind speed calculation device 100 of this embodiment is the same up to the first decimal place as the wind speed calculated by the conventional technology. Therefore, in practice, the wind speed calculation device 100 of this embodiment can calculate the wind speed with high accuracy, just like the conventional technology.

[0106] In the above, the boundary value calculation unit 2 performs low-rank approximation to calculate the velocity potential u Γi The interior point calculation unit 4 calculates the wind speed q' by performing a low-rank approximation, but this is not limited to the above. For example, the boundary value calculation unit 2 calculates the velocity potential u Γi Alternatively, the interior point calculation unit 4 may perform a low-rank approximation to calculate the wind speed q'.

[0107] As described above, according to this embodiment, the wind speed calculation device 100 calculates the velocity potential u at the boundary Γ based on the shape of the area Ω, the position and normal wind speed q of the ventilation equipment VF installed at the boundary Γ with the outside of the area Ω, and the position and normal wind speed q of the ventilation opening VP installed at the boundary Γ. Γi Then, the wind speed calculation device 100 calculates the velocity potential u Γi The wind speed calculation device 100 also performs low-rank approximation to calculate the wind speed q' at a given position in the region Ω. Γi or calculates the wind speed q' by performing a low-rank approximation. Therefore, the wind speed calculation device 100 can calculate the wind speed q' with high accuracy in a short time according to the complex conditions of the domain Ω.

[0108] <Program> The above-mentioned wind speed calculation device 100 can be realized by a computer 101. A program for functioning as the wind speed calculation device 100 may be provided. The program may be stored in a storage medium or provided through a network. FIG. 9 is a block diagram showing a schematic configuration of a computer 101 each functioning as the wind speed calculation device 100. Here, the computer 101 may be a general-purpose computer, a dedicated computer, a workstation, a PC (Personal Computer), an electronic notepad, or the like. The program instructions may be program code, code segments, or the like for performing necessary tasks.

[0109] 9, the computer 101 includes a processor 110, a read only memory (ROM) 120, a random access memory (RAM) 130, a storage 140, an input unit 150, a display unit 160, and a communication interface (I / F) 170. Each component is connected to each other so as to be able to communicate with each other via a bus 180. The processor 110 is specifically a central processing unit (CPU), a micro processing unit (MPU), a graphics processing unit (GPU), a digital signal processor (DSP), a system on a chip (SoC), or the like, and may be configured by a plurality of processors of the same type or different types.

[0110] The processor 110 controls each component and executes various arithmetic processing. That is, the processor 110 reads a program from the ROM 120 or the storage 140, and executes the program using the RAM 130 as a working area. The processor 110 controls each component and executes various arithmetic processing according to the program stored in the ROM 120 or the storage 140. In the above-described embodiment, the program according to the present disclosure is stored in the ROM 120 or the storage 140.

[0111] The program may be stored in a storage medium readable by the computer 101. By using such a storage medium, the program can be installed in the computer 101. Here, the storage medium in which the program is stored may be a non-transitory storage medium. The non-transitory storage medium is not particularly limited, and may be, for example, a CD-ROM, a DVD-ROM, or a USB (Universal Serial Bus) memory. In addition, the program may be in a form that is downloaded from an external device via a network.

[0112] The ROM 120 stores various programs and various data. The RAM 130 temporarily stores programs or data as a working area. The storage 140 is configured with an HDD (Hard Disk Drive) or an SSD (Solid State Drive) and stores various programs including an operating system and various data.

[0113] The input unit 150 includes one or more input interfaces that receive an input operation from a user and acquire information based on the user's operation. For example, the input unit 150 is a pointing device, a keyboard, a mouse, etc., but is not limited to these.

[0114] The display unit 160 includes one or more output interfaces that output information. For example, the display unit 160 is a display that outputs information as a video or a speaker that outputs information as a sound, but is not limited to these. Note that, if the display unit 160 is a touch panel type display, it also functions as the input unit 150.

[0115] The communication interface (I / F) 170 is an interface for communicating with an external device.

[0116] The following supplementary notes are further disclosed regarding the above embodiment.

[0117] (Additional note 1) A control unit is provided, The control unit is Calculating a velocity potential at the boundary based on the shape of an area within the structure, the position and normal wind speed of a ventilation device installed at the boundary with the outside of the area, and the position and normal wind speed of a ventilation opening installed at the boundary; Calculating a wind speed at a given location within the region based on the velocity potential; A wind speed calculation device in which the control unit performs a low-rank approximation to calculate the velocity potential, or the control unit performs a low-rank approximation to calculate the wind speed at the specified position. (Additional note 2) The control unit performs the low-rank approximation to calculate the wind speed at the specified position. (Additional note 3) The control unit generates a model indicating nodes at the boundary, generates boundary conditions according to a shape of the area, a position of the ventilation equipment and a wind speed in a normal direction, and a position of the ventilation opening and a wind speed in a normal direction, and performs the low-rank approximation by calculating an interpolation matrix obtained by decomposing a matrix included in a determinant obtained by discretizing a boundary integral equation based on the boundary conditions, to calculate the velocity potential. (Additional note 4) The control unit calculates the wind speed at the predetermined position by performing low-rank approximation using the interpolation matrix based on the boundary condition and the velocity potential. (Additional note 5) 6. The wind speed calculation device according to any one of appended claims 1 to 5, wherein the control unit and the boundary value calculation unit determine whether the wind speed at the predetermined position is equal to or lower than a reference wind speed. (Additional note 6) A boundary value calculation step of calculating a velocity potential at the boundary by performing a low-rank approximation based on the shape of the area inside the structure, the position and normal wind speed of a ventilation equipment installed at the boundary with the outside of the area, and the position and normal wind speed of a ventilation opening installed at the boundary; an interior point calculation step of calculating a wind speed at a predetermined position within the region based on the velocity potential; Including, A wind speed calculation method, wherein the boundary value calculation step comprises a step of calculating the velocity potential by performing a low-rank approximation, or the interior point calculation step comprises a step of calculating the wind speed at the specified position by performing a low-rank approximation. (Additional note 7) A non-transitory storage medium storing a program executable by a computer, the non-transitory storage medium storing the program causing the computer to function as the wind speed calculation device according to any one of claims 1 to 5.

[0118] All publications, patent applications, and standards mentioned in this specification are herein incorporated by reference to the same extent as if each individual publication, patent application, and standard was specifically and individually indicated to be incorporated by reference.

[0119] Although the above-mentioned embodiment has been described as a representative example, it is obvious to those skilled in the art that many modifications and substitutions can be made within the spirit and scope of the present disclosure. Therefore, the present invention should not be interpreted as being limited by the above-mentioned embodiment, and various modifications or changes can be made without departing from the scope of the claims. For example, it is possible to combine multiple building blocks shown in the configuration diagram of the embodiment into one, or to divide one building block. [Explanation of symbols]

[0120] 1 Condition memory section 2 Boundary value calculation section 3 Calculation result storage section 4 Interior point calculation section 11 Shape memory section 12 Ventilation equipment memory section 13 Ventilation vent memory section 21 Shape reading section 22 Model Generation Unit 23 Ventilation information reading section 24 Boundary condition generator 25 Tree structure generator 26 Interpolation matrix calculation unit, 27 High-speed matrix vector multiplication execution unit 28 Velocity potential calculation section 31 Model memory section 32 Boundary condition storage section 33 Tree structure memory section 34 Interpolation matrix storage unit 35 Velocity potential memory 41 Model Loading Section 42 Boundary condition reading section 43 Tree structure reading part 44 Interpolation matrix reading part 45 Velocity potential reading section 46 Extraction of interior point calculation points 47 Wind speed calculation section 48 Judgment section 100 Wind speed calculation device 101 Computer 110 Processor 120 ROM 130 RAM 140 Storage 150 Input section 160 Output section 170 Communication Interface 180 Bus

Claims

1. A boundary value calculation unit that calculates a velocity potential at the boundary based on the shape of an area within the structure, the position and normal wind speed of a ventilation equipment installed at the boundary with the outside of the area, and the position and normal wind speed of a ventilation opening installed at the boundary; an interior point calculation unit that calculates a wind speed at a predetermined position within the region based on the velocity potential; Equipped with A wind speed calculation device in which the boundary value calculation unit calculates the velocity potential by performing a low-rank approximation, or the interior point calculation unit calculates the wind speed at the specified position by performing a low-rank approximation.

2. The wind speed calculation device according to claim 1 , wherein the interior point calculation unit performs the low-rank approximation to calculate the wind speed at the predetermined position.

3. 3. The wind speed calculation device according to claim 1, wherein the boundary value calculation unit generates a model indicating nodes at the boundary, generates boundary conditions according to a shape of the region, a position of the ventilation equipment and a wind speed in a normal direction, and a position of the ventilation opening and a wind speed in a normal direction, and performs the low-rank approximation by calculating an interpolation matrix obtained by decomposing a matrix included in a determinant obtained by discretizing a boundary integral equation based on the boundary conditions, to calculate the velocity potential.

4. The wind speed calculation device according to claim 3 , wherein the interior point calculation unit calculates the wind speed at the predetermined position by performing low-rank approximation using the interpolation matrix based on the boundary condition and the velocity potential.

5. The wind speed calculation device according to claim 1 , wherein the boundary value calculation unit determines whether the wind speed at the predetermined position is equal to or lower than a reference wind speed.

6. A boundary value calculation step of calculating a velocity potential at the boundary by performing a low-rank approximation based on the shape of the area inside the structure, the position and normal wind speed of a ventilation equipment installed at the boundary with the outside of the area, and the position and normal wind speed of a ventilation opening installed at the boundary; an interior point calculation step of calculating a wind speed at a predetermined position within the region based on the velocity potential; Including, A wind speed calculation method, wherein the boundary value calculation step comprises a step of calculating the velocity potential by performing a low-rank approximation, or the interior point calculation step comprises a step of calculating the wind speed at the specified position by performing a low-rank approximation.

7. A program for causing a computer to function as the wind speed calculation device according to any one of claims 1 to 5.

Images