Measuring instrument, measuring instrument control method, and program

Abstract

A sensor unit 10 performs measurement continuously, and outputs actual measured values of the measurement results with a predetermined cycle. A control unit 21 performs: first prediction processing (S14, S19) for predicting, on the basis of m actual measured values, the measurement result n cycles after the last measurement value; sensor control processing (S17, S18, S22) for stopping the operation of the sensor unit 10, after the sensor unit 10 has output the corresponding actual measured values, for a period from when the actual measured values were measured until the measured value a further n cycles later is measured, in accordance with a result obtained by comparing the predicted value with the actual measured value; and second prediction processing (S20) for predicting the measurement results in the period during which the operation of the sensor unit 10 is stopped, on the basis of the m actual measured values. This makes it possible to provide a low-power measuring instrument with which it is possible to obtain results equivalent to those obtained when the sensor unit is operating constantly.

Description

Measuring instrument, measuring instrument control method, and program

[0001] The present invention relates to a measuring instrument, a method for controlling the measuring instrument, and a program.

[0002] In recent years, various devices have become increasingly miniaturized and portable. For example, there is a growing demand for battery-powered water quality measuring devices that measure the pH and electrical conductivity of water. Reducing power consumption and extending battery life are important issues for battery-powered measuring devices. Conventional measuring devices have reduced power consumption by, for example, turning off or dimming the display screen.

[0003] Patent Literature 1 describes a sensor control device including a sensor, a prediction unit, a transmission unit, and a change unit. In this sensor control device, the prediction unit uses one of a plurality of prediction models to obtain a predicted value of the sensor measurement result. The transmission unit transmits the actual measurement value to a server device when the difference between the actual measurement value and the predicted value is equal to or greater than a threshold. The change unit changes the prediction model depending on the transmission frequency of the actual measurement value. This sensor control device can reduce power consumption by reducing the frequency of transmitting the actual measurement value to the server device.

[0004] JP 2016-115289 A

[0005] Some measuring devices perform continuous measurements and output the measurement results at relatively short intervals (e.g., one second). In such measuring devices, the sensor unit must be constantly operating. However, if the sensor unit is constantly operating, the power consumption of the measuring device increases and the battery of the measuring device is quickly depleted.

[0006] The present invention has been made in consideration of the above circumstances, and its object is to provide a low-power measuring instrument that can obtain results equivalent to those obtained when the sensor unit is constantly operating.

[0007] (1) A measuring device of the present invention includes a sensor unit that continuously performs measurements and outputs actual measurement values ​​of the measurement results at a predetermined cycle, and a control unit that controls the sensor unit. The control unit performs a first prediction process that calculates a first predicted value, which is a predicted value of the measurement result at a first measurement timing that is a first time later than the measurement timing of the last actual measurement value of the plurality of actual measurement values, based on a plurality of actual measurement values ​​output from the sensor unit; a sensor control process that, after the sensor unit outputs the first actual measurement value that is the actual measurement value of the measurement result at the first measurement timing, stops operation of the sensor unit during a period from the measurement at the first measurement timing to the measurement at a second measurement timing that is further delayed by the first time by the first measurement timing, depending on a result of comparing the first predicted value with the first actual measurement value; and a second prediction process that calculates a predicted value of the measurement result during the period during which the sensor unit stops operation, based on the plurality of actual measurement values ​​output from the sensor unit.

[0008] According to the above-described measuring device, the operation of the sensor unit is stopped between the first measurement timing and the second measurement timing in accordance with the comparison result between the predicted value of the measurement result at the first measurement timing and the actual measured value, thereby reducing the operation frequency of the sensor unit and reducing the power consumption of the sensor unit. Furthermore, during the period when the sensor unit is stopped, the control unit performs a second prediction process to obtain a predicted value of the measurement result of the sensor unit. Therefore, the power consumption of the measuring device can be reduced while obtaining results similar to those obtained when the sensor unit is constantly operating.

[0009] (2) Preferably, in the sensor control process, the control unit may stop operation of the sensor unit during the period when a difference between the first predicted value and the first actual measured value is less than a threshold value.

[0010] (3) Preferably, in the sensor control process, the control unit may operate the sensor unit during the period when a difference between the first predicted value and the first actual measured value is equal to or greater than the threshold value.

[0011] (4) Preferably, the threshold value may be a value based on a standard deviation of the plurality of actual measurement values ​​output from the sensor unit.

[0012] (5) Preferably, in the first prediction process, the control unit may calculate a regression curve that approximates changes over time in the actual measurement values ​​based on the plurality of actual measurement values ​​output from the sensor unit, and calculate the first predicted value based on the calculated regression curve.

[0013] (6) Preferably, in the first prediction process, the control unit may obtain a quadratic curve as the regression curve by a least squares method based on three or more of the actual measurement values ​​output from the sensor unit.

[0014] (7) Preferably, the control unit has a prediction model obtained in advance by machine learning, and in the first prediction process, the first predicted value may be obtained by providing the plurality of actual measurement values ​​output from the sensor unit to the prediction model.

[0015] (8) Preferably, the measuring instrument further includes a display unit, and the control unit may further perform processing to calculate a reliability of the first predicted value and cause the display unit to display a display corresponding to the calculated reliability.

[0016] (9) Preferably, the measuring instrument further includes a communication unit, and the control unit may further control the communication unit to transmit the actual measurement value output from the sensor unit and the predicted value calculated by the control unit to an information processing device having a display unit.

[0017] (10) Preferably, the measuring device further includes a battery storage section, the sensor section operates by receiving power from a battery stored in the battery storage section, and the control section may stop the operation of the sensor section by stopping the power supply from the battery stored in the battery storage section during the sensor control process.

[0018] (11) A method for controlling a measuring instrument of the present invention is a method for controlling a measuring instrument equipped with a sensor unit, comprising the steps of: operating the sensor unit to perform continuous measurements and output actual measurement values ​​of the measurement results at a predetermined cycle; calculating a first predicted value, which is a predicted value of the measurement result at a first measurement timing that is a first time later than the measurement timing of the last actual measurement value of the plurality of actual measurement values, based on the plurality of actual measurement values ​​output from the sensor unit; after the sensor unit outputs the first actual measurement value, which is the actual measurement value of the measurement result at the first measurement timing, stopping the operation of the sensor unit during a period from when the measurement was performed at the first measurement timing to when a measurement is performed at a second measurement timing that is further later than the first measurement timing by the first time, depending on a result of comparing the first predicted value with the first actual measurement value; and calculating a predicted value of the measurement result during the period during which the sensor unit is stopped operating, based on the plurality of actual measurement values ​​output from the sensor unit.

[0019] (12) A program of the present invention, in a measuring instrument including a sensor unit and a computer, causes the computer to execute the following steps: operating the sensor unit to perform continuous measurements and output actual measurement values ​​of the measurement results at a predetermined cycle; calculating a first predicted value, which is a predicted value of the measurement result at a first measurement timing that is later by a first time than the measurement timing of the last actual measurement value of the plurality of actual measurement values, based on the plurality of actual measurement values ​​output from the sensor unit; stopping operation of the sensor unit during a period from when the measurement was performed at the first measurement timing to when a measurement is performed at a second measurement timing that is further later by the first time than the first measurement timing, based on a result of comparing the first predicted value and the first actual measurement value, after the sensor unit outputs the first actual measurement value, which is the actual measurement value of the measurement result at the first measurement timing; and calculating a predicted value of the measurement result during the period during which the sensor unit is stopped operating, based on the plurality of actual measurement values ​​output from the sensor unit.

[0020] According to the present invention, it is possible to reduce the power consumption of the measuring device while obtaining results equivalent to those obtained when the sensor unit is constantly operating.

[0021] FIG. 1 is a block diagram showing the configuration of a water quality measuring instrument 1 according to an embodiment of the present invention. FIG. 2(A) is a diagram for explaining a first prediction process performed by the control unit 21 of the water quality measuring instrument 1. FIG. 2(B) is a diagram showing the case where m = n = 5 in FIG. 2(A). FIG. 3 is a flowchart showing the operation of the control unit 21. FIG. 4(A) is a diagram showing an example of the display screen 41 of the display unit 25 of the water quality measuring instrument 1. FIG. 4(B) is a diagram showing an example of a window 42 displayed on the display unit 33 of the information processing device 30. FIG. 5 is a diagram showing an example of a quadratic regression curve determined by the control unit 21. FIG. 6 is a diagram showing a first example of operation of the control unit 21. FIG. 7 is a diagram showing a second example of operation of the control unit 21. FIG. 8 is a continuation of FIG. 7. FIG. 9(A) is a block diagram showing the control unit 21 of a water quality measuring instrument according to a first modified example. FIG. 9(B) is a diagram schematically showing the prediction process performed by the control unit 21 of the water quality measuring instrument according to the first modified example. FIG. 10 is a block diagram showing the configuration of a water quality measuring instrument 5 according to a second modified example.

[0022] A water quality measuring instrument 1 according to an embodiment of the present invention will be described below. The water quality measuring instrument 1 is an example of a measuring instrument that measures the characteristics of water. The water quality measuring instrument 1 measures, for example, pH, ORP (Oxidation Reduction Potential), ion concentration, electrical conductivity, salinity, DO (Dissolved Oxygen), etc. It goes without saying that the embodiment described below is merely an example of the present invention, and that the embodiment can be modified as appropriate without departing from the spirit and scope of the present invention.

[0023] 1, the water quality measuring instrument 1 includes a sensor unit 10 and a main body unit 20. The sensor unit 10 includes a sensor circuit 11, a controller 12, a connection unit 13, a PWM signal generating circuit 14, and a cable 15. The main body unit 20 includes a control unit 21, an input unit 24, a display unit 25, a communication unit 26, a connection unit 27, and a battery storage unit 28. The control unit 21 includes a CPU 22 and a memory 23.

[0024] One end of the cable 15 is fixedly connected to the connection portion 13 of the sensor unit 10. The other end of the cable 15 is detachable from the connection portion 27 of the main body unit 20. By attaching the other end of the cable 15 to the connection portion 27, the sensor unit 10 and the main body unit 20 are connected so that they can communicate with each other via the cable 15. This puts the water quality measuring instrument 1 into an operable state. The cable 15 has power wiring and signal wiring inside. A sensor unit 10 exists for each measurement object. The user of the water quality measuring instrument 1 connects the sensor unit 10 corresponding to the measurement object to the main body unit 20.

[0025] The water quality measuring instrument 1 is communicably connected to an external information processing device 30. The information processing device 30 includes a CPU 31, a memory 32, a display unit 33, and a communication unit 34. The information processing device 30 is, for example, a smartphone, a tablet terminal, or a personal computer.

[0026] In the sensor unit 10, the controller 12 controls the operation of the sensor unit 10. The connection unit 13 is an interface circuit for connecting to the main body unit 20. The PWM signal generation circuit 14 receives power from the main body unit 20 and generates a PWM signal. The sensor circuit 11 performs measurements using the generated PWM signal as a drive signal. The sensor unit 10 performs measurements continuously and outputs the actual measurement results at a predetermined period. The predetermined period (hereinafter referred to as the "measurement period") is a time short enough that a user recognizes it as continuous. The measurement period is, for example, one second. The measurement period may be less than one second, or may be two or three seconds. The sensor unit 10 may output the actual measurement results as an analog signal or as a digital value.

[0027] In the main body unit 20, the control unit 21 controls the sensor unit 10 and the main body unit 20. The memory 23 is a working memory for the CPU 22. The memory 23 stores data (not shown) necessary for the operation of the CPU 22, a control program 29, and the like. The CPU 22 executes the control program 29, causing the control unit 21 to perform the operations described below. The control program 29 may be stored in a computer-readable storage medium. A computer-readable storage medium is a non-transitory medium. Non-transitory media include recording media such as ROM, RAM, EEPROM, hard disk, SSD drive, etc., as well as CD-ROM, DVD-ROM, etc. Furthermore, non-transitory media are also tangible media. On the other hand, electrical signals that carry programs downloaded from a server on the Internet, etc., are not included in non-transitory computer-readable storage media.

[0028] The input unit 24 has a function of inputting instructions from the user. The input unit 24 is, for example, a button or key provided on a housing that houses the main body 20. The display unit 25 displays a display screen including the measurement results, etc. The display unit 25 displays the measurement results digitally, for example. The display unit 25 is, for example, a liquid crystal display or an organic EL display.

[0029] The communication unit 26 is a circuit for communicating with the information processing device 30. The communication unit 26 of the main body 20 and the communication unit 34 of the information processing device 30 communicate in accordance with a wireless communication standard such as Wi-Fi (registered trademark) or Bluetooth (registered trademark). Alternatively, the communication unit 26 and the communication unit 34 may be connected to a mobile phone communication network.

[0030] The connection unit 27 is an interface circuit for connecting to the sensor unit 10. The battery housing 28 houses the battery 2. The battery 2 may be a dry battery or a rechargeable battery. The sensor unit 10 and the main body 20 operate by receiving power from the battery 2 housed in the battery housing 28.

[0031] [Outline of First Prediction Process] As described above, the sensor unit 10 continuously performs measurements and outputs actual measurement values ​​at a predetermined cycle. Hereinafter, the measurement cycle of the sensor unit 10 (more specifically, the cycle of outputting the measurement results) is defined as C, where m is an integer equal to or greater than 3 and n is an integer equal to or greater than 2. Based on the m actual measurement values ​​output from the sensor unit 10, the control unit 21 calculates a predicted value of the measurement result at a measurement timing that is n times the measurement cycle C later than the measurement timing of the last of the m actual measurement values. Hereinafter, this process will be referred to as the "first prediction process." The first prediction process is performed immediately after the last actual measurement value is obtained (after m actual measurement values ​​are obtained).

[0032] In Figure 2, white circles represent m actual measurement values ​​other than the last one, double circles represent the last one, dashed circles represent measurement results that have not yet been obtained, and black circles represent predicted values. As shown in Figure 2(A), the control unit 21 uses a first prediction process to obtain a predicted value based on m actual measurement values. This predicted value is a measurement result at a measurement timing that is n times the measurement period C later than the measurement timing of the last actual measurement value among the m actual measurement values. This predicted value can also be considered a prediction of the measurement result n values ​​after the last actual measurement value. For example, when m = n = 5, the control unit 21 uses the first prediction process to predict the measurement result five values ​​after the fifth actual measurement value based on the five actual measurement values ​​output from the sensor unit 10 (see Figure 2(B)).

[0033] [Operation of control unit 21] The operation of the control unit 21 will be described below with reference to Fig. 3. When the water quality measuring instrument 1 is powered on, the control unit 21 performs the operation shown in Fig. 3. The control unit 21 continues the operation shown in Fig. 3 until the water quality measuring instrument 1 is powered off.

[0034] When the water quality measuring instrument 1 is powered on, the control unit 21 first starts supplying power to the sensor unit 10 (S11). In S11, the control unit 21 starts supplying power from the battery 2 housed in the battery housing unit 28, thereby starting operation of the sensor unit 10. To start the supply of power from the battery 2, the control unit 21 controls, for example, a switch provided on the power wiring connecting the battery housing unit 28 and the sensor unit 10 to a conductive state. After the control unit 21 executes S11, power is supplied from the battery 2 to the sensor circuit 11, the controller 12, and the PWM signal generating circuit 14, and the sensor unit 10 continuously performs measurements and outputs actual measurement values ​​of the measurement results at a measurement period C.

[0035] Next, the control unit 21 receives and displays the actual measurement value output from the sensor unit 10 (S12). In S12, the control unit 21 digitally displays the actual measurement value received from the sensor unit 10 on the display unit 25. As a result, the display unit 25 displays, for example, a display screen 41 shown in Fig. 4(A). The display screen 41 digitally displays the measurement result (here, pH) together with an icon indicating the remaining battery level, the water temperature, and the predicted value reliability (described below).

[0036] Next, the control unit 21 determines whether or not m actual measurement values ​​have been received from the sensor unit 10 (S13). If the control unit 21 determines that m actual measurement values ​​have not yet been received (S13: No), the process proceeds to S12, and if the control unit 21 determines that m actual measurement values ​​have been received (S13: Yes), the process proceeds to S14.

[0037] In the latter case, the control unit 21 predicts the nth measurement result by a first prediction process (S14). In S14, the control unit 21 obtains a predicted value of the measurement result at a measurement timing that is n times the measurement period C later than the measurement timing of the last actual measurement value of the m actual measurement values. Next, the control unit 21 receives and displays the actual measurement values ​​output from the sensor unit 10 (S15). S15 is the same process as S12.

[0038] Next, the control unit 21 determines whether or not n new actual measurement values ​​have been received from the sensor unit 10 (S16). If the control unit 21 determines that n new actual measurement values ​​have not been received (S16: No), the process proceeds to S15. If the control unit 21 determines that n new actual measurement values ​​have been received (S16: Yes), the process proceeds to S17.

[0039] In the latter case, the control unit 21 determines whether the difference (more specifically, the absolute value of the difference) between the predicted value obtained in the first prediction process and the actual measurement value last received from the sensor unit 10 is less than a threshold value Th (S17). The control unit 21 performs S17 after the sensor unit 10 outputs the actual measurement value corresponding to the predicted value obtained in the first prediction process. Note that, when the predicted value is a value predicted from a measurement result at a certain measurement timing, the actual measurement value corresponding to the predicted value is a value obtained by actually measuring the measurement result at the same measurement timing.

[0040] If the control unit 21 determines that the difference between the predicted value and the actual measurement value is equal to or greater than the threshold value Th (S17: No), the control unit 21 proceeds to S14. In this case, the predicted value is far from the actual measurement value, and the predicted value cannot be used in place of the actual measurement value, so the control unit 21 operates the sensor unit 10.

[0041] If the control unit 21 determines in S17 that the difference between the predicted value and the actual measurement value is less than the threshold value Th (S17: Yes), the control unit 21 proceeds to S 18. In this case, the predicted value is close to the actual measurement value and can be used instead of the actual measurement value, so the control unit 21 temporarily stops the operation of the sensor unit 10 as follows.

[0042] The control unit 21 stops the power supply to the sensor unit 10 (S18). In S18, the control unit 21 stops the power supply from the battery 2 housed in the battery housing unit 28, thereby stopping the operation of the sensor unit 10. To stop the power supply from the battery 2, the control unit 21 controls the switch to a non-conductive state, for example. After the control unit 21 executes S18, the power supply from the battery 2 to the sensor circuit 11, the controller 12, and the PWM signal generating circuit 14 is stopped, and the sensor unit 10 stops operating.

[0043] Next, the control unit 21 predicts the nth measurement result using a first prediction process (S19). S19 is the same process as S14. Next, the control unit 21 predicts (n-1)th measurement results using a second prediction process (S20). In S20, the control unit 21 calculates a predicted value of the measurement result of the sensor unit 10 during a period when the sensor unit 10 is not operating, between the predicted value calculated in S19 and the predicted value calculated previously using the first prediction process.

[0044] Next, the control unit 21 waits while displaying the (n-1) predicted values ​​(S21). In S21, the control unit 21 displays the (n-1) predicted values ​​calculated in S20 on the display unit 25, one for each measurement period C. The control unit 21 waits in S21 until the elapsed time since the sensor unit 10 last measured an actual value becomes (n-1) times the measurement period C.

[0045] Next, the control unit 21 resumes the power supply to the sensor unit 10 (S22). In S22, the control unit 21 resumes the power supply from the battery 2 housed in the battery housing unit 28, thereby restarting the operation of the sensor unit 10. To resume the power supply from the battery 2, the control unit 21 controls the switch to a conductive state, for example. After the control unit 21 executes S22, the sensor unit 10 continuously performs measurements and outputs the actual measurement values ​​of the measurement results at a measurement period C.

[0046] Next, the control unit 21 receives and displays the actual measurement value output from the sensor unit 10 (S23). S23 is the same process as S12 and S15. Next, the control unit 21 proceeds to S17.

[0047] While the sensor unit 10 is operating, the display unit 25 displays the actual measurement values ​​of the measurement results output from the sensor unit 10 (S12, S15, S23). While the sensor unit 10 is stopped, the display unit 25 displays the predicted values ​​of the measurement results obtained by the control unit 21 through the second prediction process (S21). Therefore, the user can understand the measurement results of the water quality measuring instrument 1 by looking at the actual measurement values ​​and predicted values ​​displayed on the display unit 25. Furthermore, the display unit 25 displays the predicted values ​​only when it is determined in S17 that the predicted values ​​are close to the actual measurement values. Therefore, even if the operation of the sensor unit 10 is stopped and the predicted values ​​are displayed on the display unit 25 instead of the actual measurement values, the same results as when the sensor unit 10 is constantly operating can be obtained.

[0048] In the above description, the control unit 21 stops the operation of the sensor unit 10 by stopping the supply of power to the sensor unit 10 in S18. Alternatively, the control unit 21 may stop the operation of the sensor unit 10 in S18 by using another method. For example, the control unit 21 may stop the supply of power to the sensor circuit 11, control the controller 12 to enter a sleep mode, stop the supply of power to the PWM signal generating circuit 14, or stop communication between the connection units 13 and 27 in S18. The operation of the sensor unit 10 can also be stopped by these methods.

[0049] [Other Operations of Control Unit 21] In S17, in addition to the process of determining the difference between the predicted value and the actual measured value, the control unit 21 also performs a process of calculating the reliability of the predicted value and displaying a screen corresponding to the calculated reliability on the display unit 25. The display screen 41 shown in Fig. 4(A) digitally displays that the reliability of the predicted value is 95%.

[0050] The reliability indicates the degree to which the predicted value differs from the actual measurement value. The reliability can be calculated, for example, by subtracting the absolute value of the ratio of the difference between the predicted value and the actual measurement value to the actual measurement value from 100%. In this case, the reliability is 100% when the predicted value matches the actual measurement value, and 90% when the predicted value is 110% of the actual measurement value. The control unit 21 may calculate the reliability of the predicted value using a method other than the above. Furthermore, the display unit 25 may digitally display the reliability of the predicted value as a numerical value, or may switch the type, shape, size, etc. of an icon indicating the reliability of the predicted value.

[0051] The control unit 21 also controls the communication unit 26 to perform a process of transmitting the actual measurement values ​​output from the sensor unit 10 and the predicted values ​​obtained in the first prediction process and the second prediction process to the information processing device 30. The control unit 21 may perform the transmission process each time an actual measurement value or a predicted value is obtained, or may transmit a plurality of actual measurement values ​​and predicted values ​​together.

[0052] In the information processing device 30, the communication unit 34 receives the actual measurement values ​​and predicted values ​​transmitted from the water quality measuring instrument 1. The CPU 31 displays a screen on the display unit 33 that includes a graph showing the changes in the actual measurement values ​​and the predicted values ​​received by the communication unit 34. For example, the display unit 33 displays a screen including a window 42 shown in FIG. 4(B). The window 42 includes a curve approximating the temporal changes in the actual measurement values ​​and a triangle (dot pattern) with the most recent actual measurement value (double circle), the most recent predicted value (black circle), and the most recent actual measurement value (hatched circle) used to calculate the most recent predicted value as its vertices. The curve shows the trend of changes in the measurement results obtained by the sensor unit 10. The triangle shows the degree to which the predicted value and the actual measurement value differ.

[0053] [Details of Operation of Control Unit 21] The first prediction process and the second prediction process performed by the control unit 21 will be described in detail with reference to Fig. 5. Here, as an example, it is assumed that the measurement period C of the sensor unit 10 is 1 second and m = n = 5. In the first prediction process, the control unit 21 first calculates a quadratic regression curve that approximates the temporal change of the actual measurement values ​​based on five actual measurement values.

[0054] 5, the five actual measurement values ​​are designated Y1 to Y5, and the measurement timings of the respective actual measurement values ​​are designated X1 to X5. The control unit 21 uses the least squares method to generate a quadratic curve Y=PX as a regression curve based on the five value pairs (Xi, Yi) (i is an integer between 1 and 5). ２ +QX+R, where P is a number other than 0, and Q and R are arbitrary numbers. Note that methods for determining a quadratic regression curve based on a set of multiple values ​​are well known, and therefore will not be described here.

[0055] Next, the control unit 21 calculates a predicted value of the measurement result at a measurement timing 5 seconds later than the measurement timing of the fifth actual measurement value based on the calculated quadratic regression curve. When the measurement timing 5 seconds later than the measurement timing of the fifth actual measurement value is designated as Xe, the control unit 21 calculates a predicted value of the measurement result at a measurement timing 5 seconds later than the measurement timing of the fifth actual measurement value based on the calculated quadratic regression curve. ２ Ye, which is obtained when Xe is substituted for X in +QX+R, is set as the predicted value of the measurement result at a measurement timing 5 seconds later than the measurement timing of the fifth actual measurement value.

[0056] In the second prediction process, the control unit 21 predicts four measurement results based on the quadratic regression curve obtained in the first prediction process. The measurement times 1, 2, 3, and 4 seconds later than the measurement time of the fifth actual measurement value are designated as Xa, Xb, Xc, and Xd, respectively. The control unit 21 predicts four measurement results based on the quadratic regression curve obtained in the first prediction process. ２ Ya, Yb, Yc, and Yd obtained by substituting Xa, Xb, Xc, and Xd, respectively, for X included in +QX+R are used as predicted values ​​of the measurement results at each measurement timing.

[0057] In the first prediction process in S14, the intervals between the five measurement timings X1 to X5 are all 1 second (see FIG. 6, which will be described later). On the other hand, in the first prediction process in S19, the intervals between the five measurement timings X1 to X5 are 1 second or 5 seconds (see FIGS. 7 and 8, which will be described later). The control unit 21 calculates a quadratic regression curve on the assumption that the intervals between the measurement timings may vary.

[0058] In S17, the control unit 21 determines whether the difference between the predicted value and the actual measurement value is less than a threshold value Th. The threshold value Th referenced at this time is, for example, a value based on the standard deviation σ of the five actual measurement values ​​used to calculate the predicted value. For example, the threshold value Th is k times the standard deviation σ (k is a positive number). The coefficient k is a value such as 1, 2, or 3. For example, when k=3, the control unit 21 determines in S17 whether the difference between the predicted value and the actual measurement value is less than three times the standard deviation σ of the five actual measurement values.

[0059] 6 to 8, an example of operation of the control unit 21 will be described. In the following example of operation, the interval between times t1, t2, etc. is equal to the measurement cycle C of the sensor unit 10, and m = n = 5. A white circle indicates an actual measurement value, a double circle indicates a new actual measurement value, a black circle indicates a predicted value obtained by the first prediction process, a square indicates a predicted value obtained by the second prediction process, and a cross indicates that an actual measurement value cannot be obtained.

[0060] In the first operation example shown in Fig. 6, the difference between the predicted value and the actual measured value is assumed to be equal to or greater than the threshold value Th. From time t1 to t5, the actual measured value is measured at each time. The actual measured value is stored in the memory 23. At time t5, a predicted value at time t10 is predicted based on the five actual measured values ​​from time t1 to t5 by the first prediction process in S14. The obtained predicted value is stored in the memory 23.

[0061] From time t6 to t10, actual values ​​are measured at each time. At time t10, in S17, it is determined that the difference between the predicted value for time t10 predicted at time t5 and the actual value measured at time t10 is equal to or greater than threshold value Th. Therefore, in the first prediction process in S14, a predicted value for time t15 is predicted based on the five actual values ​​from time t6 to t10.

[0062] From time t11 to t15, actual values ​​are measured. At time t15, in S17, it is determined that the difference between the predicted value for time t15 predicted at time t10 and the actual value measured at time t15 is equal to or greater than threshold value Th. Therefore, in the first prediction process in S14, a predicted value for time t20 is predicted based on the five actual values ​​from time t11 to t15.

[0063] 7 and 8, the difference between the predicted value and the actual measured value is assumed to be less than the threshold value Th. The second operation example is the same as the first operation example until the actual measured value at time t10 is obtained (see FIG. 7).

[0064] Then, at time t10, in S17, it is determined that the difference between the predicted value at time t10 predicted at time t5 and the actual value measured at time t10 is less than threshold Th. Therefore, in S18, power supply to the sensor unit 10 is stopped. Next, in a first prediction process in S19, a predicted value at time t15 is predicted based on five actual measurement values ​​from times t6 to t10. Next, in a second prediction process in S20, predicted values ​​from times t11 to t14 are predicted based on five actual measurement values ​​from times t6 to t10.

[0065] From time t11 to t14, the sensor unit 10 is not operating, so no actual measurement values ​​can be obtained. Therefore, from time t11 to t14, in S21, the predicted values ​​for times t11 to t14 predicted at time t10 are displayed in sequence.

[0066] Before time t15, power supply to the sensor unit 10 is resumed in S22. At time t15, an actual measurement value at time t15 is measured. At time t15, it is determined in S17 that the difference between the predicted value at time t15 predicted at time t10 and the actual measurement value measured at time t15 is less than threshold Th. Therefore, power supply to the sensor unit 10 is stopped in S18. Next, a first prediction process in S19 predicts a predicted value at time t20 based on five actual measurement values ​​from times t7 to t10 and t15. Next, a second prediction process in S20 predicts predicted values ​​from times t16 to t19 based on five actual measurement values ​​from times t7 to t10 and t15. From time t16 to t19, the sensor unit 10 is inactive. Therefore, from time t16 to time t19, in S21, the predicted values ​​for times t16 to t19 predicted at time t15 are displayed in sequence.

[0067] Similarly, at time t20, it is determined that the difference between the predicted value and the actual measured value is less than threshold Th, power supply to the sensor unit 10 is stopped, and the first prediction process and second prediction process are performed. From time t21 to t24, four predicted values ​​are displayed in sequence. Thereafter, power supply to the sensor unit 10 is resumed, and at time t25, the actual measured value at time t25 is measured. This is similar to the process from time t25 to t30.

[0068] However, at time t20, it is determined in S17 that the difference between the predicted value at time t20 predicted at time t15 and the actual value measured at time t20 is less than threshold value Th. In the first prediction process in S19, a predicted value at time t25 is predicted based on five actual measurement values ​​measured from times t8 to t10, t15, and t20. In the second prediction process in S20, four predicted values ​​from times t21 to t24 are predicted based on the same five actual measurement values. Also, at time t25, it is determined in S17 that the difference between the predicted value at time t25 predicted at time t20 and the actual measurement value measured at time t25 is less than threshold value Th. In the first prediction process in S19, a predicted value at time t30 is predicted based on five actual measurement values ​​measured at times t9, t10, t15, t20, and t25. In the second prediction process in S20, four predicted values ​​from time t26 to time t29 are predicted based on the same five actual measured values.

[0069] In the above description, the control unit 21 is an example of a computer. The control program 29 is an example of a program. The m actual measurement values ​​are an example of a plurality of actual measurement values. n times the measurement period C is an example of a first time. A measurement timing that is n times the measurement period later than the measurement timing of the last actual measurement value of the m actual measurement values ​​is an example of a first measurement timing. A measurement timing that is further n times the measurement period later than this measurement timing is an example of a second measurement timing. The predicted value obtained in S17 is an example of a first predicted value. The actual measurement value corresponding to the predicted value obtained in S17 is an example of a first actual measurement value. S17, S18, and S22 are examples of sensor control processing.

[0070] [Effects of the embodiment] As described above, the water quality measuring device 1 (measuring device) according to the embodiment of the present invention comprises a sensor unit 10, a control unit 21, a display unit 25, a communication unit 26, and a battery storage unit 28. The control unit 21 performs a first prediction process (S14, S19) to calculate a predicted value (first predicted value) of the measurement result at a measurement timing (first measurement timing) that is n times the measurement period C (first time) later than the measurement timing of the last actual measurement value of the m actual measurement values, based on the m actual measurement values ​​(multiple actual measurement values) output from the sensor unit 10; a sensor control process (S17, S18, S22) to stop the operation of the sensor unit 10 during the period from when the measurement was performed at the first measurement timing to when the measurement was performed at a measurement timing (second measurement timing) that is n times the measurement period C later than the first measurement timing, based on the result of comparing the first predicted value and the first actual measurement value after the sensor unit 10 outputs the actual measurement value (first actual measurement value) that corresponds to the predicted value calculated in the first prediction process; and a second prediction process (S20) to calculate a predicted value of the measurement result during the period when the sensor unit 10 is stopped operating, based on the m actual measurement values ​​output from the sensor unit 10.

[0071] In addition, the control method for the water quality measuring instrument 1 according to an embodiment of the present invention includes the steps of operating the sensor unit 10 to perform continuous measurements and output actual measurement values ​​of the measurement results at a measurement period C (S11), calculating a first predicted value based on the m actual measurement values ​​output from the sensor unit 10 (S14, S19), stopping the operation of the sensor unit during the period from when the measurement is performed at the first measurement timing to when the measurement is performed at the second measurement timing (S17, S18, S22) after the sensor unit 10 outputs the first actual measurement value, depending on the result of comparing the first predicted value with the first actual measurement value, and calculating a predicted value of the measurement result during the period when the sensor unit 10 is stopped operating, based on the m actual measurement values ​​output from the sensor unit 10 (S20).

[0072] The control program 29 is a program that causes the control unit 21 to execute the four steps included in the above method in the water quality measuring instrument 1 (measuring instrument) that includes the sensor unit 10 and the control unit 21 (computer).

[0073] Therefore, according to the water quality measuring instrument 1, the control method for the water quality measuring instrument 1, and the control program 29, the operation of the sensor unit 10 is stopped between the first measurement timing and the second measurement timing in accordance with the comparison result between the predicted value and the actual measurement value of the measurement result at the first measurement timing, thereby reducing the operation frequency of the sensor unit 10 and reducing the power consumption of the sensor unit 10. Furthermore, during the period when the operation of the sensor unit 10 is stopped, the control unit 21 performs the second prediction process, thereby obtaining a predicted value of the measurement result of the sensor unit 10. Therefore, the power consumption of the water quality measuring instrument 1 can be reduced while obtaining results similar to those obtained when the sensor unit 10 is constantly operating.

[0074] Furthermore, in S17 and S18, if the difference between the first predicted value and the first actual measured value is less than the threshold value Th, the control unit 21 stops the operation of the sensor unit 10 during the above-mentioned period. This reduces the operation frequency of the sensor unit 10, thereby reducing the power consumption of the sensor unit 10 and the power consumption of the water quality measuring instrument 1. Furthermore, in S17, if the difference between the first predicted value and the first actual measured value is equal to or greater than the threshold value Th, the control unit 21 operates the sensor unit 10 during the above-mentioned period. This allows the actual measured value of the measurement result to be obtained during the above-mentioned period. Furthermore, the threshold value Th is a value based on the standard deviation σ of the m actual measured values ​​output from the sensor unit 10. By using such a threshold value Th, it is possible to suitably switch whether or not to operate the sensor unit 10 between the first measurement timing and the second measurement timing.

[0075] Furthermore, in the first prediction process, the control unit 21 calculates a regression curve (see FIG. 5 ) that approximates the temporal change of the actual measurement values ​​based on the m actual measurement values ​​output from the sensor unit 10, and calculates a predicted value (first predicted value) for the nth time period based on the calculated regression curve. Therefore, it is possible to calculate a regression curve based on the m actual measurement values ​​output from the sensor unit 10, and to calculate an appropriate first predicted value based on the calculated regression curve. Furthermore, in the first prediction process, the control unit 21 calculates a quadratic curve as a regression curve by the least squares method based on five actual measurement values ​​(three or more actual measurement values) output from the sensor unit 10. Therefore, it is possible to calculate a quadratic regression curve based on three or more actual measurement values, and to calculate an appropriate first predicted value based on the calculated quadratic regression curve.

[0076] Furthermore, the control unit 21 further performs a process of calculating the reliability of the first predicted value in S17 and causing the display unit 25 to display a display according to the calculated reliability (see FIG. 4A). Therefore, the user can recognize the degree of reliability of the predicted value calculated in the first prediction process by looking at the display content of the display unit 25.

[0077] The control unit 21 also controls the communication unit 26 to transmit the actual measurement values ​​output from the sensor unit 10 and the predicted values ​​calculated by the control unit 21 to an information processing device 30 having a display unit 33. The information processing device 30 graphically displays the actual measurement values ​​and predicted values ​​received from the water quality measuring instrument 1 on the display unit 33 (see FIG. 4(B)). Therefore, the user can easily recognize the changes over time in the measurement results of the sensor unit 10 by looking at the contents displayed on the display unit 33 of the information processing device 30.

[0078] Furthermore, the sensor unit 10 operates by receiving power from the battery 2 housed in the battery housing 28, and in S18 the control unit 21 stops the operation of the sensor unit 10 by stopping the power supply from the battery 2 housed in the battery housing 28. Therefore, in a battery-powered water quality measuring instrument 1, the frequency of operation of the sensor unit 10 can be reduced, reducing the power consumption of the sensor unit 10 and extending the life of the battery 2 of the water quality measuring instrument 1.

[0079] [Modifications] Various modifications can be made to the water quality measuring instrument 1 according to this embodiment. For example, as shown in FIG. 9A , the control unit 21 of a water quality measuring instrument according to a first modification may have a prediction model 45 previously obtained through machine learning. The prediction model 45 is obtained by performing machine learning using a large amount of data related to the measurement target in an information processing device other than the water quality measuring instrument 1. The information processing device obtains the prediction model 45 by performing machine learning using a method such as linear regression, logistic regression, support vector machine, decision tree, random forest, neural network, naive Bayes, k-nearest neighbor algorithm, or generative adversarial network. For example, the information processing device obtains the prediction model 45 by performing deep learning using a neural network.

[0080] The obtained prediction model 45 is stored in the memory 23 of the water quality measuring instrument 1. As shown in Fig. 9(B) , in the first prediction process at S14 and S19 and the second prediction process at S20, the control unit 21 obtains a predicted value by providing the actual measurement value output from the sensor unit 10 and timing information of the actual measurement value to the prediction model 45. The water quality measuring instrument of the first modified example can use a machine learning prediction model to obtain a first predicted value that is suitable as the nth-nth predicted value.

[0081] 10 , the water quality measuring instrument 5 according to the second modification includes a sensor unit 50 and a main body unit 60. The main body unit 60 is obtained by removing the connection unit 27 from the main body unit 20 of the water quality measuring instrument 1. The sensor unit 50 is obtained by removing the connection unit 13 and cable 15 from the sensor unit 10 of the water quality measuring instrument 1 and adding a battery housing unit 52 that houses the battery 3.

[0082] More specifically, the sensor unit 50 includes a sensor circuit 11, a controller 12, a PWM signal generating circuit 14, and a battery housing unit 52. The main body unit 60 includes a control unit 21, an input unit 24, a display unit 25, a communication unit 26, and a battery housing unit 28. The control unit 21 includes a CPU 22 and a memory 23.

[0083] The main body 60 and the control unit 21 operate by receiving power from a battery 2 housed in a battery housing 28 provided in the main body 60. The sensor unit 50 operates by receiving power from a battery 3 housed in a battery housing 52 provided in the sensor unit 50. The sensor unit 50 and the main body 60 are not connected by a cable. The communication unit 51 of the sensor unit 50 transmits and receives data between the communication unit 26 of the main body 60 and the communication unit 34 of the information processing device 30 in accordance with wireless communication standards such as Wi-Fi (registered trademark) and Bluetooth (registered trademark). Alternatively, the communication unit 51 may be connected to a mobile phone communication network.

[0084] In the water quality measuring instrument 5, the control unit 21 also performs the first prediction process, sensor control process, and second prediction process described above. In the sensor control process, the control unit 21 controls the communication unit 26 to send and receive data to and from the communication unit 51, thereby stopping the power supply from the battery 3 housed in the battery housing unit 52 and stopping the operation of the sensor unit 50. Therefore, the water quality measuring instrument 5 can also reduce its power consumption while obtaining results similar to those obtained when the sensor unit 50 is constantly operating.

[0085] In the above description, the water quality measuring instruments 1 and 5 are connected to the information processing device 30, but the connection configuration of the water quality measuring instruments is not limited to this. For example, the water quality measuring instruments may be connected to a server via a communication device or a communication network and transmit measurement values ​​to the server. This allows cloud computing to be performed on the measurement results from the water quality measuring instruments.

[0086] The measuring device according to the modified example may measure properties other than those of water. The measuring device according to the modified example does not necessarily have to be battery-powered. The control unit of the measuring device according to the modified example may perform operations other than those described above. For example, the control unit of the measuring device according to the modified example may calculate a cubic or higher order regression curve based on multiple actual measurement values ​​in the first prediction process. Furthermore, the number m of actual measurement values ​​used to calculate a predicted value in the first prediction process may be any number equal to or greater than 3. Furthermore, the number n indicating how many subsequent measurement results are predicted in the first prediction process may be any number equal to or greater than 2. Furthermore, the control unit of the measuring device according to the modified example may determine whether to stop the operation of the sensor unit according to criteria other than those described above.

[0087] Furthermore, in the water quality measuring instruments 1 and 5, the time from the measurement of the last actual measurement value to the first measurement in the first prediction process and the time from the first measurement to the second measurement in the sensor control process are both n times the measurement period C, i.e., fixed values. In the measuring instrument according to the modified example, the time from the measurement of the last actual measurement value to the first measurement in the first prediction process and the time from the first measurement to the second measurement in the sensor control process are not fixed values, but may be switchable between multiple values. For example, these two times may be alternately switched between 5 and 10 times the measurement period C. Furthermore, these two times may be different values.

[0088] REFERENCE SIGNS LIST 1, 5... Water quality measuring instrument (measuring instrument) 2, 3... Battery 10, 50... Sensor unit 20, 60... Main body unit 21... Control unit (computer) 25... Display unit 26... Communication unit 28... Battery storage unit 29... Control program 30... Information processing device 45... Prediction model

Claims

1. A measuring instrument comprising: a sensor unit that continuously performs measurements and outputs actual measurement values ​​of the measurement results at a predetermined cycle; and a control unit that controls the sensor unit, wherein the control unit performs a first prediction process that determines, based on a plurality of actual measurement values ​​output from the sensor unit, a first predicted value that is a predicted value of the measurement result at a first measurement timing that is a first time later than the measurement timing of the last actual measurement value of the plurality of actual measurement values; a sensor control process that, after the sensor unit outputs the first actual measurement value that is the actual measurement value of the measurement result at the first measurement timing, stops operation of the sensor unit during a period from when the measurement was performed at the first measurement timing to when a measurement is performed at a second measurement timing that is further delayed by the first time by the first measurement timing; and a second prediction process that determines, based on the plurality of actual measurement values ​​output from the sensor unit, a predicted value of the measurement result during the period when the sensor unit is stopped operating.

2. The measuring instrument according to claim 1, wherein the control unit stops operation of the sensor unit during the period when the difference between the first predicted value and the first actual measured value is less than a threshold value during the sensor control process.

3. The measuring instrument according to claim 2, wherein the control unit operates the sensor unit during the period when the difference between the first predicted value and the first actual measured value is equal to or greater than the threshold value during the sensor control process.

4. The measuring instrument according to claim 2 or 3, wherein the threshold value is a value based on the standard deviation of the plurality of actual measurement values ​​output from the sensor unit.

5. The measuring instrument of claim 1, wherein in the first prediction process, the control unit calculates a regression curve that approximates the temporal changes in the actual measurement values ​​based on the multiple actual measurement values ​​output from the sensor unit, and calculates the first predicted value based on the calculated regression curve.

6. The measuring instrument according to claim 5, wherein the control unit, in the first prediction process, uses the least squares method to determine a quadratic curve as the regression curve based on three or more of the actual measurement values ​​output from the sensor unit.

7. The measuring instrument of claim 1, wherein the control unit has a prediction model obtained in advance by machine learning, and in the first prediction process, the first predicted value is obtained by providing the plurality of actual measurement values ​​output from the sensor unit to the prediction model.

8. The measuring instrument according to claim 1, further comprising a display unit, wherein the control unit further performs processing to determine the reliability of the first predicted value and cause the display unit to display a display corresponding to the determined reliability.

9. The measuring instrument according to claim 1, further comprising a communication unit, wherein the control unit controls the communication unit to further perform processing to transmit the actual measurement value output from the sensor unit and the predicted value calculated by the control unit to an information processing device having a display unit.

10. A measuring instrument as described in claim 1, further comprising a battery storage section, wherein the sensor section operates by receiving power from a battery stored in the battery storage section, and wherein the control section, during the sensor control process, stops the operation of the sensor section by stopping the power supply from the battery stored in the battery storage section.

11. A method for controlling a measuring instrument equipped with a sensor unit, comprising the steps of: operating the sensor unit to perform continuous measurements and output actual measurement values ​​of the measurement results at a predetermined cycle; calculating a first predicted value, which is a predicted value of the measurement result at a first measurement timing that is a first time later than the measurement timing of the last actual measurement value of the multiple actual measurement values, based on the multiple actual measurement values ​​output from the sensor unit; stopping operation of the sensor unit during a period from when the measurement was performed at the first measurement timing to when a measurement is performed at a second measurement timing that is further later than the first measurement timing by the first time, after the sensor unit outputs the first actual measurement value, which is the actual measurement value of the measurement result at the first measurement timing, based on a result of comparing the first predicted value with the first actual measurement value; and calculating a predicted value of the measurement result during the period during which the sensor unit is stopped operating, based on the multiple actual measurement values ​​output from the sensor unit.

12. A program for causing the computer to execute the following steps in a measuring instrument equipped with a sensor unit and a computer: operating the sensor unit to perform continuous measurements and output actual measurement values ​​of the measurement results at a predetermined cycle; calculating a first predicted value, which is a predicted value of the measurement result at a first measurement timing that is a first time later than the measurement timing of the last actual measurement value of the multiple actual measurement values, based on the multiple actual measurement values ​​output from the sensor unit; stopping the operation of the sensor unit during the period from when the measurement was performed at the first measurement timing to when a measurement is performed at a second measurement timing that is further later than the first measurement timing by the first time, after the sensor unit outputs the first actual measurement value, which is the actual measurement value of the measurement result at the first measurement timing, based on a result of comparing the first predicted value with the first actual measurement value; and calculating a predicted value of the measurement result during the period during which the sensor unit is stopped operating, based on the multiple actual measurement values ​​output from the sensor unit.

Images