Spatial control system, spatial control method, and program

Abstract

The present invention provides a spatial control system, etc., that can suppress the decrease in the accuracy of predicting the spatial state of the target space. [Solution] The spatial control system 1 comprises a model calculation unit 406 that generates a reduced model of the target space 10 based on the results of fluid analysis of the target space 10 which is the object of spatial control; an assimilation unit 408 that predicts the spatial distribution of the target space 10 by performing assimilation processing using the reduced model generated by the model calculation unit 406 and environmental data obtained by sensing the target space 10; and an information processing unit 410 that searches for control candidates when spatially controlling the target space 10 based on the spatial distribution of the target space 10.

Description

[Technical Field]

[0001] This disclosure relates to a spatial control system, a spatial control method, and a program. [Background technology]

[0002] Conventionally, spatial control systems that predict the spatial conditions within a building equipped with air conditioning equipment are known. Patent Document 1 discloses a system that predicts the performance degradation of air conditioning equipment from a performance trend database that stores, in association with "partial load rate - cumulative operating time" (the cumulative operating time relative to the partial load rate of the air conditioning equipment over a predetermined period) and "partial load rate - COP (Coefficient of Performance) decline rate," and generates an operating pattern that optimizes long-term efficiency. [Prior art documents] [Patent Documents]

[0003] [Patent Document 1] Japanese Patent Publication No. 2005-203544 [Overview of the project] [Problems that the invention aims to solve]

[0004] Conventional systems can only predict spatial conditions from candidate operating patterns of air conditioning equipment stored in a performance trend database, which leads to a problem of reduced accuracy in predicting the spatial conditions of the target space subject to spatial control.

[0005] This disclosure provides a spatial control system, etc., that can suppress a decrease in the accuracy of predicting the spatial state of a target space. [Means for solving the problem]

[0006] A spatial control system according to one aspect of the present disclosure includes: a model calculation unit that generates a reduced model of a target space based on the results of a fluid analysis of the target space to be controlled; an assimilation unit that predicts the spatial distribution of the target space by performing assimilation processing using the reduced model generated by the model calculation unit and environmental data obtained by sensing the target space; and an information processing unit that searches for control candidates for spatial control of the target space based on the spatial distribution of the target space.

[0007] A spatial control method according to one aspect of the present disclosure includes the steps of: generating a reduced model of a target space based on the results of a fluid analysis of the target space to be controlled; predicting the spatial distribution of the target space by performing an assimilation process using the reduced model and environmental data obtained by sensing the target space; and searching for control candidates for spatial control of the target space based on the spatial distribution of the target space.

[0008] A program relating to one aspect of this disclosure is a program for causing a computer to execute the spatial control method described above.

[0009] Furthermore, the general or specific aspects of this disclosure may be implemented as a system, method, integrated circuit, computer program, or recording medium such as a computer-readable CD-ROM, or as any combination of a system, method, integrated circuit, computer program, and recording medium. [Effects of the Invention]

[0010] The spatial control system and the like of this disclosure can suppress a decrease in the prediction accuracy of spatial control of the target space. [Brief explanation of the drawing]

[0011] [Figure 1] This is a block diagram showing the configuration of the spatial control system in the embodiment. [Figure 2] This figure shows an example of a target space that is subject to spatial control. [Figure 3] This is a diagram showing an example of data set and input to a spatial state prediction device when performing fluid analysis. [Figure 4] This is a diagram conceptually showing a reduced model and a control reduced model after assimilation processing as a result of CFD analysis. [Figure 5] This is a diagram showing an example of information regarding the operation pattern of equipment. [Figure 6] This is a diagram showing an example of performance data of partial load ratio, COP, and power consumption. [Figure 7] This is a diagram showing another example of performance data of partial load ratio, COP, and power consumption. [Figure 8] This is a diagram showing an example of operating conditions derived by control optimization. [Figure 9] This is a diagram showing an example of operating conditions that can actually be operated on equipment. [Figure 10] This is a diagram showing an example of Pareto solutions of control candidates. [Figure 11] This is a diagram showing an example of a target space within a building. [Figure 12] This is a flowchart showing the preliminary preparation for space control in an embodiment. [Figure 13] This is a flowchart showing the space control method of an embodiment. Embodiments for Carrying Out the Invention

[0012] Hereinafter, embodiments and the like will be described with reference to the drawings. All of the embodiments and the like described below show comprehensive or specific examples. The numerical values, shapes, materials, components, arrangement positions and connection forms of the components, steps, order of steps, etc. shown in the following embodiments and the like are merely examples and are not intended to limit the present disclosure. In addition, among the components in the following embodiments and the like, the components not described in the independent claims are described as optional components.

[0013] Furthermore, each figure is a schematic diagram and not necessarily a strictly accurate representation. In addition, the same reference numerals are used for substantially identical components in each figure, and redundant explanations may be omitted or simplified. Also, even when the same object is shown in each figure, the scale may be changed for convenience.

[0014] (Embodiment) [Configuration of the spatial control system] The configuration of the spatial control system in this embodiment will be described with reference to the figures.

[0015] Figure 1 is a block diagram showing the configuration of the spatial control system 1 in an embodiment.

[0016] The spatial control system 1 is a system that predicts the state quantities of the space within a building and searches for control candidates for spatial control within the building based on these predictions.

[0017] The target space 10, which is the subject of spatial control, is a space partitioned by walls in a building such as a house, office, shop, public facility, entertainment facility, art museum, museum, factory, or warehouse. The state quantities of space are physical quantities that indicate the state of the space, such as the temperature distribution, humidity distribution, wind speed (including wind direction) distribution, gas concentration distribution, and PM2.5 (particulate matter) distribution of the space.

[0018] The spatial control system 1 comprises a spatial state prediction device 400, one or more sensors 200, and one or more devices 300. The spatial state prediction device 400, the sensors 200, and the devices 300 can communicate with each other via a network 100.

[0019] Figure 2 shows an example of a target space 10 that is subject to spatial control.

[0020] Sensor 200 is a device that detects state quantities and boundary conditions at a predetermined location in the target space 10. State quantities at a predetermined location include, for example, temperature, humidity, wind speed, gas concentration, and PM2.5 levels at that location. Boundary conditions at a predetermined location are physical quantities that indicate the external environment affecting the state quantities of the space, or the state of the boundary region between the space and the outside. External environment affecting the state quantities of the space includes, for example, outside temperature, exterior wall temperature, outside humidity, outside wind speed, outside gas concentration, outside PM2.5 levels, and solar radiation. The state of the boundary region between the space and the outside is, for example, the opening area (or opening angle) of a door or window installed in the building. Sensor 200 includes, for example, a thermometer, hygrometer, anemometer, gas concentration meter, PM2.5 measuring instrument, pyranometer, door sensor, and window sensor. Sensor 200 is installed in the target space 10 based on the results of the measurement position optimization by the measurement position optimization unit 411, which will be described later.

[0021] The detection information (sensing information) detected by the sensor 200 is transmitted to the spatial state prediction device 400 via the network 100.

[0022] The device 300 is a device that forms the environment of the target space 10, and is, for example, an air conditioner, a ventilation fan, an air purifier, a circulator, or a gas diffusion device. The device 300 is installed within the target space 10 of the building, or in the boundary area between the target space 10 and the outside. The device 300 also transmits operating information, including current operating conditions and past operating history, to the spatial state prediction device 400. The operating conditions of the device 300 include physical quantities such as the temperature, humidity, airflow rate, and airflow direction of the air sent out from the device 300. If the device 300 is an air conditioner, the operating information may include information on the set air temperature, discharge airflow rate, intake airflow rate, fan rotation speed, and power supply amount to the heat exchanger. If the device 300 is a gas diffusion device that releases diffusing substances such as hypochlorous acid, disinfectant ions, or fragrances, the operating information may include information on the released gas concentration, release amount, and release direction of the diffusing substance.

[0023] As shown in Figure 1, the spatial state prediction device 400 comprises a fluid calculation information acquisition unit 402, an indoor information acquisition unit 403, a control information acquisition unit 404, an information processing unit 410, a model calculation unit 406, an assimilation unit 408, and a storage unit 416. The information processing unit 410 includes a measurement position optimization unit 411, a partial load rate acquisition unit 412, and a COP acquisition unit 414. The fluid calculation information acquisition unit 402, the indoor information acquisition unit 403, and the control information acquisition unit 404 may also be included in the information processing unit 410.

[0024] The fluid calculation information acquisition unit 402 acquires the simulation results of a fluid analysis performed using the structural information of the target space 10 and the control capability information of the equipment 300. In this example, the fluid analysis is a CFD (Computational Fluid Dynamics) analysis.

[0025] The structural information of the target space 10 includes information about the shape, size, and arrangement of the target space 10, as well as information about objects placed within the target space 10, such as desks and partitions, and the positions of those objects. Information about the shape of the target space 10 is, for example, data obtained by 3D modeling the target space 10 and then converting it into a point cloud using the finite volume method. The structural information of the target space 10 is input in advance into the fluid calculation information acquisition unit 402 by the user of the target space 10. Alternatively, the structural information of the target space 10 may be obtained by imaging the target space 10 using a camera provided on the device 300.

[0026] The control capability information for equipment 300 indicates the spatial control capability of equipment 300. For example, if equipment 300 is an air conditioning unit, the control capability information for equipment 300 includes information such as the airflow rate, airflow direction, and outlet temperature of the air conditioning unit. The control capability information for equipment 300 is pre-input into the fluid calculation information acquisition unit 402 by a user or other person using the target space 10.

[0027] Figure 3 shows an example of data that is set and input to the spatial state prediction device 400 when performing fluid analysis.

[0028] Figure 3 shows information regarding the settings of the model used for fluid analysis, specifically the turbulence model and steady / transient information. Figure 3 also shows an example of structural information for the target space 10, including information on the computational domain / mesh size, wall boundary conditions, and thermal boundary conditions. Furthermore, Figure 3 shows flow boundary data, an example of control capability information for equipment 300. This figure includes flow boundary data such as the airflow rate at the exhaust ports of two air conditioners A1 and A2, the intake volume and intake temperature at the intake ports, the outlet temperature, airflow rate, and airflow direction for each air conditioner A1 and A2, and information on surface pressure and inflow temperature at two locations in the gaps (clearances) of the target space 10.

[0029] The spatial state prediction device 400 performs fluid analysis based on structural information of the target space 10 and control capability information of the equipment 300, and outputs the simulation results of the fluid analysis.

[0030] The model calculation unit 406 generates a reduced model of the target space 10 based on the results of the fluid analysis of the target space 10. For example, the model calculation unit 406 generates a reduced model using proper orthogonal decomposition (POD) from the simulation results of the fluid analysis acquired by the fluid calculation information acquisition unit 402. Proper orthogonal decomposition is a decomposition method that extracts low-dimensional components from given multidimensional data.

[0031] Figure 4 conceptually shows the reduced model and controlled reduced model after assimilation treatment, based on the results of CFD analysis.

[0032] The procedure for generating the reduced model is described below. First, the physical quantities (or state variables) of the target space 10 obtained from the CFD analysis are converted into one-dimensional information arranged in grid point order, and a matrix is ​​created with the number of data points. Then, the mean difference of the created matrix is ​​converted into a square matrix, modes consisting of eigenvalues ​​and eigenvectors are obtained, and the fluid field of the target space 10 is reconstructed by assigning POD coefficients to the obtained modes.

[0033] In this example, a reduced model of 10 modes, shown in Figure 4(b), is generated from the results of CFD analysis for 100 cases, as shown in Figure 4(a).

[0034] In reality, the operating conditions include 5 levels for the air conditioner's outlet temperature, 2 levels for the airflow, and 2 levels for the exhaust angle. There is also a case where the air conditioner is turned off. These combinations exist for each of the two air conditioners, and furthermore, there are 2 levels for the temperature of the walls and drafts as boundary conditions. Executing all possible combinations results in (5 × 2 × 2 + 1). 2 Since ×2 = 882, we would have to analyze 882 cases, which would increase the computation time required to predict the spatial state. Therefore, we extract 100 cases from the 882 cases at arbitrary intervals and replace the analysis, which includes the other 782 cases between those 100 cases, with a 10-mode reduced model.

[0035] A reduced model is a model that can maintain the essential behavior in spatial control of the target space 10 while reducing the model's dimensionality. By using a reduced model, it is possible to reduce the analysis time and data volume of spatial control.

[0036] The assimilation unit 408 predicts the spatial distribution of the target space 10 by performing assimilation processing using the reduced model generated by the model calculation unit 406 and environmental data obtained by sensing the target space 10. For example, the assimilation unit 408 uses the reduced model generated by the model calculation unit 406 and the environmental data of the target space 10 acquired by the indoor information acquisition unit 403 to determine appropriate initial values, boundary values, and parameters so that phenomena occurring in the spatial distribution of the target space 10 can be reproduced. As a result, the assimilation unit 408 generates a reduced model after assimilation processing.

[0037] The measurement position optimization unit 411 searches for the optimal position of the sensor 200 in order to predict the spatial distribution by assimilation processing with high accuracy. However, the locations where the sensor 200 can be installed in the target space 10 are constrained by the fact that they must be in places that do not interfere with human activity and are free of furniture. Therefore, using the reduced model generated by the model calculation unit 406, the sensor sensitivity when the sensor 200 is installed at each coordinate is quantified, and while understanding the spatial state, the trade-off relationship between the number of sensors 200 and the installation location of the sensors 200 is obtained. This optimizes the number of sensors 200 and the installation location of the sensors 200 in the target space 10. An example of an optimization algorithm used is the multi-objective genetic algorithm NSGA-2 (Elitist Non-dominated Sorting Genetic Algorithm).

[0038] The aforementioned assimilation unit 408 acquires the above-mentioned environmental data by changing the position of the sensor 200 installed in the target space 10, and predicts the spatial distribution before and after the change by performing assimilation processing using the environmental data before and after the change in the position of the sensor 200. The information processing unit 410 determines the position of the sensor 200 to an appropriate position based on the spatial distribution before and after the change.

[0039] The control information acquisition unit 404 acquires the control conditions of each simulation result acquired by the fluid calculation information acquisition unit 402.

[0040] The information processing unit 410 generates a control reduction model by performing machine learning using the reduced model after assimilation processing and the control conditions from the simulation results acquired by the control information acquisition unit 404 (see Figure 4(c)). The control reduction model is a model that predicts the spatial state using the control conditions as input values. The information processing unit 410 generates a prediction model of the POD coefficients using the control conditions as input by learning the POD coefficients of the physical quantities (state variables) of the target space 10 in the reduction model and the control conditions from the simulation results using a Gaussian process distribution. Hereafter, the control reduction model may be referred to as the trained reduction model.

[0041] In this way, the information processing unit 410 takes the control conditions in the fluid analysis as input values ​​and generates a trained reduced model by training the reduced model after the assimilation process. Furthermore, the information processing unit 410 predicts the spatial distribution of the target space 10 by performing an assimilation process using the trained reduced model and environmental data, and searches for control candidates for spatially controlling the target space 10 based on this spatial distribution. Searching for control candidates means extracting an operating pattern that is appropriate for spatial control from among various operating patterns. The information processing unit 410 searches for control candidates so that the target space 10 approaches the target spatial distribution.

[0042] The memory unit 416 stores the trained reduced model. The memory unit 416 also stores information regarding physical quantities (state variables) such as temperature, and spatial control (such as air quality control), as desired by the user. The memory unit 416 may also store information regarding the target environmental state and spatial distribution of the target space 10.

[0043] The following describes an example in which the information processing unit 410 acquires information on the partial load factor, power consumption, and COP of the device 300, and searches for control candidates based on this information.

[0044] The indoor information acquisition unit 403 acquires environmental data of the target space 10 from the sensor 200 whose position has been determined by the measurement position optimization unit 411. The indoor information acquisition unit 403 also acquires the operating patterns of the equipment 300 that performs spatial control (such as air quality control) of the target space 10.

[0045] Figure 5 shows an example of information regarding the operating pattern of equipment 300.

[0046] Figure 5 shows the set temperature and set airflow for air conditioners A1 and A2 in chronological order.

[0047] The indoor information acquisition unit 403 acquires information regarding the power consumption of the equipment 300 based on the operating pattern of the equipment 300.

[0048] The power consumption is calculated by the following (Equation 1).

[0049] Power consumption = W fan + W const ···(Equation 1)

[0050] W fan : Heat quantity required for the rotation of the fan, W const : Heat quantity required to maintain the room temperature

[0051] In addition, W fan is derived from the relational expression between the set air volume in the blowing state and the measured current value. Since the suction temperature and the blowing temperature in the steady state are constant, W const is calculated by the following (Equation 2).

[0052] W const = J(T out - T inlet )···(Equation 2)

[0053] T out : Blowing temperature of the air conditioner, T inlet : Suction temperature of the air conditioner

[0054] J can be calculated by, for example, the regression equation between the measured current value per second and (T out - T inlet ).

[0055] Information regarding the power consumption is output to the information processing unit 410.

[0056] The partial load ratio acquisition unit 412 acquires the partial load ratio of the device 300. The partial load ratio of the device 300 is calculated by the following (Equation 3).

[0057] Partial load ratio = (Output of the air conditioning device) / (Rated output of the air conditioning device)···(Equation 3)

[0058] The output of the air conditioning equipment is calculated from the supply airflow rate and the difference between the intake and discharge enthalpies. For example, the partial load factor acquisition unit 412 calculates the output of the air conditioning equipment in the current spatial distribution by acquiring the intake and discharge enthalpies from the spatial distribution predicted by the assimilation unit 408.

[0059] The amount of heat output from the air conditioning equipment is calculated using the following formula (Equation 4).

[0060] Heat quantity = (Specific heat of air × Air density × Supply air volume [CMH] × Intake / Outtake enthalpy difference [kJ / kg]) / 3600 [(kJ / h) / kW] ... (Equation 4)

[0061] The COP acquisition unit 414 derives the COP (Coefficient of Performance) of the equipment 300. COP is a coefficient used to check the energy efficiency of heating and cooling equipment, also known as the coefficient of performance (or operating coefficient). COP is calculated using the following equation (Equation 5).

[0062] COP = (Output of air conditioning equipment) / (Power consumption of air conditioning equipment) ... (Equation 5)

[0063] The memory unit 416 stores information regarding power consumption acquired from the indoor information acquisition unit 403, the partial load rate acquired from the partial load rate acquisition unit 412, and the COP acquired from the COP acquisition unit 414.

[0064] Furthermore, calculations related to power consumption, partial load factor, and COP do not necessarily have to be performed inside the spatial state prediction device 400, but may be performed on an external computer. The indoor information acquisition unit 403 may acquire information regarding the calculation results of power consumption from an external computer. The partial load factor acquisition unit 412 may acquire information regarding the calculation results of the partial load factor from an external computer. The COP acquisition unit 414 may acquire information regarding the calculation results of COP from an external computer.

[0065] The information processing unit 410 acquires information on the partial load ratio, power consumption, and COP (Coefficient of Performance) of the spatial control device 300, and searches for control candidates using at least one of power consumption and COP as evaluation indicators. A lower power consumption is desirable, and a higher COP is desirable.

[0066] Figure 6 shows an example of actual data for partial load factor, COP, and power consumption. Figure 7 shows another example of actual data for partial load factor, COP, and power consumption.

[0067] Figure 6 shows the actual data for air conditioner A1, and Figure 7 shows the actual data for air conditioner A2. In both Figures 6 and 7, the horizontal axis represents the partial load factor, the left vertical axis represents the COP, and the right vertical axis represents power consumption. The scale of the right vertical axis is set so that the actual power consumption data is generally smaller than the actual COP data.

[0068] As shown in Figures 6 and 7, the COP increases as the partial load ratio increases, but conversely decreases if the partial load ratio becomes too high. Also, power consumption remains similar within a predetermined range after the partial load ratio increases slightly, but increases accordingly if the partial load ratio becomes too high.

[0069] The information processing unit 410 searches for control candidates based on the actual data shown in Figures 6 and 7. For example, the information processing unit 410 defines the partial load ratio at which the difference between COP and power consumption is maximized as the optimal partial load ratio (see Figures 6 and 7), and identifies the operating pattern that achieves this partial load ratio as a control candidate.

[0070] Furthermore, the information processing unit 410 may derive the power consumption when the COP is at its maximum, and if this power consumption is within an acceptable range, it may consider the partial load ratio corresponding to the maximum COP as the optimal partial load ratio and select an operating pattern to achieve this partial load ratio as a control candidate.

[0071] In this way, the information processing unit 410 derives the partial load ratio at which the difference between power consumption and COP is maximized, and searches for control candidates that can maintain that partial load ratio.

[0072] The information processing unit 410 may also derive a partial load ratio when the difference between power consumption and COP is greater than or equal to a predetermined threshold, and the change in the difference between power consumption and COP in response to a change in the partial load ratio is within a predetermined range, and search for a control candidate that can maintain that partial load ratio.

[0073] Furthermore, when searching for control candidates, the information processing unit 410 obtains a spatial distribution by reconstructing the space using a multi-objective genetic algorithm from the predicted POD coefficients, and optimizes the control conditions when operating the equipment 300.

[0074] Figure 8 shows an example of operating conditions derived through control optimization. Figure 9 shows an example of operating conditions that can actually be operated with the device 300.

[0075] The operating conditions obtained through control optimization are selected from continuous values, whereas the operating conditions that can actually be operated are predetermined by the specifications of the equipment 300. Therefore, it is necessary to convert the operating conditions derived through control optimization into operating conditions that can actually be operated to form an operating pattern. Control conditions 1 and 2 shown in Figure 9 are control conditions obtained by converting the operating conditions derived through control optimization into operating conditions that can actually be operated.

[0076] Figure 10 shows an example of a Pareto solution for a control candidate.

[0077] In Figure 10, the horizontal axis represents comfort, and the vertical axis represents power consumption. Comfort refers to, for example, the percentage of total operating time during which the target spatial distribution can be achieved.

[0078] Control condition 1 shows the comfort and power consumption results when two air conditioners are running, and control condition 2 shows the comfort and power consumption results when one air conditioner is running. In this figure, control optimization has extracted two control conditions that have equivalent comfort indicators but significantly different power consumption. The information processing unit 410 lists these two control conditions as control candidates and outputs them.

[0079] For example, the spatial control system 1 may display the listed control candidates on an information terminal owned by the user (e.g., a smartphone or tablet). Alternatively, the spatial control system 1 may operate the equipment 300 using the control candidate with the highest priority among the listed control candidates.

[0080] The spatial control system 1 of this embodiment includes a model calculation unit 406 that generates a reduced model of the target space 10 based on the results of fluid analysis of the target space 10 which is the target of spatial control; an assimilation unit 408 that predicts the spatial distribution of the target space 10 by performing assimilation processing using the reduced model generated by the model calculation unit 406 and environmental data obtained by sensing the target space 10; and an information processing unit 410 that searches for control candidates when spatially controlling the target space 10 based on the spatial distribution of the target space 10.

[0081] In this way, by generating a reduced model of the target space 10, performing assimilation processing using the reduced model and environmental data, and predicting the spatial distribution of the target space 10, it is possible to suppress a decrease in the accuracy of predicting the spatial state of the target space 10. Furthermore, by searching for control candidates when spatially controlling the target space 10 based on the spatial distribution of the target space 10 described above, it is possible to appropriately search for control candidates.

[0082] [Spatial control method] A spatial control method in this embodiment will be described.

[0083] Figure 11 shows an example of a target space 10 within a building.

[0084] Figure 11 shows a view of the target space 10 from above. In Figure 11, a thermometer, a hygrometer, and an anemometer are installed in the target space 10 as sensors 200 for detecting state quantities within the target space 10. In addition, air conditioners A1 and A2, and multiple ventilation fans are installed in the target space 10 as equipment 300 that create the environment of the target space 10.

[0085] The structural information of the target space 10 includes, for example, room layout information and information on the position and shape of furniture, while the control capability information of the equipment 300 includes, for example, operational information such as airflow, air direction, and outlet temperature of air conditioning equipment (air conditioners A1 and A2). The spatial state prediction device 400 uses this information to perform a fluid analysis simulation.

[0086] Figure 12 is a flowchart showing the preliminary preparations for spatial control in the embodiment.

[0087] Figure 12 shows the necessary preparatory steps before performing spatial control. The simulation will be conducted under multiple control conditions.

[0088] First, the spatial state prediction device 400 sets the control range for the control factors (step S100). For example, the spatial state prediction device 400 sets the operating limit of the equipment 300 as the control range for each calculation condition and sets an arbitrary numerical interval. The reason for setting an arbitrary numerical interval is that running the simulation for all control patterns would result in an enormous amount of computation.

[0089] The spatial state prediction device 400 then determines the control pattern to be calculated (step S102). For example, the spatial state prediction device 400 performs random sampling on a Latin superlattice to determine the control pattern to be calculated.

[0090] The spatial state prediction device 400 executes a simulation using the control pattern determined in step S102 and obtains the calculation results (step S104).

[0091] The spatial state prediction device 400 generates a reduced model using intrinsic orthogonal decomposition based on the physical quantities for each operating condition obtained from the calculation results in step S104 (step S106).

[0092] Furthermore, the spatial state prediction device 400 optimizes the position of the sensors 200 in the target space 10 (step S108). For example, the spatial state prediction device 400 quantifies the sensor sensitivity when the sensors 200 are installed at each coordinate in the target space 10 from the reduced model generated in step S106, and determines the number and position of the sensors 200 by obtaining the trade-off relationship between the number of sensors and the understanding of the airflow state. Note that step S108 can be omitted if the sensor sensitivity is already sufficient.

[0093] The spatial state prediction device 400 performs data assimilation processing during commissioning and predicts the spatial state of the target space 10 (step S110). For example, in order to obtain the optimal partial load rate, the spatial state prediction device 400 performs commissioning under different control conditions and different ambient temperature conditions, and obtains information on the state quantities of the target space 10 from the sensor 200 whose position was determined in step S108. The state quantities are at least one of temperature, humidity, airflow rate, and wind direction. The spatial state prediction device 400 performs data assimilation processing using the state quantities obtained in step S110 and the reduced model generated in step S106 to predict the spatial state under each control condition.

[0094] Next, the spatial state prediction device 400 generates a control reduction model (step S112). For example, the spatial state prediction device 400 uses the reduction model generated in step S106 and the control conditions obtained in the calculation results in step S104 to generate a control reduction model, which is a prediction model of the POD coefficient with the control conditions as input values. The control reduction model is an example of a trained reduction model.

[0095] Furthermore, the spatial state prediction device 400 sets constraints on the partial load ratio (step S114). For example, the spatial state prediction device 400 uses the spatial distribution for each predicted control condition to plot the actual data of the partial load ratio and COP of the equipment 300, and the actual data of the partial load ratio and power consumption of the equipment 300, and extracts the optimal partial load ratio. The derived optimal partial load ratio is imposed as a constraint when searching for control candidates. Note that step S114 may be executed simultaneously with step S112.

[0096] Figure 13 is a flowchart showing a spatial control method in an embodiment.

[0097] Figure 13 shows the steps performed by the spatial control system 1.

[0098] The spatial control system 1 acquires control information for the current equipment 300 in the target space 10 (step S200). The control information includes control conditions as shown in Figure 5, as well as time-series data indicating the operating time of each control pattern.

[0099] The spatial control system 1 acquires state variables of the target space 10 (step S202). Specifically, the spatial control system 1 acquires environmental data from the sensor 200 whose position was determined in step S108 of Figure 12 via the network 100.

[0100] The spatial control system 1 predicts the spatial state of the target space 10 (step S204). For example, the spatial control system 1 takes the control information acquired in step S200 as input values ​​and predicts the spatial distribution using the control reduction model generated in step S112 in Figure 12.

[0101] Furthermore, the spatial control system 1 obtains the partial load factor and COP of the equipment 300 using the predicted spatial distribution (step S206). The spatial control system 1 also obtains the power consumption of the equipment 300.

[0102] The spatial control system 1 searches for control candidates based on the partial load factor, COP, and power consumption obtained in step S206 (step S208).

[0103] The spatial control system 1 outputs information regarding the extracted control candidates (step S210). For example, the spatial control system 1 may select the optimal partial load rate operating pattern as a control candidate and display the control candidate on the user's information terminal. For example, the spatial control system 1 may operate and control the equipment 300 using the optimal partial load rate operating pattern.

[0104] By performing the steps described above, it becomes possible to search for control candidates that balance comfort and energy efficiency.

[0105] (summary) Examples of spatial control systems, etc., relating to one aspect of this disclosure are given below.

[0106] The spatial control system 1 of Example 1 includes a model calculation unit 406 that generates a reduced model of the target space 10 based on the results of fluid analysis of the target space 10 which is the object of spatial control; an assimilation unit 408 that predicts the spatial distribution of the target space 10 by performing assimilation processing using the reduced model generated by the model calculation unit 406 and environmental data obtained by sensing the target space 10; and an information processing unit 410 that searches for control candidates when spatially controlling the target space 10 based on the spatial distribution of the target space 10.

[0107] In this way, by performing assimilation processing using a reduced model of the target space 10 and environmental data, and predicting the spatial distribution of the target space 10, it is possible to suppress a decrease in the accuracy of predicting the spatial state of the target space 10. Furthermore, by searching for control candidates when spatially controlling the target space 10 based on the spatial distribution of the target space 10 described above, it is possible to appropriately search for control candidates.

[0108] The spatial control system 1 in Example 2 is the spatial control system described in Example 1, wherein the information processing unit 410 takes control conditions in fluid analysis as input values, generates a trained reduced model by training the reduced model after assimilation processing, predicts the spatial distribution of the target space 10 by performing assimilation processing using the trained reduced model and environmental data, and searches for control candidates based on the spatial distribution.

[0109] In this way, by performing assimilation processing using a trained reduction model and environmental data, and predicting the spatial distribution of the target space 10, it is possible to suppress a decrease in the accuracy of predicting the spatial state of the target space 10.

[0110] The spatial control system 1 in Example 3 is the spatial control system described in Example 1, and the information processing unit 410 may search for control candidates so that the target space 10 approaches the target spatial distribution.

[0111] In this way, by searching for control candidates in a manner that approaches the target spatial distribution, the search for control candidates can be performed appropriately.

[0112] The spatial control system 1 of Example 4 is a spatial control system described in any of Examples 1 to 3, wherein the assimilation unit 408 acquires environmental data by changing the position of a sensor 200 provided in the target space 10, and predicts the spatial distribution before and after the change by performing assimilation processing using the environmental data before and after the change of the position of the sensor 200, and the information processing unit 410 determines the position of the sensor 200 based on the spatial distribution before and after the change.

[0113] According to this, the sensor 200 can be placed in an appropriate position within the target space 10. This makes it possible to suppress a decrease in the accuracy of predicting the spatial state of the target space 10.

[0114] The spatial control system 1 in Example 5 is the spatial control system described in Example 2, wherein the information processing unit 410 may acquire information on the partial load factor, power consumption, and COP of the equipment 300 that performs spatial control, and search for control candidates using at least one of the power consumption and COP as an evaluation index.

[0115] In this way, by using power consumption and at least one of COP as evaluation indicators to search for control candidates, the search for control candidates can be carried out appropriately.

[0116] The spatial control system 1 in Example 6 is the spatial control system described in Example 5, and the information processing unit 410 may derive the partial load ratio when the difference between power consumption and COP is maximized, and search for control candidates that can maintain said partial load ratio.

[0117] According to this, for example, information on the optimal partial load ratio can be obtained, and control candidates can be appropriately searched for.

[0118] The spatial control system 1 in Example 7 is the spatial control system described in Example 5, and the information processing unit 410 may derive a partial load ratio when the difference between power consumption and COP is greater than or equal to a predetermined threshold, and the change in the difference between power consumption and COP in response to a change in the partial load ratio is within a predetermined range, and search for a control candidate that can maintain the said partial load ratio.

[0119] This allows us to obtain information on partial loading rates that are resilient to disturbances and to appropriately search for control candidates.

[0120] The spatial control method of Example 8 includes the steps of: generating a reduced model of the target space 10 based on the results of a fluid analysis of the target space 10, which is the object of spatial control; predicting the spatial distribution of the target space 10 by performing assimilation processing using the reduced model and environmental data obtained by sensing the target space 10; and searching for control candidates for spatial control of the target space 10 based on the spatial distribution of the target space 10.

[0121] In this way, by performing assimilation processing using a reduced model of the target space 10 and environmental data, and predicting the spatial distribution of the target space 10, it is possible to suppress a decrease in the accuracy of predicting the spatial state of the target space 10. Furthermore, by searching for control candidates when spatially controlling the target space 10 based on the spatial distribution of the target space 10 described above, it is possible to appropriately search for control candidates.

[0122] The program in Example 9 is a program that causes a computer to execute the spatial control method described in Example 8.

[0123] This makes it possible to realize a spatial control method that can suppress a decrease in the accuracy of predicting the spatial state of the target space 10.

[0124] (Other embodiments) The spatial control system and the like in this disclosure have been described above based on embodiments, but this disclosure is not limited to these embodiments. As long as they do not deviate from the spirit of this disclosure, various modifications to the embodiments that a person skilled in the art could conceive of, and other forms constructed by combining some of the components of the embodiments, are also included within the scope of this disclosure.

[0125] The above example shows a configuration in which the spatial state prediction device 400 and the equipment 300 are connected via a communication network, but this is not the only configuration. For example, the spatial state prediction device 400 may be located inside the equipment 300. In this case, the sensor 200 may also be located inside the equipment 300 and connected to the spatial state prediction device 400 from within the equipment 300.

[0126] The spatial state prediction device in the above embodiment is implemented as a hardware configuration consisting of a non-volatile memory where the program is stored, a volatile memory which is a temporary storage area for executing the program, input / output ports, a communication interface, and a processor that executes the program. Each component of the spatial state prediction device is implemented by a processor that executes the program stored in memory. The spatial state prediction device may be implemented by a stationary PC (Personal Computer), a mobile terminal such as a smartphone or tablet, a dedicated computer, a server (for example, a cloud server), or a combination thereof.

[0127] Furthermore, each component may be implemented by being composed of dedicated hardware or by executing a software program suitable for each component. Each component may also be implemented by a program execution unit such as a CPU or processor reading and executing a software program recorded on a recording medium such as a hard disk or semiconductor memory.

[0128] Furthermore, the order in which each step in the flowchart is performed is illustrative for the purpose of specifically illustrating this disclosure, and may be in a different order. Also, some of the above steps may be performed simultaneously (in parallel) with other steps, and some of the above steps may not be performed.

[0129] Furthermore, the division of functional blocks in the block diagram is just one example; multiple functional blocks can be implemented as a single functional block, a single functional block can be divided into multiple parts, or some functions can be moved to other functional blocks. In addition, the functions of multiple functional blocks with similar functions can be processed in parallel or time-sharing by a single piece of hardware or software.

[0130] Furthermore, the spatial state prediction device according to the above embodiment may be implemented as a single device or as a plurality of devices. When the spatial state prediction device is implemented as a plurality of devices, the individual components of the spatial state prediction device may be distributed among the plurality of devices in any manner. When the spatial state prediction device is implemented as a plurality of devices, the communication method between the plurality of devices is not particularly limited and may be wireless communication or wired communication. In addition, wireless communication and wired communication may be combined between the devices.

[0131] Furthermore, each component described in the above embodiment may be implemented as software, or typically as an integrated circuit (LSI). These may be individually integrated onto a single chip, or some or all of them may be integrated onto a single chip. Here, we refer to them as LSIs, but depending on the degree of integration, they may also be called ICs, system LSIs, super LSIs, or ultra LSIs. Moreover, the method of integrated circuit implementation is not limited to LSIs; it may also be implemented using dedicated circuits (general-purpose circuits that execute dedicated programs) or general-purpose processors. After LSI manufacturing, a programmable FPGA (Field Programmable Gate Array) or a reconfigurable processor that allows for the reconfiguration of the connections or settings of circuit cells inside the LSI may be used. Furthermore, if an integrated circuit implementation technology that replaces LSIs emerges due to advances in semiconductor technology or other derived technologies, it is naturally possible to integrate the components using that technology.

[0132] A system LSI is a highly functional LSI manufactured by integrating multiple processing units onto a single chip. Specifically, it is a computer system consisting of a microprocessor, ROM (Read Only Memory), RAM (Random Access Memory), and other components. The ROM stores the computer program. The system LSI achieves its function by operating according to the computer program, with the microprocessor performing its operations.

[0133] Furthermore, one aspect of this disclosure may be a computer program that causes a computer to perform each characteristic step included in the method of using a battery pack.

[0134] Furthermore, for example, the program may be a program to be executed by a computer. Also, in one aspect of this disclosure, such a program may be recorded on a computer-readable non-temporary recording medium. For example, such a program may be recorded on a recording medium and distributed or made available. For example, by installing the distributed program on a device having another processor and having that processor execute the program, it becomes possible to have that device perform the above-mentioned processes. [Industrial applicability]

[0135] This disclosure can be applied to a spatial control system that can suppress a decrease in the predictive accuracy of spatial control of a target space. [Explanation of Symbols]

[0136] 1. Spatial control system 10 Target space 100 Networks 200 sensors 300 equipment 400 Spatial State Prediction Device 402 Fluid calculation information acquisition section 403 Indoor information acquisition department 404 Control Information Acquisition Unit 406 Model Calculation Unit 408 Assimilation Section 410 Information Processing Unit 411 Measurement position optimization unit 412 Partial load factor acquisition part 414 COP Acquisition Department 416 Storage section

Claims

1. A model calculation unit that generates a reduced model of the target space based on the results of fluid analysis of the target space which is the object of spatial control, An assimilation unit predicts the spatial distribution of the target space by performing assimilation processing using the reduced model generated by the model calculation unit and environmental data obtained by sensing the target space. An information processing unit searches for control candidates for spatially controlling the target space based on the spatial distribution of the target space, A spatial control system equipped with [the following features].

2. The information processing unit takes the control conditions in the fluid analysis as input values, generates a trained reduced model by training the reduced model after the assimilation process, predicts the spatial distribution of the target space by performing an assimilation process using the trained reduced model and the environmental data, and searches for control candidates based on the spatial distribution. The spatial control system according to claim 1.

3. The information processing unit searches for control candidates so that the target space approaches the target spatial distribution. The spatial control system according to claim 1.

4. The assimilation unit acquires environmental data by changing the position of the sensor provided in the target space, and predicts the spatial distribution before and after the change by performing the assimilation process using the environmental data before and after the change in the sensor position. The information processing unit determines the position of the sensor based on the spatial distribution before and after the change. A spatial control system according to any one of claims 1 to 3.

5. The information processing unit acquires information regarding the partial load rate, power consumption, and COP (Coefficient of Performance) of the equipment performing the spatial control, and searches for control candidates using at least one of the power consumption and COP as evaluation indicators. The spatial control system according to claim 2.

6. The information processing unit derives the partial load ratio at which the difference between the power consumption and the COP is maximized, and searches for a control candidate that can maintain that partial load ratio. The spatial control system according to claim 5.

7. The information processing unit derives the partial load ratio when the difference between the power consumption and the COP is greater than or equal to a predetermined threshold, and the change in the difference between the power consumption and the COP in response to the change in the partial load ratio is within a predetermined range, and searches for a control candidate that can maintain the said partial load ratio. The spatial control system according to claim 5.

8. The steps include generating a reduced model of the target space based on the results of a fluid analysis of the target space that is the subject of spatial control, The steps include predicting the spatial distribution of the target space by performing assimilation processing using the reduced model and environmental data obtained by sensing the target space, The steps include: searching for control candidates for spatially controlling the target space based on the spatial distribution of the target space; A spatial control method including

9. A program for causing a computer to execute the spatial control method described in claim 8.

Images