JP5601258B2 - Environmental monitoring device - Google Patents

Description

    The present invention relates to an environmental monitoring device.

  An environmental monitoring apparatus that monitors the concentration of a substance to be measured in the environment includes a sensor that measures a change in the weight of a substance to be measured attached to a vibrator, for example, a crystal vibrator, as a resonance frequency fluctuation of the vibrator, such as a QCM sensor (Quarts). Crystal Microbalance sensors) are widely used.

  Since the vibrator used as a sensor of such an environmental monitoring device is directly exposed to an environmental substance, for example, the atmosphere, it is susceptible to a large temperature fluctuation in the environment. Since the resonance frequency of the vibrator has temperature dependence, it depends on the temperature of the vibrator as well as the weight change of the adhered substance. For this reason, in order to accurately measure the amount of the substance to be measured, the temperature correction of the vibrator is indispensable.

  However, the temperature dependence of the resonance frequency of the vibrator is often different for each vibrator, and it is necessary to measure individual temperature correction coefficients for each vibrator. Conventionally, the temperature correction coefficient of the vibrator is measured by measuring the temperature dependence of the resonance frequency of the vibrator before incorporating the vibrator into the environmental monitoring device or after using the temperature monitoring device before using the environmental monitoring device. It was calculated by.

Japanese Patent Laid-Open No. 06-058319 JP 2004-294356 A

  Conventionally, the temperature correction coefficient of the vibrator has been calculated by performing a temperature test in advance before using the environmental monitoring device and measuring the temperature dependence of the resonance frequency of the vibrator. In such a temperature test, a vibrator or an environmental monitoring device is housed in a thermostat, the temperature is changed stepwise, the resonance frequency is measured while being held at a constant temperature, and the relationship between the temperature and the resonance frequency is measured. It was necessary and it took considerable man-hours.

  Furthermore, it takes a long time to reach a steady temperature after changing the temperature of the thermostatic chamber. In particular, the temperature test for environmental monitoring devices is performed in a narrow temperature range within ± 10 ° C from near room temperature, so the amount of change in temperature is small, and the time required to reach a steady temperature after raising or lowering the temperature of the thermostatic chamber. Is long. For this reason, it takes a long time to measure the temperature correction coefficient, and there is a problem that the manufacturing cost of not only the vibrator but also the environmental monitoring device increases.

  The present invention provides an environment monitoring apparatus that does not require a temperature test for calculating a temperature correction coefficient in an environment monitoring apparatus that observes a change in the weight of a substance to be measured attached to the vibrator as a change in the resonance frequency of the vibrator. The purpose is to do.

  According to one aspect of the present invention for solving the above-described problem, a resonator whose resonance frequency changes due to adhesion of a substance to be measured, and the resonance frequency at a first measurement time including a plurality of measurement times. A resonance frequency measuring device to be measured, a temperature sensor that measures an environmental temperature at a second measurement time including a plurality of measurement times, and a resonance frequency that associates the measured resonance frequency with the first measurement time. A storage device that stores the measured ambient temperature as time series data of the ambient temperature associated with the second measurement time, and the ambient temperature stored in the storage device. From the series data, the first time calculation unit that calculates the first time when the environmental temperature becomes the first predetermined temperature, and the time series data of the resonance frequency stored in the storage device, the first time calculation unit time A first resonance frequency calculation unit that calculates the resonance frequency of the first time as a first resonance frequency, and a conversion unit that converts the resonance frequency at the first time into an adhesion amount of the substance to be measured. Provided as an environmental monitoring device.

  According to the present invention, the resonance frequency when the environmental temperature becomes the first predetermined temperature is extracted from the measured time series data of the environmental temperature and the resonance frequency, and the substance to be measured is extracted from the extracted resonance frequency. Convert the amount of adhesion. Since this adhesion amount is always converted from the resonance frequency measured at a predetermined environmental temperature, it is not affected by fluctuations in the environmental temperature. Therefore, an environmental monitoring device that does not require a temperature test for calculating the temperature correction coefficient is provided.

1 is a configuration diagram of an environment monitoring device according to a first embodiment of the present invention. Functional configuration diagram of the environment monitoring apparatus of the first embodiment of the present invention Flowchart of the first embodiment of the present invention Measurement data of the first embodiment of the present invention Environmental temperature time series data according to the first embodiment of the present invention (part 1) Environmental temperature time-series data according to the first embodiment of the present invention (part 2) Resonant frequency time-series data of the first embodiment of the present invention (part 1) Resonant frequency time series data of comparative example The figure explaining the data processing in 1st Embodiment of this invention (the 1) The figure explaining the data processing in 1st Embodiment of this invention (the 2) Functional configuration diagram of the environment monitoring apparatus of the second embodiment of the present invention The flowchart of the environmental monitoring apparatus of 2nd Embodiment of this invention. The figure explaining the data processing in 2nd Embodiment of this invention (the 1) The figure explaining the data processing in 2nd Embodiment of this invention (the 2) The figure explaining the data processing in the modification of 2nd Embodiment of this invention Functional configuration diagram of the environment monitoring device of the third embodiment of the present invention The flowchart of the environmental monitoring apparatus of 3rd Embodiment of this invention. The figure explaining the data processing in 3rd Embodiment of this invention

  1st Embodiment of this invention is related with the environmental monitoring apparatus which outputs the adhesion amount of the to-be-measured substance at the time when environmental temperature corresponds to predetermined temperature as observation data.

  FIG. 1 is a configuration diagram of the environment monitoring apparatus according to the first embodiment of the present invention, and shows the main hardware configuration of the environment monitoring apparatus.

  Referring to FIG. 1, the environment monitoring apparatus 100 according to the first embodiment includes a computer in which a CPU 120, a memory 110, a storage device 30, and an output device 41 are connected to a bus 111, and an observation unit 20 connected to the bus 111. And have.

  The observation unit 20 includes a temperature sensor 2 that measures an environmental temperature, and a vibrator 1 whose resonance frequency changes due to adhesion of a substance to be measured included in the environment.

  The temperature sensor 2 only needs to be capable of measuring the environmental temperature. For example, a semiconductor temperature sensor that converts temperature into voltage or current can be used. In addition, it is possible to detect a change in temperature as a fluctuation in the resonance frequency by using a vibrator whose temperature dependence of the resonance frequency is known. The temperature sensor 2 is driven by a temperature measurement circuit 22 connected to the bus 111, and the measured temperature data is output to the bus 111.

  The vibrator 1 has an electrode formed on the surface of the piezoelectric plate. The resonance frequency of the vibrator 1 changes in accordance with the weight of the substance to be measured adsorbed on the surface of the piezoelectric plate or the electrode or by a chemical reaction. In the first embodiment, as the vibrator 1, a quartz crystal vibrator that performs AT-cut thickness-shear vibration with small temperature dependence of the resonance frequency is used. In such a crystal resonator, the temperature characteristic of the resonance frequency changes due to a slight shift in the cut direction. For this reason, the temperature characteristics of the individual vibrators 1 are different.

  It should be noted that the vibrator 1 used in the first embodiment is one in which the amount of the substance to be measured varies depending on the concentration in the environment, even if the substance to be measured accumulates and adheres. Also good. The former vibrator includes, for example, a corrosion sensor that measures the amount of corrosion of an electrode caused by a corrosive substance in the environment, and the latter vibrator includes, for example, an organic gas sensor that measures the concentration of an organic chemical substance in the environment. In the following description of the first embodiment, the vibrator 1 to which the substance to be measured is accumulated and attached, that is, the vibrator 1 to which the substance to be measured is accumulated and attached irreversibly will be described. The same applies to the vibrator 1 in which the amount of the substance to be measured changes reversibly according to the concentration in the environment.

  The vibrator 1 is driven by a resonance frequency measuring device 21 connected to the bus 111 to measure the resonance frequency, and the measured resonance frequency is output to the bus 111. The resonance frequency measuring device 21 is not particularly limited as long as the resonance frequency of the vibrator 1 can be measured. For example, the resonance frequency measuring device 21 is constituted by a Colpitts oscillation circuit including the vibrator 1 and a frequency measurement circuit for measuring the oscillation frequency. Can do. The vibrator 1 is preferably detachably connected to the resonance frequency measuring device 21 so as to be exchangeable.

  The memory 110 stores a program executed by the CPU 120. As this memory 110, for example, a semiconductor memory including either or both of ROM and RAM can be used. The storage device 30 stores the measured environmental temperature and resonance frequency data in association with the measurement time. As the storage device 30, it is preferable to use a storage device capable of storing a large amount of data with low power consumption, such as a magnetic disk device or a nonvolatile semiconductor memory (for example, a USB memory or an SD memory using a flash memory). .

  The output device 41 outputs various data, for example, observation data or data that has been subjected to various data processing, to an external device in accordance with an instruction from the CPU 120.

  The CPU 120 executes a program (including the program 10 in FIG. 2) stored in the memory 110, and controls operations of the observation unit 20, the storage device 30, and the output device 41 according to the program.

  FIG. 2 is a functional configuration diagram of the environment monitoring apparatus according to the first embodiment of the present invention, and shows main functions of the environment monitoring apparatus 100.

  Referring to FIG. 2, in the environment monitoring apparatus 100 of the first embodiment, when the CPU 120 executes the program 10, the time generation unit 11, the first time calculation unit 12a, and the first resonance frequency calculation unit 13a. And the following each function which the conversion part 17 has is implement | achieved.

  The time generation unit 11 generates a signal p2 representing the temperature measurement time and a signal p1 representing the resonance frequency measurement time. Based on the time series data D1 of the environmental temperature stored in the storage device 30, the first time calculation unit 12a calculates the first time t1 when the environmental temperature becomes a given constant temperature (first predetermined temperature). calculate. The first resonance frequency calculation unit 13a calculates the resonance frequency at the first time t1 (hereinafter referred to as “first resonance frequency f1”) based on the time series data D2 of the resonance frequency in the storage device 30. The conversion unit 17 converts the first resonance frequency f1 into the adhesion amount of the substance to be measured. Hereinafter, the functions of these units will be described in detail with reference to the flowchart of the program 10.

  FIG. 3 is a flowchart of the first embodiment of the present invention, showing the steps executed by the environment monitoring apparatus 100 according to the program 10 of FIG.

  Referring to FIGS. 2 and 3, in step S <b> 11, time generation unit 11 generates signal p <b> 2 rising at environmental temperature measurement time tp <b> 2 and signal p <b> 1 rising at resonance frequency measurement time tp <b> 1, via bus 111. Output to the observation unit 20. In the first embodiment, as the signals p1 and p2, pulses that occur at a predetermined time interval, for example, an hour interval and rise at the same timing are generated.

  Next, in step S12, when the observation unit 20 receives the signal p1, the observation unit 20 drives the resonance frequency measuring device 21 to measure the resonance frequency of the vibrator 1. Further, when the signal p2 is received, the temperature measuring circuit 22 is driven to measure the environmental temperature sensed by the temperature sensor 2. In the first embodiment, since the signals p1 and p2 rise at the same time, the resonance frequency and the environmental temperature are measured at the same time. Of course, the signals p1 and p2 having different rise times may be generated, and the resonance frequency and the environmental temperature may be measured at different times.

  Next, in step S13, the environmental temperature and resonance frequency data measured in step S12 are stored in the storage device 30. At this time, the measured environmental temperature is stored in association with the measurement time tp2, and stored as time series data D1 of the environmental temperature associated with the measurement time tp2. Further, the measured resonance frequency is stored in association with the measurement time tp1, and is stored as time series data D2 of the resonance frequency associated with the measurement time tp1.

  Next, in step S14, it is determined whether or not an observation period given in advance, for example, an observation period of one month has ended. If it is determined that the observation period has not ended, the routine of steps S11 to S14 is repeated. If it is determined that the process has ended, the next step S15 is executed. Accordingly, when it is determined in step S14 that the observation period has ended, the storage device 30 stores time series data D1 and D2 of the environmental temperature and resonance frequency measured over the entire observation period.

  Independently of the above-described steps S11 to S14, the first predetermined temperature Tc1 is set in step S18. An arbitrary temperature can be given as the first predetermined temperature Tc1. As the first predetermined temperature Tc1, for example, it can be set to room temperature, that is, 25 ° C., which is a reference for the temperature characteristics of the normal AT-cut crystal resonator 1. Thereby, the temperature dependence of the vibrator 1 is reduced, and the amount of adhesion is accurately measured. Moreover, it can also be set as the average value of the environmental temperature measured during the arbitrary period in the observation period, for example, the period of 3 days to 1 month, or the median value of the maximum temperature and the minimum temperature. Thereby, as will be described later, since the adhesion amount is calculated at many points in time, fluctuations in the adhesion amount can be calculated at precise time intervals.

  Next, in step S15, the first time calculation unit 12a calculates a time t1 (i) (hereinafter referred to as “first time t1”) at which the environmental temperature matches the first predetermined temperature Tc1. Here, i is a natural number. In step S15, first, the time series data D1 of the environmental temperature is read from the storage device 30.

  FIG. 4 shows measurement data according to the first embodiment of the present invention, in which time series data D1 and D2 of the environmental temperature and the resonance frequency stored in the storage device 30 are displayed in a graph. In addition, time series data D1 of environmental temperature is displayed in the upper part of FIG. 4, and time series data D2 of resonance frequency is displayed in the lower part. The resonance frequency is indicated by the frequency fluctuation range from the observation start time. FIG. 5 is time series data (No. 1) of the environmental temperature according to the first embodiment of the present invention, and details of the data for 1 to 3 days from the start of measurement in the time series data D1 of the environmental temperature in FIG. Represents. FIG. 6 is time series data (No. 2) of the environmental temperature according to the first embodiment of the present invention. Of the time series data D1 of the environmental temperature in FIG. Represents.

  Referring to FIG. 4, the measured environmental temperature fluctuates substantially in a daily cycle. On the other hand, the observed resonance frequency of the vibrator 1 decreases stepwise in a daily cycle. The extent to which the fluctuation of the resonance frequency is influenced by the environmental temperature that fluctuates in the daily cycle is not clear since the temperature correction coefficient of the vibrator 1 is unknown.

  2 and 5, the first time calculation unit 12a approximates the environmental temperature time-series data D1 read from the storage device 30 by an approximate expression, for example, a line graph. Next, first times t1 (1) to t1 (11) at which the approximate expression intersects the first predetermined temperature Tc1 are calculated. That is, the time at which the approximate expression of the environmental temperature coincides with the first predetermined temperature Tc1 is obtained as the first time t1 (1) to t1 (11).

  Referring to FIG. 6, the time series data of the environmental temperature for 23 to 25 days after the start of measurement does not coincide with the first predetermined temperature Tc1 = 25 ° C. Accordingly, during this period, the number of first times t1 (i) becomes zero. Thus, when the number of first times t1 (i) is zero or small, the calculation time point of the first resonance frequency, which will be described later, decreases, and only data with a rough time interval can be obtained. Therefore, the first predetermined temperature Tc1 is changed to the first predetermined temperature Tc1 ′, and the number of first times t1 ′ (i) at which the approximate expression of the environmental temperature intersects the first predetermined temperature Tc1 ′ ( It is preferable to increase the number of intersections). In the first embodiment, by setting the first predetermined temperature Tc1 ′ to Tc1 ′ = 28 ° C., the first times t1 ′ (1) to t1 ′ (14) in three days from 23rd to 25th ( 14) was calculated. The first predetermined temperature Tc1 ′ may be preset together with the first predetermined temperature Tc1, and for example, the observation is started so that the number of first times t1 ′ (i) increases with reference to the observation result. From the 23rd day to the 25th day, it may be set to the median value of the average environmental temperature or the maximum and minimum temperature for 3 days.

  Next, in step S <b> 16, the first resonance frequency calculation unit 13 a approximates the resonance frequency time-series data D <b> 2 read from the storage device 30 with an approximate expression, for example, a line graph. Then, the resonance frequency at the first time t1 (i) calculated by the first time calculation unit 12a and, if necessary, at the first time t1 ′ (i) is calculated from the approximate expression as described below. .

  FIG. 7 is time series data (part 1) of the resonance frequency according to the first embodiment of the present invention. Among the measurement data according to the first embodiment of the present invention shown in FIG. 4, the resonance stored in the storage device 30. The data measured in 1 to 3 days of frequency time-series data D2 are represented in a graph. FIG. 8 is time-series data of the resonance frequency of the comparative example, and represents the temperature-corrected resonance frequency. 7 to 8, the resonance frequency is displayed by the frequency fluctuation width (frequency difference) from the resonance frequency at the start of measurement.

  In this comparative example, the temperature correction coefficient of the vibrator 1 of the environment monitoring apparatus 100 of the first embodiment is measured in advance, and the measurement result of the first embodiment is temperature corrected using the temperature correction coefficient. . The temperature correction coefficient was −7 Hz / ° C. (−0.28 ppm / ° C.).

  Referring to FIG. 7, the resonance frequency of the vibrator 1 decreases approximately in a step shape. However, there is an increase in some sections. The vibrator 1 used here is of a type in which the substance to be measured accumulates irreversibly, and the resonance frequency must always be constant or decreased if the temperature is constant. Therefore, it is assumed that this increase is not due to the adhesion of the substance to be measured, but is caused by fluctuations in the environmental temperature. Referring to FIG. 8, the resonance frequency of the comparative example corrected using the temperature correction coefficient measured in advance fluctuates stepwise, and no increase is observed. This indicates that it is difficult to accurately measure the adhesion amount of the substance to be measured if the measurement data of the resonance frequency is directly used without temperature correction.

  FIG. 9 is a diagram (part 1) for explaining data processing in the first embodiment of the present invention, and shows a data processing method executed by the first resonance frequency calculation unit 13a. 9 represents the time series data D1 of the environmental temperature shown in FIG. 6, and the filled small circle in the lower part of FIG. 9 represents the resonance frequency shown in FIG. The series data D2 is displayed.

  Referring to FIG. 9, the first resonance frequency calculation unit 13a calculates the resonance frequency f1 (i) of the vibrator 1 at the first time t (i) calculated by the first time calculation unit 12a. Specifically, the time series data D2 of the resonance frequency read from the storage device 30 is approximated by an approximate expression such as a line graph. Then, the value of the approximate expression at the first time t (i), that is, the resonance frequency f1 (i) at the point where the line graph intersects the line representing the first time t (i) (one-dot chain line in FIG. 9). , And calculated as the first resonance frequency f1 (i). In other words, the first resonance frequency f1 (i) is calculated by linearly complementing the resonance frequencies measured at the measurement times before and after the first time t (i). Of course, it may be calculated using another approximation method.

  The calculated first resonance frequency f1 (i) is indicated by a large-diameter white circle (◯) denoted by f1 (1) to f1 (11) in FIG. Here, f1 (1) to f1 (11) represent the first resonance frequencies f1 (1) to f1 (11) at the first times t (1) to t (11), respectively.

  In the first embodiment, the adhesion amount of the substance to be measured is calculated using the calculated first resonance frequency f1 (i). That is, the line graph G1 (thick solid line) connecting the circles representing the first resonance frequency f1 (i) in FIG. 9 is regarded as the temperature-corrected resonance frequency (variation width thereof), and converted into the amount of adhesion. To do.

  As described above, the first resonance frequency f1 (i) approximates the resonance frequency at the first time t1 (i), that is, the resonance frequency when the environmental temperature reaches the first predetermined temperature. In other words, all the first resonance frequencies f1 (i) are measured at the same temperature (first predetermined temperature Tc1). That is, there is no difference in environmental temperature during measurement between the first resonance frequencies f (i). For this reason, the first resonance frequency f1 (i) is not affected by temperature fluctuations, and varies only due to fluctuations in the amount of adhesion of environmental substances. For this reason, even if the fluctuation | variation of 1st resonance frequency f1 (i) is converted into the fluctuation | variation of the adhesion amount of a to-be-measured substance, the exact adhesion amount which is not influenced by environmental temperature fluctuation | variation is calculated. As described above, in the environment monitoring apparatus 100 according to the first embodiment, it is possible to accurately measure the variation in the adhesion amount of the substance to be measured by using the measured resonance frequency of the vibrator 1 as it is without correcting the temperature. it can.

  In the section (between times) of t1 (1) to t1 (2), t1 (3) to t1 (6) and t1 (9) to t1 (10) in FIG. is there. In the vibrator 1 to which the substance to be measured is accumulated irreversibly and attached, the section where the fluctuation range of the resonance frequency is small means that the non-measurement substance is attached at a low rate and the concentration of the non-measurement substance is low. In the section in which the fluctuation range of the resonance frequency is small, the calculated fluctuation width df of the first resonance frequency (variation width from the observation start time) is df = 0 Hz and f1 (f1 (1) to f1 (2). 3) to f1 (6), df = −155 Hz, and f1 (9) to f1 (10), df = −317 Hz. This fluctuation range df coincides with the fluctuation range of the resonance frequency shown in FIG. 8 in which the temperature is corrected with 25 ° C. as the reference temperature. Further, a line graph G1 (thick solid line) connecting the first resonance frequencies f1 (i) in FIG. 9 with a straight line closely approximates the temperature-corrected comparative example data shown in FIG. In this manner, the temperature-corrected data can be approximated using the first resonance frequency without performing temperature correction.

  In order to more accurately approximate the data whose temperature has been corrected by the line graph G1, it is desirable to increase the time at which the first resonance frequency is calculated and shorten the time interval. Therefore, it is preferable to select the first predetermined temperature Tc1 so that the interval between the first times t1 (i) is short and large.

  FIG. 10 is a diagram (part 2) for explaining data processing in the first embodiment of the present invention, and shows data processing for 23 to 25 days from the start of measurement performed by the first resonance frequency calculation unit 13a. . The white circle at the top of FIG. 10 is the time series data D1 of the environmental temperature shown in FIG. 6, and the circle of small diameter at the bottom is the vibration measured from the 23rd to 25th days of FIG. The time series data D2 of the resonance frequency of the child 1 is displayed.

  Referring to FIG. 10, the first time t1 ′ (i) when the first predetermined temperature Tc1 ′ is set to 28 ° C. is 23 days to 25 days as described with reference to FIG. First times t1 ′ (1) to t1 ′ (14) at 14 time points in three days are calculated. The resonance at the first time t1 ′ (i) from the time-series data D2 of the resonance frequency read from the storage device 30 using the same method as the calculation of the first resonance frequency described above with reference to FIG. The frequency f1 ′ (i) was calculated as the first resonance frequency f1 ′ (i). The result was displayed by a large-diameter white circle (◯) marked with f1 ′ (i) in FIG. 10 and a line graph G1 ′ (thick solid line) connecting the circles.

  Since the first resonance frequency f1 ′ (i) is the resonance frequency at the same environmental temperature as the first resonance frequency f1 (i) described above, the amount of adhesion can be accurately determined without being affected by the environmental temperature. To reflect. Therefore, by converting the first resonance frequency f1 '(i) into the adhesion amount of the substance to be measured, it is possible to accurately measure the variation in the adhesion amount without correcting the temperature of the vibrator.

  The first resonance frequencies f1 (i) and f1 ′ (i) described above are resonance frequencies at the first predetermined temperatures Tc1 and Tc1 ′ having different environmental temperatures, and therefore, the first resonance frequencies f1 ( i) The data of f1 ′ (i) cannot be directly connected without temperature correction. However, there is a measurement period in which the first resonance frequencies f1 (i) and f1 ′ (i) overlap, and the first resonance frequencies f1 (i) and f1 ′ (i) smoothly change within the measurement period. If there is, both data can be easily connected. For example, a certain frequency is added to the resonance frequency of one graph, and both graphs are smoothly connected within the overlapping period. In this case, both the graphs can be connected by correcting the temperature using the value obtained by dividing the added frequency by the temperature difference between the first predetermined temperatures Tc1 and Tc1 'as a temperature correction coefficient.

  However, it is not always possible to connect both graphs smoothly. For example, the time interval between the first times t1 (i) and t1 '(i) is long, and the concentration variation of the substance to be measured, that is, the variation of the resonance frequency during this time is large. In this case, since the first resonance frequency f1 (i) and the first resonance frequency f1 '(i) fluctuate rapidly within the measurement time interval, they cannot be connected smoothly. Even in such a case, if the temperature correction coefficient can be calculated as in the second or third embodiment described later, the connection can be made with temperature correction.

  Next, referring again to FIG. 3, in step S <b> 17, the conversion unit 17 converts the frequency fluctuation amount df of the first resonance frequencies f <b> 1 (i) and f <b> 1 ′ (i) into the adhesion amount of the substance to be measured. As is well known, this conversion is performed by multiplying the frequency variation df by a conversion factor measured in advance. Note that the measurement of the conversion coefficient is performed in a short time because it is sufficient to measure the amount of fluctuation of the resonance frequency caused by the weight fluctuation of the substance adhering to the vibrator at an arbitrary constant temperature, like the measurement of the temperature correction coefficient. It doesn't take a long time. The converted adhesion amount is output as a result from the output device 41, and stored in the storage device 30 if necessary.

  The second embodiment of the present invention relates to an environment monitoring apparatus that automatically performs temperature correction using a vibrator 1 in which an adhering substance accumulates irreversibly.

  FIG. 11 is a functional configuration diagram of the environment monitoring apparatus according to the second embodiment of the present invention, and represents main functions of the environment monitoring apparatus 200 according to the second embodiment of the present invention. The hardware configuration of the second embodiment is the same as that of the environment monitoring apparatus 100 of the first embodiment shown in FIG.

  Referring to FIG. 11, the environment monitoring apparatus 200 according to the second embodiment performs the second time calculation unit 12b, the second resonance frequency calculation unit 13b, the first and second minute fluctuations by executing the program 10. The functions of the area extraction units 14a and 14b, the correction coefficient calculation unit 15, and the temperature correction unit 16 are realized. Furthermore, the environment monitoring apparatus 200 of the second embodiment has all the functions of the environment monitoring apparatus 100 of the first embodiment illustrated in FIG.

  FIG. 12 is a flowchart of the environment monitoring apparatus according to the second embodiment of the present invention, and shows steps executed by the environment monitoring apparatus 200 according to the program 10.

  Referring to FIG. 12, in the environment monitoring apparatus 200 of the second embodiment, first, in step S21, the time generation unit 11 generates signals p2 and p1 that indicate the measurement times of the environmental temperature and the resonance frequency, respectively. The function and operation of the time generation unit 11 are the same as the function and operation of the time generation unit 11 executed in step S11 of the first embodiment.

  Next, in step S22, the observation unit 20 receives the signal p2 and causes the temperature measurement circuit 22 to measure the ambient temperature by the temperature sensor 2, and receives the signal p1 to set the resonance frequency of the vibrator 1 to the resonance frequency measuring device 21. Let me measure. Next, in step S23, these observation data are stored in the storage device 30 as time series data D1 of ambient temperature and time series data D2 of resonance frequency associated with the measurement time.

  Next, in step S24, the CPU 120 determines whether or not the given observation period has ended, and if not, repeats steps S21 to S24. When the observation period ends, the next step S25 is executed. The functions executed by steps S21 to S24 described above are the same as steps S11 to S14 of the first embodiment described with reference to FIG. Further, the time series data D1 of the environmental temperature and the time series data D2 of the resonance frequency are the same as in the first embodiment.

  Independently of steps S11 to S14, in step S30, the first predetermined temperature Tc1 is set in the same manner as in step S18 of the first embodiment, and if necessary, the first predetermined temperature Tc1 ′ is set. . The setting of the first predetermined temperatures Tc1 and Tc1 'is the same as in the first embodiment.

  In the second embodiment, a second predetermined temperature Tc2 is further set in step S30. The second predetermined temperature Tc2 is given as a temperature different from the first predetermined temperature Tc1 by a temperature difference ΔT ° C. The temperature difference ΔT may be either a positive or negative temperature difference, and is selected to be a magnitude at which the temperature dependence of the resonance frequency of the vibrator 1 is measured, for example, 1 ° C. to several ° C. Note that if the absolute value of the temperature difference ΔT is too large, it is difficult to calculate the second time in step S25 described later. Conversely, if it is too small, the difference between the resonance frequencies at the first and second predetermined temperatures Tc1 and Tc2 becomes small, and it becomes difficult to accurately calculate the temperature correction coefficient in step S28.

  Next, in step S25, the first and second time calculation units 12a and 12b calculate the first time t1 and the second time t2, respectively.

  FIG. 13 is a diagram (part 1) for explaining data processing in the second embodiment of the present invention, and shows a data processing method for one to three days from the start of observation. FIG. 14 is a diagram (part 2) for explaining data processing in the second embodiment of the present invention, and represents a data processing method for 23 to 25 days from the start of observation. 13 and FIG. 14 are the time series data D1 of the environmental temperature and the first and second times t1 (i) and t2 (i) calculated therefrom, and the bottom is the time series data D2 of the resonance frequency. The first and second resonance frequencies f1 (i) and f2 (i) calculated therefrom are shown.

  In step S25, the first and second time calculation units 12a and 12b first store the environmental temperature time-series data D1 from the storage device 30 (similar to the environmental temperature time-series data D1 of the first embodiment). Is read.

  Next, referring to the upper part of FIG. 13, the first time calculation unit 12a calculates the first time t1 (i) when the time series data D1 of the environmental temperature coincides with the first predetermined temperature Tc1. In FIG. 13, the calculated first time t1 (i) is displayed with t1 (1) to t1 (11) in order from the shortest elapsed time. Here, the first predetermined temperature Tc1 was set to 25 ° C., and the first time t1 (i) was calculated by the same approximation method as in the first embodiment.

  Next, the second time calculation unit 12b calculates a second time t2 (i) at which the time series data D1 of the environmental temperature matches the second predetermined temperature Tc2. The calculated second time t2 (i) is displayed with second times t2 (1) to t2 (2) in order of elapsed time in FIG. The second predetermined temperature Tc2 was set to 24 ° C., and the second time t2 (i) was calculated using the same approximation (linear interpolation) used for the calculation of the first time.

  Referring to the upper part of FIG. 14, the first time calculation unit 12a sets the temperature to 25 ° C. when processing the time series data D1 of the environmental temperature for 23 days to 25 days when the average temperature of the environmental temperature increases. The first predetermined temperature Tc1 is changed to the first predetermined temperature Tc1 ′ set to 28 ° C. as in the first embodiment. That is, the first predetermined temperature Tc1 ′ = 28 ° C. Then, a first time t1 '(i) is calculated at which the time series data D1 of the environmental temperature for 23 days to 25 days coincides with the first predetermined temperature Tc1' = 28 ° C. This calculation method is performed in the same manner as the calculation of the first time t1 (i) shown in FIG. As a result, 14 time points from the first time f1 '(1) to f1' (14) shown in FIG. 4 were calculated in 23 days to 25 days.

  Further, the second time calculation unit 12b calculates the second time t2 ′ (i) that intersects the second predetermined temperature Tc2 ′ with respect to the time series data D1 of the environmental temperature for 23 days to 25 days. . When the second predetermined temperature Tc2 'was set to 27 ° C, the second times t2' (1) to t2 '(2) at two time points were calculated.

  Next, in step S26, the first and second resonance frequency calculators 13a and 13b respectively perform the first and second resonance frequencies f1 (i) and f2 at the first times t1 (i) and t2 (i). (I) is calculated by the following procedure.

  First, with reference to FIG. 13, the processing of observation data for 1 to 3 days from the start of observation will be described.

  Referring to the lower part of FIG. 13, first resonance frequency calculator 13 a first reads time series data D <b> 2 of the resonance frequency from storage device 30. In FIG. 13, the read time series data D2 of the resonance frequency is displayed as a line graph connecting small circles filled with circles.

  Next, using the resonance frequency time-series data D2, the resonance frequencies (first resonance frequencies f1 (1) to f1 (11)) at the first times t1 (1) to t1 (11) are set to the first resonance. Calculated as frequency f1. The first resonance frequency f1 (i) was calculated by approximation and complementation similar to the calculation of the first resonance frequency f1 (i) of the first embodiment. That is, the resonance frequency at the point where the time series data D1 of the resonance frequency intersects with a line (vertical one-dot chain line in FIG. 13) representing the first time t1 (i) is calculated.

  In FIG. 13, the calculated first resonance frequencies f1 (1) to f1 (11) are displayed by white circles with a large diameter, and a line graph G1 connecting the circles is displayed. The first resonance frequency f1 is calculated in the same manner as the calculation of the first resonance frequency f1 (i) in the first embodiment. Therefore, the first resonance frequencies f1 (1) to f1 (11) and the line graph G1 shown in FIG. 13 are the first resonance frequencies f1 (1) to f1 (11) of the first embodiment shown in FIG. And it is the same as the line graph G1.

  Next, the second resonance frequency calculation unit 13b uses the resonance frequency time-series data D2 read from the storage device 30, and uses the resonance frequency (f2 (1) to f2 (2) at the second time t2 (i). This second resonance frequency f2 (i) was also calculated in the same manner as the first resonance frequency f1 (i), that is, a line graph representing the resonance frequency time series data D1 The resonance frequency at a point that intersects the line representing the time t2 (i) of 2 (a vertical alternate long and short dash line passing through the time t2 (i) in Fig. 13) is calculated, and Fig. 13 shows the calculated second resonance frequency. f2 (1) to f2 (2) are displayed by double quadrilaterals and connected by a thick straight line.

  Next, processing of observation data for 23 to 25 days from the start of observation will be described with reference to FIG.

  Referring to the lower part of FIG. 14, second resonance frequency calculation unit 13 b first reads time series data D <b> 2 of the resonance frequency from storage device 30. The time-series data D2 of the resonance frequency read out in FIG. 14 is displayed as a line graph connecting small circles with filled circles.

  Next, using the resonance frequency time-series data D2, the resonance frequencies (first resonance frequencies f1 ′ (1) to f1 ′ (14)) at the first times t1 ′ (1) to t1 ′ (14) are obtained. It calculated as 1st resonance frequency f1 '. The first resonance frequency f1 '(i) is calculated by approximation and complementation similar to the calculation of the first resonance frequency f1 (i) of the first embodiment. That is, the resonance frequency at the point where the resonance frequency time-series data D1 intersects the line (vertical one-dot chain line in FIG. 14) representing the first time t1 '(i) is calculated.

  In FIG. 14, the calculated first resonance frequencies f1 ′ (1) to f1 ′ (14) are displayed by white large-diameter circles, and a line graph G1 connecting the circles is displayed. The first resonance frequency f1 'is calculated in the same manner as the calculation of the first resonance frequency f1 of the first embodiment. Therefore, the first resonance frequencies f1 ′ (1) to f1 ′ (14) and the line graph G1 ′ shown in FIG. 14 are the first resonance frequencies f1 ′ (1) to f1 of the first embodiment shown in FIG. It is the same as '(14) and line graph G1'.

  Next, the second resonance frequency calculation unit 13b uses the resonance frequency time-series data D2 read from the storage device 30, and uses the resonance frequency (f2 ′ (1) to f2 ′ (2) at the second time t2 ′. The second resonance frequency f2 ′ (i) is also calculated in the same manner as the first resonance frequency f1 ′ (i), that is, the time series data D2 of the resonance frequency is the second resonance frequency f2 ′ (i). A resonance frequency is calculated at a point that intersects the line representing the time t2 ′ (i) (a vertical alternate long and short dash line passing through the time t2 ′ (i) in FIG. 14), and the calculated second resonance frequency is shown in FIG. f2 ′ (1) to f2 ′ (2) are displayed by double quadrilaterals and connected by a thick straight line.

  Next, in step S27, with reference to FIG. 13, the first minute fluctuation region extraction unit 14a first resonance frequency f1 (i) at the first times t1 (i) and t1 (i + 1) adjacent to each other. And f1 (i + 1) are compared. If the difference Δf is within a preset minute fluctuation frequency range, for example, within ± 1 Hz, the first time t1 (i) to the next first time t1 (i + 1) Extracted as one minute fluctuation region. Referring to FIG. 13, as the first minute fluctuation range, the first resonance frequencies f1 (1) to f1 (2), the first resonance frequencies f1 (3) to f1 (6), The first resonance frequency f1 (9) to f1 (10) is extracted. Since the vibrator 1 according to the second embodiment is a sensor that detects the cumulative value of the adhesion amount, these first minute fluctuation regions are in a time zone in which the adhesion amount does not increase or the increase rate of the adhesion amount is small. It means that there is.

  Further, in step S27, the second minute fluctuation extracting unit 14b refers to FIG. 13 for the observation data for one to three days from the start of observation, for example, and the second times t2 (i) and t2 adjacent to each other. The second resonance frequencies f2 (i) and f2 (i + 1) at (i + 1) are compared. If the difference Δf is within a preset minute fluctuation frequency range, for example, within ± 1 Hz, the second time t2 (i) to the next second time t2 (i + 1) 2 is extracted as a minute fluctuation region. Referring to FIG. 13, the second resonance frequency f2 (1) to f2 (2) was extracted as the second minute fluctuation region.

  Similarly, in step S27, the first minute fluctuation region extraction unit 14a refers to the observation data for 23 days to 25 days from the start of observation, for example, referring to FIG. 14, and the first times t1 ′ (i) adjacent to each other. And the first resonance frequencies f1 ′ (i) and f1 ′ (i + 1) at t1 ′ (i + 1) are compared. If the difference Δf is within the range of the minute fluctuation frequency, the section is extracted as the first minute fluctuation region. Referring to FIG. 14, as such a minute fluctuation range, between the first resonance frequencies f1 ′ (1) to f1 ′ (3) and between the first resonance frequencies f1 ′ (8) to f1 ′ (9). , And between the first resonance frequencies f1 ′ (12) to f1 ′ (13).

  Further, in step S27, the second minute fluctuation extracting unit 14b refers to FIG. 14, and the second resonance frequency f2 ′ (i) at the second times t2 ′ (i) and t2 ′ (i + 1) adjacent to each other. ) And f2 ′ (i + 1). If the difference Δf is within the range of the minute fluctuation frequency, the period from the second time t2 ′ (i) to the next second time t2 ′ (i + 1) is extracted as the second minute fluctuation region. To do. Referring to FIG. 14, the second resonance frequency f2 '(1) to f2' (2) was extracted as the second minute fluctuation region.

  Next, in step S28, the correction coefficient calculation unit 15 extracts an overlapping period in which the first minute variation region and the second minute variation region overlap. Then, based on the difference between the first resonance frequency f1 and the second resonance frequency f2 in the overlapping period, the temperature correction coefficient κ of the vibrator 1 is calculated by the procedure described below.

  With reference to FIGS. 13 and 14, the correction coefficient calculation unit 15 first detects an overlapping period in which the first minute variation region and the second minute variation region extracted in step S27 overlap.

  Specifically, the time at both ends of each section of the first minute variation area and the time at both ends of each section of the second minute variation area are sequentially compared, and the first and second minute variation areas are obtained. Detect overlapping overlapping periods. For example, such an overlapping period is detected as a section from time t2 (1) to t2 (2) in FIG. 13, and as a section from time t2 '(1) to t2' (2) in FIG. In FIG. 13 and FIG. 14, the entire second minute fluctuation region is included in the first minute fluctuation region. However, the present invention is not limited to this, and there may be a section where the first and second minute fluctuation regions partially overlap.

Next, the correction coefficient calculation unit 15 calculates the first resonance frequency f1 (or f1 ′) and the second resonance frequency f2 (or f2 ′) in a section (overlap period) where the first and second minute fluctuation regions overlap. The frequency difference δf = f1−f2 (or f1′−f2 ′) is calculated. And the temperature correction coefficient κ is
κ = δf / ΔTc (1)
Calculated by Here, ΔTc is a temperature difference between the first predetermined temperature Tc1 (or Tc1 ′) and the second predetermined temperature Tc2 (or Tc2 ′), and ΔTc = Tc1−Tc2 (or ΔTc = Tc1′−Tc2 ′). ).

  As described above, each of the first and second minute fluctuation ranges is a section where the deposition rate of the substance to be measured is zero to slow, and the amount of deposition hardly varies in this section. That is, the first and second resonance frequencies f1, f2 (or f1'-f2 ') in the overlapping period correspond to the resonance frequencies for the same adhesion amount. Since the resonance frequency of the vibrator 1 depends on the temperature and the amount of adhesion, the difference between the first and second resonance frequencies in this overlapping period is the difference in environmental temperature at the time of measurement, that is, the first predetermined temperature Tc1 and the second predetermined frequency. This is due to the difference from the predetermined temperature Tc2. Therefore, the temperature correction coefficient κ is calculated by dividing the difference between the first and second resonance frequencies in the overlap period by the difference between the first and second predetermined temperatures according to the equation (1).

  Referring to FIG. 13, in the overlapping period between times t2 (1) to t2 (2), the first resonance frequency f1 represented by a thick straight line connecting f1 (5) and f1 (6) in FIG. The fluctuation range was constant at −155 Hz. On the other hand, the fluctuation range of the second resonance frequency f2 represented by a thick straight line connecting f2 (1) and f2 (2) was constant at −149 Hz. Accordingly, δf = f1−f2 = −6 Hz is calculated. Since Tc1 = 25 ° C. and Tc2 = 24 ° C., δTc = Tc1−Tc2 = 1 ° C. From this, the correction coefficient κ = −6 Hz / ° C. was calculated using one equation.

  Similarly, in the overlapping period between times t2 ′ (1) to t2 ′ (2) in FIG. 14, the first resonance frequency f1 ′ is expressed by f1 ′ (12) and f1 ′ (13) in FIG. It was represented by a thick straight line, and the fluctuation range of the resonance frequency was −3241 Hz. On the other hand, the fluctuation range of the second resonance frequency f2 'represented by a thick straight line connecting f2' (1) and f2 '(2) in FIG. 14 was -3234 Hz. Therefore, δf = f1′−f2 ′ = − 7 Hz is calculated. Further, since Tc1 ′ = 28 ° C. and Tc2 ′ = 27 ° C., ΔTc = Tc1′−Tc2 ′ = 1 ° C. From this, the correction coefficient κ = −7 Hz / ° C. is calculated using one equation.

  In the second embodiment, the temperature correction coefficient κ = 6 Hz / ° C. and the average value of κ = 7 Hz / ° C. calculated from the data of the first to third days and the 23rd to 25th days from the observation start = 6.5 Hz / The time series data D2 of the resonance frequency measured during the observation period of one month was corrected using ° C. as a temperature correction coefficient. This temperature correction coefficient κ = 6.5 Hz / ° C. is in good agreement with the actually measured temperature correction coefficient 7 Hz / ° C. in the comparative example.

  In FIG. 14, the time series data of the temperature-corrected resonance frequency of the comparative example is displayed with double circles (◎). In the minute fluctuation ranges (t1 ′ (1) to t1 ′ (3), t1 ′ (8) to t1 ′ (9), t1 ′ (12) to t1 ′ (13)), the first predetermined temperature Tc1 ′ = The first resonance frequencies f11 ′ (1) to f1 ′ (3), f1 ′ (8) to f1 ′ (9), and f1 ′ (12) to f1 ′ (13) ′ measured at 28 ° C. are It was 18 Hz to 20 Hz lower than the temperature-corrected resonance frequency of the comparative example in the minute fluctuation range, and the average was 18.6 Hz in these three sections. This frequency difference of 18.6 Hz is a temperature difference of 3 ° C. between the first predetermined temperature tc1 ′ at which the first resonance frequency ft1 ′ is measured = 28 ° C. and the correction temperature of 25 ° C. of the comparative example. The temperature correction coefficient κ is approximately equal to 19.5 Hz multiplied by 6.5 Hz / ° C. This indicates that an accurate temperature correction coefficient κ is obtained in the second embodiment.

  As described above, according to the second embodiment, the temperature correction coefficient of the resonance frequency of the vibrator 1 is calculated from the difference between the first and second resonance frequencies in the minute fluctuation range without performing a temperature test. be able to.

After step S28, in step S29, the conversion unit 17 applies the frequency fluctuation amount df of the resonance frequency of the vibrator 1 that has been temperature-corrected using the above-described temperature correction coefficient κ, in the same manner as in step S17. Note that the magnitude of the minute fluctuation frequency is set to be sufficiently smaller than the frequency difference Δf of the resonance frequency caused by the temperature difference between the first predetermined temperature Tc1 and the second predetermined temperature Tc2. By setting in this way, it is possible to make the fluctuation of the resonance frequency in the first minute fluctuation region mainly depend on the change of the environmental temperature, and to reduce the influence of the fluctuation of the adhesion amount. For this reason, the error of the temperature correction coefficient resulting from the fluctuation of the adhesion amount is reduced.

  In the second embodiment described above, the first predetermined temperature Tc1 'is set to 28 ° C, and the second predetermined temperature Tc2' is set to 24 ° C. A second predetermined temperature Tc2 ″ having a different temperature can be added to the second predetermined temperature Tc2 ′.

  FIG. 15 is a diagram for explaining data processing in a modification of the second embodiment of the present invention, and illustrates data processing performed by adding a second predetermined temperature Tc2 ″ to the second embodiment.

  Referring to FIG. 15, in this modification, the second predetermined temperature Tc2 ″ is set, and the time series data D1 of the environmental temperature coincides with the second predetermined temperature Tc2 ″ (for example, Tc2 ″ = 33 ° C.). Second times t2 ″ (1) to t2 ″ (4) are calculated. The second times t2 ″ (1) to t2 ″ (4) are calculated by the second resonance frequency calculation unit 13b in step S25. It is calculated using the same data processing as the calculation of the second times t2 ′ (1) to t2 ′ (2).

  In this modification, other data processing, for example, calculation of the first and second times t1 ′ and t2 ′, calculation of the first and second resonance frequencies f1 ′ and f2 ′, and first and second minute fluctuations are performed. The calculation of the area and the calculation of the overlap period are processed in the same manner as in the second embodiment. In the following, processes added to the second embodiment will be mainly described with respect to modifications.

  The second resonance frequency calculator 13b according to the modification calculates second resonance frequencies f2 ″ (1) to f2 ″ (4) at the second times t2 ″ (1) to t2 ″ (4). And the 2nd minute fluctuation region extraction part 14b of a modification compares the adjacent 2nd resonance frequency, and extracts the area where the frequency difference is smaller than a given minute fluctuation frequency as a 2nd minute fluctuation region. . These processes are performed in the same manner as steps S25 to S27 of the second embodiment.

  Next, the correction coefficient calculation unit 15 extracts the second minute fluctuation region (f2 ″ (3) to f2 ″ in FIG. 15) extracted based on the second resonance frequencies f2 ″ (1) to f2 ″ (4). 4) and an overlap period (period t2 ″ (3) to t2 ″ (4) in FIG. 15) in which the first minute fluctuation region overlaps is extracted. This extraction is performed in the same manner as in step S27.

  Next, the correction coefficient calculation unit 15 calculates the first resonance frequencies f1 ′ (12) to f1 ′ (14) in the extracted overlapping period t2 ″ (3) to t2 ″ (4) and the overlapping period t2 ″ ( 3) A frequency difference δf from the second resonance frequency f2 ″ (3) to f2 ″ (4) at t2 ″ (4) is calculated. Then, the temperature correction coefficient κ is calculated by κ = δf / (Tc1′−Tc2 ″).

  Referring to FIG. 15, the fluctuation amount of the first resonance frequency f1 ′ in the overlap period t2 ″ (3) to t2 ″ (4) is −3241 Hz, and the fluctuation amount of the second resonance frequency f2 ″ is −3278 Hz. The frequency difference δf is calculated as δf = f1′−f2 ″ = 37 Hz. Further, from Tc1 ′ = 28 ° C. and Tc2 ″ = 33 ° C., (Tc1′−Tc2 ″) = − 5 ° C. is calculated. Therefore, the temperature correction coefficient κ is calculated as κ = −7.4 Hz / ° C. This temperature correction coefficient κ = −7.4 Hz / ° C. is in good agreement with the temperature correction coefficient κ = −7 Hz / ° C. of the comparative example.

  In the modification of the second embodiment, the temperature correction coefficients κ = −6 Hz / ° C. and κ = −7 Hz / ° C. obtained in the second embodiment, and the temperature correction coefficient κ = −7.4 Hz / degree obtained in this modification. The average value of -6.8 Hz / ° C was used as the temperature correction coefficient κ, and the time series data D2 of the resonance frequency was temperature corrected.

  According to the above-described modification, two second predetermined temperatures Tc2 ′ and Tc2 ″ that are lower and higher than the first predetermined temperature Tc1 ′ are set. Therefore, the environmental temperature is the first predetermined temperature Tc1 ′. Since the second time is calculated regardless of which side is higher or lower, the number of the second times is increased, and precise data processing is possible, and the first predetermined temperature Tc1 ′ and the first time 2 is as large as 5 ° C., so a precise temperature correction coefficient κ is required.

  The third embodiment of the present invention relates to an environment monitoring apparatus using a vibrator 1 to which an adhering substance adheres reversibly.

  FIG. 16 is a functional configuration diagram of the environment monitoring apparatus according to the third embodiment of the present invention, and represents main functions of the environment monitoring apparatus 300 according to the third embodiment of the present invention.

  Referring to FIG. 16, the environment monitoring apparatus 300 according to the third embodiment is similar to the first embodiment in that the observation unit 20, the storage device 30, and the time generation unit measure the environmental temperature and the resonance frequency of the vibrator 1. 11, a first time generation unit 12 a, a first resonance frequency calculation unit 13 a, a conversion unit 17, and an output unit 18. These have the same functions as those of the first embodiment except for the vibrator 1 and operate in the same manner. Therefore, in order to avoid duplication of explanation, the main points different from the first embodiment will be described below.

  The environment monitoring apparatus 300 according to the third embodiment further includes a third minute fluctuation region extraction unit 14c, a temperature dependence detection unit 15c, and a temperature correction unit 16c. In the vibrator 1 used in the third embodiment, the substance to be measured adheres reversibly, and therefore, the resonance frequency reversibly changes according to the concentration of the substance to be measured in the environment.

  FIG. 17 is a flowchart of the environment monitoring apparatus according to the third embodiment of the present invention, and shows the steps executed in accordance with the program 10.

  Referring to FIGS. 16 and 17, in step S41, time generation unit 11 generates pulses p1 and p2 that indicate the measurement time. In step S42, the observation unit 20 measures the environmental temperature and the resonance frequency of the vibrator 1 in synchronization with the pulses p1 and p2. Next, in step S43, the measured environmental temperature and resonance frequency are stored in the storage device 30 as time series data D1 and D2 of the environmental temperature and resonance frequency associated with the measurement time. Next, in step S44, it is evaluated whether or not a predetermined observation period has elapsed, and when it is determined as no, the process is repeated from step S41 again. If it is determined that the observation period has ended, the next step S45 is executed.

  The operations in these steps S41 to S44 are the same as those in steps S11 to S14 in the first embodiment. Therefore, when step S45 is executed, the time series data D1 and D2 of the environmental temperature and the resonance frequency for the entire observation period are stored.

  Next, apart from steps S41 to S44, a first predetermined temperature Tc1 (see FIG. 18) is set in step S51. Further, in step S45, the first time calculation unit 12a calculates a first time t1. Steps S51 and S45 are the same as steps S18 and S15 of the first embodiment.

  FIG. 18 is a view for explaining data processing in the third embodiment of the present invention. The upper part of FIG. 18 shows time series data D1 of environmental temperature in a part of the observation period, and the lower part of FIG. 18 shows time series of resonance frequencies. Data D2 and temperature-corrected resonance frequency time series data D′ 2 are shown.

  Referring to the upper part of FIG. 18, in step S45, the first time calculation unit 12a reads the time series data D1 of the environmental temperature from the storage device 30. Next, the time series data D1 of the environmental temperature is approximated by an approximate expression, for example, a line graph, and times t1 (1) to t1 (4) at which the approximate expression matches the first predetermined temperature Tc1 are calculated. This step S45 is the same as step S15 of the first embodiment.

  In step S46, the first resonance frequency calculation unit 13a approximates the time series data D2 of the resonance frequency read from the storage device with an approximate expression, for example, a line graph. Then, the approximate expression values at the first times t1 (1) to t1 (4) are calculated as the first resonance frequencies f1 (1) to f1 (4). The calculated first resonance frequencies f1 (1) to f1 (4) are indicated by a large-diameter white circle (◯) in the lower part of FIG. The process of step S46 is performed in the same manner as step S16 of the first embodiment.

  Next, in step S47, the third minute fluctuation region extraction unit 14c compares the first resonance frequencies f1 (1) to f1 (4) adjacent to each other, and the absolute value of the frequency difference is a minute fluctuation with a predetermined value. A frequency band, for example, a time zone smaller than 1 Hz is extracted as a third minute fluctuation region. Here, with reference to the large-diameter white circle in the lower part of FIG. 18, the first resonance frequencies f1 (2), f1 (3), and f1 (4) that make the difference in resonance frequency within 1 Hz are included. The section, that is, the time zone from the first time t1 (2) to t1 (4) was extracted as the third minute fluctuation region. The extraction process in step S47 is the same as the first minute fluctuation region extraction method in step S27 of the second embodiment. Note that the above-described third minute fluctuation region is extracted for the entire observation period, and in many cases, a plurality of third minute fluctuation regions are extracted.

  Next, in step S48, the temperature dependence detector 15c selects one third minute fluctuation region, and whether or not the resonance frequency observed in the third minute fluctuation region follows the primary expression of the environmental temperature. Is detected.

  Specifically, first, an arbitrary observation time to (i) in the third minute fluctuation region is selected with reference to the upper part of FIG. Next, the environmental temperature To at the selected observation time to (i) is extracted from the time series data D1 of the environmental temperature. Next, referring to the lower part of FIG. 18, the resonance frequency fo (i) measured at the observation time to (i) is extracted from the resonance frequency time-series data D2.

  Next, the frequency difference (fo (i) −) between the resonance frequency fo (i) and the first resonance frequency in the third minute fluctuation region, for example, f1 (2) (= f1 (3) = f1 (4)). f1 (2)) is obtained. On the other hand, a temperature difference (To (i) −Tc1) between the environmental temperature To (i) and the first predetermined temperature Tc1 at the observation time to (i) is obtained.

Next, the ratio γ = frequency difference (fo (i) −f1 (2)) / temperature difference (To (i) −Tc1) is set to all the observation times to (i) in the selected third minute fluctuation region. Is calculated. If the ratio γ is constant throughout the selected third minute fluctuation range, the resonance frequency fo is a linear expression of the environmental temperature T,
Resonance frequency fo = γT + constant (2)
It is determined to have temperature dependence according to On the contrary, if the ratio γ is not constant, it is determined that the resonance frequency fo is influenced by the concentration fluctuation of the substance to be measured in addition to the environmental temperature. The basis for this determination is as follows.

  The resonance frequency of the resonator 1 according to the third embodiment generally varies depending on the linearity depending on the temperature and the concentration of the substance to be measured. In the third minute fluctuation region, the first resonance frequencies f1 (2) to f1 (4) are all measured under the same environmental temperature (first predetermined temperature Tc1) and are not affected by the temperature fluctuation. . Accordingly, the fact that the first resonance frequencies f1 (2) to f1 (4) coincide with each other within the minute fluctuation frequency means that the first time when the first resonance frequencies f1 (2) to f1 (4) are measured. It shows that the concentration of the substance to be measured is constant at t1 (2), t1 (3), and t1 (4). This strongly estimates that the measured substance concentration is constant within the third minute fluctuation range.

  Based on this estimation, it is assumed that the concentration of the substance to be measured is constant within the first minute fluctuation range. Furthermore, it is assumed that the concentration of the substance to be measured does not vary in proportion to the ambient temperature. Normally, it is extremely rare for the concentration of a substance to be measured in the environment to increase or decrease in proportion to the environmental temperature, and in many cases this assumption holds.

  Under these two assumptions, if Formula 2 is established within the third minute fluctuation range, the fluctuation of the first resonance frequency f1 within the third minute fluctuation range is expressed by Formula 2. In addition, it follows changes in the environmental temperature. This is because it is assumed that there is no change in the concentration of the substance to be measured within the third minute fluctuation range.

  For the reasons described above, in many cases, when the two formulas are established, the resonance frequency of the vibrator 1 may have a temperature dependency according to the two formulas. However, the assumption that the concentration of the substance to be measured does not fluctuate accurately in proportion to the environmental temperature within the third minute fluctuation region may not be satisfied. For example, this is a case where the concentration of the substance to be measured changes according to a linear expression of the environmental temperature. Therefore, a plurality of areas are extracted from the first minute fluctuation range described above at time intervals, for example, several days to several weeks, and it is confirmed that the two equations are established within the plurality of first minute fluctuation areas. It is preferable. Thereby, it is possible to avoid erroneous determination due to the concentration of the substance to be measured changing in proportion to the linear expression of the environmental temperature only for one period.

  Furthermore, the first resonance frequency f1 (for example, f1 (2), f1 (3), f1 (4)) in the third minute fluctuation range described above is equal to the resonance frequency when the concentration of the substance to be measured is zero. Only in such a case, the determination in step S48 may be performed. Usually, the conversion factor from the resonance frequency to the measured substance concentration is measured in advance by measuring the relationship between the measured substance concentration and the resonance frequency at the first predetermined temperature Tc1. For this reason, the resonance frequency corresponding to the measured substance concentration of zero is often known in advance.

  In this case, if the measured resonance frequency is shifted to a higher frequency side than the resonance frequency corresponding to the measured substance concentration of zero, it can be determined that the deviation depends only on the environmental temperature. This is because the concentration cannot be less than zero, i.e., cannot be negative, and the measured resonance frequency does not exceed the known resonance frequency due to the concentration change. Therefore, there is a section in which the resonance frequency measured in the third minute fluctuation range is shifted to a higher frequency side than the resonance frequency corresponding to the measured substance concentration of zero, and the third minute fluctuation including this section. When Formula 2 is established for the region, it is understood that the resonance frequency varies according to Formula 2.

  Thus, by extracting a section where the first resonance frequency is equal to the resonance frequency corresponding to zero of the substance to be measured, the concentration of the substance to be measured changes in proportion to the primary expression of the environmental temperature. The erroneous determination that occurs can be avoided.

  Next, in step S49, the temperature correction unit 16c calculates the ratio γ as the temperature correction coefficient γ. Next, in step S50, it is determined whether or not steps S48 and S49 have been executed for all the third minute fluctuation ranges. If it is determined that the determination is negative, in step S52, the next (on the later measurement time side) 3 minute fluctuation regions are extracted, and Steps S48 and S49 are repeated again. If it is determined that steps S48 and S49 have been executed for all the third minute regions, the next step S53 is executed.

  Next, in step S53, the temperature correction unit 16c corrects the temperature of the time series data D2 of the number of resonance frequencies using the ratio γ as the temperature correction coefficient γ. At this time, it is preferable to use the average value of the ratio γ calculated in the plurality of third minute fluctuation ranges as the temperature correction coefficient γ in order to avoid an accidental error.

  Here, the average value of the ratio κ was calculated as κ = −6.5 Hz / ° C. within the first minute fluctuation region. A white triangle (Δ) at the bottom of FIG. 18 represents time-series data D′ 2 of the resonance frequency after temperature correction. The time series data D2 'of the resonance frequency after temperature correction is corrected to a substantially constant value within the first minute fluctuation region. On the other hand, in the time zone before and after that, that is, the time zone in which the temperature is higher than the first predetermined temperature Tc1, the time series data D2 ′ of the resonance frequency after the temperature correction is the resonance frequency before the correction as the environmental temperature becomes higher. The time series data D2 is corrected so that the frequency increases. The magnitude of the correction coefficient and the direction of correction matched well with the previously measured correction coefficient of the vibrator 1.

  According to the third embodiment, in an environmental monitoring apparatus using a vibrator whose resonance frequency varies in proportion to the environmental concentration of the substance to be measured as a sensor, temperature correction can be performed without obtaining a temperature correction coefficient in advance. An environmental monitoring device is realized.

  Instead of the temperature sensors of the first to third embodiments described above, an oscillator having temperature dependency, for example, an AT-cut quartz oscillator that is cut off, is used as a temperature sensor oscillator for measuring the environmental temperature. It can also be used. In this vibrator, the relationship between temperature and resonance frequency may not be known.

  In the operating temperature range of the vibrator, the resonance frequency usually changes in proportion to the environmental temperature (according to a linear equation of temperature). Therefore, the resonance frequency of the temperature sensor vibrator can be used as an index of the environmental temperature. That is, the resonant frequency of the temperature sensor vibrator can be used instead of the temperature. However, the absolute value of temperature is not known.

  When the temperature sensor vibrator is used as the temperature sensor, the resonance frequency of the temperature sensor vibrator is used instead of the temperature. Specifically, two arbitrarily selected resonance frequencies are set in place of the first and second predetermined temperatures. This resonance frequency does not need to be a resonance frequency that exactly corresponds to the first and second predetermined temperatures, and the resonance frequency is determined as a frequency that can be roughly expected at the first and second predetermined temperatures. If it is enough. In this manner, all temperatures including the environmental temperature are replaced with the resonance frequency, and the temperature correction coefficient κ ′ is calculated in the same manner as in the first to third embodiments.

According to this method, the unit of the temperature correction coefficient κ ′ is dimensionless (Hz × Hz ⁻¹ ), and the temperature correction coefficient κ in units of Hz × ° C. ⁻¹ is not calculated. However, in the first to third embodiments described above, the adhesion amount δW of the substance to be measured is
[delta] W = [alpha] * ((f-fo)-[kappa] * (T-To)) (3)
Is converted as Here, α is a conversion coefficient for converting a frequency difference into an adhesion amount, f is a resonance frequency f of the vibrator 1 at an arbitrary time t, fo is a resonance frequency of the vibrator 1 at the observation start time, and T is an environment at the time t. Temperature and To are environmental temperatures at the start of observation.

The terms T and To representing the temperature in Equation 3 are replaced with the resonance frequency F of the temperature sensor vibrator at time t and the resonance frequency Fo of the temperature sensor vibrator at the observation start time, respectively. From this, equation 3 is expressed as δW = α × ((f-fo) −κ ′ × (F-Fo)) (4)
And transformed. From the four equations, an accurate adhesion amount is calculated based on the resonance frequency difference (F-Fo) of the temperature sensor resonator. In addition, the absolute value of temperature and a temperature difference are not known by this method. However, since the adhesion amount is calculated, there is no particular problem as an environmental monitoring device.

  In this way, by using a temperature sensor vibrator that measures the ambient temperature with a digital circuit as a change in the resonance frequency of the resonator, the manufacturing cost can be reduced compared to a temperature sensor composed of an analog circuit that requires precise adjustment of the circuit. Can be reduced.

  When the present invention is applied to an environmental monitoring device that monitors the concentration of corrosive substances or organic gases in the environment, it is not necessary to perform temperature correction in advance, so that a low-cost environmental monitoring device can be realized.

DESCRIPTION OF SYMBOLS 1 Vibrator 2 Temperature sensor 10 Program 11 Time generation part 12a 1st time calculation part 12b 2nd time calculation part 13a 1st resonance frequency calculation part 13b 2nd resonance frequency calculation part 14a 1st minute fluctuation range extraction Unit 14b second minute fluctuation region extraction unit 14c third minute variation region extraction unit 15 correction coefficient calculation unit 15c temperature dependence detection unit 16, 16c temperature correction unit 17 conversion unit 18 output unit 41 output device
DESCRIPTION OF SYMBOLS 20 Observation part 21 Resonance frequency measuring device 22 Temperature measurement circuit 30 Memory | storage device 100,200,300 Environmental monitoring apparatus 110 Memory 111 Bus | bath 120 CPU
f1 (i), f1 ′ (i) first resonance frequency f2 (i), f2 ′ (i), f2 ″ (i) second resonance frequency t1 (i), t1 ′ (i) first time t2 (i), t2 ′ (i), t2 ″ (i) Second time Tc1, Tc1 ′ First predetermined temperature Tc2, Tc2 ′, Tc2 ″ Second predetermined temperature

Claims (7)

A first vibrator whose resonance frequency changes due to adhesion of a substance to be measured;
A resonance frequency measuring device for measuring a resonance frequency of the first vibrator at a first measurement time including a plurality of measurement times;
And Rousset capacitors to measure the variation value that varies in response to the environmental temperature at the second measurement time including a plurality of measurement times,
The measured resonance frequency of the first vibrator is stored as time series data of the resonance frequency associated with the first measurement time, and the measured variation value is stored as the second measurement time. A storage device for storing time series data of associated variation values ;
A first time calculation unit for calculating a first time when the fluctuation value becomes a first predetermined value from the time-series data of the fluctuation value stored in the storage device;
A first resonance frequency calculation unit that calculates the resonance frequency at the first time as a first resonance frequency from time series data of the resonance frequency stored in the storage device;
A conversion unit for converting the first resonance frequency at the first time into the amount of the substance to be measured;
An environmental monitoring device comprising: The first vibrator includes a vibrator to which a substance to be measured in the environment accumulates and attaches irreversibly,
A first minute fluctuation range extraction unit that extracts a period in which the calculated time fluctuation of the first resonance frequency is smaller than a given minute fluctuation frequency as a first minute fluctuation range;
A second time calculation unit for calculating a second time at which the fluctuation value becomes a second predetermined value different from the first predetermined value , from the time-series data of the fluctuation values stored in the storage device; ,
A second resonance frequency calculating unit that calculates the resonance frequency at the second time as a second resonance frequency from time series data of the resonance frequency stored in the storage device ;
A second minute fluctuation region extraction unit that extracts a period in which the calculated time variation of the second resonance frequency is smaller than the given minute fluctuation frequency as a second minute fluctuation region;
Correction to calculate the compensation coefficient a difference between the first and second resonant frequency divided by the difference between the first and second predetermined values in the overlap period in which the first and second micro fluctuation region is superimposed A coefficient calculation unit;
An auxiliary Tadashibu that correct auxiliary time-series data of the resonant frequency using the front Kiho positive coefficient,
The environmental monitoring apparatus according to claim 1, further comprising: The first vibrator includes a vibrator whose resonance frequency varies in proportion to the concentration of the substance to be measured in the environment.
From the time-series data of the resonance frequency stored in the storage device, a third minute period in which the difference between the first resonance frequencies at the first measurement times adjacent to each other is smaller than a given minute fluctuation frequency. A third minute fluctuation region extraction unit that extracts the fluctuation region;
Said third of said resonant frequency fo of the minute fluctuation region is, Yi you determine whether changes in accordance with an equation fo = [gamma] T + constant of the fluctuation value T patency detection unit,
Wherein when it is determined that varies according to the primary type, with γ proportional coefficient of the linear expression as the correction factor, and complement Tadashibu that correct auxiliary time-series data of the resonant frequency,
The environmental monitoring apparatus according to claim 1, further comprising: The average value of the correction coefficients calculated from the time series data of the resonance frequency and the time series data of the fluctuation values observed during a plurality of observation periods is used as a correction coefficient. The environmental monitoring device described. The average value of the time series data of observed the variation value during a given observation period, environmental monitoring device according to any one of claims 1 to 3, characterized in that said first predetermined value. The sensor, in response to the ambient temperature as a second oscillator for changing the resonant frequency,
Environment monitoring device according to any one of claims 1 to 5, characterized in <br/> to the fluctuation value and the resonance frequency of the second oscillator. The sensor is a temperature sensor, and the fluctuation value is a temperature.
The environment monitoring apparatus according to claim 1, wherein

Images