Flow sensor and correction device and correction method thereof, flow control device, and storage medium

Abstract

The present invention provides a correction device for a flow sensor, which can improve the accuracy of the linearity between the sensor output value (x) and the flow value (y) of the actual flow sensor and can correct the flow characteristic function compared with the past, and comprises: a sensitivity correction value function storage unit (22), which stores a sensitivity correction value function (m(x)), which sets at least a part of the sensitivity correction value (m) to a different value according to the sensor output value (x) output from the flow sensor (CM) in an initial state in which the sensitivity coefficient is set to an initial value; a sensitivity setting unit (23), which sets the sensitivity coefficient based on the initial value and the sensitivity correction value function (m(x)) and adjusts the sensitivity of the flow sensor (CM); and a function calculation unit (4), which calculates a corrected flow characteristic function (g(x)) based on a function correction value (n) determined according to the flow value (y) output from the flow sensor whose sensitivity is corrected by the sensitivity correction value function (m(x)) and a standard flow characteristic function (f(x)).

Description

Technical Field

[0001] The present invention relates to a correction device for a flow rate sensor for correcting a flow rate characteristic function for converting a sensor output value x into a flow rate value y in the flow rate sensor, etc. Background Art

[0002] For example, a thermal flow sensor is configured to measure flow rate by utilizing a predetermined relationship between flow rate and heat taken from an object. Specifically, two resistor elements are provided on a flow path as a sensor mechanism, and voltage is controlled in such a manner that the temperature of each resistor element is kept constant, and a sensor output value x having a predetermined relationship with the flow rate is calculated based on each voltage value at this time (see Patent Document 1).

[0003] Here, in order to convert each sensor output value x into a flow value y, a flow characteristic function is used in which the input is set to the sensor output value x and the output is set to the flow value y. This flow characteristic function is affected by the instrument error of the flow sensor, and each flow sensor has a unique flow characteristic function. Therefore, if, for example, a standard flow characteristic function f(x) obtained by averaging the sensor characteristics of multiple flow sensors is directly used, an error will occur between the calculated flow value y and the actual flow value y. Therefore, a correction is performed to make the standard flow characteristic function f(x) close to the actual flow characteristic function F(x) of the actual flow sensor.

[0004] For example, Figure 6 As shown in (a), the standard flow characteristic function f(x) substantially maintains linearity from the zero point to the span point.

[0005] Then, if Figure 6 As shown in (b), in an actual flow sensor, the sensor output value x when a standard gas such as N2 flows at a specified flow value is measured as a sample value. For the specified flow value, for example, the range flow value at the range point of the standard flow characteristic function f(x) is set. In addition, the sample value is input into the standard flow characteristic function f(x) to calculate the calculated flow value, and the slope is calculated by multiplying the standard flow characteristic function f(x) by a function correction value n, which is the value of dividing the specified flow value by the calculated flow value, to obtain an adjusted final flow characteristic function h(x).

[0006] However, if Figure 6 As shown in (c), when the actual flow characteristic function F(x) has poor linearity, if the above correction is performed, the final flow characteristic function h(x) outputs a flow value y that is smaller than the actual flow value y between the zero point and the span point. That is, a flow error occurs outside the span flow value.

[0007] In addition, when the inventors studied the reasons for the deterioration of the linearity of the actual flow characteristic function F(x), they believed that the following reasons were responsible. Figure 7 As shown in (a), in order to make the fluid flow through the capillary P of the sensor mechanism, the thermal flow sensor CM uses the diverting element DE to divert part of the fluid from the bypass flow channel BL. Figure 7 As shown in (b), the flow diverter element DE is formed by stacking a plurality of thin circular plates DE2 having a plurality of tiny holes DE1 in the axial direction. Such tiny holes DE1 are manufactured by, for example, etching, but it is difficult to make all holes of the same size, and deviations will occur in the processing accuracy anyway. In addition, since a plurality of thin circular plates DE2 are stacked, when a combination of thin circular plates DE2 is selected in a manner that well eliminates the processing accuracy, the flow sensor shows characteristics close to the standard flow characteristic function f(x). On the other hand, if a plurality of thin circular plates DE2 with poor processing accuracy are accidentally selected, the characteristics of the flow sensor CM will have a large nonlinearity.

[0008] In the case of a thermal flow sensor with extremely poor characteristics, even if the above-mentioned conventional correction for approaching the actual flow characteristic function F(x) is performed, it is still impossible to meet the accuracy standard of the thermal flow sensor, for example. Therefore, if the flow divider element DE with poor characteristics is installed, the yield rate related to flow adjustment of the mass flow controller as a whole will deteriorate.

[0009] Prior art literature

[0010] Patent Document 1: Japanese Patent Publication No. 2013-134234 Summary of the invention

[0011] The present invention has been made in view of the above-mentioned problems, and an object of the present invention is to provide a correction device that can improve the accuracy of the linearity between the sensor output value x and the flow value y of the actual flow sensor and can correct the flow characteristic function.

[0012] The correction device of the present invention corrects the flow characteristic function of a flow sensor, wherein the flow sensor comprises: a sensor mechanism that outputs an output signal corresponding to the flow rate of a fluid flowing through a flow channel; a sensor output calculation unit that calculates a sensor output value x based on a value represented by the output signal and a sensitivity coefficient; a function storage unit that stores a flow characteristic function in which an input is set to the sensor output value x and an output is set to the flow value y; and a flow calculation unit that calculates the flow value y based on the sensor output value x calculated by the sensor output calculation unit and the flow characteristic function. The correction device is characterized in that it comprises: a sensitivity correction value function storage unit that stores a sensitivity correction value function m(x), wherein the sensitivity correction value function m(x) is calculated based on the flow sensor in an initial state from which the sensitivity coefficient is set to an initial value. a sensitivity setting unit, which sets the sensitivity coefficient based on an initial value and the sensitivity correction value function m(x), and adjusts the sensitivity of the flow sensor; a function calculation unit, which calculates a corrected flow characteristic function g(x) based on a function correction value n and a standard flow characteristic function f(x), wherein the function correction value n is determined according to a flow value y output from the flow sensor whose sensitivity has been corrected by the sensitivity correction value function m(x); and a function correction unit, which stores the corrected flow characteristic function g(x) calculated by the function calculation unit or a final flow characteristic function h(x) based on the corrected flow characteristic function g(x) as a flow characteristic function in the function storage unit.

[0013] In addition, the correction method of the present invention corrects the flow characteristic function of a flow sensor, wherein the flow sensor comprises: a sensor mechanism that outputs an output signal corresponding to the flow rate of a fluid flowing through a flow channel; a sensor output calculation unit that calculates a sensor output value x based on a value represented by the output signal and a sensitivity coefficient; a function storage unit that stores a flow characteristic function in which an input is set to a sensor output value x and an output is set to a flow value y; and a flow calculation unit that calculates a flow value y based on the sensor output value x calculated by the sensor output calculation unit and the flow characteristic function, wherein the correction method comprises: storing a sensitivity correction value function m(x), wherein the sensitivity correction value function m(x) is calculated based on the sensitivity coefficient being set to an initial value in an initial state. The method comprises the following steps: setting the sensitivity coefficient based on the initial value and the sensitivity correction value function m(x), and adjusting the sensitivity of the flow sensor; calculating the corrected flow characteristic function g(x) based on the function correction value n and the standard flow characteristic function f(x), wherein the function correction value n is determined according to the flow value y output by the flow sensor whose sensitivity has been corrected by the sensitivity correction value function m(x); and storing the corrected flow characteristic function g(x) calculated by the function calculation unit or the final flow characteristic function h(x) based on the corrected flow characteristic function g(x) as the flow characteristic function in the function storage unit.

[0014] According to such a scheme, when the actual flow characteristic function F(x) of the flow sensor in the initial state in which the sensitivity coefficient is set to the initial value has poor linearity, it can be changed to a new sensitivity-adjusted actual flow characteristic function G(x) having good linearity based on the sensitivity correction value function m(x) in which at least a part of the sensitivity correction value m is set to a different value according to the sensor output value x. Therefore, the corrected flow characteristic function g(x) can be roughly consistent with the sensitivity-adjusted actual flow characteristic function G(x) as a whole, and flow error can be avoided across the entire range of the sensor output value x.

[0015] Furthermore, by using the sensitivity correction value function m(x), a different sensitivity coefficient can be set for each sensor output value x, so that, for example, a fine sensitivity correction can be made such that the sensitivity correction value m is larger where the error is large and the sensitivity correction value m is smaller where the error is small. Therefore, even a flow sensor having a characteristic that cannot be completely corrected when the sensitivity coefficient is corrected at a certain ratio without depending on the sensor output value x can be brought within the accuracy standard. As a result, the flow characteristic function can be corrected and the standard accuracy can be achieved, so that, for example, the yield rate of adjustment related to the flow characteristic of the entire flow control device in which the flow sensor is assembled can be improved.

[0016] In the corrected flow sensor storing the final flow characteristic function h(x), in order to prevent the sensor output value x or the flow value y from greatly exceeding the specified range, and to improve the linearity of the actual flow characteristic function G(x) after sensitivity adjustment so as to easily make the corrected flow characteristic function g(x) consistent, it is preferred that the sensitivity correction value function m(x) is set so that the smaller the sensor output value x, the larger the sensitivity correction value m.

[0017] In order to be able to set the sensitivity correction value function m(x) consistent with the characteristics of each flow sensor before correction based on the actual measurement value, it is preferred that the correction device further comprises: a sensitivity adjustment data storage unit, which stores a plurality of sets of sensitivity adjustment data, the sensitivity adjustment data being composed of a group of a sensitivity reference flow value F1 actually flowing in the flow channel, a sensor output value x output from the flow sensor in the initial state corresponding to the sensitivity reference flow value F1, and a flow value y output from the flow sensor in the initial state corresponding to the sensitivity reference flow value F1, that is, a flow value S1 before sensitivity adjustment; and a correction function output unit, which calculates the sensitivity correction value m corresponding to each sensor output value x according to the difference between the sensitivity reference flow value F1 and the flow value y based on the plurality of sets of sensitivity adjustment data, and calculates the sensitivity correction value function m(x) according to the plurality of sets of sensor output values ​​x and sensitivity correction values ​​m and stores them in the sensitivity correction value function storage unit.

[0018] In order to make the corrected flow characteristic function g(x) consistent with the actual flow characteristic function G(x) after sensitivity adjustment in the entire range of the sensor output value x, it is preferred that the correction device also has a data storage unit for range adjustment, the data storage unit for range adjustment stores range adjustment data, the range adjustment data is composed of a group of a range reference flow value F2 actually flowing in the flow channel and a flow value y output from the flow sensor whose sensitivity is corrected by the sensitivity correction value function m(x), and the function calculation unit calculates the corrected flow characteristic function g(x) based on the range adjustment data, taking the value obtained by dividing the range reference flow value F2 by the flow value y as the function correction value n and multiplying it by the standard flow characteristic function f(x).

[0019] In order to output the span flow value, which is the most important flow value y, from the flow sensor without error and to further improve reliability, it is preferred that the span reference flow value F2 is the span of the flow value y set in the flow sensor.

[0020] When the inventors conducted intensive research, they first discovered that the magnitude of the measurement error caused by the type of fluid flowing through the flow channel is related to the sensitivity correction value m. In this way, in order to correct the errors caused by the difference in gas types by sensitivity correction, it is preferred that the function correction unit includes: a gas type correction value calculation unit, which calculates the gas type correction coefficient k based on the thermal conductivity λ of the fluid and the sensitivity correction value m of the specified sensor output value x; and a final output unit, which multiplies the gas type correction coefficient k by the corrected flow characteristic function g(x) to calculate the final flow characteristic function h(x), and stores the final flow characteristic function h(x) in the function storage unit.

[0021] As a specific embodiment of calculating the correction according to the gas type from the known value in a manner that omits the operation of actually flowing the fluid in the individual flow sensor to obtain the actual measurement value and obtain the gas type correction coefficient k, it can be cited that the gas type correction value calculation unit calculates the function of the sensitivity correction value m of the specified sensor output value x, that is, the slope a(m), and calculates the gas type correction coefficient k based on the following formula:

[0022] k=a(m)×(1 / λ)^(1 / 2)+b

[0023] Here, λ is the thermal conductivity of the fluid flowing in the flow channel, and b is a predetermined intercept.

[0024] If the flow sensor is configured so as to output the flow value y based on the stored final flow characteristic function h(x) by the correction device of the present invention, the flow error can be reduced in the entire range of the sensor output value x, and the correct flow value y can be output. As a result, for example, the yield rate of adjustment related to the flow characteristic of the entire flow control device in which the flow sensor is assembled can be improved.

[0025] According to the flow rate control device including the flow rate sensor of the present invention, the flow rate control accuracy in the range of the sensor output value x can be made very high.

[0026] In order to be able to enjoy the same effect as the correction device of the present invention by updating the program used in the existing correction device, it is sufficient to use a program storage medium storing a correction device program, wherein the correction device program corrects the flow characteristic function of a flow sensor, and the flow sensor comprises: a sensor mechanism that outputs an output signal corresponding to the flow rate of a fluid flowing through a flow channel; a sensor output calculation unit that calculates a sensor output value x based on a value represented by the output signal and a sensitivity coefficient; a function storage unit that stores a flow characteristic function that sets an input to a sensor output value x and an output to a flow value y; and a flow calculation unit that calculates a flow value y based on the sensor output value x calculated by the sensor output calculation unit and the flow characteristic function, wherein the program storage medium is characterized in that the correction device program enables a computer to function as a sensitivity correction value function storage unit, a sensitivity setting unit, a function calculation unit, and a function correction unit, and the sensitivity correction value function storage unit The storage unit stores a sensitivity correction value function m(x), wherein the sensitivity correction value function m(x) sets at least a part of the sensitivity correction value m to a different value according to a sensor output value x outputted from the flow sensor in an initial state in which the sensitivity coefficient is set to an initial value. The sensitivity setting unit sets the sensitivity coefficient based on the initial value and the sensitivity correction value function m(x), and adjusts the sensitivity of the flow sensor. The function calculation unit calculates a corrected flow characteristic function g(x) based on a function correction value n and a standard flow characteristic function f(x). The function correction value n is determined according to a flow value y outputted from the flow sensor whose sensitivity has been corrected by the sensitivity correction value function m(x). The function correction unit stores the corrected flow characteristic function g(x) calculated by the function calculation unit or a final flow characteristic function h(x) based on the corrected flow characteristic function g(x) as a flow characteristic function in the function storage unit.

[0027] The correction device program may be a program distributed electronically or a program stored in a program storage medium such as a CD, a DVD, a HDD, or a flash memory.

[0028] Thus, according to the correction device of the flow sensor of the present invention, for example, the sensitivity correction value function m(x) can be used to finely adjust the sensitivity of the actual flow characteristic function F(x) with poor linearity in the initial state for each sensor flow value x. As a result, the actual flow characteristic function G(x) after sensitivity adjustment can be made very linear, and the corrected flow characteristic function g(x) can be made consistent with the function G(x) based on the standard flow characteristic function f(x) and the function correction value n, so that the flow error can be reduced across the substantially entire area of ​​the range of the sensor output value x. In addition, by using the sensitivity correction value function m(x), it is possible to correct a flow sensor that could not be corrected in the past, for example, it is possible to improve the yield rate of adjustment related to the flow characteristic of the entire flow control device as an assembly of the flow sensor. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] Figure 1 It is a schematic diagram showing a correction device according to one embodiment of the present invention.

[0030] Figure 2 It is a diagram showing the sensitivity adjustment result of the same embodiment.

[0031] Figure 3 1 is a diagram showing an example of the sensitivity correction value function m(x) in the sensitivity adjustment step of the same embodiment.

[0032] Figure 4 It is a diagram showing the result of function correction in the same embodiment.

[0033] Figure 5 1 is a diagram showing the relationship among the gas type, flow rate error, and sensitivity correction value m according to the same embodiment.

[0034] Figure 6 This is a diagram showing a conventional correction method.

[0035] Figure 7 This is a schematic diagram showing the structure of a thermal flow sensor.

[0036] Description of Reference Numerals

[0037] 100 Correction Device

[0038] 1 Sensitivity adjustment data storage section

[0039] 21 Correction function output section

[0040] 22 Sensitivity correction value function storage unit

[0041] 23. Sensitivity setting section

[0042] 3 Data storage section for range adjustment

[0043] 4 Function Computing Department

[0044] 5 Function Correction Department

[0045] 51 Gas type correction value calculation unit

[0046] 52 Final output section DETAILED DESCRIPTION

[0047] An embodiment of the present invention will be described with reference to the drawings.

[0048] like Figure 1 As shown, the flow correction device 100 of the present embodiment is used, for example, to correct a thermal flow sensor by reflecting an instrument error at the time of factory shipment and reducing the flow error in the entire range of the sensor output value x.

[0049] Here, the thermal flow sensor to be corrected constitutes a part of the thermal mass flow controller MFC. Figure 1 The device shown in the figure comprises: a valve V provided for a flow channel CH; a flow sensor CM for measuring a flow rate in the flow channel; and a valve control unit VC for controlling the opening of the valve V based on an output from the flow sensor CM, and these parts are integrated in one housing.

[0050] The details of the flow sensor CM to be corrected will be described.

[0051] like Figure 1 As shown, the flow sensor CM includes: a sensor mechanism M1 that outputs an output signal according to the flow rate of the fluid flowing through the flow channel CH; and a calculation unit CAL that calculates the flow rate based on the output from the sensor mechanism M1.

[0052] The sensor mechanism M1 includes: a U-shaped capillary tube, which is arranged to branch off from the flow path of the fluid flow and merge again; and a pair of resistance wires Ru and Rd, which are wound around the capillary tube. Each resistance wire Ru and Rd is connected to a constant temperature circuit (not shown) so that its temperature becomes a certain temperature. Because the fluid flowing through the capillary tube carries heat from the resistance wire Ru on the upstream side to the resistance wire Rd on the downstream side, the upstream voltage applied to the resistance wire Ru on the upstream side becomes a higher value than the downstream voltage applied to the resistance wire Rd on the downstream side. Because the difference between the upstream voltage and the downstream voltage has a prescribed relationship with the flow rate of the fluid, the flow rate is calculated in the calculation unit CAL based on these values. In this embodiment, the sensor mechanism M1 outputs an output signal representing the value of the following calculated value FI based on the upstream voltage and the downstream voltage.

[0053] FI=(Vu-Vd) / (Vu+Vd)

[0054] Here, FI is a value represented by an output signal of the sensor mechanism M1, Vu is an upstream voltage, and Vd is a downstream voltage.

[0055] The calculation unit CAL is, for example, a computer having a CPU, a memory, an A / D converter, a D / A converter, etc., and by executing a flow calculation program stored in the memory, each device cooperates to realize at least the functions of a sensor output calculation unit M2, a function storage unit M3, and a flow calculation unit M4.

[0056] The sensor output calculation unit M2 calculates the sensor output value x based on the value represented by the output signal output from the sensor mechanism M1 and the set sensitivity coefficient. More specifically, the sensor output calculation unit M2 is configured to multiply the set sensitivity coefficient amp by the value FI represented by the output signal of the sensor mechanism M1. In addition, the sensitivity coefficient amp is set to (100%) as an initial value. In addition, the initial value mentioned here is a value temporarily set in the flow sensor CM in a state before the flow error of the flow sensor CM is corrected by the flow correction device 100 of this embodiment.

[0057] That is, the sensor output calculation unit M2 performs the following calculation.

[0058] x=FI×amp

[0059] Here, x is the sensor output value, FI is the value represented by the output signal of the sensor mechanism M1, and amp is the sensitivity coefficient. In this embodiment, amp is not a fixed value, and amp can be set to different values ​​according to the sensor output x in the initial state where amp is set to 1, that is, the value of FI.

[0060] The function storage unit M3 stores a predetermined flow characteristic function in which the input is set to the sensor output value x and the output is set to the flow value y. In addition, in the present embodiment, a standard flow characteristic function f(x) is stored in the function storage unit M3 as a default flow characteristic function. The standard flow characteristic function f(x) is obtained by actually measuring the relationship between the sensor output value x and the flow value y in a plurality of flow sensors having the same structure as the thermal flow sensor to be corrected, and averaging the flow characteristic functions produced respectively. That is, the standard flow characteristic function f(x) is obtained by averaging the instrument errors that occur accidentally in each thermal flow sensor.

[0061] The flow rate calculation unit M4 is configured to calculate the flow rate value y based on the sensor output value x calculated by the sensor output calculation unit M2 and the flow rate characteristic function stored in the function storage unit M3 at that time.

[0062] Next, flow rate control of the mass flow controller MFC by the valve control unit VC will be described.

[0063] The valve control unit VC is configured to switch between two types of feedback control. That is, the valve control unit VC can perform either of the sensor output value control of feeding back the sensor output value x and the flow value control of feeding back the flow value y. The sensor output value control is a mode mainly used to obtain the value necessary for correction by the correction device 100. In this mode, the opening of the valve V is controlled in such a way that the deviation between the target sensor output value x assigned as the target value and the sensor output value x output from the sensor output calculation unit M2 becomes smaller. On the other hand, the flow value control is a mode mainly used for actual flow control. In this mode, the opening of the valve V is controlled in such a way that the deviation between the target flow value assigned as the target value and the flow value output from the flow calculation unit M4 becomes smaller.

[0064] In addition, when the correction device 100 corrects the thermal flow sensor CM as the correction object, a reference flow sensor BM serving as a reference is provided on the upstream side or downstream side of the mass flow controller MFC. The reference flow sensor BM is, for example, a pressure flow sensor that guarantees the value of its output with a predetermined accuracy. That is, the correction device 100 corrects the thermal flow sensor CM as the correction object based on the sensor output value x output from the thermal flow sensor as the correction object, the flow value y, and the flow value y indicated by the reference flow sensor BM serving as a reference.

[0065] Next, the details of the correction device 100 are described. The flow characteristic function initially set as a mathematical formula in the function storage unit M3 in the flow sensor CM to be corrected is recorded as a standard flow characteristic function f(x) to distinguish it from the actual flow characteristic function originally possessed by the flow sensor CM to be corrected, that is, the actual flow characteristic function F(x) before sensitivity adjustment and the actual flow characteristic function G(x) after sensitivity adjustment. In addition, the actual flow characteristic function F(x) before sensitivity adjustment and G(x) after sensitivity adjustment are unknown functions, reflecting the instrument error of each flow sensor CM to be corrected. The purpose of the correction performed by the correction device 100 is to make the standard flow characteristic function f(x) set in the flow sensor CM to be corrected close to the actual flow characteristic function G(x) after sensitivity adjustment. The correction of this embodiment first improves the linearity of the actual flow characteristic function F(x) by changing the sensitivity coefficient set in the sensor output calculation unit M2. Thereafter, correction is performed to make the standard flow characteristic function f(x) consistent with the actual flow characteristic function G(x) after sensitivity adjustment.

[0066] More specifically, the correction device 100 is implemented, for example, in a computer including a CPU, a memory, an input / output device, etc., by executing a correction device program stored in the memory and causing various devices to cooperate with each other, thereby achieving its functions.

[0067] That is, the correction device 100 at least functions as a sensitivity adjustment data storage unit 1, a correction function output unit 21, a sensitivity correction value function storage unit 22, a sensitivity setting unit 23, a range adjustment data storage unit 3, a function calculation unit 4, and a function correction unit 5.

[0068] The details of each part will be described together with the correction method using the correction device 100. For correction, when the flow value y is obtained from the flow sensor CM to be corrected, the fluid flowing in the flow path is, for example, a standard gas such as nitrogen.

[0069] In addition, as a premise, the standard flow characteristic function f(x) is as follows Figure 2 (a) shows a function that ensures approximately linearity within the range of the sensor output value x.

[0070] For such a standard flow characteristic function f(x), the actual flow characteristic function F(x) in the initial state sometimes has poor linearity, for example, the flow value y corresponding to each sensor output value x becomes larger. The following is an explanation based on the case where the actual flow characteristic function F(x) in the initial state is a function that is convex upward relative to the standard flow characteristic function f(x).

[0071] First, the mass flow controller MFC is operated by feedback control of the sensor output value x, and the flow control is performed in such a way that the target sensor output value x is output from the flow sensor CM to be corrected in the initial state. Here, the flow control is performed in such a way that a plurality of sensor output values ​​x of x1, x2, and x3 are obtained relative to the range as the target sensor flow value y. At this time, each flow value actually flowing in the flow channel CH is measured by the reference flow sensor BM and stored in the sensitivity adjustment data storage unit 1 as the sensitivity reference flow value F1 (F1(x1), F1(x2), F1(x3)). In addition, the flow value y output from the flow sensor CM to be corrected when the flow flows in the flow channel CH at the sensitivity reference flow value F1 (F1(x1), F1(x2), F1(x3)) is also stored in the sensitivity adjustment data storage unit 1 as the flow value S1 (S1(x1), S1(x2), S1(x3)) before sensitivity adjustment. Specifically, the sensitivity adjustment data storage unit 1 stores a sensor output value x obtained from the flow sensor CM and a flow value S1 before sensitivity adjustment, and a sensitivity reference flow value F1 obtained from the reference flow sensor BM as a plurality of sensitivity adjustment data each forming a pair of these.

[0072] In other words, the sensitivity adjustment data storage unit 1 stores the following information: Figure 2 (a) shows values ​​measured at multiple points on the actual flow rate characteristic function F(x) before sensitivity adjustment and multiple corresponding points on the standard flow rate characteristic function f(x).

[0073] The sensitivity setting unit 23 sets the sensitivity coefficient of the flow sensor CM to be corrected to a state where the initial value is multiplied by the sensitivity correction value m, and adjusts the sensitivity of the flow sensor CM. In this embodiment, the sensitivity setting unit 23 sets the state where different sensitivity correction values ​​m are multiplied according to the sensor output value x.

[0074] More specifically, the correction function output unit 21 calculates the sensitivity correction value m for each sensitivity adjustment data according to the difference between the sensitivity reference flow value F1 and the flow value S1 before sensitivity adjustment. For example, the larger the difference, the larger the sensitivity correction value m is calculated. The relationship between the difference and the sensitivity correction value m is determined experimentally, for example. In this embodiment, Figure 3 As shown in the figure, based on the three sensitivity correction values ​​m calculated corresponding to the three sensor output values ​​x, the sensitivity correction value function m(x) is calculated by linear interpolation using a quadratic function. The correction function output unit 21 stores the calculated sensitivity correction value function m(x) in the sensitivity correction value function storage unit 22. The sensitivity setting unit 23 uses the sensitivity correction function m(x) to change the sensitivity coefficient of the flow sensor CM to be corrected by multiplying the sensitivity correction value m corresponding to each sensor output x.

[0075] like Figure 2 As shown in (b), the actual flow rate characteristic function F(x) in the initial state is changed to the actual flow rate characteristic function G(x) in the state after sensitivity adjustment in which the sensitivity is m(x) times for each sensor output value x in the sensor output value axis direction.

[0076] In this example, the actual flow characteristic function F(x) in the initial state with poor linearity expands in the sensor output value axis direction, and the increase in the flow value y per unit sensor output value x becomes smaller. Here, since the sensitivity correction value m corresponding to each sensor output x is set separately, the actual flow characteristic function G(x) after sensitivity adjustment becomes a function with improved linearity compared with the actual flow characteristic function F(x) in the initial state. On the other hand, for the standard flow characteristic function f(x), even if the sensitivity coefficient is changed, the shape of its graph does not change and remains in the original state. Therefore, the linearity of the actual flow characteristic function G(x) after the sensitivity is adjusted using the sensitivity correction value function m(x) becomes approximately the same as the linearity of the standard flow characteristic function f(x).

[0077] Next, a correction procedure for the flow sensor CM that is the correction target in the sensitivity-adjusted state will be further described.

[0078] like Figure 4 As shown in (a), this time, the mass flow controller MFC after sensitivity adjustment performs flow control in such a way that the sensor output value x becomes 100%, and the flow measured by the reference flow sensor BM at this time, i.e., the range reference flow value F2, and the flow value y output from the correction object flow sensor CM in the state after sensitivity adjustment, i.e., the sensitivity adjusted flow value S2, are stored in the data storage unit 3 for range adjustment. Using these two flow values, the standard flow characteristic function f(x) is enlarged and reduced in the y-axis direction. In other words, the range correction is performed in such a way that the standard flow characteristic function f(x) is roughly consistent with the actual flow characteristic function G(x) in the state after sensitivity adjustment within the range of the sensor output value x.

[0079] More specifically, the function calculation unit 4 calculates the corrected flow characteristic function g(x) obtained by multiplying the function correction value n by the standard flow characteristic function f(x). Here, the function correction value n is a value obtained by dividing the range reference flow value F2 by the sensitivity adjusted flow value S2. The corrected flow characteristic function g(x) calculated in this way is as follows: Figure 4 As shown in (b), the actual flow characteristic function G(x) in the state after sensitivity adjustment is roughly consistent with the sensor output value x of 0 to 100%. In this way, the calculated corrected flow characteristic function g(x) can reduce the flow error in the overall range of the sensor output value x when the nitrogen gas used for correction flows in the flow channel. In addition, the function correction value n is not limited to the value calculated based on the flow value y when the sensor output value x is 100%. For example, it can also be determined based on the flow value y when the sensor output value x is 80% or other values. In addition, as described above, the determined sensitivity correction value function m(x) and function correction value n can be used not only when calculating the flow rate of nitrogen gas actually flowing in the flow channel for correction, but also when calculating the flow rate of other types of fluids.

[0080] Finally, the function correction unit 5 stores the calculated corrected flow characteristic function g(x) or the final flow characteristic function h(x) based on the corrected flow characteristic function g(x) as the flow characteristic function in the function storage unit M3 of the flow sensor CM. In the present embodiment, the function correction unit 5 stores the corrected flow characteristic function g(x) in the function storage unit M3 as the flow characteristic function used when the standard gas, i.e., nitrogen, flows.

[0081] On the other hand, in the above-mentioned correction method, when a gas different from the nitrogen used as the standard gas flows through the flow channel CH, the corrected flow characteristic function g(x) calculated based on the standard gas is not completely consistent with the actual flow characteristic function G(x) in the state after sensitivity adjustment, resulting in a flow error.

[0082] Then, the function correction unit 5 stores the final flow rate characteristic function h(x) in which the flow rate error due to the gas type generated when the corrected flow rate characteristic function g(x) is set in the function storage unit M3.

[0083] Specifically, the function correction unit 5 comprises: a gas type correction value calculation unit 51, which calculates a gas type correction coefficient k based on the thermal conductivity λ of the fluid and a sensitivity correction value m of a specified sensor output value x; and a final output unit 52, which multiplies the gas type correction coefficient k by the corrected flow characteristic function g(x) to calculate the final flow characteristic function h(x), and stores the final flow characteristic function h(x) in the function storage unit M3.

[0084] The gas type correction value is described in detail. The flow sensor CM has various instrument errors, and the actual flow characteristic function F(x) in the initial state deviates greatly from the standard flow function f(x), such as Figure 5 As shown in the figure (a), the larger the value set as the sensitivity correction value m, the greater the flow error caused by the gas type. In other words, as a flow divider, when the processing accuracy is high and the sensitivity correction value m is small, it is difficult to cause a flow error even if the gas type changes. The inventors have found out through intensive research that: Figure 5 As shown in (a), when the thermal conductivity of the gas type is set to λ, the flow error generated is proportional to (1 / λ)^(1 / 2), and its slope is as follows: Figure 5 As shown in (b), it is proportional to the size of the sensitivity correction value m.

[0085] The gas type correction value calculation unit 51 is configured based on such knowledge. For example, for the sensitivity correction value m(10%) when the sensor output value x is 10%, the gas type correction value calculation unit 51 calculates the sensitivity correction value m(10%) based on a function with a slope a(m) prepared in advance based on data obtained from a plurality of flow sensors CM. Figure 5 The slope a of the approximate straight line of the flow rate error at the span value of (1 / λ)^(1 / 2) in the graph (a) of FIG. 5 is obtained. The gas species correction value calculation unit 51 uses the calculated slope a to calculate the gas species correction coefficient k according to the following formula.

[0086] k=a(m)×(1 / λ)^(1 / 2)+b

[0087] Here, λ is the thermal conductivity of the fluid flowing through the flow channel, and b is a predetermined intercept. In this example, b is a value obtained by substituting the thermal conductivity λ of nitrogen, which is a standard gas, into k = 0. By doing so, the calculated gas type correction coefficient k can be set without performing actual measurement separately.

[0088] According to the correction device 100 configured in this way, by adjusting the sensitivity coefficient, the actual flow rate characteristic function F(x) itself, which is the object of fitting the standard flow rate characteristic function f(x), can be changed, and the span correction can be performed while improving the linearity.

[0089] When adjusting the sensitivity coefficient, different sensitivity adjustment values ​​m are set according to the value of the sensor output value x. Therefore, even if the flow sensor CM has a characteristic with extremely poor linearity, it is possible to set the actual flow characteristic function G(x) after sensitivity adjustment that is easy to fit the standard flow characteristic function f(x).

[0090] Therefore, the accuracy of the linearity between the sensor output value x and the flow value y can be improved compared with the past, and a corrected flow characteristic function g(x) can be obtained that is substantially consistent with the actual flow characteristic function G(x) after sensitivity adjustment within the range of 0 to 100% of the standard range of the sensor output value x. As a result, for example, the yield rate of adjustment related to the flow characteristic of the mass flow controller MFC as a whole in which the flow sensor CM is assembled can be improved.

[0091] Other embodiments will be described.

[0092] The function correction value n is not limited to the determination method shown in the embodiment. The function correction value n can also be determined using an algorithm such as the least squares method. In this case, it is sufficient to obtain at least two sets of measurement data of the sensor output value x and the flow value y and perform optimization.

[0093] The sensitivity correction value function m(x) is not limited to a quadratic function, but may be a higher-order function. In addition, the sensitivity correction value function m(x) may also be defined as a step function.

[0094] In addition, the sensitivity correction value m is not limited to being multiplied by the initial value, but may be added. Similarly, the function correction value n is not limited to being multiplied by the standard flow characteristic function f(x), but may be added. The standard flow characteristic function f(x) is not limited to being obtained experimentally as shown in the embodiment, but may be a theoretical formula, etc.

[0095] The flow sensor to be corrected is not limited to the thermal sensor described in the embodiment. The correction device and correction method of the embodiment can also be used to correct sensors based on other measurement principles.

[0096] Furthermore, various combinations and modifications of the embodiments are possible as long as they do not depart from the spirit of the present invention.

Claims

1. A correction device for correcting a flow characteristic function of a flow sensor, the flow sensor comprising: a sensor mechanism that outputs an output signal corresponding to the flow rate of a fluid flowing through a flow channel; a sensor output calculation unit that calculates a sensor output value (x) based on a value represented by the output signal and a sensitivity coefficient; and a function storage unit that stores a flow characteristic function in which an input is set to the sensor output value (x) and an output is set to a flow value (y); and a flow rate calculation unit that calculates a flow rate value (y) based on the sensor output value (x) calculated by the sensor output calculation unit and the flow rate characteristic function, The correction device is characterized by comprising: a sensitivity correction value function storage unit storing a sensitivity correction value function (m(x)) for setting at least a part of the sensitivity correction value (m) to a different value according to a sensor output value (x) outputted from the flow sensor in an initial state in which the sensitivity coefficient is set to an initial value; a sensitivity setting unit that sets the sensitivity coefficient based on an initial value and the sensitivity correction value function (m(x)) and adjusts the sensitivity of the flow sensor; a function calculation unit that calculates a corrected flow characteristic function (g(x)) based on a function correction value (n) and a standard flow characteristic function (f(x)), wherein the function correction value (n) is determined based on a flow value (y) output from the flow sensor whose sensitivity has been corrected by the sensitivity correction value function (m(x)); as well as The function correction unit stores the corrected flow characteristic function (g(x)) calculated by the function calculation unit or the final flow characteristic function (h(x)) based on the corrected flow characteristic function (g(x)) in the function storage unit as a flow characteristic function.

2. The correction device according to claim 1, characterized in that The sensitivity correction value function (m(x)) is set so that the smaller the sensor output value (x) is, the larger the sensitivity correction value (m) is.

3. The correction device according to claim 1, characterized in that The correction device also includes: a sensitivity adjustment data storage unit storing a plurality of sets of sensitivity adjustment data, the sensitivity adjustment data consisting of a set of a sensitivity reference flow value (F1) actually flowing in the flow channel, a sensor output value (x) output from the flow sensor in an initial state corresponding to the sensitivity reference flow value (F1), and a flow value (y) output from the flow sensor in an initial state corresponding to the sensitivity reference flow value (F1), i.e., a flow value (S1) before sensitivity adjustment; as well as The correction function output unit calculates the sensitivity correction value (m) corresponding to each sensor output value (x) based on the difference between the sensitivity reference flow value (F1) and the flow value (y) based on multiple groups of sensitivity adjustment data, calculates the sensitivity correction value function (m(x)) based on the multiple groups of sensor output values ​​(x) and the sensitivity correction value (m) and stores it in the sensitivity correction value function storage unit.

4. The correction device according to claim 1, characterized in that: The correction device further includes a span adjustment data storage unit, the span adjustment data storage unit storing span adjustment data, the span adjustment data consisting of a set of a span reference flow value (F2) actually flowing in the flow channel and a flow value (y) output from the flow sensor whose sensitivity has been corrected by the sensitivity correction value function (m(x)). The function calculation unit calculates a corrected flow characteristic function (g(x)) by dividing the span reference flow value (F2) by the flow value (y) as the function correction value (n) and multiplying the result by the standard flow characteristic function (f(x)).

5. The correction device according to claim 4, characterized in that: The range reference flow value (F2) is a range of flow value (y) set in the flow sensor.

6. The correction device according to claim 1, characterized in that The function correction unit comprises: a gas species correction value calculation unit that calculates a gas species correction coefficient (k) based on a thermal conductivity λ of the fluid and a sensitivity correction value (m) of a predetermined sensor output value (x); as well as The final output unit multiplies the gas type correction coefficient (k) by the corrected flow characteristic function (g(x)) to calculate the final flow characteristic function (h(x)), and stores the final flow characteristic function (h(x)) in the function storage unit.

7. The correction device according to claim 6, characterized in that The gas species correction value calculation unit calculates a slope (a(m)) which is a function of a sensitivity correction value (m) of a predetermined sensor output value (x), and calculates a gas species correction coefficient (k) based on the following formula: k=a(m)×(1 / λ)^(1 / 2)+b Here, k is a gas type correction coefficient, a(m) is a slope, λ is a thermal conductivity of a fluid flowing in the flow channel, and b is a predetermined intercept.

8. A flow sensor, characterized in that: By the correction device according to claim 1, A flow value (y) is output based on the stored final flow characteristic function (h(x)).

9. A flow control device, characterized in that: A flow sensor according to claim 8 is provided.

10. A program storage medium storing a correction device program, wherein the correction device program corrects a flow characteristic function of a flow sensor, wherein the flow sensor comprises: a sensor mechanism that outputs an output signal corresponding to the flow rate of a fluid flowing through a flow channel; a sensor output calculation unit, a sensor output value (x) is calculated based on the value represented by the output signal and the sensitivity coefficient; a function storage unit stores a flow characteristic function in which the input is set as the sensor output value (x) and the output is set as the flow value (y); and a flow calculation unit calculates the flow value (y) based on the sensor output value (x) calculated by the sensor output calculation unit and the flow characteristic function, The program storage medium is characterized in that the correction device program causes the computer to function as a sensitivity correction value function storage unit, a sensitivity setting unit, a function calculation unit, and a function correction unit. The sensitivity correction value function storage unit stores a sensitivity correction value function (m(x)) that sets at least a part of the sensitivity correction value (m) to a different value according to a sensor output value (x) output by the flow sensor in an initial state in which the sensitivity coefficient is set to an initial value. The sensitivity setting unit sets the sensitivity coefficient based on an initial value and the sensitivity correction value function (m(x)) and adjusts the sensitivity of the flow sensor. The function calculation unit calculates a corrected flow characteristic function (g(x)) based on a function correction value (n) determined according to a flow value (y) output from the flow sensor whose sensitivity has been corrected by the sensitivity correction value function (m(x)) and a standard flow characteristic function (f(x)), The function correction unit stores the corrected flow characteristic function (g(x)) calculated by the function calculation unit or a final flow characteristic function (h(x)) based on the corrected flow characteristic function (g(x)) in the function storage unit as a flow characteristic function.

11. A correction method for correcting a flow characteristic function of a flow sensor, the flow sensor comprising: a sensor mechanism that outputs an output signal corresponding to the flow rate of a fluid flowing through a flow channel; a sensor output calculation unit that calculates a sensor output value (x) based on a value represented by the output signal and a sensitivity coefficient; and a function storage unit that stores a flow characteristic function in which an input is set to the sensor output value (x) and an output is set to a flow value (y); and a flow rate calculation unit that calculates a flow rate value (y) based on the sensor output value (x) calculated by the sensor output calculation unit and the flow rate characteristic function, The correction method is characterized by comprising: storing a sensitivity correction value function (m(x)) that sets at least a portion of the sensitivity correction value (m) to a different value according to a sensor output value (x) output from the flow sensor in an initial state in which the sensitivity coefficient is set to an initial value; Setting the sensitivity coefficient based on an initial value and the sensitivity correction value function (m(x)), and adjusting the sensitivity of the flow sensor; calculating a corrected flow characteristic function (g(x)) based on a function correction value (n) determined according to a flow value (y) output from the flow sensor whose sensitivity has been corrected by the sensitivity correction value function (m(x)) and a standard flow characteristic function (f(x)); and The corrected flow rate characteristic function (g(x)) or a final flow rate characteristic function (h(x)) based on the corrected flow rate characteristic function (g(x)) is stored in the function storage unit as a flow rate characteristic function.

Images