Hot water supply system

Abstract

This hot water supply system eliminates the need to install a dedicated device in the bathroom to detect whether or not a shower is being taken, and provides an accurate estimation of whether or not a shower is being taken. [Solution] The system comprises a water heater 1 that supplies hot water to a hot water source including a shower 65 in the bathroom 6, and an estimation unit 24 that estimates whether or not the user is taking a shower by inputting predetermined data into a trained model 26. The predetermined data input into the trained model 26 includes first data relating to the time the user is not taking a bath in the bathtub 60, calculated as the difference between the entry time, which is the time from entering the room to leaving the room, and the bathing time, which is the time from bathing to leaving the bath, and second data relating to the amount of heat used by the water heater 1 for hot water supply while the user is in the room. The trained model is generated by machine learning with explanatory variable data including the first data and the second data as explanatory variables and the presence or absence of a shower as the dependent variable.

Description

Technical Field

[0001] The present invention relates to a hot water supply system.

Background Art

[0002] Some water heaters have a hot water supply function for supplying hot water to a hot water tap including a shower in a house, a bathtub filling function for filling a bathtub with hot water, and a discharging function for discharging the bathtub water. In such a water heater, there is one that has a human presence sensor in a remote control installed in a bathroom to detect the presence or absence of a person in the bathroom, and a water level sensor for detecting the water level of the hot water in the bathtub, and has a function of detecting a person taking a bath in the bathtub based on the detection result of the water level sensor.

[0003] On the other hand, Patent Document 1 describes that a temperature distribution detector composed of a thermopile array or the like is installed in a bathroom, and it is detected based on the temperature distribution detected by this temperature distribution detector that a shower bath is being performed in the bathroom.

Prior Art Documents

Patent Documents

[0004]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0005] In the case of Patent Document 1, a temperature distribution detector such as a thermopile array must be installed in the bathroom as a dedicated device for detecting the presence or absence of a shower bath.

[0006] [[ID=??]]The present invention has been made to solve the above problems, and an object thereof is to provide a hot water supply system that does not require a dedicated device for detecting the presence or absence of a shower bath to be installed in the bathroom and can accurately estimate the presence or absence of a shower bath.

Means for Solving the Problems

[0007] To achieve the above objective, a hot water supply system according to one aspect of the present invention includes an entry / exit detection unit that detects when a user enters and leaves a bathroom, an entry / exit bath detection unit that detects when a user enters and leaves a bathtub in the bathroom, a water heater that supplies hot water to destinations including showers in the bathroom and can calculate the amount of heat used for hot water supply at the time of hot water supply, and an estimation unit that estimates whether or not a user is taking a shower by inputting predetermined data into a trained model generated by machine learning. The predetermined data input into the trained model includes first data relating to the time the user is not taking a bath, calculated as the difference between entry time, which is the time from entry to exit detected by the entry / exit detection unit, and bath time, which is the time from bathing to leaving the bath detected by the entry / exit detection unit, and second data relating to the amount of heat used for hot water supply by the water heater while the user is in the bathroom. The trained model is generated by machine learning with explanatory variable data including the first data and the second data as explanatory variables and the presence or absence of showering as the objective variable.

[0008] With this configuration, the estimation unit includes, as explanatory variables, first data regarding the time the user is in the bathroom but not in the bathtub, and second data regarding the amount of heat used for hot water while in the bathroom, which have a high correlation with whether or not a shower is taken. A trained model generated by machine learning, with whether or not a shower is taken as the dependent variable, is used to estimate whether or not a shower is taken, thus enabling accurate estimation of whether or not a shower is taken. Furthermore, there is no need to install a dedicated device in the bathroom to detect whether or not a shower is taken.

[0009] The water heater includes a flow sensor for detecting the flow rate of hot water used during hot water supply, a bathroom temperature sensor for detecting the temperature inside the bathroom, and a bathroom humidity sensor for detecting the humidity inside the bathroom. The predetermined data and explanatory variable data may include at least one of the following: a third data relating to the time of entry detected by the entry / exit detection unit, a fourth data relating to the flow rate of hot water used while the user is in the bathroom, a fifth data relating to the range or maximum value of the temperature inside the bathroom while the user is in the bathroom, and a sixth data relating to the range or maximum value of the humidity inside the bathroom while the user is in the bathroom.

[0010] With this configuration, by adding at least one of the following as explanatory variables—a third data point regarding the time the user enters the bathroom, a fourth data point regarding the flow rate of hot water used while the user is in the bathroom, a fifth data point regarding the range or maximum value of the temperature change inside the bathroom while the user is in the bathroom, and a sixth data point regarding the range or maximum value of the humidity change inside the bathroom while the user is in the bathroom—a highly accurate trained model can be generated, and the presence or absence of a shower can be estimated with greater accuracy using that trained model.

[0011] The system may further include a notification device that notifies users of predetermined information corresponding to the estimation results of the estimation unit. This allows the system to notify users of information such as the effects of shower baths and bath baths, depending on the frequency of shower baths and bath baths during a predetermined period.

[0012] The system may include a server that is communicatively connected to the water heater and has the estimation unit, wherein the estimation unit generates predetermined data from data received from the water heater and inputs it into the trained model. [Effects of the Invention]

[0013] The present invention has the configuration described above and has the effect of providing a hot water supply system that can accurately estimate whether or not a shower is being taken, without requiring the installation of a dedicated device in the bathroom to detect whether or not a shower is being taken. [Brief explanation of the drawing]

[0014] [Figure 1] Figure 1 is a block diagram showing a schematic configuration of an example of a hot water supply system in this embodiment. [Figure 2] Figure 2 shows an example of a household bathing vector in this embodiment. [Figure 3] Figure 3 shows an example of a neural network that can be used as a learning model in this embodiment. [Figure 4] Figure 4 shows a schematic configuration of the estimation unit in this embodiment. [Modes for carrying out the invention]

[0015] Preferred embodiments will be described below with reference to the drawings. In the following description, the same or corresponding elements will be denoted by the same reference numerals throughout the drawings, and redundant descriptions will be omitted. Furthermore, the present invention is not limited to the embodiments described below.

[0016] (Embodiment) Figure 1 is a block diagram illustrating the schematic configuration of an example of a hot water supply system in this embodiment. The hot water supply system 100 in this embodiment comprises a water heater 1 installed in the user's residence and a server 2.

[0017] The water heater 1 comprises a water heater unit 4 and a remote control for remotely operating the water heater unit 4. The remote control includes a bathroom remote control 7 installed in the bathroom 6 and a kitchen remote control 8 installed in the kitchen, etc. The kitchen remote control 8 is equipped with a monitor 80, which is a display device, as an alarm for notifying setting information and information transmitted from the server 2. In addition to or instead of the monitor 80, the kitchen remote control 8 may be equipped with a predetermined lamp or speaker as an alarm. The bathroom remote control 7 may also be equipped with an alarm.

[0018] The hot water supply machine main body 4 incorporates a controller 5 for controlling the hot water supply machine main body 4. The controller 5, the bathroom remote controller 7, and the kitchen remote controller 8 are configured to be able to communicate with each other, for example, by wired communication. The kitchen remote controller 8 incorporates a relay device 9. The relay device 9 is set up for connection with the router 3 via a wireless LAN, and can perform wireless communication with the router 3 via the wireless LAN. This relay device 9 communicates with a management server 21 via a communication network 10 such as the router 3 and the Internet. The controller 5 and the management server 21 are communicably connected via the relay device 9. The router 3 is a wireless LAN router owned by the user of the hot water supply machine 1 and is connected to the communication network 10. Note that the relay device 9 may be configured as a device separate from the kitchen remote controller 8.

[0019] The server 2 includes a management server 21 and an analysis server 22. The management server 21 is communicatively connected to the hot water supply machine 1 via the communication network 10. The analysis server 22 is communicatively connected to the management server 21 via the communication network 10. Instead of this, the management server 21 and the analysis server 22 may be directly connected so as to be able to transmit and receive data by means of a communication cable or the like. In the example of FIG. 1, an example in which the management server 21 and the analysis server 22 are configured as separate servers is shown, but the management server 21 and the analysis server 22 may be configured as a single common server.

[0020] The hot water supply machine main body 4 is installed at a predetermined location inside or outside the user's residence, and is, for example, a combustion heating type hot water supply device (heat source machine), and includes a heat exchanger and a heating device for heating the heat exchanger. The hot water supply machine main body 4 is connected to a circulation fitting 63 attached to the bathtub 60 via external pipes 61, 62. The hot water supply machine main body 4 has a well-known configuration having a hot water supply function of supplying hot water to a hot water tap such as a shower in the kitchen or the bathroom 6 via an external pipe 64, a bathtub pouring function of supplying hot water to the bathtub 60 via the external pipes 61, 62, and a discharging function of discharging the bathtub water.

[0021] The hot water supply machine body 4 includes a heat exchanger for hot water supply, has a water inlet passage for introducing water from an external water supply source to the heat exchanger for hot water supply, and a hot water outlet passage for temperature-adjusting the hot water heated by the heat exchanger for hot water supply and sending it to an external pipe 64 connected to a hot water faucet including a shower 65. The hot water supply machine body 4 is provided with a water flow rate sensor 42 for detecting the flow rate of water flowing through the water inlet passage, a water inlet temperature sensor 43 for detecting the temperature of the water in the water inlet passage, and a hot water outlet temperature sensor 44 for detecting the temperature of the hot water in the hot water outlet passage.

[0022] Further, the hot water supply machine body 4 includes a supplementary heating heat exchanger, and a bathtub circulation flow path is formed between the supplementary heating heat exchanger and the bathtub 60. This bathtub circulation flow path has a return flow path connected to the supplementary heating heat exchanger and a circulation fitting 63 attached to the bathtub 60 via an external pipe 61, and a forward flow path connected to the circulation fitting 63 via the supplementary heating heat exchanger and an external pipe 62. In the return flow path, a circulation pump, a pressure sensor, and a water level sensor 41 configured to detect the water level of the hot water stored in the bathtub 60 are provided. Also, the bathtub circulation flow path is connected to the aforementioned hot water outlet passage via a connection flow path. When filling the bathtub 60 with hot water or the like, the hot water supply machine body 4 supplies the hot water in the hot water outlet passage to the bathtub 60 via the connection flow path and the bathtub circulation flow path.

[0023] As described above, the hot water supply machine body 4 is provided with various sensors such as a water level sensor 41, a water flow rate sensor 42, a water inlet temperature sensor 43, and a hot water outlet temperature sensor 44.

[0024] Also, the bathroom remote controller 7 is provided with a human presence sensor 71 configured by, for example, a pyroelectric infrared sensor for detecting the presence or absence of a person in the bathroom 6, a bathroom temperature sensor 72 for detecting the temperature in the bathroom 6, and a bathroom humidity sensor 73 for detecting the humidity in the bathroom 6.

[0025] The controller 5 acquires various data detected by various sensors (41-44, 71-73, etc.) and controls the operation of the water heater unit 4. For this purpose, the controller 5 includes a storage unit that stores various data and control programs, and a processing circuit which is composed of a microcontroller having a CPU and memory (RAM and ROM, etc.) and performs various calculations based on the control program. The controller 5 is configured to perform, for example, automatic bath control which supplies hot water to the bathtub 60 until the water stored in the bathtub 60 reaches a set water level or set amount and the temperature of the water stored in the bathtub 60 reaches a set temperature, and reheat control which reheats the water stored in the bathtub 60 (bathtub water) using the reheat function.

[0026] Furthermore, the controller 5 stores various data in its storage unit, with the date and time of acquisition added to the data acquired from various sensors, and transmits this data to the management server 21 via the relay device 9 at predetermined intervals.

[0027] The management server 21 is equipped with a storage device 20 and stores various data sent from the water heater 1, associating it with a water heater ID used to identify the water heater 1. Alternatively, a user ID identifying the user who owns the water heater 1 may be used as the ID for identifying the water heater 1. Although only one water heater 1 is shown in Figure 1, typically data from multiple water heaters 1 installed in each residence is sent to the management server 21. Therefore, the management server 21 stores data acquired from each water heater 1. In this embodiment, the explanation will focus on a single water heater 1.

[0028] The analysis server 22 is configured to estimate whether or not a user in the residence where the water heater 1 is installed has taken a shower, based on data acquired from the water heater 1. For this purpose, the storage device 27 of the analysis server 22 stores a trained model 26 generated by machine learning using various data acquired from the water heater 1. The analysis server 22 includes a learning unit 23 that generates a trained model by performing machine learning, which will be described later, and an estimation unit 24 that estimates whether or not a shower was taken using the trained model generated by the learning unit 23.

[0029] An example of a machine learning method using the analysis server 22 is described below. The analysis server 22 acquires various data transmitted from the water heater 1 via the management server 21 and generates explanatory variable data, which will be described later, based on this data. Then, the person in charge of labeling inputs data (labels) indicating whether or not a shower was taken into the analysis server 22, according to predetermined guidelines based on the content of the explanatory variable data. For example, data indicating that a shower was taken is entered as "1" and data indicating that a shower was not taken is entered as "0".

[0030] The learning unit 23 performs machine learning using explanatory variable data as explanatory variables and shower bathing status data as the target variable. In this process, the learning unit 23 generates a household bathing vector by creating a table of explanatory variable data and shower bathing status data based on various data of the water heater 1 collected over a predetermined period (number of days, e.g., 31 days). The explanatory variable data may include essential explanatory variables, recommended explanatory variables, and other explanatory variables. The shower bathing status data can be rephrased as the target variable data.

[0031] Figure 2 shows an example of a household bathing vector in this embodiment. Data group numbers 1, 2, 3, 4, ... are numbers assigned to data groups consisting of multiple explanatory variable data (D1 to D11) that are generated in response to each time a user enters or leaves the bathroom 6.

[0032] The entry / exit detection unit A, described later, consists of a human presence sensor 71 and a controller 5. Based on the output signal from the human presence sensor 71, the controller 5 stores the date and time when the user entered the bathroom 6 (entry date and time) and the date and time when the user left the bathroom 6 (exit date and time) as data to be sent to the management server 21. Since a user may enter the bathroom 6 and then leave for a short time for some reason and immediately return, the controller 5, in order to avoid misinterpreting such short temporary exits as exits, determines that the user has left the bathroom 6 after a predetermined time has elapsed since the human presence sensor 71 stopped detecting a person, and stores the date and time of this determination as the exit date and time.

[0033] Furthermore, the bathing entry / exit detection unit B, described later, is composed of a water level sensor 41 and a controller 5. When a person enters the bathtub 60 filled with water, the water level in the bathtub 60 rises. The water level sensor 41 can detect this change in water level based on the person entering the bathtub 60. Based on the detection result of the water level sensor 41, the controller 5 detects that a person has entered the bathtub 60 and stores the detected date and time (bathing date and time) in its storage unit as data to be transmitted to the management server 21. Also, when a person leaves the bathtub 60, the water level in the bathtub 60 decreases. The water level sensor 41 can detect this change in water level based on the person leaving the bathtub 60. Based on the detection result of the water level sensor 41, the controller 5 detects that a person has left the bathtub 60 and stores the detected date and time (bathing date and time) in its storage unit as data to be transmitted to the management server 21.

[0034] The required explanatory variables include the first data D1 and the second data D2. The first data D1 is the data of the time the user is not bathing in the bathtub 60, i.e., the non-bathing time (D1) while the user is in the bathroom 6, calculated as the difference between the time the user is in the bathroom 6 (D8), which is calculated based on the time the user is in the bathroom 6 (entry time and exit time) detected by the entry / exit detection unit A, and the time the user is in the bathtub 60 (entry time and exit time) detected by the entry / exit detection unit B. Here, the entry time (D8) is calculated as the elapsed time from the entry time to the exit time, and is the time the user is in the bathroom 6. The bathing time (D11) is calculated as the elapsed time from the bath time to the exit time, and is the time the user is in the bathtub 60. Furthermore, "while the user is in bathroom 6" refers to the period from when the user enters bathroom 6 until when they leave (the period of stay), that is, within the period from the time of entry to the time of exit.

[0035] The second data, D2, is data on the amount of heat used for hot water supply while the user is in the bathroom 6. This amount of heat used for hot water supply is calculated by the controller 5 based on measurement data from the water flow sensor 42, the inlet water temperature sensor 43, and the outlet water temperature sensor 44.

[0036] The first data D1, which represents non-bathing time, and the second data D2, which represents the amount of heat used for hot water supply, both indicate a higher probability of showering the user the more likely it was. After diligent research, the inventors of this invention have found that by combining these two types of data D1 and D2 as essential explanatory variables and inputting them into the learning unit 23, machine learning can be performed with high accuracy to determine whether or not the user of the water heater 1 took a shower as the target variable. This invention is based on this finding.

[0037] Furthermore, the recommended explanatory variables are the third data D3, the fourth data D4, the fifth data D5, and the sixth data D6, and it is preferable to input at least one of these as an explanatory variable into the learning unit 23. The third data D3 is data of the entry time obtained from the date and time (date and time) of the user's entry into the bathroom 6 detected by the entry / exit detection unit A. The fourth data D4 is data of the hot water usage flow rate, which is the hot water flow rate when hot water is supplied while the user is in the bathroom 6. The hot water usage flow rate is measured by the water flow sensor 42.

[0038] Furthermore, the fifth data point, D5, represents the range of change in bathroom temperature measured by the bathroom temperature sensor 72 while the user is in the bathroom 6, and can be calculated by subtracting the minimum value from the maximum value of the bathroom temperature during the user's stay.

[0039] Furthermore, the sixth data point, D6, represents the range of change in bathroom humidity measured by the bathroom humidity sensor 73 while the user is in the bathroom 6, and can be calculated by subtracting the minimum value from the maximum value of the bathroom humidity during the user's stay.

[0040] Other explanatory variables may include, for example, data D7 for the time of leaving bathroom 6, data D8 for the time of entering, data D9 for the time of bathing, data D10 for the time of leaving the bath, and data D11 for the time of bathing. Note that the data D7 for the time of leaving is obtained from the data for the date and time of leaving. Similarly, the data D9 for the time of bathing is obtained from the data for the date and time of bathing, and the data D10 for the time of leaving the bath is obtained from the data for the date and time of leaving the bath.

[0041] In the above, for example, data D2 to D11 are included in the various data transmitted from the controller 5 of the water heater 1 to the management server 21. The non-bathing time data D1 is calculated by the analysis server 22 using the room entry time data D8 and the bathing time data D11.

[0042] Furthermore, the dependent variable is Do, which is data indicating whether or not a shower was taken, and Do-1 to Do-4, etc., are data points of "1" or "0" assigned by the person in charge, as mentioned above.

[0043] The learning unit 23 then constructs a supervised learning model using, for example, explanatory variable data including the first data D1, the second data D2, and the third data D3 as explanatory variables, and the corresponding presence or absence of showering as the target variable. The learning model can employ one or a combination of any one of the known learning models, such as neural networks, support vector machines, decision trees, and gradient boosting decision trees.

[0044] First, the learning unit 23 performs data cleansing on the explanatory variable data and the data on whether or not showers are taken in the household bathing vector. During data cleansing, the learning unit 23 checks for outliers or missing data that fall outside a predetermined assumed range, and if outliers or missing data are found, it deletes the corresponding data (explanatory variable data and data on whether or not showers are taken).

[0045] Next, the learning unit 23 standardizes the explanatory variable data after data cleansing for each data set D1, D2, D3, ... The standardization may be, for example, normalization which scales the minimum value to 0 and the maximum value to 1, or standard deviation which scales the mean to 0 and the variance to 1. By performing standardization, the impact of differences in scale between multiple data sets (for example, between the first data set D1 and the second data set D2, etc.) on machine learning can be reduced.

[0046] Figure 3 shows an example of a neural network that can be used as a learning model in this embodiment. The neural network 25 includes an input layer N1 having m input nodes, an output layer N2 having 2 output nodes, and a hidden layer N3 between the input layer N1 and the output layer N2. The hidden layer N3 consists of one or more layers. The number of input nodes m corresponds to the number of data types in the explanatory variable data. The number of output nodes (2) corresponds to the case with and without showering.

[0047] In this type of learning model, the standardized explanatory variable data is input as explanatory variables to each input node of the input layer N1. The example in Figure 3 shows an example where three explanatory variables are input: the first data D1, the second data D2, and the third data D3 from each data set of household bathing vectors. In other words, in the example in Figure 3, the number of input nodes m is 3.

[0048] Each output node in the output layer N2 outputs the probability P1 that a shower is taken and the probability P2 that a shower is not taken, respectively, given that the explanatory variable in a given dataset is input to the input layer N1. In supervised learning, the system is trained to maximize the probability at the output node corresponding to the label (whether or not a shower is taken) assigned by the person in charge, given that the explanatory variable in a given dataset is input to the input layer N1.

[0049] In this type of supervised learning, the explanatory variables input to the input node of the input layer N1 may include recommended explanatory variables. That is, in addition to the required explanatory variables, the first data D1 and the second data D2, at least one of the third to sixth data D3, D4, D5, and D6 may be input to the input node of the input layer N1. The third data D3, the time of entry into the room, is data that is influenced by the user's lifestyle, specifically whether or not they take a shower. The values ​​of the fourth data D4, the hot water usage flow rate, the fifth data D5, the range of change in bathroom temperature, and the sixth data D6, the range of change in bathroom humidity are all considered to be larger when a shower is taken compared to when a shower is not taken. Note that the maximum value of the bathroom temperature may be used as the fifth data D5 instead of the range of change in bathroom temperature. Also, the maximum value of the bathroom humidity may be used as the sixth data D6 instead of the range of change in bathroom humidity.

[0050] Therefore, by inputting at least one of the recommended explanatory variables (data sets D3, D4, D5, D6) into the learning unit 23, in addition to the essential explanatory variables (data sets D1 and D2), machine learning with the presence or absence of showering as the target variable can be performed with greater accuracy. Furthermore, any of the other explanatory variables (data sets D7 to D11) may also be input into the learning unit 23 as explanatory variables.

[0051] Machine learning is repeatedly performed on multiple data sets, separated by data set number, using explanatory variable data and target variable data, thereby generating a trained model 26 consisting of a neural network 25. The generated trained model 26 is stored in the storage device 27 of the analysis server 22.

[0052] Next, we will describe the case in which the estimation unit 24 estimates whether or not the user is taking a shower using the trained model 26. For example, when the management server 21 receives various data from the water heater 1 at a predetermined timing, it stores the data in the storage device 20 and transmits it to the analysis server 22.

[0053] Figure 4 is a diagram showing the schematic configuration of the estimation unit 24 in this embodiment. The estimation unit 24 of the analysis server 22 includes an input unit 28, a trained model 26, and an output unit 29. The estimation unit 24 of the analysis server 22 generates necessary explanatory variable data based on various data received from the water heater 1. The necessary explanatory variable data is the same type of data as the explanatory variable data used during training (data D1, D2, D3 in the example of Figure 3). The input unit 28 of the estimation unit 24 inputs the explanatory variable data D1-x, D2-x, D3-x for data group number x (where x is an integer) to the input node of the input layer N1 of the trained model 26 stored in the storage device 27.

[0054] When explanatory variable data for the water heater 1 is input to the input node of the trained model 26, the probability P1 of showering and the probability P2 of not showering, which are assigned to each output node in the output layer N2, are output. The output unit 29 of the estimation unit 24 outputs the estimation result for whether or not showering is available, which is assigned to the output node Pi (i=1,2) that has the larger of the probabilities P1 and P2 output from each output node of the output layer N2.

[0055] The estimation results output from the estimation unit 24 are associated with the water heater ID of the explanatory variable data used for estimation and stored in the storage device 20 of the management server 21. The estimation unit 24 may estimate whether or not a shower is being taken each time it receives various data from the water heater 1, or it may estimate whether or not a shower is being taken at predetermined time intervals using various data from the water heater 1 stored in the storage device 20.

[0056] The management server 21 transmits information corresponding to the estimation results of the estimation unit 24 stored in the storage device 20 to the water heater 1. The water heater 1 displays the information corresponding to the estimation results of the estimation unit 24 on the monitor 80, which is an alarm device, to inform the user. In this case, for example, if the estimation results of the estimation unit 24 for multiple data sets over a predetermined period show a high frequency of no showers, the management server 21 may transmit information such as the effects of showers and encouragement to take showers to the water heater 1 as information corresponding to the estimation results to inform the user. This can motivate the user to take more showers. Furthermore, if the estimation results of the estimation unit 24 over a predetermined period show a high frequency of showers, or if the frequency of no showers and showers are roughly equal, the management server 21 may transmit information such as the effects of showers and baths as information corresponding to the estimation results to the water heater 1 to inform the user. This can encourage users to continue taking showers, or to continue taking both showers and baths.

[0057] In this way, information about the effects of showering and bathing can be provided to the user, for example, depending on the frequency of showering versus not showering over a predetermined period. For example, the effects of bathing include muscle relaxation due to buoyancy, improved blood circulation due to water pressure, and relaxation due to heat, all of which contribute to fatigue recovery. The effects of showering include some degree of improved blood circulation and relaxation, as well as massage and body cleansing effects.

[0058] Furthermore, the management server 21 may transmit information corresponding to the estimation result of the estimation unit 24 for a given data set to the water heater 1, and the water heater 1 may display this information on the monitor 80 for notification. In this case, for example, if the estimation result of the estimation unit 24 is no shower, the information corresponding to the estimation result may include information such as the effects of showering or a message encouraging showering. Also, if the estimation result of the estimation unit 24 is showering, the information corresponding to the estimation result may include information such as the effects of showering and the effects of bathing in a bathtub.

[0059] Furthermore, if a management application is installed on the user's smartphone or other communication terminal 11 that allows viewing the operating status of the water heater 1 or remotely controlling the water heater 1, the management server 21 may notify the management application of information corresponding to the estimation result of the estimation unit 24. In this case, the notification device is composed of a monitor or the like on the communication terminal 11.

[0060] Furthermore, if the user's home is equipped with a smart speaker capable of communicating with a cloud server that performs voice analysis processing via router 3 and communication network 10, and the smart speaker is associated with the water heater 1 and registered with management server 21, the smart speaker may be used as an alarm. In this case, management server 21 transmits information corresponding to the estimation result of estimation unit 24 to the cloud server, and the cloud server converts the information corresponding to the estimation result of estimation unit 24 into voice data and transmits it to the smart speaker. As a result, the smart speaker outputs the information corresponding to the estimation result of estimation unit 24 as voice.

[0061] In this embodiment, the estimation unit 24 includes first data D1, which is the non-bathing time when the user is in the bathroom 6 but not in the bathtub, and second data D2, which is the amount of heat used by the water heater 1 for hot water supply while the user is in the bathroom 6, as explanatory variables, and estimates whether or not a shower is taken using a trained model 26 generated by machine learning with whether or not a shower is taken as the objective variable. As a result, the presence or absence of a shower can be estimated with high accuracy. Furthermore, there is no need to install a dedicated device in the bathroom 6 to detect whether or not a shower is taken.

[0062] Furthermore, by adding at least one of the following explanatory variables—the third data D3 regarding the time the user enters bathroom 6, the fourth data D4 regarding the flow rate of hot water used while the user is in bathroom 6, the fifth data D5 regarding the range or maximum value of the temperature change inside bathroom 6 while the user is in bathroom 6, and the sixth data D6 regarding the range or maximum value of the humidity change inside bathroom 6 while the user is in bathroom 6—a highly accurate trained model can be generated, and this trained model can be used to more accurately estimate whether or not the user is taking a shower.

[0063] In the above, data D1 to D11 are data corresponding to the period from when the user enters the bathroom 6 until when they leave. For the entry time data D8, the controller 5 may use a timer to measure the entry time, which is the time from when the user enters the bathroom 6 as detected by the entry / exit detection unit A until they leave, and then include this measured entry time in the various data and send it to the management server 21. For the bathing time data D11, the controller 5 may use a timer to measure the bathing time, which is the time from when the user enters the bathtub 60 as detected by the entry / exit bathing detection unit B until they leave, and then include this measured bathing time in the various data and send it to the management server 21.

[0064] Furthermore, in this embodiment, the learned model 26 may be transmitted from the server 2 to the water heater 1, and the learned model 26 may be stored in the memory of the controller 5 of the water heater 1, so that the controller 5 uses the learned model 26 to estimate whether or not a shower is being taken. In this case, information to be notified according to the estimation result of whether or not a shower is being taken, such as the effects of showering and the effects of bathing in a bathtub, may be stored in the memory of the controller 5, and the controller 5 may display information corresponding to the estimation result of whether or not a shower is being taken on the monitor 80 of the kitchen remote control 8 or the like to notify the user. Alternatively, the estimation result of whether or not a shower is being taken may be transmitted from the controller 5 to the management server 21, and information corresponding to the estimation result of whether or not a shower is being taken may be notified on the communication terminal 11 or the aforementioned smart speaker.

[0065] From the above description, many improvements and other embodiments of the present invention will be apparent to those skilled in the art. Therefore, the above description should be interpreted as illustrative only and is provided for the purpose of teaching those skilled in the art the best mode of carrying out the invention. The details of its structure and / or function can be substantially modified without departing from the spirit of the invention. [Industrial applicability]

[0066] This invention is useful as a hot water supply system that can accurately estimate whether or not a shower is being taken, without requiring the installation of a dedicated device in the bathroom to detect whether or not a shower is being taken. [Explanation of Symbols]

[0067] 1 Water heater, 2 Servers, 6 Bathrooms, 24 Estimation units, 26 Learned models, 41 Water level sensor, 42 Water flow sensor, 60 Bathtub, 71 Motion sensor, 72 Bathroom temperature sensor, 73 Bathroom humidity sensor, 80 Monitor, 100 Hot water supply system

Claims

1. A water heater having an entry / exit detection unit that detects when a user enters and leaves a bathroom, and an entry / exit detection unit that detects when a user enters and leaves a bathtub in the bathroom, and which supplies hot water to hot water destinations including showers in the bathroom, and which is capable of calculating the amount of heat used for hot water supply at the time of hot water supply, It comprises an estimation unit that estimates whether or not the user has taken a shower by inputting predetermined data into a trained model generated by machine learning, The predetermined data input to the trained model is: The first data relating to the time not spent in the bathtub is calculated as the difference between the entry time, which is the time from entering to leaving the room as detected by the entry / exit detection unit, and the bathing time, which is the time from entering to leaving the bath as detected by the entry / exit detection unit, and This includes second data relating to the amount of heat used for hot water supply by the hot water heater while entering the bathroom, The aforementioned trained model is The data, including the first and second data, is used as explanatory variables, and the presence or absence of showering is used as the dependent variable, and the data is generated by machine learning. Hot water supply system.

2. The water heater includes a flow sensor for detecting the flow rate of hot water used during hot water supply, a bathroom temperature sensor for detecting the temperature inside the bathroom, and a bathroom humidity sensor for detecting the humidity inside the bathroom. The predetermined data and the explanatory variable data are, The system includes at least one of the following: third data relating to the time of entry detected by the entry / exit detection unit; fourth data relating to the flow rate of hot water used while the user is in the bathroom; fifth data relating to the range or maximum value of the temperature change inside the bathroom while the user is in the bathroom; and sixth data relating to the range or maximum value of the humidity change inside the bathroom while the user is in the bathroom. The hot water supply system according to claim 1.

3. The system further includes a notification device that notifies predetermined information corresponding to the estimation result of the estimation unit, The hot water supply system according to claim 1 or 2.

4. The server is connected to the water heater in a communication manner and has the estimation unit, The estimation unit generates predetermined data from the data received from the water heater and inputs it into the trained model. The hot water supply system according to claim 1 or 2.

Images