JP2013090748A - Urination information measuring apparatus - Google Patents

Abstract

PROBLEM TO BE SOLVED: To perform a measurement without keeping a user waiting and provide urination information having an accurate urine flow rate and excellent temporal responsiveness.SOLUTION: This urination information measuring apparatus includes: a urine storage means 2 for storing urine; a urine data measuring means 30 that measures a level and a weight of the urine stored in the urine storage means 2; and a urination information calculation means that calculates the urination information including a urination amount and the urine flow rate on the basis of measurement data obtained by the urine data measuring means 30. The urination information calculation means applies a prescribed vibration model to calculation data that are obtained by statistically processing the measurement data of the urine data measuring means, and processes the calculation data by a particle filter, thereby calculating the urination information.

Description

  The present invention relates to a urination information measuring device.

  Since urine is responsible for discharging unnecessary waste products from the body, medical institutions are measuring urination information such as urine output as an index for determining the state of metabolic function of the body. . As such a method, an apparatus for obtaining a change in weight or volume of urine during urination as a measurement target is used.

  Specifically, there is a weight change measurement method in which urination is continuously performed in a collection container such as a beaker placed on an electronic balance, and weight change data from the electronic balance is obtained as measurement data.

  In the volume change measurement method, urination is normally performed in a bowl of a normal toilet, and the change in the water level in the bowl at that time is measured with a pressure sensor as the head pressure change, and the pressure change data is used as measurement data. There is a toilet for measuring urination information (see, for example, Patent Document 1).

  In this device, urination information is measured by measuring the water level change due to urination of toilet bowl water set in advance at the measurement start water level, which is the predetermined initial measurement water level. By applying the relational expression, urination information such as urination volume is obtained. The change in the water level is measured by converting the pressure (water head pressure) generated by the water level of the toilet bowl water into an electrical output signal using a pressure sensor fluidly connected to the water.

  In these devices, the weight change data and pressure change data obtained at the time of measurement are not only due to actual weight changes or pressure changes, but also various signals derived from the device configuration and installation environment are noise components. As shown in FIG.

  For example, in the above-described urination information measuring toilet, the horizontal cross-sectional area of the urinal is relatively large with respect to the volume of urine to be measured, so that the water level change itself caused by urination is relatively small. It will be a thing. Therefore, in order to measure the amount of urination with high accuracy, a pressure sensor as a water level measuring means is also used with high sensitivity.

  On the other hand, the change in the water level of the water to be measured is not only due to the fluctuation due to the increase in the amount of water due to the urination originally intended for measurement, but also the fluctuation caused by the valve opening / closing operation of the water level measurement pipe for measurement preparation, It includes fluctuations caused by water surface vibration due to water entry. As a result, the output signal of the pressure sensor obtained by measurement also has a signal due to these fluctuations superimposed as noise.

  Therefore, in these apparatuses, removal of these noises is achieved by performing data processing such as averaging processing and digital filter processing on the sensor output signal as measurement data.

  In addition, when obtaining the urinary flow rate using data obtained by performing the averaging process and the digital filter process, the averaging process and the digital filter process are performed again in order to further remove noise. .

JP 2007-077755 A

  However, among these noises, for example, in the case of vibration noise having a period as low as 1 Hz or less, such as fluctuations in the stored water level in the water level measurement pipe, it has been difficult to remove by the above-described conventional processing. As a result, the rising signal of vibration noise when not urinating was mistakenly determined as a urination start state, and as a result, a measurement result such as urinating even when not urinating appeared. These urination start times and urination patterns are both important parameters for urination information, and as a result, there are cases in which an accurate diagnosis of a medical condition cannot be performed.

  In addition, when the initial measurement water level is set in advance as the urination start water level as a reference for calculating the urination amount, an error from the actual urination start water level occurs, and the sensor data of the time interval in which the urination start water level is set in advance is generated. When the average is determined, the water level obtained differs depending on the length of the time interval to be averaged. As a result, there is a problem that the measurement error of the urine output becomes large in any case. And there is an equivalent problem in measuring the water level at the end of urination.

  In addition, since vibration noise cannot be completely removed by data processing as described above, at the start of measurement, a certain amount of time has elapsed from the user's measurement start operation, and the water level fluctuation that is the source of vibration noise is at a certain level. After converging below, there was a problem that the user had to be allowed to urinate and had to wait for the user with urinary intention.

  Furthermore, since the averaging process and the digital filter process are performed when obtaining the urinary flow rate, it is impossible to measure a change in which the urinary flow rate changes suddenly in time, and as a result, an accurate symptom diagnosis cannot be performed. There was a case.

  In view of the above-described problems, the present application aims to provide a urination information measuring device that can provide urination information that can be measured without waiting for the user, the urine flow rate is accurate, and has excellent temporal response. To do.

  In order to solve this problem, the present inventor has made various studies. Conventional methods used simple averaging of sensor signals and removal of vibration noise using a digital filter, but the vibration of water in the toilet pipe is as low as 1 Hz or less, so noise removal is almost impossible. The above problems are caused by the above. The inventor who paid attention to this point creates a physical model that divides the sensor signal into a water level and a vibration term according to the amount of urination, and makes a future prediction of the vibration to clarify the urination start water level and urination start time. As a result, the inventors have come to know a technology that can be prepared with a waiting time that is less than half that of the prior art and that can provide accurate urination information.

  The present invention is based on such knowledge, urine storage means for storing urine, urine data measurement means for measuring the volume and weight of urine stored in the urine storage means, and measurement data obtained by the urine data measurement means. And a urination information calculation means for calculating urination information including urination volume and urine flow rate based on the urination information measurement device, wherein the urination information calculation means is obtained by statistically processing the measurement data of the urine data measurement means The urination information is calculated by applying a predetermined vibration model to the data for calculation and processing by a particle filter.

  In such a urination information measuring device, by applying a statistical method using processing by a particle filter to the source data, the influence of vibration noise is effectively eliminated, and the future urination is predicted by predicting the future of vibration. The starting water level and urination start time can be clarified. Therefore, according to this urination information measuring device, it is possible to obtain more accurate calculation data and urination information with a simpler arithmetic process than in the past.

  In the above-described urination information measuring device, it is preferable that the urine storage means is a bowl of a Western-style toilet, and the measurement data is water level data of stored water in the bowl. In such a urination information measuring device, since a toilet is used, post-processing of the measured urination is facilitated.

  In the urination information measuring device, it is preferable that the calculation data includes at least each water level at the start of urination or at the end of urination or the water level change rate. In such a case, an important element as urination information can be accurately obtained by simple processing.

  Further, in the case of processing by the particle filter to which the vibration model is applied in the urination information measuring device, the state of urination, that is, the vibration model applied at least before urination, during urination, and after urination can be changed to calculate urination information. preferable. By performing processing according to the state of the measurement data, the calculation data can be acquired with good followability.

  In the urination information measuring device, the vibration model is preferably an autoregressive model or a trigonometric function model. In this case, since the arithmetic processing for noise removal can be performed with a small number of variables, it is possible to use an inexpensive urine information calculating means.

  According to the present invention, since the urination start water level and the urination start time can be clearly defined, accurate urination information can be provided, and the urination start time can be easily determined even when vibration noise exists. Measurement can be performed without waiting for the person, and the urine flow rate is accurate and the urination information excellent in time response can be provided.

It is a perspective view which shows the installation state of the urination information measurement toilet in one Embodiment of the urination information measurement apparatus of this invention. It is a figure which shows the whole structure of a urination information measurement toilet bowl with the combination of sectional drawing of a toilet part, and the block diagram of a measurement function part. It is a figure which shows the structural example of a urination information measurement toilet, and represents the mode at the time of preparation operation | movement (pressure sensor output calibration). It is a figure which shows the structural example of a urination information measurement toilet, and represents the mode at the time of preparation operation | movement (measurement pipe water level switching). It is a figure which shows the structural example of a urination information measurement toilet, and represents the mode at the time of measurement operation | movement. It is a figure which shows the structural example of a urination information measurement toilet, and represents the mode of the measurement pipe | tube rehydration operation | movement at the time of the measurement start water level creation after the completion | finish of a measurement. It is a figure which shows the structural example of a urination information measurement toilet, and represents the mode at the time of installation adjustment operation (measurement line air bleeding). It is a figure which shows the structural example of a urination information measurement toilet, and represents the mode at the time of installation adjustment operation (calibration curve acquisition). It is a flowchart which shows an example of the measurement operation | movement of the urination information measurement toilet accompanying the urination information measurement. It is a flowchart which shows an example of the urination information calculation process by the urination information measurement toilet. It is a graph which shows the pressure change which superimposed the pressure change by the stored water level which rises according to the amount of urination, and the pressure change by a wave when it urinates to the urination information measurement toilet. It is a graph in the case of estimating a ripple noise using an actual sensor data estimated value. It is a graph which shows the difference value (it subtracts the data before n (DELTA) t time from the data of the present time) of sensor data. It is a graph which shows the example which judges the urination start. It is a graph which shows the example which judges the urination start by a difference. It is a graph explaining the vibration noise removal by the method using the conventional digital filter. It is a graph which shows the example of the urine flow rate when misjudging the urination of FIG. It is a graph which shows an example of the data for calculation estimated with the trend model of the present invention based on the result of having measured the pressure change by the stored water level which rises according to the amount of urination with the pressure sensor when urinating in the toilet for measuring urination information . An example of the result of estimating the urinary flow rate by the method of the present invention based on the result of measuring the pressure change due to the water level of the accumulated water surface that rises according to the amount of urination when urinating in the toilet for measuring urination information according to the present invention It is a graph which shows.

  Hereinafter, the configuration of the present invention will be described in detail based on an example of an embodiment shown in the drawings.

  The urination information measuring apparatus according to the present invention is an apparatus having urine storage means, urine data measurement means, and urination information calculation means, and enables urination information to be calculated. In the following, a case where the urination information measurement device is attached to a Western-style toilet (hereinafter referred to as a urination information measurement toilet) will be exemplified. First, an outline of the urination information measurement toilet 101 will be described and then the entire configuration will be described in detail. Then, after that, a series of measurement operations such as switching of the urinary information measuring toilet 101 accompanying the urination information measurement and the calculation theory of urination information will be described in order.

  FIG. 1 shows a state in which the urination information measuring toilet 101 in the present embodiment is installed in a measurement booth. The urination information measuring toilet 101 according to the present embodiment includes a Western-style toilet 102, a cabinet 104 that houses various functional units that operate the urine information measuring toilet 101, and a control unit 120, and a urine information measuring toilet 101. An operation / display unit 122 for operating and referring to measurement results. The urination information measuring toilet 101 may be further provided with a urine component measuring device for measuring the concentration of a specific component in urine and a sanitary washing device for washing a user's local area, both of which are urinated. It is not a necessity for information measurement. The Western-style toilet 102 is made of earthenware, and a resin toilet seat 110 is rotatably attached to an upper portion thereof, and is provided with a toilet cleaning function unit similar to a normal toilet.

  On the wall of the measurement booth where the urine information measuring toilet 101 is installed (for example, a private room partitioned in the toilet), the urine information measuring unit remote control 134 is provided as an operation / display unit 122 for operating the urine information measuring toilet 101. The sanitary washing device remote control 132 for operating the sanitary washing device is also provided, but a remote control having both functions may be installed.

  The urination information measuring unit remote controller 134 is operated when the subject enters the toilet and measures urination information. The urination information measuring unit remote control 134 indicates that the urination information measurement toilet 101 indicates a measurement start switch and indicates that the urination has ended. A urination end switch is provided. The urine information measuring unit remote controller 134 may be provided with a reading unit such as a personal authentication switch or an ID card in order to manage the measurement data by individual. In addition, the measurement result may be disclosed to the subject on the display unit of the urination information measuring unit remote control 134 or may be disclosed to the subject using the printer 141. Furthermore, assuming that the urination information measurement toilet 101 in the present embodiment is installed in a medical institution, the nurse operates the urination information measurement unit remote control 134 to collect a plurality of subject data at a predetermined time, It is also possible to retrieve data from a printer.

  In FIG. 2, the whole structure of the urination information measurement toilet 101 of this embodiment is shown with the combination of sectional drawing of a toilet part, and the block diagram of a measurement function part.

  A Western-style toilet 102 constituting the urine information measuring toilet 101 includes a toilet cleaning mechanism similar to a normal Western toilet, and is connected to a sewage pipe (not shown) via a drain socket 9. In the sewage pipe, the trapped water 4 is formed by the trap part 5, and the sanitation is taken into consideration so that odors and sanitary pests generated in the sewage pipe do not enter the toilet.

  The Western-style toilet 102 has a bowl 2 where a subject excretes, a rim water discharge nozzle 7 installed at the top of the bowl 2 for discharging water toward the inner surface of the bowl, and a jet water discharge nozzle installed at the bottom for discharging water toward the trap section 5. 8 are formed.

  The water level setting means 6 includes a rim water supply means 71 and a jet water supply means 81 for supplying water to the rim water discharge nozzle 7 and the jet water discharge nozzle 8 respectively, and further a water replenishing means 91 for supplying a small amount of water to the jet water discharge nozzle 8. It has.

  In the urination information measuring toilet 101, after the measurement of the accumulated water level for measuring the amount of urination is completed, the inner surface of the bowl 2 is washed by the water discharged from the rim water discharge nozzle 7, and thereafter, from the jet water discharge nozzle 8. The stored water 4 is sent out to the sewer pipe together with the excrement by a siphon phenomenon that occurs inside the trap portion 5 due to the discharged water.

  The supply flow path from the jet water supply means 81 to the jet water discharge nozzle 8 has a branching portion 82, and a pressure guiding water path 83 to the stored water level measuring means 14 is branched and connected. The pressure guiding channel 83 is for transmitting the water level head of the stored water 4 to the accumulated water level measuring means 14 in order to measure the accumulated water level. It is branched and connected to the water supply path to the bowl 2 when the measurement start water level is formed.

  In the urine information measuring toilet 101, after the discharge of the accumulated water is once completed, the measurement start water level which is the initial water level of the measurement start is controlled by controlling the supply amount of the water supply from the rim water supply means 71 and the zette water supply means 81. Form roughly. Then, by switching to the water supply from the replenishing means 91 and supplying a small amount of water into the bowl 2, the water level of the stored water 4 is accurately set to the measurement start water level necessary for the measurement.

  A sewage pressure fluctuation measuring means 16 is connected to a drain socket 9 serving as a connection portion with a sewage pipe (not shown) through a sewage pipe connection port 10. The water level setting means 6, the stored water level measuring means 14, the operation / display unit 22, and the sewage pressure fluctuation measuring means 16 are connected to a common control unit 15, respectively for controlling the water supply operation, measuring the stored water level, Control of various operations such as operation reception processing of the subject and sewage pressure measurement is performed.

  The control unit 15 includes a measurement value correcting unit 17 and a urination information calculating unit 18. The measurement value correction means 17 corrects the water level measurement value of the stored water level measurement means 14 to the state where there is no sewage pressure fluctuation by the sewage pressure measurement value of the sewage pressure fluctuation measuring means 16. The corrected information regarding the stored water level is converted into the change information of the stored water amount based on the calibration relationship between the water level and the stored water amount, and the urine information and the urine flow rate are calculated from the stored water amount change information by the urine information calculating means 18. The urination information is calculated.

  The urination information measuring toilet 101 further includes various measurement results calculated by the urination information calculation means 18 and measurement environment information such as personal authentication results and measurement times, in addition to the subject, doctors, nurses, and the like who use the measurement data. External output means (not shown) that communicates with other medical personnel and facility managers who manage the operation of the urination information measuring toilet.

  In this embodiment, the water level detecting means of the stored water level measuring means 14 is set in the measuring pipe 31 and the measuring pipe 31 that are in fluid communication with the stored water 4 in the bowl through a pipe line, and is proportional to the water level change of the stored water. A pressure sensor S1 (to be described later) is used. In the case of this configuration, since the water level proportional to the water level of the stored water 4 can be measured at a location separated from the bowl 2, the overall configuration of the apparatus is simplified. As another example of the water level detection means, there is one that measures the surface position of the accumulated water 4 by a non-contact ultrasonic displacement sensor or the like, and since it does not contact the accumulated water in the bowl that becomes dirty water, it operates with high reliability. Can be expected.

  Further, the stored water level measuring means 14 includes a vibration isolating means 14a for removing the influence of minute vibrations such as ripples of the stored water 4 on the water level measurement, and a calibration means 14b for performing a calibration operation of the output of the pressure sensor S1. Has been.

  The sewage pressure fluctuation measuring means 16 measures the amount of influence of the pressure fluctuation in the sewage pipe on the level of the stored water. In this embodiment, the pressure sensor is used to directly measure the pressure fluctuation in the sewage pipe. However, as another example, a configuration may be adopted in which the amount of change in the stored water level caused by the pressure fluctuation in the sewage pipe is measured using a water level sensor.

  Next, a configuration example of the urination information measuring device will be described. FIG. 3 and the like schematically show configurations of piping, sensors, and the like in the urination information measuring toilet 101 of the present embodiment.

  In FIG. 3, a replenishing tank 60 is a tank that temporarily stores water supplied from a water supply source (not shown) for supplying water to each part in accordance with the operation of the urination information measuring toilet 101. Is connected to the jet water discharge nozzle 8 by another pipes 36 and 37, and is in communication with the stored water 4. A pipe 34 communicating with the drain tank 61 is branched and connected to the pipe 36, and a drainage pump P2 is provided on the way. Each pipe is appropriately provided with on-off valves V1 to V5 for forming a predetermined pipe line corresponding to each operation.

  The measurement pipe 31 is for forming the same water level as the stored water 4 in the bowl 2 through the pipes (pressure guiding paths) 36 and 37 and the jet water discharge nozzle 8, and the pressure in contact with the water in the measurement pipe 31. The sensor S1 outputs a signal proportional to the water level. In the present embodiment, the water level of the stored water 4 is estimated by measuring the water level in the measuring pipe 31 using the pressure sensor S1, and the measuring pipe 31 is connected to the water level of the stored water level measuring means 14 together with the pressure sensor S1. Functions as a measuring means. Reference sign S <b> 2 is a pressure sensor that functions as the sewage pressure fluctuation measuring means 16 by measuring the pressure in the drain tank 61 through the pipe 35 connected to the sewage pipe connection port 10.

  The urine storage means of the urination information measuring device in the present invention has a function of temporarily storing urine. For example, in the case of this embodiment, the bowl 2 of the urination information measuring toilet 101 functions as a urine storage means (see FIG. 2 and the like). The urine of the subject is temporarily stored in a water storage part formed at the bottom of the bowl 2. When the bowl 2 of the Western-style toilet is used as the urine storage means of the urination information measuring device as in this embodiment, after the measurement is completed, the post-processing can be easily performed simply by performing a toilet cleaning operation during normal urination / defecation. Urination after the measurement can be performed).

  The urine data measuring means in the present invention has a function of measuring the urine water level and weight stored in the urine storage means. For example, in the case of the present embodiment, since the water level formed by the stored urine in the bowl 2 (urine storage means) that stores the urine in a mixed state with the stored water is measured, the measurement tube described above is used. 31, the pressure sensor S <b> 1 constitutes the urine data measuring means 30.

  Further, based on the measurement data obtained by the urine data measurement means 30, the urination information calculation means 18 calculates urination information including the amount of urination and the urine flow rate. As will be described in detail later, in this embodiment, urination is performed using such urination information calculation means 18 and using calculation data obtained by applying a predetermined vibration model to measurement data and processing by a particle filter. Information is to be calculated.

  Subsequently, an operation of the urination information measurement toilet 101 associated with the urination information measurement, such as switching the pipelines, will be described (see FIGS. 3 to 10).

  When the subject performs a measurement start operation by, for example, depressing the measurement start switch of the urination information measuring unit remote control 134 (step S101 shown in FIG. 9), the measurement tube 31 is opened at a predetermined height before actual output measurement. The output of the pressure sensor S1 is measured with the water level up to the end (step S102). Next, output calibration of the pressure sensor S1 is performed based on the measured value at this time (step S103).

<Preparation operation 1 (pressure sensor output calibration)>
Here, in order to absolutely calibrate the output voltage of the pressure sensor S1, the water in the replenishing tank 60 is supplied and overflowed from the open end of the atmosphere provided at a predetermined height of the measuring pipe 31. A calibration water column with a constant height (known) up to the open end of the atmosphere is created, and the water level converted output value of the pressure sensor S1 is calculated based on the measured value of the output voltage of the constant water pressure (water head pressure) generated by the calibration water column. Calibrate (see Figure 3). The water in the replenishing tank 60 is sent to the measuring pipe 31 through the pipes 32 and 33 by the water supply pump P1. The water overflowing from the open end of the measurement pipe 31 flows into the drain tank 61, and the portion exceeding the predetermined water level is drained into the drain socket 9 through the pressure guiding path 35 and the sewage pipe connection port 10. In this embodiment, the pressure sensor output calibration described above is performed every time measurement is performed, but the process of creating a calibration water column that requires time is completed in order to reduce the waiting time at the start of measurement by the subject. When the measurement is started, only the steps of output measurement (step S102) and calibration (step S103) that can be performed in a relatively short time are performed.

<Preparation operation 2 (switching water level of measuring tube)>
Next, the pipe is switched so that the measurement pipe 31 and the bowl 2 communicate with each other, and the water level in the measurement pipe 31 is set to the same level as the water level of the stored water 4 in the bowl 2 (step S105). That is, the open / close valve V5 is closed, the open / close valve V2 is opened, and the pressure guiding path 36 and the pressure guiding path 37 are communicated to bring the measuring tube 31 and the bowl 2 into communication (see FIG. 4). At this time, since the water levels before communication are different from each other, when the measurement tube 31 and the bowl 2 communicate with each other, the water level of the measurement tube 31 vibrates in the process of shifting to the same water level, and this vibration becomes the output value of the pressure sensor S1. It will appear. In addition, in FIG. 4 etc., the pipe | tube (pressure guide path) of the connected state is shown by the thick line.

<Measurement operation>
When the preparatory operation is completed, the temporal change in the water level in the urination information measuring toilet 101 is measured from the urination start to the end using the pressure sensor S1 (step S105). During this time, vibration occurs on the water surface of the bowl 2 with urination. This vibration is also transmitted to the measuring tube 31 communicating with the bowl 2, and this vibration appears in the measured value of the pressure sensor S1. The change in the water level measured here is converted into a change in the amount of stored water based on the calibration relationship between the water level and the water amount based on the bowl shape stored in advance. The change in the amount of accumulated water per unit time is the urine flow rate _St (mL / s), and the urine flow rate integrated value during urination is the urine output Vu (mL). Calculation data is created from the measurement result (step S106), and urination information is calculated (step S107). As in the present embodiment, at least at the start of urination or at the end of urination, the calculation data including the water level (water surface height) and the water level change rate is used, so that an important element as urination information can be simplified. It is possible to obtain accurately by processing. The above measurement operation is continued until the subject inputs urination end from the urination end switch of the urination information measuring unit remote control 134 (step S108).

<Toilet bowl cleaning and subsequent rehydration operation>
When the measurement operation is finished, the water supply is sequentially performed from the rim water supply means 71 and the jet water supply means 81 in the same manner as a normal toilet, and the contents including the excrement in the bowl 2 are discharged to the outside of the toilet. Again, a toilet cleaning operation is performed to form water in the bowl 2 (step S109). Thereafter, the stored water in the replenishing tank 60 is replenished from the jet water discharge nozzle 8 by the water supply pump P1 to create a measurement start water level in the bowl 2 (step S110). As an example, the measurement start water level is preferably as low as possible with a sealing depth of 50 mm or more. Therefore, in this embodiment, in order to form accurately at the said water level, after supplying water by the toilet bowl washing operation, the replenishing stored water is supplied from the replenishing tank 60 toward the jet water discharge nozzle 8 (see FIG. 6).

  When the measurement start water level is created, the water supply pipe is switched to the pipe to the measurement pipe 31 by closing the open / close valve V2 and opening the open / close valve V3 (step S111). After the pipe switching, as described above, the water in the replenishing tank 60 is supplied to create a calibration water column having a constant height (known) in the measurement pipe 31 (step 112).

  That is, the water in the replenishing tank 60 is sent to the measurement pipe 31 through the pipes 32 and 33 by the water supply pump P1. Then, the water overflowing from the open end of the atmosphere provided at the predetermined height of the measurement pipe 31 flows into the drain tank 61, and the portion exceeding the predetermined water level passes through the pressure guide path 35 and the sewage pipe connection port 10. By draining into the drain socket 9, a water column having a height up to the open end of the atmosphere is accurately created in the measuring tube 31. When the measurement tube 31 is filled with water up to the open end of the atmosphere and the next measurement preparation is completed, a series of urination information measurement operations is terminated.

  In addition, the operation | movement (installation adjustment operation | movement) performed only once only at the time of field installation of the urination information measurement toilet 101 is shown below for reference (refer FIG. 7, FIG. 8).

<Installation adjustment (air venting)>
The water in the replenishing tank 60 and the bowl 2 is passed through the pressure guide path 36 etc. so that the remaining air in the pipe does not hinder the water level measurement, and the air in the pressure measurement pipe is discharged (see FIG. 7). .

<Installation adjustment operation (calibration curve acquisition)>
In order to learn a calibration relationship, which is the relationship between the water level of the stored water 4 in the bowl 2 and the amount of stored water, A: a fixed amount of the stored water 4 in the bowl 2 is sucked with the drainage pump P2, and B: a measuring tube with the pressure sensor S1. The pressure measurement of the water column in 31 is repeated a predetermined number of times to obtain a regression equation that is a calibration relationship between the amount of accumulated water and the measured pressure value (water level) (see FIG. 8).

  Next, the theory of calculating urination information will be described (see FIGS. 10 to 19).

FIG. 18 shows sensor data as a result of measuring the pressure change due to the water level of the accumulated water surface rising according to the amount of urination when urinating into the bowl 2 of the urination information measuring toilet 101 of the present embodiment. When the urination state is continuously measured with a large-area container such as the bowl 2 as the urine storage means in the present invention, a highly sensitive sensor is used because the water level change in the container due to urine inflow is slight. However, a high-sensitivity sensor has a phenomenon that the noise of the output signal increases, resulting in a large amount of noise as shown in FIG. From such sensor data, the true value of the water level is used as a sensor data estimated value, which is called calculation data. In order to obtain the calculation data, estimation is performed using a trend model with a particle filter. _{Now, y di = y di-1} + w ydi is calculated for the data at the time i in When the primary trend model and y _di (1)
In the second-order trend model, y _di = 2y _di-1 -y _di-2 + w _ydi (2)
It can be expressed. Here, _wydi is a normal distribution function having an average of 0 and a variance _σydi .

  In this way, as shown in FIG. 18, it is possible to obtain calculation data that is an estimated value of the true value of the water level due to urine. There is an averaging process such as moving average for obtaining calculation data. However, using the trend model has better temporal follow-up to changes in urinary flow rate than using moving average.

FIG. 11 shows a pressure change in which the pressure change due to the level of the accumulated water level that rises according to the amount of urination and the pressure change due to the ripple are superimposed, and a pressure change due to the level of the accumulated water that rises according to the amount of urine. It is a graph which shows only two cases. As indicated by the solid line in FIG. 11, low-frequency vibrations of 1 Hz or less appear in the sensor estimated value due to the fluctuation of pressure due to opening and closing of the piping valve of the measurement system and the urination of the stored water surface. In FIG. 11, the sensor output in the case where this low frequency vibration does not exist (which is impossible in reality but assumed to be virtual) is displayed by a broken line. If there is no noise as in this broken line data, the urinary flow rate is urinary flow rate = (current urine volume−urine volume before n period) / (nΔt)
Can be easily obtained. Here, Δt is a sampling period.

  However, the actual sensor has wave noise and other noises, and the urine flow rate cannot be easily obtained as described above.

FIG. 12 is a diagram for estimating ripple noise using actual calculation data. The solid line in FIG. 12 is the actual sensor output, and the broken line is the estimated value of the ripple noise. Since the ripple noise is a damped oscillation, the autoregressive model is y _di = ΣA _K y _di-K + w _ydi (3)
It becomes. here
y _di ： Data for calculation at time i
A _K is a coefficient of the regression model. In particular, a vibration model such as a wave is a second-order autoregressive model y _di = 2r _i cos θ _i y _di-1 −r ² y _di-2 + Y _O (4)
Can be easily estimated using a particle filter (hereinafter referred to as PF).

Here, Y ₀ ; vibration center value r _i ; value 0 to 1 related to amplitude intensity at time i θ _i ; vibration frequency at time i.

It should be noted that the following trigonometric function model (two types) is used as an expression representing the damped oscillation instead of the expression (4)
y _di = A _i cos (θ _i t + φ _i ) + Y _O (5)
y _di = Ae ⁻ τ ^t cos (θ _i t + φ _i ) + Y _O (6)
Etc. may be used.

here,
A: amplitude intensity τ; attenuation constant φ: phase.

  A simple model with the smallest number of parameters for the particle filter is obtained by using equation (4). Hereinafter, the case where the equation (4) is used will be described.

  In the case of the urination information measuring toilet 101, the frequency of the ripple vibration is a natural vibration, and the frequency can be identified experimentally, and it can be confirmed that the frequency is about 0.5 Hz. At this frequency, if the user (subject) takes the data for about 1 second after pressing the measurement start switch to measure urine, various data necessary for estimating the ripple noise such as vibration frequency and amplitude intensity at that time are taken. Information can be calculated, and thereafter, using these pieces of information, it is possible to estimate the sensor output when it is assumed that urination does not start. That is, measurement preparation is completed in about one second after the user tries to measure urine. Conventionally, a filter such as a digital filter is used to separate a component due to urination, which is a direct current component, and a ripple noise component, which is a vibration component. However, the wave component to be separated is very close to a direct current component at a low frequency, and it has been impossible to completely separate them. Therefore, even if the user presses the measurement start switch to display the intention of urine measurement, a non-measurement time of about 5 seconds was provided so that the squealing noise is settled. However, there was a problem that the undulating portion was erroneously detected as urination.

The example of FIG. 12 is an example of a method for obtaining the urination start time by making a future prediction of the center of vibration Y ₀ that is the water level before urination and the ripples. In FIG. 12, r, θ, and Y ₀ are first obtained from PF using equation (4) and calculation data before urination. Next, r, θ, and Y ₀ and calculation data before urination are substituted into the equation (4) to predict the future of the ripple. This future prediction result is shown by a broken line in FIG. The rippling actual sensor data estimates first threshold difference e _y is given as y _d B ₁ (the first threshold value B ₁ represents the estimation error over the magnitude of the value using the PF and future predicted y _u0 components If it exceeds (set to), it is determined that urination starts, and that time is set as urination start time T _S. This utilizes the fact that the future prediction of vibration and the calculation data when not urinating are within an error smaller than this predetermined threshold.

Further, as described above, in FIG. 12, the center of vibration Y ₀ which is the water level before urination is also obtained. Similarly, since the center of vibration Y _e can be obtained after urination is finished, the amount of urination V _u shown in FIG. 10 can be easily obtained if V _u = Y ₁ −Y _e .

Further, after the start of urination, it is determined that urination has ended when the urine flow rate (second threshold B ₂ shown in FIG. 10) continues for a predetermined time interval (third threshold monitoring time W ₇ shown in FIG. 10) or more. If the time is the urination end time, the urination start time T _S is known, and therefore the urination time T _U can be easily calculated. Note that it is desirable to end the future prediction of the wave component using the equation (4) when it is determined that urination starts, in order to save the operating energy of the urine information calculation microcomputer in the control unit 15.

FIG. 13 shows the difference value of the calculation data (subtract data n times before from the current time data). When the difference is made in this way, the center of vibration vibrates with the numerical value related to the urine flow rate as the vibration center. For example, if the difference interval is 1 second, the center of vibration is the urine flow rate at that time. Now, the difference interval n △ t, the difference value at time i when the _{Y di Y di -S i n △} t = 2rcosθ i (Y di-1 -S i-1 n △ t) -r 2 (Y _di-2 -S _i-2 nΔt) (7)
As described above, the urine flow rate S _i at time i can be obtained.

  The method of FIG. 13 uses a difference value of calculation data to simplify the calculation, but causes a time delay due to the difference. In other words, there is a drawback that results in a slightly slower movement with respect to the actual time change of the urine flow rate.

On the other hand, the method of FIG. 14 estimates the urine flow rate S _i at time i using the calculation data itself, not the difference of the calculation data as shown in FIG. However, the urine flow rate without time delay is obtained. In FIG.
y _di = V _di−2 + (S _i + S _i−1 ) Δt + 2rcos θ _i (y _di−1 −V _di−2 −S _i−1 Δt) −r ² (y _di−2 −V _di−2 ) ( 8)
V _di = V _di-2 + (S _i + S _i-1 ) Δt (9)
And using the PF, the estimation may be performed by the same method as in FIG.
Here, V _di is the amount of urine at time i.
In the estimation of the urine flow rate described above with reference to FIGS. 13 and 14, it is desirable to start the calculation after the urination start determination shown in FIGS. 12 and 15 and end the calculation when the urination end determination is made.

FIG. 15 shows an example of determining the start of urination based on the difference. If the calculation data is differentiated in the same manner as in FIG. 13, the urine flow rate is 0 before the start of urination, and therefore vibration data centered on 0 is obtained. That y _di = 2r (4) eliminating the Y ₀ of formula _{i cosθ i y di-1 -r} 2 y di-2 (10)
It can be expressed by a simple expression. In this way, as in FIG. 12, the future prediction of vibration is performed assuming that there is no urination, and when the difference e _{y from} the difference value of the actual sensor data estimated value is equal to or greater than the first threshold value B ₁ set in advance, Judge that urination started. Although the method of FIG. 15 is easy to calculate, there is a drawback that the amount of urination cannot be obtained because the water level Y ₀ before the start of urination is not estimated.

The method described with reference to FIG. 15 is preferably terminated when it is determined that urination starts. Further, the method of FIG. 15 can be applied by the same method when estimating the water level Y _e after the end of urination, as described in FIG. If the water level Y ₀ at the start of urination and the water level Y _e at the end of urination can be estimated in this way, the urine output V _U can be easily estimated from the difference. As described above, by changing the application method of the vibration model before urination, during urination, and after urination, it is possible to efficiently obtain urination information to be known according to the urination status.

  FIG. 16 is a diagram for explaining vibration noise removal by a conventional method using a digital filter. In FIG. 16, the sensor output (measurement data) subjected to digital filter processing is indicated by a broken line. Even if it is processed with a digital filter, vibration with a vibration frequency of 1 Hz or less cannot be completely removed. For this reason, as shown in FIG. 16, assuming the time when the vibration is settled, a urination prohibition period corresponding to the time is set in advance (currently, for example, 5 seconds). This is because when a large vibration remains, it is erroneously determined that urination has occurred even though urination has not occurred. Meanwhile, users are forced to endure urination. Nevertheless, depending on the degree of occurrence of vibration noise, a large vibration may remain even after the urination prohibition period has passed, and as shown in FIG. 16, the still water level estimated value indicated by the one-dot chain line is larger than the urination start threshold (two-dot chain line). May be a signal. When this happens, it is erroneously determined that urination has occurred even though urination has not occurred.

  FIG. 17 is an example of the urine flow rate when urination is erroneously determined in the method using the conventional digital filter of FIG. The part where a small signal is first output is a misjudgment.

On the other hand, in this embodiment, the urination information calculation unit 18 is used, and the calculation data obtained by processing the urine data measurement unit 30 with the PF by applying a trend model as an example of statistical processing to the measurement data, Since urination information is calculated by applying a predetermined vibration model and processing by PF, when urinating to the urination information measuring toilet 101, the pressure sensor S1 measures the pressure change due to the level of the water level rising according to the amount of urination. As a result, data as shown in FIG. 19 is obtained. As can be seen from the figure, according to the present embodiment, it is possible to easily determine the urination start time even when vibration noise is present, so the urination start water level Y ₀ and the urination start time T _S become clear. , Accurate urinary flow rate (urination information) can be obtained.

  Here, an example of the urination information calculation processing procedure based on the above-described theory will be described (see FIG. 10).

After the measurement start switch is pressed, Once it takes the output signal y from the pressure sensor S1 (step S501), using the sensor signal y to estimate the sensor data y _d such as described above (step S502). Then, from the sensor data estimate y _d, static water level y ₀ at the start of urination, vibration frequency theta, to estimate the vibration intensity r (step S503), further, a still water level y _0, the vibration frequency theta, a vibration intensity r using historical sensor data y _d _ _p the current time to estimate the waving signal y _u0 when the not started urination (step S504). In the steps so far, the water level is estimated by PF using an autoregressive model as the vibration model, but the calculation data used for the estimation is data before the start of urination, and the estimation result is also urinated. There is a feature that it is estimated under the condition that it is not. As a result, if the test subject starts urination during estimation, a calculation data value different from the estimation result appears, and it is possible to easily determine the start of urination even if vibration noise exists.

Next, rippling the sensor data estimate y _d is a difference e _y of the signal y _u0 whether the first threshold value B ₁ or more, it is determined (step S505). While repeating the determination unless B ₁ or more, if the B ₁ or more, to determine the urination start time T _S (step S506), then using the method described with reference to FIG 13, the urine flow rate S _t Estimate (step S507). Thereafter, urine flow rate S _t is determined whether the second threshold value B ₂ is smaller than the state continues for the third threshold monitoring time W _T or more (step S508), the process returns to step S507 if not continue more than W _T, if continued W _T more than the process proceeds to step S509. In the steps so far, a vibration model is constructed with an autoregressive model (hereinafter referred to as a differential autoregressive model) using the difference value of the calculation data, and the urine flow rate _St is estimated with PF. Such is the most important voiding information in the section of this step by the urine flow rate S _t it can be easily estimated.

  In step S509, it is determined that urination has ended, and sensor data estimation and urine flow rate estimation are terminated (step SP509). Thereafter, the static water level Ye at the end of urination is estimated (step S510), and the urination volume Vu and the urination time Tu are calculated (step S511). Steps S510 and S511 are the same as those described in step 503, and it is possible to easily estimate the static water level Ye when there is no urination by using the sensor data estimated value after determining that urination has ended. Thereafter, initialization is performed (step S512), and a series of calculation processing ends.

  In an example of the urination information calculation process described above, the vibration model applied before, during, and after urination is changed to obtain urination information that is particularly important in that situation. As described above, it is preferable to perform the process according to the state of the measurement data because the calculation data can be easily obtained.

  The above-described embodiment is an example of a preferred embodiment of the present invention, but is not limited thereto, and various modifications can be made without departing from the scope of the present invention. For example, in the present embodiment, the autoregressive model or the trigonometric function model is exemplified as the vibration model. However, in the autoregressive model or the trigonometric function model in this case, when determining the urination start water level or the urination end water level, urination is performed. It is also preferable to use the start water level or the urination end water level, the vibration frequency and the amplitude intensity as parameters constituting the particles of the particle filter. In such a case, it is possible to accurately determine the measurement start water level and the measurement end water level necessary for knowing urination information, and as a result, it is possible to more accurately determine the urine output.

  Alternatively, when the autoregressive model or trigonometric function model is a vibration model, when calculating the urination start time or the urination end time, the calculation data is estimated using the vibration frequency and the amplitude intensity and the water level at the start of measurement, The time when the difference between the estimation result and the calculation data becomes greater than or less than a preset threshold value may be set as the urination start time or the urination end time, respectively. By doing so, it is possible to accurately know the urination start time, and as a result, there is no problem that urination occurs without urination. In addition, an accurate urination time can be determined.

  Furthermore, when the autoregressive model or trigonometric function model is used as the vibration model, when obtaining the urine flow rate, the particle size of the particle filter is composed of the measurement start water level, the water level rising component due to urination, and the component due to vibration as the vibration model. It may be a parameter. In such a case, it is possible to obtain an accurate urinary flow rate that is excellent in following the time change of the actual urinary flow rate.

  In the above description, the urine data measuring means has been described as an embodiment of the invention that measures the volume of urine using the water level of a container of a certain size as a parameter. Is applicable.

  In other words, even when measuring the weight of urine discharged into the container, the measurement data may vibrate due to vibrations during operation of the device and vibrations due to urine urination during urination. Even in that case, it is possible to obtain accurate urination information by applying the present invention and applying the vibration model to the calculation data obtained by statistically processing the measurement data and processing it by the particle filter. Become.

2: Bowl (urine storage means)
18: Urination information calculation means 30: Urine data measurement means

Claims (5)

Urine storage means for storing urine,
Urine data measurement means for measuring the volume and weight of urine stored in the urine storage means;
Urination information calculation means for calculating urination information including urination volume and urine flow rate based on measurement data obtained by the urine data measurement means;
In a urination information measuring device having
The urination information calculation means calculates the urination information by applying a predetermined vibration model to the calculation data obtained by statistically processing the measurement data of the urine data measurement means and processing the data by a particle filter. A urination information measuring device characterized by the above.   The urine information measuring apparatus according to claim 1 or 2, wherein the urine storage means is a bowl of a Western-style toilet, and the measurement data is water level data of stored water in the bowl.   The urination information measuring device according to any one of claims 1 to 3, wherein the calculation data includes at least each water level or a water level change rate at the start or end of urination.   2. The urination information is calculated by changing a state of urination, that is, at least a vibration model to be applied before urination, during urination, and after urination, when processing by a particle filter to which a vibration model is applied. Urination information measuring device.   The urination information measuring device according to claim 1, wherein the vibration model is an autoregressive model or a trigonometric function model.