Detection Devices and Measurement Systems

Abstract

A detection device and a measurement system are provided that can accurately calculate force based on detected strain, regardless of the characteristics of a strain-flexing part. [Solution] The detection device 100 of the embodiment corresponds to the relationship between strain and force occurring in the strain-flexible part 12 with an overall model M consisting of multiple basic models E connected in parallel, each of which is a series combination of a spring element and a Coulomb friction element, and has a setting unit 141 in which a proportionality coefficient and a limit value are set as individual parameters representing each basic model E, an input unit 142 that inputs a signal from at least one detection element 13 that detects the strain in the strain-flexible part 12, and a force calculation unit 143 that calculates a force dependent on the calculation of each basic model E based on the strain from the input unit 142 and the parameters set for each basic model E, and estimates the force applied to the strain-flexible part 12 by summing up the forces dependent on the calculation of each basic model E.

Description

[Technical Field]

[0001] The present invention relates to a detection device and a measurement system. [Background technology]

[0002] Measurement systems such as force sensors and torque meters detect the strain of an object to which force is applied, and calculate the force (e.g., torque) applied to the object based on this strain. For example, there are measurement systems that consist of a detection element that detects strain due to tensile or compressive strain using a strain gauge attached to a strain-generating part, and a detection device that calculates force from this strain.

[0003] A strain-causing part is a mechanical element that detects strain in an object to be measured. It may be the object to be measured itself, or a dedicated component that is attached in series to the object to be measured. Alternatively, a subordinate strain-causing part may be configured in parallel without interfering with the strain that occurs in the object to be measured, which is the main strain-causing part. A strain gauge is attached to a strain-causing part and is a detection element whose resistance value changes depending on the strain in the strain-causing part. By configuring a bridge circuit using strain gauges, strain can be detected based on changes in the output voltage of the bridge circuit. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Japanese Patent Application Laid-Open No. 2001-330525 Summary of the Invention [Problem to be solved by the invention]

[0005] In order to accurately calculate the force using such a measurement system, it is preferable that the strain of the strain-flexible part and the force are proportional to each other. More precisely, the measurement accuracy depends on the mechanical characteristics resulting from the structure, materials, environmental conditions, etc., including the method of connecting the strain-flexible part to the measurement target.

[0006] However, when the measurement object is used as a strain-generating part, the strain-generating part does not necessarily have ideal characteristics for measurement. For example, a power transmission shaft of a vehicle is designed for the purpose of transmitting power, so it only needs to have the performance required as a transmission shaft, and does not necessarily have ideal characteristics for torque measurement. In other words, the relationship between torque and strain is linear. only If a simple conversion using a constant is performed without showing a straight line, the measurement accuracy will decrease. Furthermore, in the process of increasing the load and then returning to the no-load state, hysteresis occurs, in which the forward and backward movements are separated, which is also a problem that reduces the measurement accuracy.

[0007] An object of the present invention is to provide a detection device and a measurement system that can accurately calculate force based on detected strain, regardless of the characteristics of the strain-flexing part. [Means for solving the problem]

[0008] In order to achieve the above-mentioned objectives, the detection device of the present invention has a setting unit that corresponds an overall model in which multiple basic models, each of which is a series combination of a spring element and a Coulomb friction element, are connected in parallel to the relationship between the strain and force occurring in the strain-flexible part, and in which a proportionality coefficient and a limit value are set as individual parameters representing each basic model; an input unit that receives a signal from at least one detection element that detects the strain in the strain-flexible part; and a force calculation unit that calculates a force dependent on the calculation of each basic model based on the strain output by the input unit and the parameters set for each basic model, and adds up the forces dependent on the calculation of each basic model to estimate the force applied to the strain-flexible part.

[0009] In addition, in the measurement system of the present invention, a measurement device having components outside the housing of the detection device and input / output units for external devices is connected to components inside the housing of the detection device via wireless communication.

[0010] In the measurement system of the present invention, the detection device and a measurement device having an input / output unit for an external device are connected via wireless communication. [Effects of the Invention]

[0011] According to the present invention as described above, whether the strain-causing part is the object to be measured itself, or is attached in series to the object to be measured, or a subordinate strain-causing part is configured in parallel without interfering with the strain occurring in the object to be measured, which is the main measuring part, the force transmitted by the object to be measured can be accurately measured based on the detected strain, regardless of the characteristics of the strain-causing part. [Brief explanation of the drawings]

[0012] [Figure 1] FIG. 1 is a side view illustrating an example of the configuration of a measurement system according to an embodiment. [Figure 2] FIG. 1 is a block diagram illustrating a detection device and a measurement device according to an embodiment. [Figure 3] 1A is an example of a graph showing the linear characteristics of a measurement object, FIG. 1B is an example of a graph showing the nonlinear characteristics, and FIG. 1C is an example of a graph showing the hysteresis characteristics. [Figure 4] (A) is a schematic diagram showing the configuration of the basic model, (B) is an example of a graph showing the relationship between force and displacement of the basic model, and (C) is an example of a graph corresponding to the occurrence and removal of strain when the relationship between strain and torque is expressed using the basic model. [Figure 5] FIG. 1 is a schematic diagram showing an overall model. [Figure 6] Examples of graphs (A) to (C) show the behavior when the relationship between strain and torque is expressed using three different basic models, and an example of a graph (D) shows the behavior when the relationship between strain and torque is expressed using the overall model. [Figure 7] 10 is a table showing specific examples of output distortion from an input section of a detection device relative to a load torque. [Figure 8] 8 is a graph showing strain versus load torque corresponding to FIG. 7. [Figure 9]9 is a graph showing the gradient of each section that the overall model should represent in FIG. 8. [Figure 10] 8 is a table showing the calculation results of the proportionality coefficient and limit value of each basic model based on the strain detected in the forward path corresponding to the increase in load torque in FIG. 7. [Figure 11] 8 is a table showing the calculation results of the proportionality coefficient and limit value of each basic model based on the detected strain during the return path corresponding to the decrease in load torque in FIG. 7. [Figure 12] 5 is a flowchart illustrating an example of a torque calculation process according to the present embodiment. [Figure 13] 13 is a flowchart showing a torque calculation process shown by each basic model in FIG. 12. DETAILED DESCRIPTION OF THE INVENTION

[0013] An embodiment of the present invention (hereinafter referred to as "the present embodiment") will be described with reference to the drawings. The present embodiment is applied to a measurement system S, and a detection device 100 and a measurement device 200 that constitute the measurement system S. A measurement method that performs the processing described below and a program that causes a computer to execute the processing are also aspects of the present invention. Note that the measurement device 200 is not shown in FIG. 1, and everything other than the detection element 13 of the detection device 100 is not shown in FIG. 2.

[0014] [composition] As shown in Figures 1 and 2, measurement system S is a system that calculates the force (e.g., torque) applied to a measurement object from the strain occurring in a strain-flexible part. Here, a measurement system S will be described in which a subordinate strain-flexible part is connected in parallel to a main strain-flexible part that is the measurement object. In other words, the subordinate strain-flexible part is arranged so that it can indirectly detect the strain of the measurement object. However, this does not exclude a general measurement method in which the strain-flexible part provided by measurement system S is connected in series to a transmission system. Note that in the following description, torque may be used as a derived physical quantity of force (a derived unit including the unit N), but in the present invention, torque can also be interpreted as force. Measurement system S has detection device 100 and measurement device 200.

[0015] [Detection device] The detection device 100 detects strain in a strain-flexible part and calculates the force applied to the measurement object based on the detected strain. The detection device 100 is a torque sensor that detects the transmitted torque of a power transmission shaft SF, such as a vehicle drive shaft, that transmits power from a drive source. As described above, it is preferable that the measurement object used as the main strain-flexible part exhibits a linear relationship between torque and strain (see FIG. 3(A)). However, many measurement objects, including the power transmission shaft SF, exhibit nonlinearity that cannot be ignored (see FIG. 3(B)). Furthermore, hysteresis, in which the forward and backward movements corresponding to the increase and decrease in torque are separated, is also observed (see FIG. 3(C)).

[0016] The detection device 100 has a gripping part 11, a strain-flexing part 12, a detection element 13, a detection processing part 14, and a power supply 15, all of which are housed in a housing 10. The housing 10 of this embodiment is attached so as to cover the power transmission shaft SF. Note that no additional processing is required on the measurement target in order to attach the housing 10.

[0017] (gripping part) The gripping portion 11 is an element that is directly fixed to the measurement target. The gripping portion 11 of this embodiment grips the power transmission shaft SF by contacting it at two points spaced apart in the axial direction.

[0018] (Strain part) The strain-flexing part 12 is the measurement object itself or an element that reflects the strain of the measurement object. In this embodiment, the strain-flexing part 12 is attached to the measurement object via the holding part 11. More specifically, the strain-flexing part 12 is a plate that connects the two holding parts 11, and is an element whose influence on the strain occurring in the measurement object is negligible.

[0019] (detector element) The detection element 13 is an element that detects strain in the strain-flexible part 12. The detection element 13 is affixed to the strain-flexible part 12. At least one detection element 13 is required, but multiple detection elements 13 may be provided. The resistance value of the detection element 13 changes depending on the strain in the strain-flexible part 12, so by configuring a Wheatstone bridge circuit, it is possible to sensitively detect strain as a voltage change. In this embodiment, the strain-flexible part 12 is attached to the measurement object via the grip part 11, and the detection element 13 is arranged so as to be able to indirectly detect strain in the measurement object. More specifically, the detection element 13 is not directly affixed to the power transmission shaft SF of the vehicle, but is affixed to the strain-flexible part 12 attached to the power transmission shaft SF.

[0020] Furthermore, the strain-flexing part 12 and the detection element 13 attached thereto rotate together with the power transmission shaft SF. In other words, the strain-flexing part 12, together with the measurement target, is movable relative to the measurement device 200, but the connection between the detection device 100 and the measurement device 200 is wireless, so communication is possible.

[0021] (Detection processing unit) The detection processing unit 14 performs processing necessary for detecting a force in the detection device 100. The detection processing unit 14 is configured with a processing circuit such as a processor that operates according to a program, a storage medium such as a memory that stores programs and data, and the like, but the implementation of the hardware and program can be changed in various ways.

[0022] As shown in FIG. 2, the detection processing unit 14 includes a setting unit 141, an input unit 142, a force calculation unit 143, a control unit 144, and a communication unit 145.

[0023] <Settings section> The setting unit 141 sets a proportionality coefficient and a limit value as individual parameters representing each basic model in the overall model. The overall model is now described. The overall model is a model in which multiple basic models are connected in parallel. As shown in FIG. 4(A), basic model E is a model in which a spring element α and a Coulomb friction element β are connected in series. In the initial stage of increasing displacement, as shown in FIG. 4(B), basic model E exhibits linear behavior with force based on the spring constant (proportionality coefficient) of the spring element α as an ideal spring. Once the maximum static friction force (limit value) is exceeded, the Coulomb friction element β begins to slip, and the force remains constant. For example, if the relationship between strain and torque in the strain-flexible part 12 follows one basic model E, in a graph with strain on the horizontal axis and torque on the vertical axis, after a positive strain is induced in the power transmission shaft SF and then the strain is removed, the strain-flexible part 12 behaves as shown in FIG. 4(C).

[0024] In this way, once the upper limit is exceeded, the torque does not increase, and when the strain is removed from that point, the spring element α reacts, behaving as if to release the stored torque. The torque continues to drop as the strain decreases, down to the lower limit, which is zero. Physically, the lower limit should be the value obtained by negating the sign of the upper limit, but for ease of handling and practical convenience, it is made to coincide with the left end of the hysteresis loop.

[0025] As shown in Fig. 5, the overall model M applied to this embodiment is a model (Masing model) in which multiple basic models E are connected in parallel. If the basic models E constituting this overall model M represent the relationship between strain and torque, then in the initial stage when strain occurs, the torque corresponding to the spring elements α of all the basic models E also increases linearly. However, as the strain further increases, the torque indicated by the spring elements α of each basic model E gradually exceeds the limit value of the Coulomb friction element β, and the torque indicated by each basic model E becomes constant. Therefore, after the torque indicated by all the spring elements α exceeds the limit value of the Coulomb friction element β, the torque indicated by the overall model M no longer changes with increasing strain.

[0026] Furthermore, in a graph with strain on the horizontal axis and torque on the vertical axis, when positive strain occurs, the behavior of multiple basic models E is as shown in Figures 6(A) to 6(C), respectively. When multiple basic models E are connected in parallel, the behavior is as shown in Figure 6(D). The dashed line in Figure 6(D) indicates the behavior when positive strain is removed during its occurrence. In this way, by using the overall model M, it is possible to express behavior that mimics the hysteresis loop exhibited by the actual strain-flexible portion 12. Note that, for convenience, the behavior of the model when the hysteresis loop closes upon the occurrence and removal of positive strain is shown here, but it is also possible to express hysteresis loops for both positive and negative strains using the overall model M. In this case, the proportionality coefficient and limit value may be set separately in the force calculation unit 143 according to the positive and negative signs of the strain occurring in the strain-flexible portion 12.

[0027] On the other hand, as shown in the example of Figure 8 (the vertical axis is strain and the horizontal axis is load torque), when a load torque is applied to the measurement object and then removed, the strain detected by the strain-generating part 12 draws a curve that deviates from the regression line (dotted line) with an intercept of zero in the load increase section (forward path) and the load decrease section (return path).

[0028] By appropriately setting the parameters of the proportionality coefficient representing the spring element α and the limit value at which the Coulomb friction element β begins to slip, it becomes possible to represent (estimate) the behavior of an actual measurement object using the overall model M. Such proportionality coefficient and limit value can be set in advance based on the results of actual measurements of the behavior of an actual measurement object. Furthermore, the proportionality coefficient and limit value may be set separately in response to an increase and a decrease in the strain occurring in the strain-flexing part 12.

[0029] The proportional coefficient and limit value are set, for example, as follows. (a) Using a torque loading device, a predetermined number of load torques are applied to the power transmission shaft SF in sequence, and the detection device 100 measures the strain corresponding to each of the plurality of load torques. From the results, manually calculated parameters are input through the operation unit 250 of the measuring device 200 described later and set in the setting unit 141 of the detection device 100.

[0030] (b) A learning mode is prepared in the detection device 100 or the measurement device 200, and the operation unit 250 of the measurement device 200 instructs the detection device 100 that a predetermined number of load torques will be applied from the torque load device to the power transmission shaft SF. The detection device 100 then automatically determines that the detected strain is stable, collects multiple torque-strain pairs, and calculates the same number of parameters of the basic model E as the obtained multiple torque-strain pairs, and stores them in the setting unit 141 of the detection device 100.

[0031] <Input section> The input unit 142 receives a signal corresponding to the strain, i.e., a voltage value, from at least one detection element 13 that detects the strain of the strain-generating part 12. The input unit 142 of this embodiment includes at least an ADC (analog-to-digital converter), and further includes one or more of a bridge circuit, an amplifier circuit, an offset adjustment circuit, and a filter circuit, with the ADC being disposed after passing through these circuits. In addition, a temperature compensation circuit may be incorporated into the bridge circuit.

[0032] <Force calculation section> The force calculation unit 143 calculates a force dependent on the calculation of each basic model E based on the strain from the input unit 142 and the parameters set for each basic model E, and estimates the force applied to the strain-flexing part 12 by adding up these forces. The force calculation unit 143 first calculates the force represented by each basic model E based on the proportionality coefficient and limit value set for each basic model E by the setting unit 141 and the strain from the input unit 142. Next, the force represented by each basic model E is added up to estimate the force applied to the strain-flexing part 12 as the entire model M. In other words, a value equivalent to the force applied to the strain-flexing part 12 is calculated by adding up the forces.

[0033] The force calculation unit 143 has a memory unit 143a, a differential strain calculation unit 143b, a differential force calculation unit 143c, a determination unit 143d, an accumulated force calculation unit 143e, and an adder unit 143f. The memory unit 143a stores strain and force. The stored strain can include the input strain as well as the differential strain described below. The stored force can include the force calculated by each basic model E as well as the differential force described below.

[0034] The differential strain calculation unit 143b calculates the difference between the input strain and the strain (previous strain) stored in the memory unit 143a. Initially, the previous strain saved in the memory unit 143a may be read out and restarted, or the previous strain may be reset. If the previous strain has been reset, the input strain becomes the differential strain. The differential force calculation unit 143c calculates the value of the differential force for each basic model E based on the differential strain and the proportionality coefficient.

[0035] The determination unit 143d determines whether the value obtained by adding the differential force value to the force of each basic model E stored in the storage unit 143a (the force accumulated up to the previous time) exceeds the corresponding limit value (true), or whether the value does not exceed the limit value (false). Initially, the force accumulated up to the previous time stored in the storage unit 143a may be read out and restarted, or the accumulated force may be reset. If the accumulated force has been reset, it is determined whether the differential force exceeds the limit value. Note that exceeding the limit value includes cases where the force exceeds the upper limit value in the case of an increase in force, and cases where the force falls below the lower limit value in the case of a decrease in force.

[0036] If the determination unit 143d determines that the force does not exceed the limit value, that is, false, the accumulated force calculation unit 143e accumulates the differential force to the force (the force accumulated up to the previous time) stored in the storage unit 143a and stores the value in the storage unit 143a. That is, the differential force is added to the stored force of each basic model E, and the stored force is updated as the force represented by that basic model E. If the determination unit 143d determines that the force exceeds the limit value, that is, true, the accumulated force calculation unit 143e adopts the limit value.

[0037] The summing unit 143f sums up the accumulated forces corresponding to each basic model E, thereby estimating the force applied to the strain-flexing part 12.

[0038] <Control unit> The control unit 144 controls input of parameters such as a proportional coefficient and a limit value from the communication unit 145 to the setting unit 141, and input of parameters from the setting unit 141 to the force calculation unit 143, etc.

[0039] <Communications Department> The communication unit 145 transmits a voltage value representing the force obtained by the force calculation unit 143 to the measurement device 200. The communication unit 145 also receives parameters such as a proportionality coefficient and a limit value transmitted from the measurement device 200. In this embodiment, the detection device 100 and the measurement device 200, which are physically separated from each other, are connected by wireless communication to enable communication.

[0040] (power supply) The power supply 15 is a secondary battery that supplies power to the input unit 142 , the force calculation unit 143 , the setting unit 141 , the control unit 144 and the communication unit 145 .

[0041] [Measuring equipment] The measuring device 200 is a device that outputs the force detected by the detecting device 100 to an external device, such as a recording device.

[0042] The measuring device 200 is configured with a processing circuit such as a processor that operates according to a program, a storage medium such as a memory that stores the program and data, etc., but the implementation of the hardware and the program can be changed in various ways. The measuring device 200 has a communication unit 210, a calculation unit 220, an input / output unit 230, a display unit 240, an operation unit 250, and a control unit 260.

[0043] (Communications Department) The communication unit 210 communicates information with the detection device 100. For example, the communication unit 210 receives a voltage value representing a force transmitted from the detection device 100. The communicated information also includes a proportional coefficient, a limit value, and the like set in the setting unit 141.

[0044] (calculation section) The calculation unit 220 converts the received voltage value into a signal that can be input to an external recording device, etc. At this time, the calculation unit 220 may perform secondary calculations. Specifically, calculations such as multiplication by a constant, multiplication by another measured physical quantity such as rotation speed, and low-pass filtering are possible.

[0045] (input / output section) The input / output unit 230 inputs and outputs information from and to the outside. For example, the input / output unit 230 outputs a converted and calculated voltage value to an external recording device or the like. In this embodiment, a so-called analog voltage is output as an output signal to the outside, but the measuring device 200 may include a recording device. In that case, the detecting device 100 may be configured to include some or all of the elements of the measuring device 200.

[0046] (Display) The display unit 240 is an output unit that displays information on a display screen so that the user can visually recognize it. The display unit 240 can display the proportional coefficient set in the setting unit 141, the limit value, the force calculated by the force calculation unit 143, etc.

[0047] (Operation unit) The operation unit 250 is a device for inputting information from the outside. The input here includes input via communication with the outside. The operation unit 250 includes input devices such as a keyboard, mouse, touch panel, and switch, and communication devices such as a network adapter.

[0048] (Control unit) The control unit 260 controls the calculation by the calculation unit 220 and the display on the display unit 240 in response to input from the operation unit 250 .

[0049] [Processing Procedure] The processing procedure of this embodiment as described above will be described with reference to the figures, as well as specific formulas, examples of parameters and graphs, and flowcharts showing the processing flow of each device.

[0050] (Actual strain measurement) First, a static torque load device is used to measure the strain in response to the load torque of the measurement target. For example, this device is attached to the power transmission shaft SF, which is the measurement target, and after applying a load up to the rated torque, the strain corresponding to multiple load torques is detected while returning to no load.

[0051] FIG. 7 is a table simulating strain at 10 measurement points (200, 400, 600, 800, 1000, 800, 600, 400, 200, and 0 [N·m]) ranging from a load torque of 0 [N·m] to a rated torque of 1000 [N·m], expressed as voltage values ​​[V] corresponding to the output of the input unit 142 (11 points if the initial load torque of 0 [N·m] is counted). Note that the actual output of the input unit 142 is a digital signal resulting from ADC conversion of the detected voltage, but for convenience, it is represented here as an analog voltage value before conversion. Also, for ease of understanding, the measurement points in FIG. 7 are spaced equally apart. However, this is merely an example, and the measurement point spacing can be arbitrary. For example, if the hysteresis curve exhibits large changes, more accurate correction can be achieved by narrowing the measurement point spacing. Furthermore, while FIG. 7 shows the same load torque measurement points for both the forward and backward movements, this is not a limitation.

[0052] In FIG. 7, the corrected distortion [V] is a value obtained by correcting each detected value of the output distortion [V] so that its starting point is aligned with the origin. In this way, a circuit or process that performs offset correction of the detected value is also included in the detection device 100. The regression line [V] is a least-squares line created based on the corrected distortion [V]. The distortion error [V] is the difference between the corrected distortion [V] at each detection point and the regression line [V]. Ideally, this distortion error [V] should be zero.

[0053] Figure 8 shows a graph of strain against load torque [N·m], with the corrected strain [V] shown as a solid line and the regression line [V] shown as a dotted line.

[0054] (Identification of proportionality coefficients) If the torque applied to the measurement object is T and the corresponding strain is γ, then this torque T can be considered to be distributed to each basic model E as Ti in the following equation (1). JPEG0007780050000002.jpg19143

[0055] Here, if the proportionality coefficient corresponding to the spring constant of each basic model E is ki and the limit value corresponding to the maximum static friction force is Li, then the strain γ can be expressed as the following equation (2) according to Hooke's law. JPEG0007780050000003.jpg22143Here, for convenience, we consider the case where the domain of definition ranges from zero to the positive rated torque, as shown in Figure 8. However, even when dealing with positive and negative rated torque, the domain can be similarly expressed as from the lower limit (negative rated torque) to Li. Furthermore, the following explanation can also be applied to positive and negative rated torque.

[0056] Since each of the multiple sections separated by the detection points contains one or more basic models E functioning as springs, the slope Ai of each section shown in Fig. 9 can be obtained by the following formula (3). Note that element numbers outside the range are treated as ki = 0. JPEG0007780050000004.jpg20143

[0057] When the torque is gradually increased from the starting point, basic model E also has the effect of Coulomb friction, and as each basic model E reaches its limit value one by one, it can be considered as expressing the broken line shown in the graph in Figure 9. From the behavior of basic model E, which remains constant once it reaches this limit value, the relationship in equation (4) below becomes clear. Here, each Ai (here, A1 to A5) can be found from the measurement results, and from this, the proportionality coefficient ki corresponding to each basic model E can be obtained. JPEG0007780050000005.jpg45143

[0058] In actual applications, strain is input from the detection element 13 of the detection device 100, and the purpose of the detection device 100 is to determine the corresponding torque. Therefore, if the basic model E is expressed with the input (horizontal axis) as strain and the output (vertical axis) as torque, it becomes as shown in the following equation (5). JPEG0007780050000006.jpg38143

[0059] (Identification of limit values) The basic model E also requires a limit value for the Coulomb friction element β that corresponds to the onset of slippage. The constant Li that represents the limit value is calculated using the above formula (5). However, it is determined from Ti, not the input load torque T at the measurement point (end point of each section) during actual measurement. The limit value remains constant outside the range of each section.

[0060] Ki is a proportionality coefficient corresponding to each interval, and can be calculated from the previously calculated ki using the following equation (6). JPEG0007780050000007.jpg48143

[0061] An example of calculating the proportionality coefficient ki for the forward path using the simulated data shown in Figure 7 is shown in Figure 10, and an example of calculating the proportionality coefficient for the return path is shown in Figure 11. As above, each Ai (A'1 to A'5 in Figure 9) for the return path can be found from the measurement results, and the proportionality coefficient ki can be obtained from this. In Figure 11, ki_ave is the average value when strain increases and decreases.

[0062] The proportionality coefficient and the limit value may be set separately in the force calculation section 143 in response to an increase and a decrease in the force applied to the strain-flexing section 12, respectively.

[0063] (Torque calculation) The process of calculating torque from strain detected by the detection device 100 based on the proportionality coefficient and limit value determined as described above will be described with reference to the flowcharts of Figures 12 and 13. The proportionality coefficient and limit value are input in advance via the communication unit 145 and set in the setting unit 141.

[0064] 12, initially, the storage unit 143a is reset (step S101). Then, the strain detected by the detection device 100 is input and acquired by the input unit 142 (step S102). The differential strain calculation unit 143b calculates the difference between the current strain and the previous strain stored in the storage unit 143a as the differential strain (step S103). Since the strain is reset initially, the current strain becomes the differential strain as is.

[0065] Next, the force calculation unit 143 calculates the torque of each basic model E using the above formula (4) (step S104). The flow of this torque calculation will be explained with reference to the flowchart in FIG. 13. The differential force calculation unit 143c calculates the differential torque based on the differential strain and the proportionality coefficient (step S201). The determination unit 143d determines whether or not the value obtained by adding the differential torque to the accumulated torque up to the previous time exceeds a limit value (step S202). Note that, for convenience, step S201 in FIG. 13 only indicates whether or not the upper limit value is exceeded.

[0066] If it is determined that the limit value is not exceeded (YES in step S202), the accumulated torque calculation unit 143e adds the differential torque to the accumulated torque (step S203), and the storage unit 143a stores the added value as the accumulated torque (step S204).If it is determined that the limit value is exceeded (NO in step S202), the accumulated torque stored in the storage unit 143a is set as the limit value (step S205), and this limit value is stored as the accumulated torque (step S204).

[0067] The above calculation is continued (step S104) until the accumulated torques of all basic models E have been calculated (YES in step S105). Once the accumulated torques of all basic models E have been calculated (NO in step S105), the summing unit 143f calculates the total torque by summing the accumulated torques of each basic model E as shown in the above equation (6) (step S106). Furthermore, the storage unit 143a stores the strain input this time (step S107). Note that the above processing may be performed by repeatedly calling the processing from S102 to S107 for repeated measurements.

[0068] [effect] (1) The detection device 100 of this embodiment corresponds to the relationship between the strain and force occurring in the strain-flexible part 12 an overall model M in which multiple basic models E, each of which is a series combination of a spring element α and a Coulomb friction element β, are connected in parallel, and includes a setting unit 141 in which a proportionality coefficient and a limit value are set as individual parameters representing each basic model E; an input unit 142 that inputs a signal from at least one detection element 13 that detects the strain in the strain-flexible part 12; and a force calculation unit 143 that calculates a force dependent on the calculation of each basic model E based on the strain from the input unit 142 and the parameters set for each basic model E, and estimates the force applied to the strain-flexible part 12 by summing up the forces dependent on the calculation of each basic model E.

[0069] In this way, by calculating the force based on the proportionality coefficient and limit value that represent each basic model E and the detected strain, and adding these up, the force applied to the strain-flexing part 12 can be estimated, including the hysteresis characteristics. This makes it possible to calculate the force with high accuracy, regardless of the characteristics of the object to be measured. In other words, by taking into account the nonlinear characteristics and hysteresis characteristics of the object to be measured and improving the linearity of the conversion, high-precision measurement can be achieved.

[0070] For example, in a feedback control system based on torque measurement, even if strain occurring in the power transmission shaft SF being measured is detected, if hysteresis characteristics are present, the measured torque output will continue to be output when the system is returned to an unloaded state after applying a load, and therefore the control output cannot be stopped unless exceptional processing is performed near zero.

[0071] In this embodiment, the effects of nonlinearity and hysteresis of the object to be measured can be suppressed, ensuring the accuracy of the torque meter, thereby stabilizing control using the measurement system S. Therefore, in addition to improving the accuracy of the measurement values, it is also possible to choose to use inexpensive but poorly-characterized strain-flexible materials, which contributes to reducing development and manufacturing costs.

[0072] Generally, for an object consisting of many mechanical elements, such as a steering mechanism consisting of a steering wheel, steering shaft, gear mechanism, etc., some kind of model can be associated with each of the contact mechanisms between the mechanical elements, and an overall model M that shows the unique characteristics of the overall configuration can be expressed by analyzing the final output of the forces transmitted via the multiple mechanical elements, and this can be incorporated as an additional control that does not cause a sense of discomfort to people.

[0073] However, the inventors of the present invention have not been limited to such conventional methods, but have come up with a new idea of ​​using an entire model M consisting of multiple basic models E to measure the forces acting on a single member (for example, a power transmission shaft SF of a vehicle) or an integrated structure as the measurement target. Moreover, they have come up with a previously unseen technology of using this not only for analyzing and reproducing forces but also for correcting measurement values ​​based on detected values, thereby achieving the effect of improving the accuracy of measurement values.

[0074] (2) The force calculation unit 143 includes a memory unit 143a that stores strain and force, a differential strain calculation unit 143b that calculates a differential strain, which is the difference between the input strain and the strain stored in the memory unit 143a, a differential force calculation unit 143c that calculates the value of the differential force in each basic model E based on the differential strain and a proportionality coefficient, and a differential force calculation unit 143c that calculates whether the value obtained by adding the differential force value to the force of each basic model E stored in the memory unit 143a exceeds a limit value (true) or does not exceed the limit value (false). The device has a determination unit 143d that determines whether the force of each basic model E is equal to or greater than the force of the basic model E stored in the memory unit 143a, and if the determination unit 143d determines that the force is false, adds the differential force to the force of each basic model E stored in the memory unit 143a and updates the force stored in the memory unit 143a as the force represented by the basic model, and if the determination unit 143d determines that the force is true, sets the force of each basic model E stored in the memory unit 143a as a limit value, and an addition unit 143f that estimates the total force applied to the strain-flexing part 12 by adding up the forces corresponding to each basic model E.

[0075] Therefore, even for objects where the strain and force do not have a completely linear relationship, it is possible to calculate the force with high overall accuracy by calculating the force taking into account the accumulation and release of force based on the operating range for each basic model E. In addition, the previous strain can be stored and used for the next calculation.

[0076] (3) The proportional coefficient and limit value are set separately in the force calculation unit 143 in response to an increase and a decrease in the strain occurring in the strain-flexing part 12. The proportional coefficient and limit value are also set separately in the force calculation unit 143 in response to the positive or negative sign of the strain occurring in the strain-flexing part 12. This makes it possible to calculate the force with high accuracy in the forward and backward paths, which have hysteresis and nonlinearity corresponding to the increase and decrease in force.

[0077] For example, by switching between referencing when strain is increasing and decreasing, it is possible to express more realistic characteristics rather than a point-symmetric figure such as the parallelogram of the basic hysteresis shape shown in Figure 4(C).Also, by switching between whether strain is positive or negative, it is possible to easily express characteristics that include symmetry about the y-axis.In other words, it is possible to correct not only hysteresis characteristics but also nonlinear characteristics.

[0078] (4) The housing 10 is designed to be attachable to the object to be measured without any additional processing. Therefore, even if the object to be measured rotates, such as a vehicle's power transmission shaft SF, making it difficult to directly attach the detection element 13, or even if the strain-flexing part 12 cannot be connected in series to the power transmission shaft SF, calculations can be performed with reduced errors caused by the object to be measured, as in this embodiment, enabling highly accurate torque measurement. Furthermore, even if the strain-flexing part 12 can be connected in series to the power transmission shaft SF, expensive materials with excellent linear characteristics in the elastic range are generally used for the strain-flexing part 12, but by using this method, it is possible to select inexpensive materials with poorer linear characteristics.

[0079] [Variations] This embodiment is not limited to the above-described aspects, and also includes the following modifications.

[0080] (1) As shown in the example above, the forward and backward paths corresponding to the increase and decrease in torque (force) are often not point-symmetric. In this case, different proportional coefficients and limit values ​​can be selected for the forward and backward paths, as described above, or they can be calculated separately and then combined by taking the average value (see ki_ave in Figure 11). This has the effect of reducing data variability associated with instability of the torque load device and the object being measured when acquiring data used to identify the parameters of basic model E. This instability also includes behavior caused by stochastic deformation of the material.

[0081] (2) The setting unit 141 may set a time-dependent coefficient in addition to the proportional coefficient and limit value as an individual parameter representing each basic model E. In this case, the force calculation unit 143 calculates a time-dependent coefficient based on the limit value using a time-dependent coefficient corresponding to the elapsed time since the determination unit 143d of an arbitrary basic model became true, and uses the time-dependent coefficient instead of the limit value, and / or calculates a time-dependent proportional coefficient based on the proportional coefficient and uses the time-dependent proportional coefficient instead of the proportional coefficient. The time-dependent coefficient can be prepared, for example, as a lookup table or a function using the elapsed time since the determination unit 143d of an arbitrary basic model became true as a parameter. Alternatively, the residence time of an arbitrary load may be used as a parameter. This allows the creep phenomenon to be reproduced.

[0082] Alternatively, a temperature measurement unit for detecting temperature may be provided, and a temperature-dependent coefficient may be set in the setting unit 141 in addition to a proportionality coefficient and a limit value as individual parameters representing each basic model E. In this case, the force calculation unit 143 calculates a temperature-corrected limit value based on a limit value according to the difference between the temperature detected by the temperature measurement unit and a fixed reference temperature, and uses the temperature-corrected limit value instead of the limit value, and / or calculates a temperature-corrected proportionality coefficient based on a proportionality coefficient and uses the temperature-corrected proportionality coefficient instead of the proportionality coefficient. This makes it possible to represent the temperature characteristics of the measurement object. However, before starting to calculate the differential force for each basic model E, the force calculation unit 143 must calculate the equivalent differential strain from the force represented by the basic model E stored in the memory unit 143a and the proportionality coefficient used last time, and recalculate the force represented by the basic model when the corrected proportionality coefficient and / or corrected limit value to be applied are used.

[0083] (3) In the above embodiment, the strain of the measurement object is indirectly detected by attaching the strain-flexing part 12 with the detection element 13 attached to the measurement object, but the detection element 13 may be directly attached to the measurement object to detect strain in the detection device 100. In this case, the strain-flexing part 12 becomes the measurement object itself, allowing for calculations with reduced errors due to the characteristics of the measurement object.

[0084] (4) The measurement object is not limited to the power transmission shaft SF of a vehicle, but can be applied to any member or structure. It is not limited to a single member, and the measurement object may be a structure in which multiple members are fixedly or movably connected. Furthermore, although it is suitable for measuring movable objects such as the power transmission shaft SF of a vehicle, it is not limited to this, and the measurement object may also be a stationary member.

[0085] (5) The connection between the detection device 100 and the measurement device 200 is not limited to wireless and may be wired, or may be connected via a communication network, whether wired or wireless. Optical communication is also included in the communication. In the above embodiment, the setting unit 141, the input unit 142, and the force calculation unit 143, along with the detection element 13, are all housed in the housing 10. However, the housing 10 may also house some of the components, namely the setting unit 141, the input unit 142, and the force calculation unit 143. Furthermore, components outside the housing 10 may be provided in the measurement device 200. In other words, the measurement device 200 may have some of the functions of the detection device 100. For example, the force calculation unit 143 may be provided in the measurement device 200. In this way, even when some components are provided in the housing 10 and other components are provided outside the housing 10, the entire device can be considered as the detection device 100. The components housed in the housing 10 and the components outside the housing 10 may be connected by wireless or wired communication, or may be connected via a communication network, whether wired or wireless. For example, a measuring device 200 including components outside the housing 10 of the detecting device 100 and an input / output unit 230 for an external device may be connected to the components inside the housing 10 of the detecting device 100 via wireless communication.

[0086] (6) The above-described detection device 100, measurement device 200, and measurement system S have been described as an example of torque measurement, but are not limited to this and can be widely applied as a force sensor that measures the force applied to a measurement object.

[0087] [Other embodiments] The above describes embodiments of the present invention and modifications of each part, but these embodiments and modifications are presented as examples and are not intended to limit the scope of the invention. These novel embodiments described above can be embodied in various other forms, and various omissions, substitutions, modifications, and combinations can be made without departing from the spirit of the invention. These embodiments and modifications are included within the scope and spirit of the invention, and are also included in the invention described in the claims. [Explanation of symbols]

[0088] 10. Cabinet 11 Gripping part 12 Strain part 13 Detector element 14 Detection processing section 15 Power supply 100 Detection device 141 Setting section 142 Input section 143 Force calculation section 143a Storage section 143b Differential strain calculation section 143c Differential force calculation section 143d Judgment section 143e Accumulation force calculation unit 143f Totalization section 144 Control Unit 145 Communications Department 200 Measuring Equipment 210 Communications Department 220 Arithmetic section 230 Input / output section 240 Display section 250 Operation section 260 Control Unit E Basic Model M overall model S Measurement System SF Power Transmission Shaft α spring element β Coulomb friction element

Claims

1. a setting section in which an overall model in which a plurality of basic models, each of which is a series combination of a spring element and a Coulomb friction element, are connected in parallel corresponds to the relationship between the strain and force occurring in the strain-generating section, and in which a proportional coefficient and a limit value are set as individual parameters representing each basic model; an input unit that inputs a signal from at least one detection element that detects the strain of the strain-flexing part; a force calculation unit that calculates a force depending on the calculation of each basic model based on the strain from the input unit and parameters set for each basic model, and estimates the force applied to the strain-flexing part by summing up the forces depending on the calculation of each basic model; A detection device comprising:

2. The force calculation unit a memory unit that stores the strain and force; a differential strain calculation unit that calculates a differential strain between the input strain and the strain stored in the storage unit; a differential force calculation unit that calculates a differential force value for each basic model based on the differential strain and a proportionality coefficient for each basic model; a determination unit that determines whether a value obtained by adding the force of each basic model stored in the storage unit to the value of the differential force exceeds the limit value (true) or does not exceed the limit value (false); an accumulated force calculation unit that, when the determination unit determines that the answer is false, adds the differential force to the force of each basic model stored in the storage unit, thereby updating the force stored in the storage unit as the force represented by the basic model, and, when the determination unit determines that the answer is true, sets the force of each basic model stored in the storage unit as a limit value; an adder that estimates a total force applied to the strain-flexing part by adding up forces corresponding to each basic model; 2. The detection device according to claim 1, further comprising:

3. 2. The detection device according to claim 1, wherein the proportional coefficient and the limit value are set separately in the force calculation unit in response to an increase and a decrease in strain occurring in the strain-flexing part.

4. 2. The detection device according to claim 1, wherein the proportionality coefficient and the limit value are set separately in the force calculation unit in accordance with the positive or negative sign of the strain occurring in the strain-flexing part.

5. a setting unit in which the proportional coefficient, the limit value, and a time-dependent coefficient are set as individual parameters representing each of the basic models; the force calculation unit that calculates a time-dependent correction limit value based on the limit value in accordance with the time elapsed since the determination by the determination unit became true and the time-dependent coefficient, and uses the time-dependent correction limit value instead of the limit value, and / or calculates a time-dependent correction proportional coefficient based on the proportional coefficient, and uses the time-dependent correction proportional coefficient instead of the proportional coefficient; 3. The detection device according to claim 2, further comprising:

6. a temperature measurement unit for detecting a temperature; a setting unit in which the proportional coefficient, the limit value, and a temperature dependency coefficient are set as individual parameters representing each of the basic models; the force calculation unit that calculates a temperature-corrected limit value based on the limit value in accordance with a difference between the temperature detected by the temperature measurement unit and a fixed reference temperature, and uses the temperature-corrected limit value instead of the limit value, and / or calculates a temperature-corrected proportional coefficient based on the proportional coefficient, and uses the temperature-corrected proportional coefficient instead of the proportional coefficient; 3. The detection device according to claim 2, further comprising:

7. a housing that houses the detection element; the housing accommodates all or part of the setting unit, the input unit, and the force calculation unit, 2. The detection device according to claim 1, wherein the housing is provided so as to be attachable to an object to be measured without any additional processing.

8. a housing that houses the detection element; the housing accommodates the setting unit, the input unit, and a part of the force calculation unit, 2. The detection device according to claim 1, wherein the components housed in the housing and the components outside the housing are connected via wireless communication.

9. 2. The detection device according to claim 1, wherein the strain-flexing part is provided so as to be able to indirectly detect strain of the measurement object.

10. A measurement system characterized in that a measurement device having components outside the housing of the detection device and input / output units for external devices is connected to components inside the housing of the detection device described in claim 8 via wireless communication.

11. 10. A measurement system comprising: a measuring device having an input / output section for an external device connected to the detection device according to claim 1 via wireless communication.

Images