JP2009210352A - Liquid sample dispensing device - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a liquid sample dispensing device can dispense accurately a necessary amount of liquid sample into each reaction container. <P>SOLUTION: This liquid sample dispensing device for dispensing the liquid sample held by a dispensing nozzle as much as each prescribed amount into a plurality of reaction containers, is equipped with a discharge operation storage part for storing in time series, data related to operation during discharge operation of an actuator for dispensation; a normal discharge operation storage part for storing in time series, data related to operation during normal discharge operation of the actuator for dispensation; and a discharge operation determination part for performing multivariate analysis of the data related to the operation during the discharge operation of the actuator for dispensation and the data related to the operation during the normal discharge operation of the actuator for dispensation, and determining existence of an abnormality of the actuator for dispensation based on the analysis result. <P>COPYRIGHT: (C)2009,JPO&INPIT

Description

  The present invention relates to a liquid sample dispensing apparatus that dispenses a required amount of a liquid sample into a reaction container with high accuracy, for example.

  The sample dispensing apparatus of Conventional Example 1 and the automatic analyzer using the same connect a pressure sensor to a dispensing flow path system including a sample probe and a dispensing syringe, and output values of the pressure sensor during the sample dispensing operation. A multi-item analysis (Mahalanobis distance) using the obtained pressure sensor output values as items, and comparing with a threshold value to determine whether dispensing has been performed normally (for example, patents) Reference 1).

  FIG. 13 is a schematic configuration diagram of a sample dispensing apparatus according to an embodiment of Conventional Example 1.

  The sample probe 101 is connected to a dispensing syringe 103 via a tube 102, and the inside thereof is filled with a liquid. The dispensing syringe 103 includes a cylinder 103a and a plunger 103b, and a dispensing syringe driving means 104 is connected to the plunger 103b. The cylinder 103a is fixed, and the plunger 103b is driven up and down by the dispensing syringe driving means 104, thereby performing the sample dispensing operation. A sample probe driving means 105 is connected to the sample probe 101, so that the sample probe 101 can be moved to a predetermined position. The dispensing syringe driving unit 104 and the sample probe driving unit 105 are controlled by the control unit 106.

  Now, when the sample probe 101 is lowered by the sample probe driving means 105 and reaches the liquid of the sample 108 in the sample container 107, the dispensing syringe 103 performs a suction operation by the dispensing syringe driving means 104. It is assumed that air (segmented air) has been sucked in advance before the sample probe 101 reaches the sample 108 liquid so that the liquid filled in the sample probe 101 and the sample 108 do not mix. When the sample suction operation ends, the sample probe 101 moves to the sample discharge position, and the dispensing syringe 103 performs the discharge operation.

  After completion of the dispensing, the sample probe 101 can be cleaned by flowing the cleaning water 111 in the water supply tank 110 at a high pressure by the water supply pump 109 as necessary. The switching is performed by the electromagnetic valve 112, which is controlled by the control unit 106.

  A pressure sensor 113 as a means for detecting a dispensing abnormality, which is the main purpose of Conventional Example 1, is connected to a dispensing flow path system including a sample probe 101, a tube 102, and a dispensing syringe 103 via a branch block 114. ing. Here, the pressure sensor 113 is preferably connected to the sample probe 101 side as much as possible in order to detect the pressure fluctuation at the opening of the sample probe 101 with high sensitivity.

  The output signal of the pressure sensor 113 is amplified by the amplifier 115 and converted into a digital signal by the A / D converter 116. Further, the output of the A / D converter 116 is sent to the microcomputer 117, and it is determined whether the dispensing has been normally performed.

  FIG. 14 is a process flow diagram of dispensing abnormality detection according to the embodiment of Conventional Example 1.

  The internal pressure of the dispensing flow path system including the sample probe at the time of sample dispensing constantly changes including minute changes, and the output of the pressure sensor also changes accordingly. Conventional example 1 takes in pressure values that change every moment along the time series, and uses a set of these taken pressure values (hereinafter referred to as a time series pressure data group) as it is (S1).

  Next, the Mahalanobis distance is obtained from the time-series pressure data group obtained as described above (S2). Here, the Mahalanobis distance is a method of multivariate analysis, and it is a scale for measuring whether a certain test object belongs to a reference group (hereinafter referred to as a reference space). In Conventional Example 1, a time-series pressure data group when dispensing is normally performed is used as a reference, and this group becomes a reference space.

  The Mahalanobis distance when dispensing is normally performed shows a value in the vicinity of 1, whereas if abnormal, the distance takes a value much larger than 1. Using this, the normality / abnormality of dispensing is determined by threshold determination (S3).

  As described above, the sample dispensing apparatus of Conventional Example 1 and the automatic analyzer using the same can provide a highly accurate determination result even if there is a variation in the sensitivity of the pressure sensor.

  The dispensing apparatus of Conventional Example 2 detects the pressure of the pressure transmission liquid, the dispensing pump that sends and sucks the pressure transmission liquid, the nozzle that sucks and discharges the sample liquid to be dispensed by the operation of the dispensing pump. A pressure sensor and an abnormality detection means are provided. Prior to dispensing the target liquid, the dispensing pump is preliminarily operated under the same operating conditions as when the dispensing is performed and with the tip opening of the nozzle not in contact with the liquid. The memory stores reference pressure data representing the pressure fluctuation detected by the pressure sensor during operation, and the abnormality detection means is operated by the pressure sensor when the dispensing pump actually operates to dispense the target liquid. Inhalation abnormality and ejection abnormality are detected based on the dispensing pressure data representing the detected pressure fluctuation and the reference pressure data (see, for example, Patent Document 2).

  FIG. 15 is a diagram schematically showing an embodiment of a dispensing device of Conventional Example 2.

  As shown in FIG. 15, a dispensing apparatus 201 of Conventional Example 2 has a dispensing pump 202 fixedly installed in an apparatus main body (not shown) and a sample liquid (target liquid) 300 to be dispensed from its distal end opening 231. It has a nozzle 203 for sucking and discharging, a nozzle moving mechanism 204 for moving the nozzle 203 in the horizontal direction and the vertical direction, and a control means 205 for controlling the operation of the dispensing pump 202 and the nozzle moving mechanism 204.

  The dispensing pump 202 is a syringe-type pump, and includes a cylinder 221, a piston 222 slidably installed in the cylinder 221, and a pump drive unit 223 that moves the piston 222. Although detailed illustration of the pump drive unit 223 is omitted, for example, the piston 222 is driven by a feed screw (ball screw) and a servo motor (stepping motor) that rotates the feed screw (ball screw). A space surrounded by the cylinder 221 and the piston 222 is filled with the pressure transmission liquid 400. When the piston 222 moves, the dispensing pump 202 delivers the pressure transmission liquid 400 to the nozzle 203 side or sucks it from the nozzle 203 side.

  As shown in FIG. 15, the dispensing device 201 of Conventional Example 2 is detected by a pressure sensor 209 that detects the pressure of the pressure transmission liquid 300 in the path from the dispensing pump 202 to the nozzle 203, and the pressure sensor 209. And a memory 212 that can store the pressure data, and an abnormality detection unit 213 that detects dispensing abnormality based on the pressure data detected by the pressure sensor 209.

  In the dispensing apparatus 201 of Conventional Example 2, dispensing abnormality is detected by the method described below.

[1] In order to inhale the sample liquid 300 into the nozzle 203, the control unit 205 moves the nozzle 203 above the dispensing source container 500. Next, an air layer 700 is formed at the tip of the nozzle 203 by operating the dispensing pump 202 with a predetermined amount of suction.

[2] Next, the nozzle 203 is lowered toward the dispensing source container 500. When the nozzle 203 is being lowered and the tip opening 231 is not yet in contact with the sample liquid 300 (a state in the air), the control means 205 performs the actual dispensing thereafter. The dispensing pump 202 is preliminarily operated under the same operating conditions as in FIG.

  In this preliminary operation, the dispensing pump 202 first performs a preliminary suction operation. This preliminary suction operation is performed at the driving speed and driving amount of the piston 222, which is the same as when the specimen liquid 300 is actually sucked into the nozzle 203. When this preliminary delivery operation is performed, air is sucked from the tip opening 231 of the nozzle 203. Subsequently, the dispensing pump 202 performs a preliminary delivery operation. This preliminary delivery operation is performed at the same driving speed and driving amount of the piston 222 as when the specimen liquid 300 is actually discharged from the nozzle 203. When this preliminary delivery operation is performed, air is discharged from the tip opening 231 of the nozzle 203.

[3] FIG. 16 is a graph showing a change in pressure detected by the pressure sensor 209 when the preliminary operation as described above is performed. As shown in the figure, when the dispensing pump 202 performs the preliminary suction operation or the preliminary delivery operation, the pressure transmission liquid 400 flows in the pipe 206 and the tube pipe 207. In addition to the pressure loss, the pressure detected by the pressure sensor 209 oscillates at the start and end of the suction / delivery operation due to the influence of the unsteady flow. Further, in FIG. 16, the pressure is decreasing as a general tendency with the passage of time due to a change in the water head pressure accompanying the lowering of the nozzle 203.

[4] The abnormality detection means 213 takes in the pressure detected by the pressure sensor 209 via the pressure sensor output circuit 214, and corrects the water head pressure as follows. The height Z of the nozzle 203 during the preliminary dispensing operation when the height at which the nozzle 203 is located at the start of the preliminary dispensing operation is used as the origin is acquired from the control means 205, and the water head difference ρgZ (ρ is the pressure transmission liquid 400). Density, g is the gravitational acceleration). Further, simply it may be corrected by the difference between the pressure P ₃ at the end the pressure P ₁ at the time of the preliminary dispensing start. Then, by correcting this water head pressure, reference pressure data as shown in FIG. 17 is obtained.

  This reference pressure data indicates that the pressure sensor 209 is affected by the pressure loss of the pressure transmission liquid 400 in the pipe, the dynamic pressure of the flow, and the unsteady flow when the dispensing pump 202 operates in the same manner as during actual dispensing. This is data representing the effect on the pressure detected by.

The abnormality detecting means 213 stores (stores) the reference pressure data in the preliminary suction operation in the memory 212 as the preliminary suction pressure data D ₁ and the reference data in the preliminary delivery operation as the preliminary delivery pressure data D _2. . Preliminary during suction pressure data _{D 1} is a set of multiple values acquired at a predetermined time interval (sampling time), i.e. _{D 1i (i = 0,1,2, ···} , n) is composed of Similarly, pre-delivery time pressure data _{D 2} is composed of a _{D 2i (i = 0,1,2, ···} , m). The measurement start points (D ₁₀ and D ₂₀ ) for D ₁ and D ₂ may be set at or slightly before the pump operation start time with reference to the pump operation start.

[5] When the descending nozzle 203 supplies the sample liquid 300, the nozzle 203 is stopped. At this time, the liquid level may be detected by a known method. Next, the dispensing pump 202 is operated for main suction, and the sample liquid 300 is actually sucked into the nozzle 203.

[6] When inhaling a sample solution 300 into the nozzle 203, the abnormality detecting means 213, suction time pressure data detected by the pressure sensor 209 (dispense time pressure data) at the D ₃ preliminary suction pressure data D ₁ is obtained at the same measurement start point and measurement interval as those of 1 (see FIG. 18). Then, the abnormality detecting means 213 is calculated based on the difference data DA of the preliminary suction pressure during data D ₁ and the suction pressure during data D ₃ in the following formula.
DA _i = D _3i −D _1i + ΔP (i = 0, 1, 2,..., N)
Here, [Delta] P, _{P 1} the pressure of the pre-suction operation start time, when the pressure at the start suction of the sample solution 300 was set to _{P 3,} a ΔP ₌ P 3 -P _1, by using this [Delta] P, the difference in water head pressure at the time of the preliminary suction and pressure data D ₁ and the suction pressure during data D ₃ can be corrected. ΔP may be directly calculated based on the above-described equation ΔP = P ₃ −P ₁ using the value detected by the pressure sensor 209, or the height of the nozzle 203 and the inhalation of the sample liquid 300 at the start of the preliminary suction operation A difference ΔZ with respect to the height of the nozzle 203 at the time may be acquired from the control means 205 and calculated as ΔP = ρgΔZ.

When the inhalation operation of the sample liquid 300 is normal, the difference data DA is, for example, as shown in the graph in FIG. The abnormality detection means 213 detects an inhalation abnormality based on the difference data DA. For example, it is determined that if the minimum value DA _min difference data DA is less than the threshold value P _T1, it is determined that the suction clogging occurs, not sucking jam otherwise occur. Similarly, by using another threshold value or recognizing the fluctuation pattern of the difference data DA, it is possible to detect an inhalation abnormality such as a shortage of inhalation amount (short sample) or bubble inhalation.

Difference data DA, so is obtained by subtracting the pre-suction time pressure data D ₁ from the actual intake pressure during data D _3, removed the influence of the pressure loss of the pressure transmitting fluid 400 and fluid dynamic pressure in the pipe Therefore, only the pressure change due to the nozzle 203 inhaling the sample liquid 300 is accurately represented. Therefore, by making a determination based on the difference data DA, the abnormality detection unit 213 can accurately determine the presence or absence of the inhalation abnormality as described above.

  The abnormality detection means 213 may make the determination by comparing the difference data DA itself with the threshold value as described above, or may use another known technique to determine the primary partial value of the difference data DA. Inhalation abnormalities such as suction clogging, inhalation amount shortage (short sample), and air bubble inhalation can be detected by calculating the second derivative value and recognizing the fluctuation pattern of DA.

[7] When the abnormality detection unit 213 detects the occurrence of an inhalation abnormality, the notification unit (not shown) notifies the operator to that effect, and discharges the already inhaled sample liquid 300 if necessary. . Thereafter, after the nozzle 203 is washed, another attempt is made to dispense the same dispensing source container 500, or another dispensing source container 500 is dispensed.

[8] If the abnormality detection unit 213 does not detect the occurrence of an abnormal inhalation, that is, if the sample liquid 300 can be normally inhaled, the control unit 205 moves the nozzle 203 above the dispensing container 600. Then, after being lowered to a predetermined height, the dispensing pump 202 is operated for main delivery, and the sample liquid 300 in the nozzle 203 is discharged into the dispensing container 600. When discharging a sample solution 300 from the nozzle 203, the abnormality detecting means 213, detected during ejection pressure data (dispense time pressure data) D ₄ The same measurement as the pre-delivery-time pressure data D ₂ to the pressure sensor 209 Acquired at the starting point and at the same measurement interval (see FIG. 20). Then, the abnormality detecting means 213, the difference data DD of the preliminary delivery-time pressure data D ₂ and the discharge-time pressure data D ₄ is calculated based on the following formula.
DD _i = D _4i −D _2i + ΔP ′ (i = 0, 1, 2,..., M)
Here, DerutaP' which, when the pressure of the pre-suction operation start time and _{P i,} the pressure at the discharge starting sample liquid 300 and P4, ΔP' ₌ a P 4 -P _i, using this DerutaP' It makes it possible to correct the difference in water head pressure of the preliminary delivery-time pressure data D ₂ and the discharge-time pressure data D _4. ΔP ′ may be directly calculated based on the equation ΔP ′ = P ₄ −P _i using the value detected by the pressure sensor 209, or the height of the nozzle 203 and the sample liquid 300 at the start of the preliminary delivery operation. The difference ΔZ ′ from the height of the nozzle 203 at the time of discharging may be obtained from the control unit 205 and calculated as ΔP ′ = ρgΔZ ′.

When the ejection operation of the sample liquid 300 is normal, the difference data DD is, for example, as shown in the graph shown in FIG. The abnormality detection unit 213 detects a discharge abnormality based on the difference data DD. For example, it is determined that if the maximum value DD _max of the differential data DD is larger than the threshold P _T2, it is determined that the discharge jam, no discharge jam otherwise occur.

Difference data DD, since minus the pre-delivery time pressure data D ₂ from the actual discharge time pressure data D _4, removed the influence of the pressure loss of the pressure transmitting fluid 400 and fluid dynamic pressure in the pipe Therefore, only the pressure change due to the discharge of the sample liquid 300 by the nozzle 203 is accurately represented. Therefore, by making a determination based on the difference data DD, the abnormality detection unit 213 can accurately determine the presence or absence of the ejection abnormality as described above.

Thus, the dispensing device of Conventional Example 2 can accurately determine and detect the presence / absence of dispensing abnormality based on the output data of the pressure sensor.
JP 2004-125780 A (page 17, FIGS. 1 and 3) Japanese Patent Laying-Open No. 2005-227102 (pages 12, 13 and 1, 5-10)

  However, in the conventional example, a pressure sensor is connected to a dispensing flow path system including a sample probe and a dispensing syringe, and the output value of the pressure sensor during the sample dispensing operation is compared with a reference pressure value. Although the abnormal dispensing is detected, the pressure sensor has a long response time and the sensitivity varies, so that there is a problem that the determination of dispensing abnormalities is delayed. In addition, there is a problem that abnormality determination of the dispensing actuator cannot be performed.

  The present invention has been made in view of such problems, and includes a discharge operation storage unit that stores operation-related data at the time of discharge operation of the dispensing actuator in time series, and normal discharge of the dispensing actuator. Normal discharge operation storage unit that stores operation-related data during operation in time series, and operation-related data and normal discharge operation during dispensing operation stored in the discharge operation storage unit in time series Provided is a liquid sample dispensing device that performs multivariate analysis on the operation-related data during the dispensing operation of the dispensing actuator stored in the storage unit, and determines whether or not there is an abnormality in the dispensing actuator based on the analysis result For the purpose.

In order to solve the above problem, the present invention is configured as follows.
The invention according to claim 1 is a dispensing nozzle for sucking and holding a liquid sample, a syringe composed of a cylinder and a plunger for generating suction / discharge pressure of the liquid sample, a flexible tube for flowing the liquid sample, A dispensing actuator that performs suction and discharge of a liquid sample from the dispensing nozzle via the syringe and the flexible tube, a dispensing actuator driving unit that drives the dispensing actuator, and the dispensing In the liquid sample dispensing apparatus, comprising a dispensing actuator control unit for controlling the actuator driving unit, and dispensing a predetermined amount of the liquid sample held by the dispensing nozzle into a plurality of reaction containers. A discharge operation storage unit for storing operation-related data in a time-series manner during the discharge operation of the actuator; A normal discharge operation storage unit that stores operation-related data in a normal discharge operation of the actuator in time series, and an operation related in the discharge operation of the dispensing actuator stored in the discharge operation storage unit in time series Multivariate analysis of the data and the operation-related data during the normal discharge operation of the dispensing actuator stored in the normal discharge operation storage unit, and the presence or absence of abnormality of the dispensing actuator is determined based on the analysis result An ejection operation determination unit is provided.

  According to a second aspect of the present invention, there is provided a dispensing nozzle for sucking and holding a liquid sample, a syringe composed of a cylinder and a plunger for generating a suction / discharge pressure of the liquid sample, a flexible tube for flowing the liquid sample, A dispensing actuator that performs suction and discharge of a liquid sample from the dispensing nozzle via the syringe and the flexible tube, a dispensing actuator driving unit that drives the dispensing actuator, and the dispensing In the liquid sample dispensing apparatus, comprising a dispensing actuator control unit for controlling the actuator driving unit, and dispensing a predetermined amount of the liquid sample held by the dispensing nozzle into a plurality of reaction containers. A discharge operation storage unit for storing operation-related data in a time-series manner during the discharge operation of the actuator; A normal discharge operation storage unit that stores operation-related data in a normal discharge operation of the actuator in time series, and an operation related in the discharge operation of the dispensing actuator stored in the discharge operation storage unit in time series The Mahalanobis distance is calculated from the above-mentioned data and the operation-related data during the normal discharge operation of the dispensing actuator stored in the normal discharge operation storage unit, and the calculated result is compared with a predetermined threshold value. Is used to determine whether or not the dispensing actuator is abnormal.

  According to a third aspect of the present invention, operation-related data during the normal discharge operation is prepared in accordance with a dispensing condition, the operation-related data during the dispensing operation of the dispensing actuator and the dispensing actuator The presence or absence of abnormality of the dispensing actuator is determined by comparing the Mahalanobis distance calculated from the operation-related data during normal ejection with a predetermined threshold value.

  The invention according to claim 4 selectively uses speed, acceleration, deceleration, and thrust during a plurality of discharge operations as operation-related data during the discharge operation of the dispensing actuator.

  When the dispensing actuator determines that the dispensing actuator is abnormal, the invention according to claim 5 calculates a difference from each operation command of operation-related data during the dispensing operation of the dispensing actuator, An operation command correction unit that uses the sum of the differences as a new operation command is provided.

  According to invention of Claim 1 and 2, the discharge operation | movement memory | storage part which memorize | stores the operation related data at the time of discharge operation of the said dispensing actuator in time series, and the time of normal discharge operation of the said actuator for dispensing Normal discharge operation storage unit that stores operation-related data in time series, operation-related data in the discharge operation of the dispensing actuator stored in the discharge operation storage unit in time series, and the normal discharge operation A multivariate analysis of the operation-related data during the discharge operation of the dispensing actuator stored in the storage unit, and a discharge operation determination unit that determines the presence or absence of an abnormality of the dispensing actuator based on the analysis result, By comparing the calculation result with a predetermined threshold value, it is possible to accurately determine whether or not the dispensing actuator is abnormal.

  According to the invention described in claim 3, since the operation-related data at the time of the normal discharge operation is prepared according to the dispensing condition, it is possible to accurately determine whether or not the dispensing actuator is abnormal under any dispensing condition. Can be determined.

  According to the invention described in claim 4, the operation-related data during the discharge operation of the dispensing actuator selectively uses the speed, acceleration, deceleration, and thrust during the plurality of discharge operations. The operating state of the dispensing actuator can always be monitored, and the presence or absence of an abnormality in the dispensing actuator can be accurately determined.

  According to the invention of claim 5, when the discharge condition determination unit determines that there is an abnormality in the dispensing actuator, a difference from each operation command of operation-related data during the discharge operation of the dispensing actuator Is calculated, and the sum of each operation command and the difference is used as a new operation command, whereby the operation when the dispensing actuator is abnormal can be corrected.

  Hereinafter, embodiments of the present invention will be described with reference to the drawings.

  FIG. 1 is a diagram showing a configuration of a liquid sample dispensing apparatus according to a first embodiment of the present invention. 1 is a liquid sample dispensing apparatus, 2 is a dispensing nozzle for sucking and holding a liquid sample, 3 is a flexible tube made of a material such as urethane or fluororesin that flows the liquid sample, and 4 is a syringe composed of a cylinder and a plunger. 5 is a cylinder, 6 is a plunger that discharges and sucks a liquid sample in the syringe, 7 is a dispensing actuator (for example, a linear motor), 8 is a dispensing actuator driving unit that drives the dispensing actuator, and 9 is Dispensing actuator control unit for controlling the dispensing actuator drive unit, 10 is a CPU in the dispensing actuator control unit, 11 is a discharge amount with respect to the stroke of the plunger 6 in the syringe 4 in the dispensing actuator control unit And a discharge condition storage unit for storing the number of discharges after suction of the liquid sample, and 12 is a discharge operation of the dispensing actuator. A discharge operation storage unit that stores time-related operation-related data, 13 is a normal discharge operation memory that stores operation-related data during a normal discharge operation of the dispensing actuator, and 14 is a discharge operation. Multi-variate analysis of operation-related data at the time of discharge operation of the dispensing actuator and operation-related data at the time of normal discharge operation stored in time series in the storage unit, and the actuator for dispensing based on the analysis result A discharge operation determination unit for determining whether there is an abnormality, 15 is a liquid sample container for supplying a liquid sample, 16 is a reaction container for discharging the liquid sample, and 17 is a liquid sample.

  The dispensing nozzle 2 is hollow, and is shaped so that the inner diameter becomes smaller toward the tip. The dispensing nozzle 2 is held in a posture perpendicular to the liquid sample 17, sucks the liquid sample 17 contained in the liquid sample container 15, and discharges the liquid sample 17 to the reaction container 16. Although not shown, the dispensing nozzle 2 sucks and discharges the cleaning water in the cleaning water container. Various general methods can be employed as a cleaning water supply method and a suction / discharge method.

  The cylinder 5 changes its volume in the syringe 4 as the plunger 6 moves. The cylinder 4 is connected to the dispensing nozzle 2 via the flexible tube 3. The plunger 6 moves upward on the surface of the syringe 4 so that the liquid sample 17 is sucked and the volume in the syringe 4 is increased, and the plunger 6 is discharged from the syringe 4 so that the volume in the syringe 4 is decreased by discharging the liquid sample 14. Move down the page. The plunger 6 is driven by a dispensing actuator 7 and is controlled by a dispensing actuator drive unit 8. The dispensing actuator drive unit 8 is connected to the dispensing actuator control unit 9 and drives the dispensing actuator 7 based on the operation command. The dispensing actuator control unit 9 includes a CPU 10 and a discharge condition storage unit 11, and commands an operation command to the dispensing actuator drive unit 8 to discharge a predetermined liquid sample 17. The CPU 10 is inputted by an operator in advance, or based on the number of samples read from the identification mark displayed on the liquid sample container 15 or the like and a desired analysis item, a required dispensing amount and a dispensing order. The dispensing actuator drive unit 8 is commanded to correspond to the above. The discharge condition storage unit 11 includes a stroke, a pressure, a nozzle diameter, a discharge speed, a liquid sample suction amount, an air suction amount, a discharge amount, the number of discharges after the liquid sample is sucked, and a suction-discharge operation pattern during the discharge. The relationship between the maximum suction speed, the suction acceleration / deceleration time, the maximum discharge speed, and the discharge acceleration / deceleration time is stored.

  Next, the operation of this embodiment will be described with reference to FIGS. FIG. 2 is a diagram illustrating a flow of the actuator control unit, and FIG. 3 is a diagram illustrating a suction state of the nozzle.

According to the procedure shown below, the liquid sample inhalation amount with respect to the discharge amount with good discharge accuracy is confirmed in advance by experiments. When the internal pressure is in a normal state where it does not decrease due to bubbles or the like, the fluid mass conservation law (inflow mass flow rate (density x flow velocity x cross-sectional area) and outflow flow rate (density x flow velocity x cross-sectional area) are equal) If it is established and the flow velocity at the tip of the nozzle is equal to or higher than the liquid running speed, the liquid can be discharged without causing a liquid pool. Therefore, if the internal pressure is kept in a normal state, for example, the maximum speed, acceleration, and deceleration, which are plunger operation patterns (trapezoid pattern), are fixed, and the air intake amount is a constant value, the plunger stroke in the syringe, liquid The relationship between the sample inhalation amount, the ejection amount, and the number of ejections after the liquid sample is aspirated is confirmed and summarized as shown in the table of FIG.
(1) First, the liquid sample inhalation amount is stored (F0).
The liquid sample suction amount with respect to the discharge amount measured in advance is stored in the discharge amount storage unit.
(2) Aspirate the washing water (F1).
When the plunger 6 in the syringe 4 is suction driven by the actuator 7, the cleaning water 18 in a cleaning water container (not shown) is sucked from the nozzle 2. Alternatively, the cleaning water may be supplied to the flexible tube and the nozzle 2 from a pump (not shown) via a valve (not shown) and another flexible tube (not shown).
(3) Air is sucked (F2).
The actuator 6 sucks and drives the plunger 6 in the syringe 4, and sucks a certain amount of air 19 from the nozzle 2 into the syringe 4.
(4) A liquid sample is aspirated (F3).
The plunger 6 in the syringe 4 is sucked and driven by the actuator 7, and the liquid sample 17 in the liquid sample container 15 is sucked from the nozzle 2 based on the liquid sample suction amount stored in advance.
(5) Discharge the liquid sample (F4)
In order to eliminate the influence of the initial meniscus at the tip of the nozzle, the actuator 7 is driven to discharge from the plunger 6 in the syringe 4 in accordance with a thrust command to the actuator 7 based on the discharge operation pattern stored in advance, and the discharge not shown from the nozzle 2 is performed. The liquid sample 17 is discharged into the container or the liquid sample container 15 for the number of times stored in advance.
(6) The liquid sample is discharged (F5).
The actuator 7 is driven to discharge from the plunger 6 in the syringe 4 by the thrust command to the actuator 7 based on the discharge operation pattern, and discharges the liquid sample 17 from the nozzle 2 to the reaction container 16.
(7) Repeat (2) to (6) for each discharge.

As described above, since the liquid sample inhalation amount with respect to the ejection amount and the number of ejections after the liquid sample aspiration are stored in advance and the liquid sample is ejected, both the CV value and the accuracy can be ejected with high accuracy.
When the average value of the discharge amount when discharging a predetermined number of times is σ, the standard deviation of the discharge amount when discharging a predetermined number of times is μ, and the target value of the discharge amount is α, the CV value and accuracy are They can be calculated by the equations (1) and (2), respectively.

In addition, it is assumed that the operator prepares the liquid sample container 15 and the reaction container 16 under the nozzle 2, but the control unit 9 synchronizes with the actuator 7 and the nozzle 2 or the liquid sample container 15 and the reaction container (not shown). The liquid sample may be automatically sucked and discharged by controlling a table or the like on which 16 is mounted.
In this embodiment, the discharge condition storage unit 11 is provided in the dispensing actuator control unit 9, but the discharge condition storage unit 11 may be provided in the dispensing actuator drive unit 8, or the dispensing actuator drive unit 8. Alternatively, the dispensing actuator driving unit 9 and the discharge condition storage unit 11 may be integrated.

  Next, a method for determining abnormal operation of the dispensing actuator will be described.

  FIG. 5 is a processing flow diagram of abnormality detection of the dispensing actuator according to the embodiment of the present invention.

  The operation-related data during the dispensing operation of the dispensing actuator is acquired from a storage unit inside the servo amplifier provided in the dispensing actuator drive unit 8 (not shown). In the present invention, operation-related data during the discharge operation of the dispensing actuator that changes for each discharge operation is read one by one along the time series, and a time-series data group related to these operations is acquired (G1). The discharge operation storage unit 12 stores operation-related data during the discharge operation of the dispensing actuator that changes for each discharge operation. The normal discharge operation storage unit 13 stores operation-related data during the normal discharge operation of the dispensing actuator.

  Next, the discharge condition determination unit 14 obtains the Mahalanobis distance from the acquired data group (G2). Here, the Mahalanobis distance is one of multivariate analyses, and is a scale for measuring whether a certain test object belongs to a reference group (hereinafter referred to as a reference space). In the present invention, the time series data group when the dispensing actuator operates normally is used as a reference, and this group becomes the reference space.

The Mahalanobis distance when the dispensing actuator operates normally is a value close to 1, but if it is abnormal, the Mahalanobis distance is a value greater than 1. Using this, the discharge condition determination unit 14 determines normality / abnormality of the dispensing actuator by threshold determination (G3).
For example, it is determined that the Mahalanobis distance is less than 1.1 is normal and 1.1 or more is abnormal. The threshold value is not necessarily limited to 1.1, and it is necessary to confirm in advance before operating the liquid sample dispensing apparatus.

  A method for acquiring a time series data group related to the operation of the dispensing actuator and a method for calculating the Mahalanobis distance will be described.

  FIG. 6 shows a method for obtaining a time-series data group related to the operation of the dispensing actuator, and is a diagram showing a method for obtaining a response speed as an example. From the time series data waveform related to the operation that changes every moment, the time series data of k points are taken as shown in FIG.

  The acquisition points of the time series data group related to the operation of the dispensing actuator may be the sample suction section, the sample discharge section, and all sections of the sample suction section and the sample discharge section.

  The interval at which the time series data group related to the operation of the dispensing actuator is taken may be constant or variable. It may be possible to narrow the capture interval at a location where the abnormality of the dispensing actuator is likely to occur, or to capture data roughly at a location where there is almost no abnormality.

  As a method for acquiring k-point time-series data groups, the method of obtaining only k-point data by preliminarily specifying the capture time of each point is initially acquired as finely as possible and thinned out to obtain k-point data. Either method of leaving data is acceptable.

  The time series data group related to the operation of the dispensing actuator acquired as described above is summarized as shown in FIG. Here, the time-series data group at each time is used as an item for obtaining the Mahalanobis distance.

  If the operation-related data at the time of the normal discharge operation of the dispensing actuator is obtained by the above operation, the time-series data group related to the operation of the reference dispensing actuator can be obtained. FIG. 8 summarizes each time series data group related to the operation of the dispensing actuator obtained when normal dispensing is performed n times, and is a reference space of n event k items.

Time series data set u ₁ of operation related dispensing actuator obtained by the _above, u _2, · · ·, the reference space of n events k items like Figure 8 where the uptake of u _k was performed n times, Average μ ₁ , μ ₂ ,..., Μ _{k for} each item
Then, standard deviations σ ₁ , σ ₂ ,..., Σ _k are obtained, and the calculation of Expression (3) is performed for normalization.

When the reference space is a matrix of n rows and k columns, and a correlation matrix of this matrix is obtained, a k × k matrix A is obtained. If the inverse matrix of the matrix A and A ^-1, the Mahalanobis distance D ² can be expressed by the equation (4).

  Of these calculations, the mean and standard deviation for each item in the reference space and the inverse matrix of the correlation matrix in the reference space are calculated in advance and stored, and the Mahalanobis distance is calculated repeatedly. These calculation processing times can be shortened.

  As described above, in this embodiment, the Mahalanobis distance in the time series data group related to the operation of the dispensing actuator can be obtained, and the presence or absence of the operation abnormality of the dispensing actuator can be determined.

  The present invention differs from Patent Document 1 and Patent Document 2 in that a discharge operation storage unit that stores operation-related data in a time-series manner during the discharge operation of the dispensing actuator and a normal discharge operation of the dispensing actuator. Normal discharge operation storage unit that stores operation-related data in time series, and operation-related data and normal discharge operation storage during the discharge operation of the dispensing actuator stored in time series in the discharge operation storage unit To provide a liquid sample dispensing device that performs multivariate analysis on the operation-related data during the dispensing operation of the dispensing actuator stored in the section and determines whether or not the dispensing actuator is abnormal based on the analysis result It is.

Since the configuration of the present embodiment is the same as that of the first embodiment, description thereof is omitted.
In this embodiment, the number and method of fetching operation-related time-series data groups during the dispensing operation of the dispensing actuator are changed for each dispensing amount, and a reference space is also provided for each dispensing amount. Liquid sample dispensing devices have minimum, maximum dispensing volume, and dispensing resolution determined by the specifications of the device. Therefore, it is possible to have a group of operation-related data and reference spaces for the dispensing operation of that number of dispensing actuators. That's fine.

  FIG. 9 summarizes each time-series data group related to the operation of the actuator for dispensing obtained when normal dispensing is performed n times, and is a reference space of n event k items, with a dispensing amount of 1 , Dispensed amount 2, dispensed amount 3,..., Prepared for each dispensed amount X.

  When dispensing a certain dispensing volume, calculate the Mahalanobis distance of the dispensing volume, compare it with a pre-stored reference space for that dispensing volume, and dispense according to the magnitude relationship with a predetermined threshold An abnormal operation of the actuator can be determined.

  As described above, in this embodiment, for each dispensing amount, the Mahalanobis distance is obtained from the time series data group related to the operation during the discharging operation of the dispensing actuator, and compared with a predetermined threshold value, thereby dispensing. It can be determined whether or not there is an abnormal operation of the actuator.

Since the configuration of the present embodiment is the same as that of the first embodiment, description thereof is omitted.
FIG. 10 summarizes the time series data groups of speed, acceleration, deceleration, and thrust of the actuator for dispensing obtained when normal dispensing is performed n times. It is a space, and is prepared for each dispensing amount 1, dispensing amount 2, dispensing amount 3, ..., dispensing amount X.

  When dispensing a certain dispensing volume, calculate the Mahalanobis distance of the dispensing volume, compare it with the pre-stored reference space for that dispensing volume, and dispense according to the magnitude relationship with a predetermined threshold It is possible to determine whether there is an abnormal operation of the actuator.

  In this way, in this embodiment, the operation-related data during the dispensing operation of the dispensing actuator is used by selectively using the speed, acceleration, deceleration, and thrust during the multiple dispensing operations. It is possible to determine whether there is an abnormal operation of the actuator.

  FIG. 11 is a diagram showing a configuration of a liquid sample dispensing apparatus according to the fourth embodiment of the present invention. In the configuration of the first embodiment, when it is determined that the dispensing actuator is abnormal, the difference between each operation command of the operation-related data during the discharge operation of the dispensing actuator is calculated, and each operation command and the difference are calculated. An operation command correction unit 20 that adds a new operation command is added.

  FIG. 12 is a diagram illustrating a flow of the operation command correction unit. A discharge operation storage unit that stores operation-related data during the discharge operation of the dispensing actuator in time series, and a discharge operation of the dispensing actuator that is stored in the discharge operation storage unit in time series The Mahalanobis distance is calculated from the operation-related data and the operation-related data stored in the normal discharge operation storage unit during the normal discharge operation of the dispensing actuator, and the calculated result is compared with a predetermined threshold value. In addition, when an abnormality of the dispensing actuator is determined, for example, when it is determined that the speed is abnormal, a new maximum speed command = current maximum speed command + (current maximum speed command−current maximum response speed) is set. If the operation of the dispensing actuator is not improved even with this setting, it is determined as abnormal, and the abnormality is presented to the user by a presentation means such as visual, auditory, or tactile sensation (not shown).

  Thus, in this example, when the discharge condition determination unit determines that there is an abnormality in the dispensing actuator, the difference between each operation command of the operation-related data during the dispensing operation of the dispensing actuator is calculated, By adding each operation command and the difference as a new operation command, it is possible to correct the operation when the dispensing actuator is abnormal.

  In addition, a discharge operation storage unit that stores operation-related data during the discharge operation of the dispensing actuator in time series, and a normal operation that stores operation-related data during the normal discharge operation of the dispensing actuator in time series Operation-related data at the time of discharge operation of the dispensing actuator stored in time series in the discharge operation storage unit and the discharge operation of the dispensing actuator stored in the normal discharge operation storage unit Since multivariate analysis is performed on the motion-related data and the presence / absence of abnormality of the dispensing actuator can be determined based on the analysis result, the present invention can be applied not only to the dispensing actuator but also to other actuators.

The figure which shows the structure of the liquid sample dispensing apparatus of the 1st Embodiment of this invention. The flowchart which shows operation | movement of the control part in the 1st Embodiment of this invention. Sectional drawing which shows the suction state of the nozzle of FIG. The table | surface which shows the relationship of the stroke of the plunger in the syringe of the 1st Embodiment of this invention, the liquid sample suction | inhalation amount, discharge amount, and the frequency | count of discharge after liquid sample suction. Processing flow chart of abnormality detection of the dispensing actuator of the first embodiment of the present invention This shows how to acquire time-series data groups related to the operation of dispensing actuators, and shows an example of response speed acquisition method. Table showing time-series data related to dispensing actuator operation The figure which summarizes each time series data group related to operation of the actuator for dispensing obtained when normal dispensing is performed n times, and shows a reference space of n events and k items A collection of time series data groups related to the operation of the dispensing actuator obtained when normal dispensing of the second embodiment of the present invention is performed n times, and a reference space for n events and k items Table prepared for each dispensing amount 1, dispensing amount 2, dispensing amount 3, ..., dispensing amount X Summarizing each time series data group of speed, acceleration, deceleration, and thrust of the actuator for dispensing obtained when normal dispensing of the third embodiment of the present invention is performed n times, This is a reference space for n events and k items, a table prepared for each dispensing amount 1, dispensing amount 2, dispensing amount 3, ..., dispensing amount X The figure which shows the structure of the liquid sample dispensing apparatus of the 4th Embodiment of this invention. The figure which shows the flow of the operation command correction | amendment part of the 4th Embodiment of this invention. The schematic block diagram of the sample dispensing apparatus of embodiment of the prior art example 1 Processing flow diagram of dispensing abnormality detection according to the embodiment of Conventional Example 1 The figure which shows typically embodiment of the dispensing apparatus of the prior art example 2 The graph which shows the change of the pressure detected by a pressure sensor, when the dispensing pump of embodiment of the prior art example 2 carries out preliminary operation | movement. The graph which shows the reference pressure data of embodiment of the prior art example 2 The graph which shows the actual pressure data at the time of inhalation of embodiment of the prior art example 2 The graph which shows the difference data of the pressure data at the time of preliminary suction of the embodiment of the prior art example 2, and the pressure data at the time of suction The graph which shows the actual discharge pressure data of embodiment of the prior art example 2 The graph which shows the difference data of the pressure data at the time of preliminary delivery and the pressure data at the time of discharge of embodiment of the prior art example 2

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Liquid sample dispensing apparatus 2 Dispensing nozzle 3 Flexible tube 4 Syringe 5 Cylinder 6 Plunger 7 Dispensing actuator 8 Dispensing actuator drive unit 9 Dispensing actuator control unit 10 CPU
11 Discharge condition storage unit 12 Discharge operation storage unit 13 Normal discharge operation storage unit 14 Discharge operation determination unit 15 Liquid sample container 16 Reaction vessel 17 Liquid sample 18 Washing water 19 Air 20 Operation command correction unit 101 Sample probe 102 Tube 103 Dispensing syringe 103a Cylinder 103b Plunger 104 Dispensing syringe driving means 105 Sample probe driving means 106 Control unit 107 Sample container 108 Sample 109 Water supply pump 110 Water supply tank 111 Washing water 112 Solenoid valve 113 Pressure sensor 114 Branch block 115 Amplifier 116 A / D converter 117 Micro Computer 201 Dispensing device 202 Dispensing pump 203 Nozzle 204 Nozzle moving mechanism 205 Control means 206 Pipe 207 Tube pipe 209 Pressure sensor 212 Memory 213 Abnormality detection means 214 Pressure sensor Output circuit 221 Cylinder 222 Piston 223 Pump drive unit 231 Tip opening 300 Sample liquid (target liquid)
400 Pressure transmission liquid 500 Dispensing source container 600 Dispensing destination container 700 Air layer

Claims (5)

A dispensing nozzle for sucking and holding a liquid sample, a syringe composed of a cylinder and a plunger for generating a suction and discharge pressure for the liquid sample, a flexible tube for flowing the liquid sample, and the syringe for sucking and discharging the liquid sample And a dispensing actuator to be implemented from the dispensing nozzle through the flexible tube, a dispensing actuator driving unit for driving the dispensing actuator, and a dispensing actuator for controlling the dispensing actuator driving unit In a liquid sample dispensing apparatus that includes an actuator control unit, and dispenses the liquid sample held by the dispensing nozzle into a plurality of reaction containers by a predetermined amount,
A discharge operation storage unit that stores operation-related data in a time-series manner during the discharge operation of the dispensing actuator;
A normal discharge operation storage unit that stores operation-related data in a time series in the normal discharge operation of the dispensing actuator;
Operation-related data at the time of discharge operation of the dispensing actuator stored in time series in the discharge operation storage unit and operation at the time of normal discharge operation of the dispensing actuator stored in the normal discharge operation storage unit A liquid sample dispensing apparatus comprising: a discharge operation determination unit that performs multivariate analysis on related data and determines whether or not the dispensing actuator is abnormal based on the analysis result. A dispensing nozzle for sucking and holding a liquid sample, a syringe composed of a cylinder and a plunger for generating a suction and discharge pressure for the liquid sample, a flexible tube for flowing the liquid sample, and the syringe for sucking and discharging the liquid sample And a dispensing actuator to be implemented from the dispensing nozzle through the flexible tube, a dispensing actuator driving unit for driving the dispensing actuator, and a dispensing actuator for controlling the dispensing actuator driving unit In a liquid sample dispensing apparatus that includes an actuator control unit, and dispenses the liquid sample held by the dispensing nozzle into a plurality of reaction containers by a predetermined amount,
A discharge operation storage unit that stores operation-related data in a time-series manner during the discharge operation of the dispensing actuator;
A normal discharge operation storage unit that stores operation-related data in a time series in the normal discharge operation of the dispensing actuator;
Operation-related data at the time of discharge operation of the dispensing actuator stored in time series in the discharge operation storage unit and operation at the time of normal discharge operation of the dispensing actuator stored in the normal discharge operation storage unit Liquid sample dispensing characterized in that it calculates Mahalanobis distance from related data and determines the presence or absence of abnormality of the dispensing actuator by comparing the calculated result with a predetermined threshold value. apparatus.   Operation-related data at the time of the normal discharge operation is prepared according to a dispensing condition, operation-related data at the time of the discharge operation of the dispensing actuator, and operation-related data at the time of normal discharge of the dispensing actuator, 3. The liquid sample dispensing apparatus according to claim 2, wherein the presence or absence of abnormality of the dispensing actuator is determined by comparing the Mahalanobis distance calculated from the above and a predetermined threshold value.   3. The liquid sample dispensing according to claim 2, wherein the operation-related data during the dispensing operation of the dispensing actuator selectively uses speed, acceleration, deceleration, and thrust during a plurality of dispensing operations. apparatus.   When the discharge condition determination unit determines that there is an abnormality in the dispensing actuator, it calculates a difference from each operation command of operation-related data during the discharge operation of the dispensing actuator, and each operation command and the difference The liquid sample dispensing device according to claim 2, further comprising an operation command correction unit that uses a sum of the values as a new operation command.