Mold occurrence prediction device, mold occurrence prediction system, mold occurrence prediction method, and mold occurrence prediction program

Abstract

A mold occurrence prediction device (100) supports a plan for mold prevention by predicting the occurrence of mold on an item surface. A mold prediction unit (110) uses, as input variables (20), a surface temperature observed value (21), a room temperature observed value (22) at the time of observation, weather data including air temperature (23) at the time of observation at a peripheral location of the item surface, and weather data including air temperature (31) past at the peripheral location. The mold prediction unit (110) inputs the input variables (20) into a mold prediction model (50), and predicts, as mold occurrence information (40), the risk of mold occurrence on the item surface in a period after the present time.

Description

Mold growth prediction device, mold growth prediction system, mold growth prediction method, and mold growth prediction program

[0001] The present disclosure relates to a mold growth prediction device, a mold growth prediction system, a mold growth prediction method, and a mold growth prediction program.

[0002] In hot and humid regions or seasons, mold is likely to grow on surfaces in buildings and facilities such as residences, stores, factories, and warehouses. Examples of surfaces in buildings and facilities include ceilings, walls, fixtures, furniture, and home appliances. Because mold growth reduces hygiene and aesthetic appeal, it is necessary to predict and suppress mold growth.

[0003] Patent Document 1 discloses a mold suppression system that estimates the mold index after a predetermined period of time based on the indoor temperature and absolute humidity, and determines the operation schedule of a humidity control device according to the progress of the mold index.

[0004] Patent No. 7261103

[0005] Mold growth occurs on the surface of an article. The technology of Patent Document 1 uses only room temperature, absolute humidity, and outdoor air temperature to predict mold growth. Therefore, the technology of Patent Document 1 cannot evaluate whether the surface of an article is suitable for mold growth. Furthermore, the technology of Patent Document 1 cannot predict where mold will grow on the surface of an article. Furthermore, the technology of Patent Document 1 uses forecast outdoor air temperature to predict mold growth. Therefore, there is a problem in that it is not possible to make long-term predictions, such as from one month or more to ten years in the future, when forecasts have not been announced.

[0006] The present disclosure aims to achieve long-term prediction of mold growth on the surface of an item.

[0007] The mold occurrence prediction device according to the present disclosure is a mold occurrence prediction device that predicts mold occurrence on the surface of an item, and includes a mold prediction unit that inputs the observed surface temperature of the item surface, the observed room temperature at the time of observation, meteorological data including the air temperature at a position surrounding the item surface at the time of observation, and meteorological data including past air temperatures at the position surrounding the item surface as input variables into a mold prediction model, and uses the mold prediction model to predict the risk of mold occurrence on the surface of the item for a period beyond the present as mold occurrence information.

[0008] In the mold growth prediction device according to the present disclosure, the mold prediction unit inputs the observed surface temperature, the observed room temperature, the temperature at the surrounding location at the time of observation, and the past temperature at the surrounding location as input variables into a mold prediction model. The mold prediction unit then predicts the risk of mold growth on the surface of an item for a future period as mold growth information. Therefore, the mold growth prediction device according to the present disclosure can realize long-term prediction of mold growth on the surface of an item.

[0009] FIG. 1 is a diagram showing an example of the configuration of a mold occurrence prediction device according to embodiment 1. FIG. 2 is a diagram showing an example of the functional configuration of a mold occurrence prediction device according to embodiment 1. FIG. 3 is a flow diagram showing an example of the operation of a mold occurrence prediction device according to embodiment 1. FIG. 4 is a diagram showing an example of the configuration of a mold occurrence prediction device according to modified example 2 of embodiment 1. FIG. 5 is a diagram showing an example of the configuration of a mold occurrence prediction device according to modified example 3 of embodiment 1. FIG. 6 is a diagram showing an example of the functional configuration of a mold occurrence prediction device according to embodiment 2. FIG. 7 is a flow diagram showing an example of the operation of a mold occurrence prediction device according to embodiment 2. FIG. 8 is a diagram showing an example of the overall configuration of a mold occurrence prediction system according to embodiment 3. FIG. 9 is a diagram showing an example of the functional configuration of a mold occurrence prediction device according to embodiment 3. FIG. 10 is a flow diagram showing an example of the operation of a mold occurrence prediction device according to embodiment 3.

[0010] The present embodiment will be described below with reference to the drawings. In each drawing, the same or corresponding parts are assigned the same reference numerals. In the description of the embodiment, the description of the same or corresponding parts will be omitted or simplified as appropriate. The arrows in the drawings mainly indicate the flow of data or the flow of processing.

[0011] Embodiment 1. *** Description of Configuration *** Figure 1 is a diagram showing an example of the configuration of a mold occurrence prediction device 100 according to this embodiment. Mold occurrence prediction device 100 is a computer. Mold occurrence prediction device 100 includes a processor 910, as well as other hardware such as a memory 921, an auxiliary storage device 922, an input interface 930, an output interface 940, and a communication device 950. Processor 910 is connected to the other hardware via signal lines and controls this other hardware.

[0012] The mold occurrence prediction device 100 includes, as functional elements, a mold prediction unit 110 and a storage unit 150. The storage unit 150 stores a mold prediction model 50 and mold occurrence information 30.

[0013] The functions of the mold prediction unit 110 are realized by software. The storage unit 150 is provided in the memory 921. The storage unit 150 may be provided in the auxiliary storage device 922, or may be provided separately in the memory 921 and the auxiliary storage device 922.

[0014] The processor 910 is a device that executes a mold growth prediction program. The mold growth prediction program is a program that realizes the functions of the mold prediction unit 110. The processor 910 is an IC that performs arithmetic processing. Specific examples of the processor 910 are a CPU, a DSP, and a GPU. IC is an abbreviation for Integrated Circuit. CPU is an abbreviation for Central Processing Unit. DSP is an abbreviation for Digital Signal Processor. GPU is an abbreviation for Graphics Processing Unit.

[0015] The memory 921 is a storage device that temporarily stores data. Specific examples of the memory 921 are SRAM and DRAM. SRAM is an abbreviation for Static Random Access Memory. DRAM is an abbreviation for Dynamic Random Access Memory. The auxiliary storage device 922 is a storage device that saves data. A specific example of the auxiliary storage device 922 is an HDD. The auxiliary storage device 922 may also be a portable storage medium such as an SD (registered trademark) memory card, CF, NAND flash, flexible disk, optical disk, compact disk, Blu-ray (registered trademark) disk, or DVD. Note that HDD is an abbreviation for Hard Disk Drive. SD (registered trademark) is an abbreviation for Secure Digital. CF is an abbreviation for CompactFlash (registered trademark). DVD is an abbreviation for Digital Versatile Disk.

[0016] The input interface 930 is a port connected to an input device such as a mouse, keyboard, or touch panel. Specifically, the input interface 930 is a USB terminal. The input interface 930 may also be a port connected to a LAN. USB is an abbreviation for Universal Serial Bus. LAN is an abbreviation for Local Area Network.

[0017] The output interface 940 is a port to which a cable of an output device such as a display is connected. Specifically, the output interface 940 is a USB terminal or an HDMI (registered trademark) terminal. Specifically, the display is an LCD. The output interface 940 is also called a display interface. HDMI (registered trademark) is an abbreviation for High Definition Multimedia Interface. LCD is an abbreviation for Liquid Crystal Display.

[0018] The communication device 950 has a receiver and a transmitter. The communication device 950 is connected to a network communication network such as a LAN, the Internet, a telephone line, or Wi-Fi (registered trademark). Specifically, the communication device 950 is a communication chip or NIC. NIC is an abbreviation for Network Interface Card.

[0019] The mold growth prediction program is executed in the mold growth prediction device 100. The mold growth prediction program is loaded into the processor 910 and executed by the processor 910. The memory 921 stores not only the mold growth prediction program but also an OS. OS is an abbreviation for Operating System. The processor 910 executes the mold growth prediction program while executing the OS. The mold growth prediction program and the OS may be stored in an auxiliary storage device 922. The mold growth prediction program and the OS stored in the auxiliary storage device 922 are loaded into the memory 921 and executed by the processor 910. Note that part or all of the mold growth prediction program may be incorporated into the OS.

[0020] Mold growth prediction device 100 may include multiple processors that replace processor 910. These multiple processors share the task of executing the mold growth prediction program. Each processor is a device that executes the mold growth prediction program in the same way as processor 910.

[0021] The data, information, signal values ​​and variable values ​​used, processed or output by the mold growth prediction program are stored in memory 921, auxiliary storage device 922, or registers or cache memory within processor 910.

[0022] The "part" of each part of mold growth prediction device 100 may be read as a "circuit," "process," "procedure," "process," or "circuitry." The mold growth prediction program causes a computer to execute each process of mold growth prediction device 100. The "process" of each process of mold growth prediction device 100 may be read as a "program," "program product," "computer-readable storage medium storing a program," or "computer-readable recording medium recording a program." Furthermore, the mold growth prediction method is a method performed by mold growth prediction device 100 executing a mold growth prediction program. The mold growth prediction program may be provided by being stored in a computer-readable recording medium. Furthermore, the mold growth prediction program may be provided as a program product.

[0023] *** Functional Overview *** FIG. 2 is a diagram showing an example of the functional configuration of a mold occurrence prediction device 100 according to this embodiment. The mold occurrence prediction device 100 is a device that predicts mold occurrence on the surface of an item. The mold prediction unit 110 uses a first prediction model 51 as a mold prediction model 50. The mold prediction unit 110 acquires input variables 20 via an input interface 930 or a communication device 950. Specifically, the input variables 20 are meteorological data including an observed value of the surface temperature of the item surface, an observed value of the room temperature at the time of observation, temperatures at surrounding locations at the time of observation, and past temperatures at surrounding locations. The mold prediction unit 110 inputs the input variables 20 into the mold prediction model 50. The mold prediction unit 110 uses the mold prediction model 50 to predict the risk of mold occurrence on the surface of an item for a future period as mold occurrence information 40. The mold prediction unit 110 acquires the mold occurrence information 40 output from the mold prediction model 50.

[0024] Specifically, the mold prediction unit 110 uses a first prediction model 51 as the mold prediction model 50. The first prediction model 51 has an observed surface temperature value 21, an observed room temperature value 22, an air temperature 23 at the time of observation at the surrounding location, and a past air temperature 31 at the surrounding location as input variables 20. When the input variables 20 are input, the first prediction model 51 outputs mold occurrence information 40.

[0025] ***Explanation of Operation*** Next, the operation of mold occurrence prediction device 100 according to this embodiment will be described. The operating procedure of mold occurrence prediction device 100 corresponds to a mold occurrence prediction method. Furthermore, the program that realizes the operation of mold occurrence prediction device 100 corresponds to a mold occurrence prediction program.

[0026] FIG. 3 is a flow diagram showing an example of the operation of the mold occurrence prediction device 100 according to this embodiment. In step S101, the mold prediction unit 110 inputs the surface temperature observation value 21, the room temperature observation value 22, the temperature at the time of observation 23 at the surrounding location, and the past temperature 31 at the surrounding location as input variables 20 to the first prediction model 51. In addition to the input variables 20, the mold prediction unit 110 may also input a prediction time period for which the risk of mold occurrence is desired to be predicted to the first prediction model 51. The first prediction model 51 is assumed to have a predetermined prediction time period for which the risk of mold occurrence is desired to be predicted. For example, the first prediction model 51 may include a first prediction model for predicting one week in the future, a first prediction model for predicting one month in the future, and a first prediction model for predicting six months in the future. When a prediction time period is input, the first prediction model corresponding to the prediction time period is used. According to this embodiment, it is possible to predict the risk of mold occurrence two or three days in the future, as well as several months or years in the future.

[0027] The surface temperature observation value 21 is the observation value of the temperature of the surface of the item for which mold growth is predicted. The room temperature observation value 22 is the observation value of the air temperature in the room where the item is placed. The air temperature 23 at the time of observation at the surrounding location is the air temperature included in the weather data around the location where the item is located. The surrounding location is the area around the location where the item is located, for example, the location where the item is placed, or the area including that location. The weather data for the surrounding location is, for example, the weather data for the area. The weather data can be obtained from an external weather database.

[0028] The past temperature 31 at the surrounding location is, for example, a temperature included in weather data for the past year or more. The past temperature 31 at the surrounding location is a temperature included in weather data for the surrounding location in the past, such as the same season, the same time, the same month, or the same date and time, more than a year ago. The past temperature 31 at the surrounding location needs to be weather data for the surrounding location in the past that corresponds to the prediction time.

[0029] Furthermore, the past weather data for the surrounding location may be time-series weather data for the past year or more. Using time-series data allows learning of environmental changes such as temperature and humidity due to day and night or season, enabling highly accurate mold occurrence predictions. As a specific example, the time-series data may be, for example, four or more data points per day, at intervals of less than six hours, with missing data of 5% or less.

[0030] It is also preferable to use weather data from the past year or more. Data from a few hours, days, or months may not provide information on seasonal fluctuations or information that is in line with the weather trends specific to the region. For this reason, a data period of at least one year is preferable. This allows for highly accurate, long-term mold occurrence predictions. Furthermore, it allows for not only air conditioning operation control, but also the installation of new air conditioning systems and the layout design of heat source equipment and items. Furthermore, using data from two or more years allows for a more average mold occurrence prediction that is not influenced solely by trends from the past year. Furthermore, using data from two or three or more years allows for mold occurrence predictions that take recent weather fluctuations into account.

[0031] Furthermore, the surface temperature observations, room temperature observations, and temperatures observed at surrounding locations do not need to be data from a single observation. Using a prediction model that uses the results of several observations at different dates and times can improve prediction accuracy. Furthermore, the room temperature observations can be measured at multiple points at different distances from the item surface. Using multiple room temperatures allows for the expression of a temperature gradient, allowing for an accurate description of the movement of air heat in the space. Furthermore, the temperature of the item surface does not need to be data from a single observation point. Linear or area distributions can be evaluated using a thermal camera and used as input variables. This also makes it possible to predict the distribution of mold growth information on the surface of an item.

[0032] Next, in step S102, the mold prediction unit 110 uses the first prediction model 51 to predict the risk of mold growth on the surface of the item for a future period as mold growth information 40. The mold prediction unit 110 acquires the mold growth information 40 output from the first prediction model 51.

[0033] The mold occurrence information 40 is the risk of mold occurrence on the surface of an item at the predicted time. For example, the mold occurrence information 40 is the risk of mold occurrence on the surface of an item one month or more into the future. Specifically, it is information on whether mold will occur on the surface of an item at the predicted time. Alternatively, the mold occurrence information 40 may be information indicating the probability of mold occurrence on the surface of an item at the predicted time. Furthermore, by making predictions using a regression model, the mold occurrence information 40 may include information indicating when mold will occur.

[0034] In this embodiment, predictions can be made for several hours, several days, or even several weeks in advance. However, predictions for several hours or several days in advance are not suitable for controlling the operation of air conditioners, or for installing new air conditioners or for designing the layout of heat source devices and items. Therefore, it is preferable that mold growth information 40 be a prediction for one month or more in the future.

[0035] ***Explanation of the Effects of the Present Embodiment*** The mold growth prediction device according to the present embodiment uses the surface temperature of an item, which is the most direct factor in mold growth, as an explanatory variable of the mold prediction model. This makes it possible to predict where mold may grow on an item. Furthermore, the mold growth prediction device according to the present embodiment uses weather data from a past period, such as the past year or more, as input to the mold prediction model. This allows for predictions that are in line with the weather trends specific to the region and for long-term predictions. The mold growth prediction device according to the present embodiment links the surface temperature and room temperature of an item, which are the factors most influential in mold growth, with the temperature in the weather data at the time of observation. This allows for information on the tendency of the item and space compared to the outside air. In addition, by inputting the temperature in the weather data from the past year or more, information on seasonal fluctuations can be added. This makes it possible to realize long-term predictions of mold growth on the surface of the item. As described above, the mold growth prediction device according to the present embodiment not only controls the operation of equipment such as air conditioners, but also allows for the installation of new air conditioners and the layout design of heat source equipment and items.

[0036] ***Other Configurations*** <Variation 1> In this embodiment, the input variables 20 used are the observed surface temperature value 21, the observed room temperature value 22, the air temperature 23 at the surrounding location at the time of observation, and the past air temperature 31 at the surrounding location. The air temperature 23 at the surrounding location at the time of observation is included in the weather data at the surrounding location at the time of observation. Furthermore, the past air temperature 31 at the surrounding location is included in the past weather data for the surrounding location.

[0037] As a first modification, the mold prediction unit 110 may include the following data in the input variables 20: Observed surface temperature Observed moisture content on the surface of an item Observed room temperature and indoor humidity Weather data including at least some combinations of temperature, humidity, precipitation, wind speed, wind direction, and air pressure at the time of observation at the surrounding location Weather data including at least some combinations of past temperature, humidity, precipitation, wind speed, wind direction, and air pressure at the surrounding location It is preferable that the weather data include at least temperature and humidity.

[0038] The mold growth prediction device of the first modification allows inputting many factors that may affect mold growth, making it possible to more accurately grasp differences in the environment, and therefore to more accurately predict the risk of mold growth.

[0039] <Modification 2> Figure 4 is a diagram showing an example of the configuration of a mold occurrence prediction device 100 according to Modification 2 of this embodiment. A storage unit 150 of mold occurrence prediction device 100 associates input variables 20 with mold occurrence information 40 and stores them as training data 60. Mold occurrence prediction device 100 also includes a learning unit 120 that generates a mold prediction model 50 by learning from training data 60.

[0040] In addition to the initial data, the storage unit 150 stores newly evaluated or accumulated observation values, past weather data, and mold occurrence risk in association with each other as learning data 60. The learning unit 120 learns this learning data 60. This enables the accuracy of mold occurrence prediction to be improved. The learning unit 120 may learn from scratch, or may use fine tuning or re-learning.

[0041] <Modification 3> In this embodiment, the functions of the mold prediction unit 110 are realized by software. As a modification 3, the functions of the mold prediction unit 110 may be realized by hardware. Specifically, the mold occurrence prediction device 100 includes an electronic circuit 909 instead of the processor 910.

[0042] FIG. 5 is a diagram showing an example of the configuration of a mold occurrence prediction device 100 according to a third modification of this embodiment. The electronic circuit 909 is a dedicated electronic circuit that realizes the functions of the mold prediction unit 110. Specifically, the electronic circuit 909 is a single circuit, a composite circuit, a programmed processor, a parallel programmed processor, a logic IC, a GA, an ASIC, or an FPGA. GA is an abbreviation for Gate Array. ASIC is an abbreviation for Application Specific Integrated Circuit. FPGA is an abbreviation for Field-Programmable Gate Array.

[0043] The functions of the mold prediction unit 110 may be realized by a single electronic circuit, or may be realized by distributing the functions across multiple electronic circuits.

[0044] As another variation, some of the functions of mold prediction unit 110 may be realized by electronic circuits and the remaining functions may be realized by software, or some or all of the functions of mold prediction unit 110 may be realized by firmware.

[0045] Each of the processor and the electronic circuit is also called a processing circuitry. That is, the functions of the mold prediction unit 110 are realized by the processing circuitry.

[0046] The above-described first to third modifications can also be applied to the embodiment described below.

[0047] Embodiment 2 In this embodiment, differences from and additions to embodiment 1 will be mainly described. In this embodiment, components having the same functions as those in embodiment 1 will be assigned the same reference numerals, and descriptions thereof will be omitted.

[0048] *** Description of Configuration *** The configuration example of the mold occurrence prediction device 100 according to this embodiment is the same as the configuration example of Fig. 1 described in embodiment 1. However, the mold prediction unit 110 uses a second prediction model 52 and a third prediction model 53 as the mold prediction model 50.

[0049] *** Functional Description *** Figure 6 is a diagram showing an example of the functional configuration of a mold occurrence prediction device 100 according to this embodiment. The mold prediction unit 110 uses a second prediction model 52 and a third prediction model 53 as a mold prediction model 50. The second prediction model 52 takes as input variables 20 an observed surface temperature value 21, an observed room temperature value 22, an air temperature 23 at the time of observation, and a past air temperature 31. When the input variables 20 are input, the second prediction model 52 outputs trends in the surface temperature and room temperature over a future period as trend data 35. The third prediction model 53 takes as input the trend data 35 output from the second prediction model 52, and outputs mold occurrence information 40.

[0050] ***Description of Operation*** Figure 7 is a flow diagram showing an example of the operation of mold occurrence prediction device 100 according to this embodiment. In step S201, mold prediction unit 110 inputs the observed surface temperature value 21, the observed room temperature value 22, the temperature at the time of observation 23 at the surrounding location, and the past temperature 31 at the surrounding location as input variables 20 into second prediction model 52. The past temperature 31 at the surrounding location is, for example, a temperature included in weather data for the past year or more.

[0051] In step S202, the mold prediction unit 110 predicts the trends in surface temperature and room temperature over a future period using the second prediction model 52. The mold prediction unit 110 acquires the trends in surface temperature and room temperature over a future period output from the second prediction model 52 as trend data 35. The trend data 35 is a predicted value of the trends in the surface temperature and room temperature of an item for a predetermined period of time or more in the future. For example, the trend data 35 is time-series data of predicted values ​​of the surface temperature and room temperature from the present time to a time point one month or more in the future. Note that the prediction period for the trend data 35 is not limited to one month, and may be two or three days, several days, several weeks, several months, or even several years or more in the future.

[0052] In step S203, the mold prediction unit 110 inputs the transition data 35 output from the second prediction model 52 to the third prediction model 53. In step S204, the mold prediction unit 110 acquires the mold occurrence information 40 output from the third prediction model 53.

[0053] The mold occurrence information 40 is the risk of mold occurrence on the surface of an item at the prediction time. Alternatively, the mold occurrence information 40 may be time-series data of the risk of mold occurrence on the surface of an item from the present to the prediction time. Specifically, the mold occurrence information 40 is information on whether mold will occur on the surface of an item from the present to the prediction time. Alternatively, the mold occurrence information 40 may be information that indicates the probability of mold occurrence on the surface of an item from the present to the prediction time in a time series. For example, the prediction time can be set appropriately to several days, several weeks, several months, or even several years or more in the future.

[0054] ***Description of Effects of the Present Embodiment*** In the mold growth prediction device according to the present embodiment, the second prediction model is input with observed surface temperature values, observed room temperature values, meteorological data at the time of observation, and meteorological data for, for example, the past year or more. The second prediction model calculates predicted values ​​of changes in the surface temperature and room temperature of an item for, for example, one month or more into the future. Then, in the mold growth prediction device, the third prediction model is input with predicted values ​​of changes in the surface temperature and room temperature of the item output from the second prediction model. The third prediction model predicts the risk of mold growth on the surface of the item for, for example, one month or more into the future. In this way, by first predicting the predicted values ​​of changes in the surface temperature and room temperature of an item, it is possible to evaluate the validity of the prediction, such as why mold is likely to grow. Furthermore, by separating the second prediction model from the third prediction model and sandwiching the prediction of changes in the surface temperature and room temperature between them, a highly generalizable mold growth prediction is realized. Furthermore, by predicting changes in the surface temperature and room temperature of an item, it is possible to analyze time-series changes, such as in which seasons mold is likely to grow or when the risk increases. The risk of mold growth on the surface of an item predicted by the mold growth prediction device according to this embodiment can be used as input for a system that proposes mold countermeasures, which will be described later. This makes it possible to analyze when and how mold is likely to grow and how to deal with it.

[0055] ***Other Configurations*** Modifications 1 to 3 described in the first embodiment can also be applied to this embodiment.

[0056] <Variation 4> The following is a case where Variation 1 described in Embodiment 1 is applied to the mold growth prediction device according to the present embodiment. The mold prediction unit inputs, as input variables, the observed values ​​of surface temperature, the observed values ​​of moisture content on the surface of an item, the observed values ​​of room temperature and indoor humidity, meteorological data including at least some combinations of temperature, humidity, precipitation, wind speed, wind direction, and air pressure at the time of observation at the surrounding location, and meteorological data including at least some combinations of past temperature, humidity, precipitation, wind speed, wind direction, and air pressure at the surrounding location into the second prediction model. The mold prediction unit obtains transition data output from the second prediction model, which is transition data of the surface temperature, moisture content on the surface of an item, room temperature, and indoor humidity over a future period.

[0057] Embodiment 3 In this embodiment, differences from and additions to Embodiments 1 and 2 will be mainly described. In this embodiment, additions to the mold occurrence prediction device 100 described in Modification 4, in which Modification 1 described in Embodiment 1 is applied to the configuration of Embodiment 2, will be particularly described. In this embodiment, components having the same functions as those in Embodiments 1 and 2 will be assigned the same reference numerals, and descriptions thereof will be omitted.

[0058] *** Description of Configuration *** Figure 8 is a diagram showing an example of the overall configuration of mold occurrence prediction system 500 according to this embodiment. Mold occurrence prediction system 500 according to this embodiment includes mold occurrence prediction device 100 and artificial intelligence 200. Mold occurrence prediction system 500 is a communication system in which mold occurrence prediction device 100 and artificial intelligence 200 communicate with each other. Mold occurrence prediction device 100 and artificial intelligence 200 are capable of inputting and outputting data or transmitting and receiving data.

[0059] The artificial intelligence 200 includes an inference unit that performs inference using a trained model. The inference unit receives a prompt 61 from the proposal unit 140 as input and outputs suppression support information 42 corresponding to the prompt 61 based on the trained model. The suppression support information 42 is information for supporting mold suppression. The artificial intelligence 200 may be a generative AI. AI is an abbreviation for Artificial Intelligence. The proposal unit 140 inputs the prompt 61 to the artificial intelligence 200, which is an external component, and instructs it to generate the suppression support information 42, thereby outputting the suppression support information 42. In this manner, various external artificial intelligence services can be utilized to output the suppression support information 42. Note that, although the artificial intelligence 200 is an external component of the mold occurrence prediction device 100 in FIG. 8, it may also be an internal component.

[0060] The artificial intelligence 200 may be configured using algorithms such as Transformer, BERT, and GPT. The artificial intelligence 200 may also be configured by combining a plurality of algorithms including these algorithms. BERT is an abbreviation for Bidirectional Encoder Representations from Transformers. GPT is an abbreviation for Generative Pre-Training.

[0061] Furthermore, mold occurrence predicting device 100 according to this embodiment includes prompt creating unit 130, proposing unit 140, and information accepting unit 160 in addition to the components described in the second embodiment.

[0062] *** Functional Description *** Figure 9 is a diagram showing an example of the functional configuration of a mold occurrence prediction device 100 according to this embodiment. The prompt creation unit 130 acquires input variables 20, a mold prediction model 50, mold occurrence information 40 output from the mold prediction model 50, image data 71 capturing an image of the area surrounding the surface of an item, and recommended room temperature and indoor humidity values ​​72. The prompt creation unit 130 then creates a prompt 61 for estimating mold occurrence factors and mitigation measures for suppressing mold occurrence. The prompt creation unit 130 may also include in the prompt 61 content for estimating an additional evaluation location 421, which is a location where mold is likely to occur and which is to be additionally evaluated.

[0063] The proposing unit 140 acquires a prompt 61 and outputs suppression support information 42 including a cause of occurrence and a suppression measure based on the prompt 61. The proposing unit 140 inputs the prompt 61 to the artificial intelligence 200 and acquires the suppression support information 42 from the artificial intelligence 200. The proposing unit 140 may output the suppression support information 42 including an additional evaluation position 421 based on the prompt 61. In this embodiment, the proposing unit 140 outputs the suppression support information 42 including the additional evaluation position 421.

[0064] ***Description of Operation*** FIG. 10 is a flow diagram showing an example of the operation of the mold occurrence prediction device 100 according to this embodiment. In step S301, the prompt creation unit 130 acquires input variables 20, a mold prediction model 50, the output of the mold prediction model 50, image data 71, and recommended values ​​72 for room temperature and indoor humidity. In this embodiment, the mold prediction model 50 acquires a second prediction model 52 and a third prediction model 53. Furthermore, the image data 71, which acquires the transition data 35 and mold occurrence information 40 as the output of the mold prediction model 50, is an ambient image linked to the position of the object surface. The recommended values ​​72 for room temperature and indoor humidity are recommended values ​​that avoid room temperature and indoor humidity that are prone to mold occurrence. This information may be input by the user, or preset recommended values ​​may be acquired.

[0065] In step S302, the prompt creation unit 130 creates a prompt 61 for estimating the causes of mold growth and prevention measures using the input variables 20, the second prediction model 52, the third prediction model 53, the transition data 35, the mold growth information 40, the image data 71, and the recommended room temperature and indoor humidity values ​​72. The prompt creation unit 130 also includes in the prompt 61 content for estimating an additional evaluation location 421 that is likely to develop mold and should be additionally evaluated.

[0066] As described above, the prompt creation unit 130 inputs the second prediction model 52 and the third prediction model 53 and their inputs and outputs into the prompt 61. This allows the prompt 61 to include information on how an artificial intelligence (AI) such as a generation AI should use these models for analysis.

[0067] In step S303, the proposing unit 140 inputs the prompt 61 to the artificial intelligence 200. In step S304, the proposing unit 140 acquires from the artificial intelligence 200 suppression support information 42, which includes factors that may cause mold growth and suppression measures. The suppression support information 42 is information that includes, for example, a presentation of specific improvement targets for temperature and humidity to reduce the risk of mold growth and measures to achieve those targets. In this embodiment, the suppression support information 42 also includes an additional evaluation position 421.

[0068] In step S305, the proposing unit 140 presents the user with the suppression support information 42. For example, the proposing unit 140 presents the suppression support information 42 to the user by displaying the suppression support information 42 on a display device.

[0069] The suppression support information 42 includes, for example, factors that cause mold growth and feasible and comprehensive suppression measures that are suitable for the surrounding environment of the mold suppression area. The suppression measures include measures to remove the factors that cause the growth. The suppression support information 42 also includes additional evaluation positions 421, which are positions on the surface of items where the risk of mold growth may increase if the surrounding environment changes after the suppression measures are implemented. Specific examples of suppression measures are as follows: Creation of an air conditioner operation plan, application of anti-mold coating, installation of air conditioners, and heat insulation of heat source equipment.

[0070] In step S306, the information receiving unit 160 determines whether there is additional information 62, which is feedback on the suppression support information 42. If there is additional information 62 from the user, the process proceeds to step S307. If there is no additional information 62 from the user, the process ends.

[0071] In step S307, the information accepting unit 160 accepts additional information 62 from the user via the input interface 930 or the communication device 950. The additional information 62 is, for example, information that the user was unable to enter when entering the input variables 20. Alternatively, the additional information 62 is information that the user has determined, based on reference to the suppression support information 42, should additionally enter regarding the mold-suppressed area or the surrounding area. For example, the additional information 62 includes an additional evaluation location 421 included in the suppression support information 42. The information accepting unit 160 may accept information from the user by manual input or voice input. Alternatively, the information accepting unit 160 may accept the additional information 62 through an interactive or chat format with the user. Alternatively, the information accepting unit 160 may have a function of automatically including the additional evaluation location 421 as additional information 62 when it determines that the suppression support information 42 includes the additional evaluation location 421.

[0072] As described above, the information receiving unit 160 receives the proposal request by the additional evaluation position 421, the constraints on the proposal for which measures cannot be taken, and the like as the additional information 62. When the information receiving unit 160 receives the additional information 62, the process returns to step S302.

[0073] In step S302, if the process returns from step S307, the prompt generator 130 generates the prompt 61, further taking into consideration the additional information 62. That is, the prompt generator 130 generates the prompt 61 based on the input variables 20, the mold prediction model 50, the transition data 35, the mold occurrence information 40, the image data 71, the recommended room temperature and indoor humidity values ​​72, and the additional information 62.

[0074] ***Description of the Effects of This Embodiment*** In the mold growth prediction device according to this embodiment, the second prediction model predicts the transition data of the item surface temperature, the moisture content on the item surface, the room temperature, and the indoor humidity for up to one month or more into the future. The third prediction model uses the transition data of the item surface temperature, the moisture content on the item surface, the room temperature, and the indoor humidity as input variables to predict the risk of mold growth on the surface of the item for up to one month or more into the future. In this way, the mold growth prediction device according to this embodiment allows for the input of many factors that may affect mold growth. Therefore, the mold growth prediction device according to this embodiment makes it possible to more accurately capture environmental differences, thereby achieving highly accurate mold growth predictions.

[0075] Furthermore, in the mold growth prediction device according to this embodiment, input variables, a prediction model specialized for mold growth prediction, and prediction results are input to a proposal unit. This allows for the presentation of specific temperature and humidity improvement targets to reduce the risk of mold growth and the proposal of measures to achieve those targets. In this case, the proposal unit is expected to utilize artificial intelligence, such as generative AI. Furthermore, the mold growth prediction device according to this embodiment can take into account recommended values ​​for room temperature and indoor humidity, as well as image data of the area around the surface of an item. Therefore, the mold growth prediction device according to this embodiment can present specific improvement targets with higher accuracy and propose measures to achieve those targets.

[0076] Furthermore, in the mold growth prediction device according to this embodiment, the suggestion unit proposes additional evaluation positions on the surface of an item that may be at a high risk of mold growth. As a result, the mold growth prediction device according to this embodiment can output positions that may be at a higher risk of mold growth than the evaluated item positions in the current environment. In this way, the mold growth prediction device according to this embodiment can optimize items and the entire space by repeating evaluation, prediction, and countermeasure proposals at the additional evaluation positions. Furthermore, by allowing input of requests for further proposals of additional evaluation positions, additional information, and constraints for proposals that do not support countermeasures, it is possible to output positions that may be at a higher risk of mold growth after countermeasures are implemented. This makes it possible to consider countermeasures for optimizing items and the entire space.

[0077] In addition, in the mold growth prediction device according to this embodiment, the suggestion unit outputs information such as the estimated cause of mold growth, proposed countermeasures, and additional evaluation locations. Based on the output information, the user can input information such as requests for further proposals for additional evaluation locations, additional information, or constraints for proposed countermeasures that are not possible, via the information receiving unit. The mold growth prediction device can then use the information received from the user to create prompts to input to the suggestion unit. In this way, the mold growth prediction device according to this embodiment can further narrow down mold growth factors by inputting additional information. Furthermore, by providing feedback on the estimated cause and countermeasures, it is possible to consider other possibilities or propose more realistic countermeasures, especially alternatives to countermeasures that are difficult to implement in the environment. Furthermore, by providing a prompt requesting an analysis of locations that may be at higher risk of mold growth after countermeasures are implemented, it is possible to consider countermeasures to optimize items and the entire space.

[0078] In the mold occurrence prediction device according to this embodiment, the additional evaluation positions output by the suggestion unit are fed back to the evaluation. In this way, the mold occurrence prediction device according to this embodiment performs evaluation based on the additional evaluation positions output by the suggestion unit, and optimizes the items and the entire space by repeating evaluation, prediction, and proposal. This repetition of evaluation, prediction, and proposal may be performed by a worker carrying this system, or may be performed by a fully automated robot equipped with this system.

[0079] In the above first to third embodiments, each part of the mold growth prediction device has been described as an independent functional block. However, the configuration of the mold growth prediction device does not have to be the same as that of the above-described embodiments. The functional blocks of the mold growth prediction device may have any configuration as long as they can realize the functions described in the above-described embodiments. Furthermore, the mold growth prediction device may not be a single device, but may be a system composed of multiple devices. Furthermore, multiple parts of the first to third embodiments may be combined and implemented. Alternatively, only one part of these embodiments may be implemented. In addition, these embodiments may be combined in any way, either as a whole or in part. In other words, in the first to third embodiments, the embodiments may be freely combined, or any component of each embodiment may be modified, or any component of each embodiment may be omitted.

[0080] The above-described embodiments are essentially preferred examples and are not intended to limit the scope of the present disclosure, the scope of application of the present disclosure, or the scope of use of the present disclosure. The above-described embodiments can be modified in various ways as needed. For example, the procedures described using flow charts or sequence diagrams may be modified as appropriate.

[0081] Various aspects of the present disclosure are summarized below as appendices.

[0082] (Supplementary Note 1) A mold occurrence prediction device for predicting mold occurrence on the surface of an item, comprising a mold prediction unit that inputs, into a mold prediction model, as input variables, an observed value of the surface temperature of the item surface, an observed value of the room temperature at the time of observation, meteorological data including the air temperatures at a peripheral position around the item surface at the time of observation, and meteorological data including past air temperatures at the peripheral position, and uses the mold prediction model to predict the risk of mold occurrence on the surface of the item for a future period as mold occurrence information. (Supplementary Note 2) The mold occurrence prediction device according to Supplementary Note 1, wherein the mold prediction unit uses a first prediction model as the mold prediction model, and the first prediction model has as input variables the observed value of the surface temperature, the observed value of the room temperature, the air temperatures at the peripheral position at the time of observation, and past air temperatures at the peripheral position, and outputs the mold occurrence information. (Appendix 3) The mold occurrence prediction device described in Appendix 2 is configured such that the mold prediction unit inputs the observed surface temperature, the observed moisture content on the surface of the item, the observed room temperature and indoor humidity, meteorological data including at least a partial combination of the temperature, humidity, precipitation, wind speed, wind direction, and air pressure at the time of observation at the surrounding location, and meteorological data including at least a partial combination of the past temperature, humidity, precipitation, wind speed, wind direction, and air pressure at the surrounding location as input variables into the first prediction model. (Appendix 4) The mold occurrence prediction device described in Appendix 1, wherein the mold prediction unit uses a second prediction model and a third prediction model as the mold prediction models, the second prediction model uses the observed value of the surface temperature, the observed value of the room temperature, the air temperature at the time of observation, and the past air temperature as input variables, and outputs the trends in the surface temperature and the room temperature over the future period as trend data, and the third prediction model inputs the trend data output from the second prediction model and outputs the mold occurrence information.(Appendix 5) The mold occurrence prediction device described in Appendix 4, wherein the mold prediction unit inputs the observed value of the surface temperature, the observed value of the moisture content on the surface of the item, the observed values ​​of the room temperature and indoor humidity, meteorological data including at least some combinations of the temperature, humidity, precipitation, wind speed, wind direction, and air pressure at the time of observation at the surrounding location, and meteorological data including at least some combinations of the past temperature, humidity, precipitation, wind speed, wind direction, and air pressure at the surrounding location as input variables into the second prediction model, and obtains the transition data output from the second prediction model, which is the transitions of the surface temperature, the moisture content on the surface of the item, the room temperature, and indoor humidity over the future period. (Supplementary Note 6) The mold occurrence prediction device according to any one of Supplementary Notes 1 to 5, comprising: a prompt creation unit that acquires the input variables, the mold prediction model, the mold occurrence information output from the mold prediction model, image data of an area surrounding the surface of the item, and the recommended room temperature and indoor humidity, and creates a prompt for estimating a cause of mold occurrence that is a factor in mold occurrence and a suppression measure for suppressing the mold occurrence, and a suggestion unit that acquires the prompt and outputs suppression support information including the cause of mold occurrence and the suppression measure based on the prompt. (Supplementary Note 7) The mold occurrence prediction device according to Supplementary Note 6, wherein the prompt creation unit creates a prompt for estimating an additional evaluation position that is a position where mold is likely to occur and is to be additionally evaluated, and the suggestion unit outputs the suppression support information including the additional evaluation position based on the prompt. (Supplementary Note 8) The mold occurrence prediction device according to Supplementary Note 6 or Supplementary Note 7, wherein the mold occurrence prediction device includes an information receiving unit that receives additional information that is feedback to the suppression support information and includes the additional evaluation position, and the prompt creation unit creates the prompt using the additional information. (Supplementary Note 9) The mold occurrence prediction device according to any one of Supplementary Note 6 to Supplementary Note 8, wherein the suggestion unit inputs the prompt to an artificial intelligence and acquires the suppression support information from the artificial intelligence.(Supplementary Note 10) The mold occurrence prediction device according to any one of Supplementary Notes 1 to 9, comprising: a storage unit that associates the input variables with the mold occurrence information and stores them as learning data; and a learning unit that generates the mold prediction model by learning the learning data. (Supplementary Note 11) The mold occurrence prediction device according to any one of Supplementary Notes 1 to 10, wherein the past weather data for the surrounding location is time-series weather data for the past one year or more. (Supplementary Note 12) A mold occurrence prediction system comprising: the mold occurrence prediction device according to any one of Supplementary Notes 1 to 11; and artificial intelligence. (Supplementary Note 13) A mold occurrence prediction system for predicting mold occurrence on the surface of an item, comprising: an input interface that acquires, as input variables, an observed value of the surface temperature of the item surface, an observed value of the room temperature at the time of observation, meteorological data including the air temperature at a position around the item surface at the time of observation, and meteorological data including the past air temperature at the position around the item surface; and a mold prediction unit that inputs the input variables into a mold prediction model and uses the mold prediction model to predict the risk of mold occurrence on the surface of the item for a period beyond the present as mold occurrence information. (Appendix 14) A mold occurrence prediction method used in a mold occurrence prediction device that predicts mold occurrence on the surface of an item, wherein a computer inputs, as input variables, the observed value of the surface temperature of the item surface, the observed value of the room temperature at the time of observation, meteorological data including the air temperature at a position around the item surface at the time of observation, and meteorological data including the past air temperature at the position around the item surface into a mold prediction model, and uses the mold prediction model to predict the risk of mold occurrence on the surface of the item for a period beyond the present as mold occurrence information.(Appendix 15) A mold occurrence prediction program used in a mold occurrence prediction device that predicts mold occurrence on the surface of an item, the mold occurrence prediction program inputs, as input variables, an observed value of the surface temperature of the item surface, an observed value of the room temperature at the time of observation, meteorological data including the air temperature at a position around the item surface at the time of observation, and meteorological data including the past air temperature at the position around the item surface into a mold prediction model, and causes the mold occurrence prediction device, which is a computer, to execute a mold prediction process using the mold prediction model to predict the risk of mold occurrence on the surface of the item for a period beyond the present as mold occurrence information.

[0083] 20 Input variables, 21 Observed surface temperature, 22 Observed room temperature, 23 Air temperature at time of observation, 31 Past air temperature, 35 Trend data, 40 Mold occurrence information, 42 Suppression support information, 421 Additional evaluation position, 60 Learning data, 61 Prompt, 62 Additional information, 50 Mold prediction model, 51 First prediction model, 52 Second prediction model, 53 Third prediction model, 71 Image data, 72 Recommended value, 100 Mold occurrence prediction device, 110 Mold prediction unit, 120 Learning unit, 130 Prompt creation unit, 140 Proposal unit, 150 Storage unit, 160 Information reception unit, 200 Artificial intelligence, 500 Mold occurrence prediction system, 909 Electronic circuit, 910 Processor, 921 Memory, 922 Auxiliary storage device, 930 Input interface, 940 Output interface, 950 Communication device.

Claims

1. In a mold growth prediction device that predicts the occurrence of mold on the surface of an object, A mold growth prediction device comprising a mold growth prediction unit that inputs observed surface temperature values ​​of the surface of the article, observed room temperature values ​​at the time of observation, meteorological data including the temperature at the time of observation at a location surrounding the surface of the article, and meteorological data including past temperatures at the location surrounding the article as input variables to a mold growth prediction model, and uses the mold growth prediction model to predict the risk of mold growth on the surface of the article in a period beyond the present as mold growth information.

2. The mold prediction unit is, The first prediction model is used as the mold prediction model mentioned above. The first prediction model described above is: The mold growth prediction device according to claim 1, wherein the observed surface temperature, the observed room temperature, the temperature at the time of observation at the surrounding location, and the past temperature at the surrounding location are used as input variables, and the device outputs the mold growth information.

3. The mold prediction unit is, The mold growth prediction device according to claim 2, wherein the observed surface temperature, the observed moisture content of the surface of the article, the observed room temperature and indoor humidity, meteorological data including at least some combinations of temperature, humidity, precipitation, wind speed, wind direction, and atmospheric pressure at the time of observation at the surrounding location, and meteorological data including at least some combinations of past temperature, humidity, precipitation, wind speed, wind direction, and atmospheric pressure at the surrounding location are input to the first prediction model as input variables.

4. The mold prediction unit is, The mold prediction models used are the second prediction model and the third prediction model. The second prediction model described above is: The observed surface temperature, the observed room temperature, the temperature at the time of observation, and the past temperature are used as input variables, and the changes in the surface temperature and room temperature over the aforementioned period are output as trend data. The third prediction model described above is: The mold occurrence prediction device according to claim 1, which inputs the transition data output from the second prediction model and outputs the mold occurrence information.

5. The mold prediction unit is, The observed surface temperature, the observed moisture content of the surface of the article, the observed room temperature and indoor humidity, meteorological data including at least some combinations of temperature, humidity, precipitation, wind speed, wind direction, and atmospheric pressure at the observation site in the surrounding area, and meteorological data including at least some combinations of past temperature, humidity, precipitation, wind speed, wind direction, and atmospheric pressure at the surrounding area are input to the second prediction model as input variables. The mold growth prediction device according to claim 4, which acquires the transition data, which is the transition of the surface temperature, the amount of moisture on the surface of the article, the room temperature, and the indoor humidity over the aforementioned period, output from the second prediction model.

6. The mold growth prediction device is A prompt generation unit acquires the aforementioned input variables, the mold prediction model, the mold occurrence information output from the mold prediction model, image data of the surrounding area of ​​the surface of the article, and the recommended values ​​for room temperature and indoor humidity, and generates prompts for estimating the factors causing mold occurrence and the suppression measures to suppress mold occurrence. A proposal unit that obtains the prompt and outputs suppression support information including the cause and suppression measures based on the prompt. A mold growth prediction device according to any one of claims 1 to 5, comprising:

7. The prompt generation unit, The aforementioned location is prone to mold growth, and a prompt is created to estimate additional evaluation locations to be further evaluated. The aforementioned proposal section is, The mold growth prediction device according to claim 6, which outputs the suppression support information including the additional evaluation position based on the prompt.

8. The mold growth prediction device is The system includes an information receiving unit that receives additional information which is feedback to the suppression support information, and which includes the additional evaluation location. The prompt generation unit, The mold growth prediction device according to claim 7, which generates the prompt using the additional information.

9. The aforementioned proposal section is, The mold growth prediction device according to claim 6, which inputs the aforementioned prompt to artificial intelligence and obtains the suppression support information from the artificial intelligence.

10. The mold growth prediction device is A storage unit that associates the input variables with the mold growth information and stores them as learning data, A learning unit that generates the mold prediction model by learning the aforementioned training data. A mold growth prediction device according to any one of claims 1 to 5, comprising:

11. The mold growth prediction device according to any one of claims 1 to 5, wherein the past weather data at the aforementioned surrounding location is time-series data of weather data for the past year or more.

12. A mold growth prediction device according to any one of claims 1 to 5, Artificial intelligence and A mold growth prediction system equipped with [feature / feature].

13. In a mold growth prediction system that predicts the occurrence of mold on the surface of an object, An input interface that acquires the following as input variables: the observed surface temperature of the article surface, the observed room temperature at the time of observation, meteorological data including the temperature at the time of observation at a location surrounding the article surface, and meteorological data including past temperatures at the location surrounding the article surface. A mold prediction unit inputs the aforementioned input variables into a mold prediction model and uses the mold prediction model to predict the risk of mold growth on the surface of the article in a period beyond the present as mold growth information. A mold growth prediction system equipped with [feature / feature].

14. In a mold growth prediction method used in a mold growth prediction device that predicts the occurrence of mold on the surface of an object, A mold growth prediction method comprising: a computer inputting observed surface temperature of the article surface, observed room temperature at the time of observation, meteorological data including the temperature at the time of observation at a location surrounding the article surface, and meteorological data including past temperatures at the location surrounding the article as input variables to a mold growth prediction model; and using the mold growth prediction model to predict the risk of mold growth on the article surface for a period beyond the present as mold growth information.

15. In a mold growth prediction program used in a mold growth prediction device that predicts the occurrence of mold on the surface of an object, A mold prediction program that inputs the observed surface temperature of the article surface, the observed room temperature at the time of observation, meteorological data including the temperature at the time of observation at a location surrounding the article surface, and meteorological data including past temperatures at the location surrounding the article as input variables into a mold prediction model, and causes a mold prediction device, which is a computer, to perform a mold prediction process that uses the mold prediction model to predict the risk of mold growth on the article surface for a period beyond the present as mold growth information.