State determination system and liquid manufacturing supply system

Abstract

To continuously and easily grasp the state of the quality of a liquid to be inspected.SOLUTION: This device has a signal intensity detector 200 for continuously detecting signal intensity associated with impurities in the inspection object liquid, and an information processor 100 for determining a liquid quality state of the inspection object liquid on the basis of the signal intensity detected by the signal intensity detector 200.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present invention relates to a state determination system and a liquid manufacturing and supply system. [Background technology]

[0002] In a system for analyzing impurities in a test liquid using an adsorbent that adsorbs the impurities, a technique has been devised that switches between passing the test liquid through the adsorbent and passing an eluent that elutes the impurities adsorbed to the adsorbent through the adsorbent (see, for example, Patent Document 1). Also, a method has been devised that analyzes particles in the test liquid by passing the test liquid through a particle capture membrane, recovering the particle capture membrane through which the test liquid has passed, and observing the recovered particle capture membrane with a scanning electron microscope (SEM) (see, for example, Patent Document 2). [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Publication No. 2022-120536 [Patent Document 2] Patent Publication No. 2021-162564 Summary of the Invention [Problem to be solved by the invention]

[0004] The above-mentioned techniques have a problem in that it is not easy to continuously monitor the state of the liquid quality of the liquid to be tested.

[0005] An object of the present invention is to provide a state determination system and a liquid manufacturing and supply system that can continuously and easily grasp the state of the liquid quality of a liquid to be inspected. [Means for solving the problem]

[0006] The state determination system of the present invention comprises: a signal intensity detection device that continuously detects signal intensities associated with impurities in the test liquid; and an information processing device that determines the state of the quality of the test liquid based on the signal strength detected by the signal strength detection device.

[0007] The liquid manufacturing and supply system of the present invention further comprises: a signal intensity detection device that continuously detects signal intensities associated with impurities in the test liquid; an information processing device that determines the state of the liquid quality of the test liquid based on the signal strength detected by the signal strength detection device; a valve unit for controlling the supply of the test object liquid from a liquid production and supply facility that produces and / or supplies the test object liquid to a use point that uses the test object liquid; and a control device that controls the valve unit based on the state of the liquid quality of the test liquid determined by the information processing device. [Effects of the Invention]

[0008] In the present invention, the state of the liquid quality of the liquid to be tested can be continuously and easily monitored. [Brief explanation of the drawings]

[0009] [Figure 1] 1 is a diagram illustrating a first embodiment of a state determination system according to the present invention. [Figure 2] 2 is a diagram illustrating an example of components included in the information processing device illustrated in FIG. 1. FIG. [Figure 3] 2 is a flowchart illustrating an example of an information processing method in the information processing device shown in FIG. [Figure 4] 10 is a flowchart illustrating another example of an information processing method in the information processing device shown in FIG. [Figure 5] 3 is a diagram for explaining a method for determining the state of the test liquid in the determination unit shown in FIG. 2. FIG. [Figure 6] FIG. 2 is a diagram showing a second embodiment of the impurity capturing system of the present invention. [Figure 7] 7 is a diagram illustrating an example of components included in the information processing device illustrated in FIG. 6. FIG. [Figure 8] 7 is a diagram illustrating an example of input and output of the learning model illustrated in FIG. 6. [Figure 9] 7 is a flowchart illustrating an example of an information processing method in the information processing device shown in FIG. 6. [Figure 10] 1 is a diagram showing an example of a liquid manufacturing and supply system to which a state determination system of the present invention is applied. [Figure 11] FIG. 10 is a diagram showing another example of a liquid manufacturing and supply system to which the state determination system of the present invention is applied. DETAILED DESCRIPTION OF THE INVENTION

[0010] Hereinafter, an embodiment of the present invention will be described with reference to the drawings. (First embodiment)

[0011] FIG. 1 is a diagram showing a first embodiment of a condition determination system of the present invention. As shown in FIG. 1, the condition determination system of this embodiment includes an information processing device 100, a signal intensity detection device 200, a metering device 300, a flow meter 310, and an on-off valve 400. A branch path 20 branches off from a main path 10 that passes a test liquid from an ultrapure water production facility to a point of use. Here, the ultrapure water production facility produces ultrapure water to be supplied to a semiconductor cleaning device, which is a point of use, and supplies the ultrapure water to the semiconductor cleaning device. In the following description, this ultrapure water will be the liquid to be tested (test liquid), and the test liquid refers to the ultrapure water supplied from the ultrapure water production facility.

[0012] The flow meter 310 measures the flow rate of the test liquid passing through the main path 10. The flow meter 310 outputs the measured flow rate values ​​in time series to the information processing device 100. The flow meter 310 in this embodiment is included in the metering device 300. Note that a pressure gauge may be provided instead of the flow meter 310.

[0013] The on-off valve 400 is an on-off valve that controls the flow of the test liquid branching from the main path 10 to the branch path 20. The on-off valve 400 may be an on-off valve whose opening degree is adjustable. The on-off valve 400 controls opening / closing and the opening degree based on a control signal from the information processing device 100 or an external instruction.

[0014] The signal strength detection device 200 is a device that continuously detects (for example, at time intervals shorter than a predetermined time, such as 1 ms intervals) the signal strength of impurities in the test liquid passing through the branch path 20. The detected signal strength is an index that corresponds to the amount of impurities contained in the test liquid passing from the branch path 20 to the signal strength detection device 200. The impurities may be, for example, ionized impurities. Examples of the signal strength detection device 200 include an analyzer using ICP mass spectrometry (ICP / ICP-MS), an analyzer using single particle inductively coupled plasma mass spectrometry (spICPMS), an analyzer using liquid chromatography mass spectrometry (LC-MS), and an analyzer using an ion chromatograph or gas chromatograph (GC-MS). When an ICP-MS is used as the signal intensity detection device 200, the liquid to be tested is introduced into argon plasma in the ICP-MS, elements contained in the liquid to be tested are ionized, and the number of ions at the m / z (mass-to-charge ratio) of the ionized elements is measured as the signal intensity using a mass spectrometer. The predetermined time here may be, for example, the measurement time interval in the quantification device 300, or a preset time. The signal intensity detection device 200 outputs time-series data of the signal intensity of the detected impurities to the information processing device 100.

[0015] The quantification apparatus 300 is an apparatus that quantitatively measures impurities (e.g., metals, ions, fine particles, organic matter, etc.) contained in the test liquid passing through the branch path 20. For example, the quantification apparatus 300 measures the particle size and number of fine particles contained in the test liquid passing through the branch path 20. Furthermore, the quantification apparatus 300 may measure (calculate) the concentration of fine particles using the measured number of fine particles. In this case, the quantification apparatus 300 may be, for example, a particle counter or an apparatus that performs measurement using a method of trapping fine particles in a membrane (e.g., the method described in JP 2021-162564 A). Furthermore, for example, the quantification apparatus 300 measures the concentration of metal ions in the test liquid passing through the branch path 20. In this case, the quantification apparatus 300 may be, for example, an apparatus or method that performs analysis using metal concentration (e.g., the method described in WO 2019-221186 A or the system described in JP 2022-120536 A). In this case, the quantitative measurement device 300 may be equipped with, for example, an ion adsorption membrane or monolithic organic porous material that adsorbs impurities (ionic metal impurities or fine particles) contained in the test liquid passing through the branch path 20, or an ion exchanger unit filled with ion exchange resin. When the quantitative measurement device 300 is equipped with an ICP / ICP-MS, spICPMS, LC-MS, GC-MS, or the like, these may also be used as the signal intensity detection device 200. The quantitative measurement device 300 outputs time-series data of the measurement results to the information processing device 100. From the viewpoint of outputting time-series data of quantitative information, the quantitative measurement device 300 is preferably a particle counter.

[0016] The information processing device 100 determines the state of the liquid quality of the test target liquid based solely on the time series data of signal strength output from the signal strength detection device 200. Alternatively, the information processing device 100 may determine the state of the liquid quality of the test target liquid based on the time series data of signal strength output from the signal strength detection device 200 and the time series data of the measurement results output from the quantification device 300. Alternatively, the information processing device 100 may determine the state of the liquid quality of the test target liquid based on the time series data of signal strength output from the signal strength detection device 200, the time series data of the measurement results output from the quantification device 300, and the time series data of the flow rate values ​​output from the flowmeter 310.

[0017] Fig. 2 is a diagram showing an example of components included in the information processing device 100 shown in Fig. 1. As shown in Fig. 2, the information processing device 100 shown in Fig. 1 has an intensity information acquisition unit 110, a quantitative information acquisition unit 120, a determination unit 130, and an output unit 140. Note that Fig. 2 shows only the main components related to this embodiment among the components included in the information processing device 100 shown in Fig. 1.

[0018] The strength information acquisition unit 110 acquires strength information indicating time-series data of signal strength output from the signal strength detection device 200. The strength information acquisition unit 110 outputs the acquired strength information to the determination unit .

[0019] The quantitative information acquisition unit 120 acquires quantitative information indicating time-series data of measurement results output from the quantification apparatus 300. The quantitative information acquisition unit 120 also acquires quantitative information indicating time-series data of flow rate values ​​output from the flow meter 310. The quantitative information acquisition unit 120 outputs the acquired quantitative information to the determination unit 130. The quantitative information acquisition unit 120 may also acquire time-series data of operation management information that manages the flow of the test liquid as quantitative information. The operation management information is information indicating the operation status of a water treatment device (hereinafter referred to as an upstream water treatment device) and a control device (hereinafter referred to as an upstream control device) that are provided upstream of the arrangement position of the flow meter 310 on the main path 10. Examples of the operation management information include information indicating the operation status (operation / stop) of the upstream water treatment device, information indicating the open / close state of an on-off valve provided as an upstream control device, and information indicating the operation status (operation / stop) of a pump provided as an upstream control device for pumping the test liquid from a storage tank or a coagulation tank. When the quantitative information acquisition unit 120 acquires the time series data of the operation management information, the quantitative information acquisition unit 120 outputs the acquired time series data of the operation management information to the determination unit 130 .

[0020] The determination unit 130 determines the state of the test target liquid based on the time-series data of signal intensity indicated by the intensity information output from the intensity information acquisition unit 110. The determination unit 130 also determines the state of the test target liquid based on the time-series data of signal intensity indicated by the intensity information acquisition unit 110, and the time-series data of the metal ion concentration and the number of fine particles and particle size indicated by the quantitative information acquisition unit 120. In this case, the determination unit 130 can also identify impurities contained in the test target liquid. The determination unit 130 also determines the state of the test target liquid based on the time-series data of signal intensity indicated by the intensity information acquisition unit 110, the time-series data of the metal ion concentration and the number of fine particles and particle size indicated by the quantitative information acquisition unit 120, and the time-series data of the flow rate value indicated by the quantitative information acquisition unit 120. Furthermore, when time-series data of the operation management information is output from the quantitative information acquisition unit 120, the determination unit 130 identifies the cause of the change in the state of the test liquid based on the time-series data of the signal intensity indicated by the intensity information output from the intensity information acquisition unit 110, the time-series data of the metal ion concentration and the number of fine particles and particle size indicated by the quantitative information output from the quantitative information acquisition unit 120, and the time-series data of the operation management information. Specific determination (identification) methods will be described later. The determination unit 130 outputs the determination result to the output unit 140. Furthermore, the determination unit 130 compares the signal intensity indicated by the intensity information output from the intensity information acquisition unit 110 with a preset threshold. If the signal intensity indicated by the intensity information output from the intensity information acquisition unit 110 exceeds the threshold, the determination unit 130 notifies the output unit 140 of this fact. Furthermore, the determination unit 130 compares the particle concentration based on the number of fine particles indicated by the quantitative information output from the quantitative information acquisition unit 120 with a preset threshold. If the concentration of the particulate matter exceeds the threshold value, the determining unit 130 notifies the output unit 140 to that effect.

[0021] The output unit 140 outputs the determination result output from the determination unit 130. This output method may be displaying on a display or the like, printing using a printing device, or transmitting to another device. Furthermore, the output unit 140 performs a predetermined output when notified by the determination unit 130 that the signal strength has exceeded a threshold. Furthermore, the output unit 140 performs a predetermined output when notified by the determination unit 130 that the concentration of fine particles has exceeded a threshold. These output methods may be displaying on a display or the like, outputting a warning sound, turning on a warning light, or transmitting to another device.

[0022] The following describes an information processing method in the information processing device 100 shown in Fig. 1. Fig. 3 is a flowchart for explaining an example of an information processing method in the information processing device 100 shown in Fig. 1.

[0023] The intensity information acquisition unit 110 acquires intensity information indicating time-series data of the signal intensity of impurities contained in the test liquid from the signal intensity detection device 200 (step S1). Furthermore, the quantitative information acquisition unit 120 acquires quantitative information indicating time-series data of the measurement results output from the quantification device 300 (step S2). Either the process of step S1 or the process of step S2 may be performed first. Next, the determination unit 130 determines the state of the test liquid based on the time-series data of the signal intensity indicated by the intensity information acquired by the intensity information acquisition unit 110 in step S1, and the time-series data and particle size of the metal ion concentration and number of fine particles indicated by the quantitative information acquisition unit 120 in step S2 (step S3).

[0024] Fig. 4 is a flowchart for explaining another example of the information processing method in the information processing device 100 shown in Fig. 1. Here, the explanation will be given taking as an example a case where the quantification device 300 is a device equipped with an ion exchanger unit that concentrates the test target liquid.

[0025] The intensity information acquisition unit 110 acquires intensity information indicating time-series data of the signal intensity of impurities contained in the test liquid from the signal intensity detection device 200 (step S11). Then, the information processing device 100 determines whether it is time for the quantification device 300 to perform quantification (step S12). The timing for performing quantification is the timing for acquiring impurities contained in the test liquid in the elution step following the concentration step of the test liquid and calculating the concentration of the acquired impurities. Because quantification cannot be performed during the concentration step, the information processing device 100 needs to determine whether it is time for the quantification device 300 to perform quantification of impurities. The information processing device 100 may determine the timing for performing the concentration step and elution step based on a pre-planned operation schedule.

[0026] When the information processing device 100 determines that it is time for the quantification device 300 to perform quantification, the quantitative information acquisition unit 120 acquires quantitative information indicating time-series data of the measurement results output from the quantification device 300 (step S13). The determination unit 130 then determines the state of the test liquid based on the time-series data of signal intensity indicated by the intensity information acquired by the intensity information acquisition unit 110 in step S11, and the time-series data and particle size of the metal ion concentration and number of fine particles indicated by the quantitative information acquisition unit 120 in step S13 (step S14).

[0027] The following describes a method for determining the state of the test liquid in the determination unit 130 shown in FIG. 2. FIG. 5 is a diagram for explaining the method for determining the state of the test liquid in the determination unit 130 shown in FIG. 2. The upper graph in FIG. 5 is time-series data showing fluctuations in water quality over time as determined from the signal strength of impurities detected by the signal strength detection device 200 shown in FIG. 1 in the test liquid passing through the branch path 20. The middle graph in FIG. 5 is time-series data showing the impurity concentration / particle count measured over time in the test liquid passing through the branch path 20 by the quantitation device 300 shown in FIG. 1. The lower graph in FIG. 5 is time-series data showing the flow rate of the test liquid passing through the main path 10 as measured by the flowmeter 310 shown in FIG. 1. The quantitation device 300 is equipped with an ion exchanger unit, and during the concentration period in which the ion exchanger unit concentrates the test liquid, the quantitation device 300 cannot measure the impurity concentration. Therefore, the sampling interval for the impurity concentration / particle count shown in the middle graph of Figure 5 is longer than the sampling interval for the water quality fluctuations (signal strength) shown in the upper graph of Figure 5 and the sampling interval for the flow rate shown in the lower graph of Figure 5.

[0028] Referring to the data enclosed by the dashed-dotted line in the measurement results shown in Figure 5, it can be seen that the timing when the fluctuations in water quality fluctuated significantly and the timing when the flow rate increased are almost the same. It can also be seen that during the concentration period of the test liquid, when the impurity concentration / particle count increased, the fluctuations in water quality fluctuated significantly and the flow rate increased. This indicates that there is a correlation between the fluctuations in water quality, changes in flow rate, and changes in impurity concentration / particle count at the timing enclosed by the dashed-dotted line. In this case, the determination unit 130 determines that the fluctuations in water quality and changes in impurity concentration / particle count of the test liquid are caused by changes in the flow rate of the test liquid.

[0029] Furthermore, when a particle meter is used as the quantification device 300, if the number of particles measured by the quantification device 300 does not change during a time (period) when the fluctuations in water quality determined from the signal strength detected by the signal strength detection device 200 fluctuate significantly, the determination unit 130 determines that the increase in the number of ions is caused by ions, not particles. Furthermore, when a particle meter is used as the quantification device 300, if the number of particles measured by the quantification device 300 increases, the particle size of the particles measured by the quantification device 300 can be used to identify the particles causing the increase.

[0030] Furthermore, if the timing of a large change in the water quality fluctuation determined from the signal strength detected by the signal strength detection device 200 corresponds to the timing of a change in the operating status of the upstream water treatment device or the upstream control device indicated in the operation management information, the determination unit 130 determines that the cause of the water quality fluctuation determined from the signal strength detected by the signal strength detection device 200 is the change in the operating status of the upstream water treatment device or the upstream control device. If the measurement value measured by the quantification device 300 is changing, the determination unit 130 determines that the change in the measurement value measured by the quantification device 300 is caused by the change in the operating status of the upstream water treatment device or the upstream control device.

[0031] Generally, for the signal strength detection device 200 to measure ion concentration, it is necessary to convert the number of detected ions into ion concentration using a pre-created calibration curve. Furthermore, the signal strength detection device 200 does not provide high analytical accuracy for ion concentration, etc. On the other hand, when using an ion exchanger unit capable of measuring ion concentration as described above, analysis using the quantification device 300 requires pretreatment such as concentration, and the sampling interval becomes long, making continuous analysis difficult. While the techniques described in Patent Documents 1 and 2 above can perform analysis (quantification) with appropriate accuracy, they are unable to continuously grasp the state of water quality. Furthermore, even if the quantification device 300 uses only the flowmeter 310 or only time-series data from operation management information, it is difficult to analyze trends in impurities.

[0032] Therefore, in this embodiment, the determination unit 130 determines the state of the test liquid using the detection results from the signal strength detection device 200, which detects continuous changes in signal strength, and the measurement results from the quantification device 300, which performs analysis with appropriate accuracy. This makes it possible to make determinations that compensate for the inconvenient aspects of the signal strength detection device 200 and the quantification device 300 described above, and to continuously and easily perform analysis of the test liquid with appropriate accuracy. (Second embodiment)

[0033] FIG. 6 is a diagram illustrating a second embodiment of a status determination system according to the present invention. As shown in FIG. 6, the status determination system according to this embodiment includes an information processing device 101, a signal strength detection device 200, a metering device 300, a flowmeter 310, an on-off valve 400, and a learning model 500. A branch path 20 branches off from a main path 10, which carries a test liquid from an ultrapure water production facility to a point of use. Here, the ultrapure water production facility produces ultrapure water to be supplied to a semiconductor cleaning device, which is a point of use, and supplies the ultrapure water to the semiconductor cleaning device. In the following description, this ultrapure water is the liquid to be tested (test liquid), and the test liquid refers to the ultrapure water supplied from the ultrapure water production facility. The signal strength detection device 200, the metering device 300, the flowmeter 310, and the on-off valve 400 are the same as those in the first embodiment.

[0034] The information processing device 101 acquires time-series data of signal strength output from the signal strength detection device 200, time-series data of measurement results output from the quantification device 300, and time-series data of operational management information. In the learning phase of the learning model 500, the information processing device 101 inputs intensity information indicating the acquired time-series data of signal strength, quantitative information indicating the time-series data of measurement results, and time-series data of operational management information to the learning model 500. In the inference phase of the learning model 500, the information processing device 101 inputs the intensity information indicating the acquired time-series data of signal strength and the time-series data of operational management information to the learning model 500, and acquires status information indicating the status of the test liquid output from the learning model 500. This status information is, for example, information indicating quantitative values ​​(e.g., particle size, number of particle, particle concentration, metal ion concentration) and flow rate values ​​of impurities (e.g., metals, ions, fine particles, organic substances, etc.) contained in the test liquid passing through the branch path 20. In addition, the information processing device 101 may input the time series data of signal strength output from the signal strength detection device 200, the time series data of measurement results output from the quantification device 300, and the time series data of flow rate values ​​output from the flow meter 310 into the learning model 500, and obtain a status signal indicating the status of the liquid to be tested output from the learning model 500.

[0035] Fig. 7 is a diagram showing an example of components included in the information processing device 101 shown in Fig. 6. As shown in Fig. 7, the information processing device 101 shown in Fig. 6 has an intensity information acquisition unit 110, a quantitative information acquisition unit 120, a determination unit 131, an output unit 140, and a learning model generation unit 151. The intensity information acquisition unit 110, the quantitative information acquisition unit 120, and the output unit 140 are the same as those included in the information processing device 100 in the first embodiment. Note that Fig. 7 shows only the main components related to this embodiment among the components included in the information processing device 101 shown in Fig. 6.

[0036] The learning model generation unit 151 generates a learning model 500 using time series data of signal strength output as strength information from the strength information acquisition unit 110, time series data of measurement results output as quantitative information from the quantitative information acquisition unit 120, and time series data of operation management information output as quantitative information from the quantitative information acquisition unit 120.

[0037] The learning model 500 is a learning model generated by performing machine learning in advance on quantitative values ​​of impurities contained in the test liquid according to time-series data of signal strength and time-series data of operational management information. The learning model 500 may be provided within the information processing device 101. The learning model 500 may have, for example, a neural network structure in which multiple neurons are interconnected. A neuron is an element that performs a predetermined calculation on multiple inputs and outputs a single value as the calculation result. The learning method for the learning model 500 may be a general method for generating a learning model, that is, a method in which quantitative values ​​of impurities contained in the test liquid are learned based on time-series data of signal strength and time-series data of operational management information.

[0038] Fig. 8 is a diagram showing an example of input and output of the learning model 500 shown in Fig. 6. As shown in Fig. 8, when time-series data of signal intensity indicated by the intensity information and time-series data of operation management information indicated by the quantitative information are input, the learning model 500 shown in Fig. 6 outputs status information including information indicating impurities contained in the test target liquid and information indicating the quantitative values ​​of the impurities contained in the test target liquid, based on a learning model generated in advance by machine learning.

[0039] The determination unit 131 determines the state of the test target liquid based on the time-series data of signal intensity indicated by the intensity information output from the intensity information acquisition unit 110. Specifically, the determination unit 131 inputs the intensity information (time-series data of signal intensity) output from the intensity information acquisition unit 110 and the quantitative information (time-series data of operation management information) output from the quantitative information acquisition unit 120 into the learning model 500 and acquires the state information output from the learning model 500. The determination unit 131 also determines the state of the test target liquid based on the time-series data of signal intensity indicated by the intensity information output from the intensity information acquisition unit 110 and the time-series data and particle size of the metal ion concentration and the number of fine particles indicated by the state information output from the learning model 500. The determination unit 131 outputs the determination result to the output unit 140. Furthermore, the determination unit 131 compares the signal intensity indicated by the intensity information output from the intensity information acquisition unit 110 with a preset threshold. If the signal strength indicated by the strength information output from the strength information acquisition unit 110 exceeds the threshold, the determination unit 131 notifies the output unit 140 to that effect.

[0040] The following describes an information processing method in the information processing device 101 shown in Fig. 6. Fig. 9 is a flowchart for explaining an example of an information processing method in the information processing device 101 shown in Fig. 6.

[0041] The intensity information acquisition unit 110 acquires intensity information indicating time-series data of signal intensities of impurities contained in the test liquid from the signal intensity detection device 200 (step S21). Furthermore, the quantitative information acquisition unit 120 acquires quantitative information indicating time-series data of operation management information output from the quantification device 300 (step S22). Either the process of step S21 or the process of step S22 may be performed first. Subsequently, the determination unit 131 inputs the intensity information acquired by the intensity information acquisition unit 110 in step S21 and the quantitative information acquired by the quantitative information acquisition unit 120 in step S22 into the learning model 500 (step S23). Then, the determination unit 131 acquires the state information output from the learning model 500 (step S24).

[0042] As described above, in this embodiment, the determination unit 131 inputs the detection results of the signal strength detection device 200 and the operation management information output from the quantification device 300 into a learning model 500 that outputs the state of the test liquid, and acquires the state information output from the learning model 500. The determination unit 131 determines the state of the test liquid based on the acquired state information. This allows for determinations that compensate for the inconvenient aspects of the signal strength detection device 200 and the quantification device 300, and allows for continuous and easy analysis of the test liquid with appropriate accuracy. Furthermore, by using the pre-trained learning model 500, determinations can be made based on statistical trends.

[0043] The following describes an embodiment in which the above-described state determination system is used. FIG. 10 is a diagram showing an example of a liquid manufacturing and supply system to which the state determination system of the present invention is applied. The embodiment shown in FIG. 10 is a system in which ultrapure water is supplied to a semiconductor cleaning device (point of use) via CP1000, a non-regenerative ion exchange device, and UF1100, an ultrafiltration device, in an ultrapure water manufacturing facility. The ultrapure water (liquid to be tested) supplied to CP1000 is supplied from a liquid manufacturing and supply facility located upstream. The liquid manufacturing and supply facility is also a facility for manufacturing ultrapure water. The dashed lines in FIG. 10 indicate the water flow path or the path of a control signal for testing the water quality of the ultrapure water, which is the liquid to be tested.

[0044] There are two flow paths through which ultrapure water is supplied to the semiconductor cleaning apparatus, and one of the flow paths is provided with an impurity removal unit 1200, so that the ultrapure water is supplied to the semiconductor cleaning apparatus via the impurity removal unit 1200. In addition, an on-off valve 2000 is provided between the CP 1000 and the UF 1100. In addition, an on-off valve 2300 is provided to control the recovery of ultrapure water from the CP 1000 to an ultrapure water recovery tank. In addition, an on-off valve 2400 is provided to control the recovery of ultrapure water from the UF 1100 to the ultrapure water recovery tank. In addition, on-off valves 2100 and 2200 are provided on each of the two flow paths for supplying ultrapure water to the semiconductor cleaning apparatus.

[0045] 1 and 6, and performs the processing described in the first and second embodiments on the ultrapure water from the CP 1000 or the ultrapure water from the UF 1100, which is the liquid to be tested. The control device 1500 is a second control device that controls the opening and closing of the on-off valves 2000, 2100, 2200, 2300, and 2400 based on the state of the liquid to be tested determined by the state determination system 1.

[0046] When the state determination system 1 determines that the concentration of particles contained in the test target liquid exceeds the threshold, the control device 1500 controls the on-off valve 2000 to a closed state. At this time, the control device 1500 controls the on-off valve 2300 to an open state. Furthermore, when the state determination system 1 determines that the concentration of particles contained in the test target liquid is equal to or lower than the threshold, the control device 1500 opens the on-off valve 2000. At this time, the control device 1500 closes the on-off valve 2300. Furthermore, when the state determination system 1 determines that the concentration of particles contained in the test target liquid exceeds the threshold, the control device 1500 controls the on-off valves 2100 and 2200 to a closed state. At this time, the control device 1500 controls the on-off valve 2400 to an open state. Furthermore, when the state determination system 1 determines that the concentration of particles contained in the test target liquid is equal to or lower than the threshold, the control device 1500 opens the on-off valves 2100 and 2200. At this time, the control device 1500 controls the on-off valve 2400 to a closed state. Note that the control device 1500 may control the on-off valve 2100 to an open state when the state determination system 1 determines that the concentration of impurities contained in the test liquid is equal to or lower than the first concentration threshold, control the on-off valve 2200 to an open state and control the on-off valve 2100 to a closed state when the state determination system 1 determines that the concentration of impurities contained in the test liquid exceeds the first concentration threshold and is equal to or lower than the second concentration threshold, and control the on-off valves 2100 and 2200 to a closed state when the state determination system 1 determines that the concentration of impurities contained in the test liquid exceeds the second concentration threshold. This utilizes the fact that even if the impurity concentration in the flow path provided with the impurity removal unit 1200 is somewhat high, the impurities contained in the ultrapure water are removed by the impurity removal unit 1200, thereby reducing the impurity concentration of the ultrapure water supplied to the semiconductor cleaning apparatus. In addition, the control device 1500 may control the open / closed state of the on-off valves 2000, 2100, 2200, 2300, and 2400 based on the signal strength acquired by the state determination system 1, the type of impurity identified by the state determination system 1, and the cause of the state change identified by the state determination system 1.For example, if the signal strength acquired by the state determination system 1, the type of impurity identified by the state determination system 1, or the cause of the state change identified by the state determination system 1 does not allow the liquid to be tested to be supplied to the semiconductor cleaning equipment, which is the point of use, the on-off valve 2000 is controlled to a closed state.

[0047] FIG. 11 is a diagram showing another example of a liquid manufacturing and supply system to which the state determination system of the present invention is applied. In the application example shown in FIG. 11, the state determination system 1, CP1000, UF1100, control device 1500, and on-off valve 2400 are the same as the state determination system 1, CP1000, UF1100, control device 1500, and on-off valve 2400 shown in FIG. 10. Ultrapure water, which is the outlet water from the UF1100, is distributed to multiple flow paths and supplied to multiple semiconductor cleaning devices connected to each flow path. Each of the multiple flow paths branches off to the state determination system 1, and the ultrapure water flowing through each flow path is treated as the test liquid in the state determination system 1, as described in the first and second embodiments. The control device 1500 selects which flow path to treat the ultrapure water flowing through by controlling the opening and closing of on-off valves 2500-1 to 2500-4 provided in each branch flow path. Similarly to the above-described process, the control device 1500 controls the opening and closing of the on-off valves 2100-1 to 2100-4 provided in each flow path based on the state of the liquid under test determined by the state determination system 1. The control device 1500 may have conditions corresponding to each of the multiple semiconductor cleaning devices, and control the opening and closing of the on-off valves 2100-1 to 2100-4 based on the state of the liquid under test determined by the state determination system 1 and the conditions corresponding to each semiconductor cleaning device.

[0048] In this way, if the condition of the test liquid determined by the state determination system 1 is such that the test liquid is not suitable for supply to the semiconductor cleaning equipment (point of use), the on-off valve is controlled to prevent the supply of ultrapure water to the semiconductor cleaning equipment, thereby preventing contamination of the semiconductor device and components within the ultrapure water facility. The liquid (water) to be measured is not limited to ultrapure water, but may also be a chemical solution such as IPA (isopropyl alcohol), PGMA (polyglycerol methacrylate), or PGMEA (propylene glycol monomethyl ether acetate). While the embodiment using a bottle to collect the eluent has been described, the collected eluent may also be directly sprayed into an analytical device for quantitative analysis. The concentration of metal impurities measured by the state determination system 1 is not particularly limited, but is preferably 100 ng / L or less, preferably 1 ng / L or less, and more preferably 0.1 ng / L or less.

[0049] Although the above description has been given by allocating each function (process) to each component, this allocation is not limited to the above. Furthermore, the configuration of the components is also not limited to the above-described embodiments, which are merely examples. Furthermore, each embodiment may be combined.

[0050] The processes performed by each of the information processing devices 100 and 101 described above may be performed by logic circuits manufactured for each device depending on the purpose. Alternatively, a computer program (hereinafter referred to as a program) describing the process contents as procedures may be recorded on a recording medium readable by each of the information processing devices 100 and 101, and the program recorded on the recording medium may be read and executed by each of the information processing devices 100 and 101. Examples of recording media readable by each of the information processing devices 100 and 101 include removable recording media such as floppy (registered trademark) disks, magneto-optical disks, DVDs (Digital Versatile Discs), CDs (Compact Discs), Blu-ray (registered trademark) Discs, USB (Universal Serial Bus) memories, and SD cards, as well as memories such as ROMs (Read Only Memory), RAMs (Random Access Memory), and HDDs (Hard Disc Drives) built into each of the information processing devices 100 and 101. The program recorded on this recording medium is read by the CPU provided in each of the information processing devices 100 and 101, and the same processing as described above is performed under the control of the CPU. Here, the CPU operates as a computer that executes the program read from the recording medium on which the program is recorded.

[0051] A part or all of the above-described embodiments can be described as, but not limited to, the following supplementary notes. (Supplementary Note 1) An intensity information acquisition unit that acquires, from a signal intensity detection device, intensity information indicating signal intensities associated with impurities continuously detected in the test liquid; and a determination unit that determines the state of the liquid quality of the test object liquid based on the signal intensity indicated by the intensity information acquired by the intensity information acquisition unit. (Appendix 2) A process of acquiring, from a signal intensity detection device, intensity information indicating signal intensities associated with impurities continuously detected in the test liquid; and determining the state of the liquid quality of the test object liquid based on the signal intensity indicated by the acquired intensity information. (Appendix 3) To the computer, acquiring, from a signal intensity detection device, intensity information indicating signal intensities associated with impurities continuously detected in the test liquid; and a procedure for determining the state of the quality of the test liquid based on the signal intensity indicated by the acquired intensity information. [Explanation of symbols]

[0052] 1. Status determination system 10 Main Route 20 Branching Paths 100,101 Information processing equipment 110 Strength information acquisition unit 120 Quantitative Information Acquisition Department 130,131 Judgment part 140 Output section 151 Learning model generation unit 200 Signal Strength Detector 300 Quantification device 310 Flowmeter 400, 2000, 2100, 2100-1 to 2100-4, 2200, 2300, 2400, 2500-1 to 2500-4 On-off valve 500 Learning Models 1000 CP 1100 UF 1200 Impurity Removal Unit 1500 Control Device

Claims

1. a signal intensity detection device that continuously detects signal intensities associated with impurities in the test liquid; and an information processing device that determines the state of the quality of the test liquid based on the signal strength detected by the signal strength detection device.

2. The state determination system according to claim 1, a quantitative measurement device for quantitatively measuring impurities contained in the test liquid; The information processing device is a state determination system that determines the state of the liquid quality of the test object liquid based on the signal strength detected by the signal strength detection device and the measurement result of the quantification device.

3. 3. The state determination system according to claim 2, The information processing device is a state determination system that identifies impurities contained in the test liquid based on the signal strength detected by the signal strength detection device and the measurement results of the quantification device.

4. 4. The state determination system according to claim 2 or 3, The information processing device acquires time series data of operational management information that manages the flow of the test target liquid, and is a state determination system that identifies the cause of a change in the liquid quality state of the test target liquid based on the signal strength detected by the signal strength detection device, the measurement results of the quantification device, and the time series data of the operational management information.

5. 5. The state determination system according to claim 4, The information processing device is a state determination system that performs machine learning in advance to generate a learning model for the quantitative values ​​of impurities contained in the test liquid based on the time series data of the signal strength and the time series data of the operation management information.

6. The state determination system according to claim 5, The information processing device is a status determination system that inputs the signal strength detected by the signal strength detection device and the time series data of the operation management information into the learning model, and determines the liquid quality status of the test target liquid based on the status information indicating the status of the test target liquid output from the learning model.

7. 3. The state determination system according to claim 1, The information processing device is a state determination system that performs a predetermined output when the signal strength detected by the signal strength detection device exceeds a preset threshold.

8. 3. The state determination system according to claim 1, The signal intensity detection device is an ICP-MS, and is a state determination system that ionizes elements contained in the test liquid and measures the number of ions in the m / z (mass-to-charge ratio) of the ionized elements as signal intensity.

9. The state determination system according to claim 1 ; a valve unit for controlling the supply of the test object liquid from a liquid production and supply facility that produces and / or supplies the test object liquid to a use point that uses the test object liquid; a control device that controls the valve unit based on the liquid quality state of the test liquid determined by the information processing device.

Images