State determination device, state determination control system, state determination method, and computer program

Abstract

To provide a state determination device that can determine the state of an industrial machine by using a database for DD-PID control and a sensor for operation control.SOLUTION: A state determination device for determining a state of an industrial machine includes: an acquiring unit that acquires a database that stores operation state data indicating the operation state of the industrial machine, control parameters for controlling the industrial machine in the relevant operation state, and state values indicating the state of the industrial machine in advance in association with each other, and operation state data obtained by observing the operation state of the industrial machine; and a processing unit that determines the state of the industrial machine based on the operation state data acquired by the acquisition unit and the operation state data and state values stored in the database.SELECTED DRAWING: Figure 1

Description

Technical Field

[0004] , , , , , ,

[0006] , , , ,

[0005] , , , , ,

[0001] The present disclosure relates to a state determination device, a state determination control system, a state determination method, and a computer program.

Background Art

[0002] In order to operate industrial machinery in an optimal state, it is necessary to grasp the state of the industrial machinery, such as the presence or absence of abnormalities. Usually, sensors are attached to the monitoring target parts to determine the state of the industrial machinery. For example, Patent Document 1 discloses a monitoring method in which vibration is monitored by an acceleration sensor provided in a movable part of an injection molding machine to detect an abnormality in the movable part.

[0003] [[ID=​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​

[0007] A state determination device relating to one aspect of the present disclosure is a state determination device for determining the state of an industrial machine, comprising: a database that stores in advance a correspondence between operating status data indicating the operating status of the industrial machine, control parameters for controlling the industrial machine in the operating status, and state values ​​indicating the state of the industrial machine; an acquisition unit that acquires operating status data obtained by observing the operating status of the industrial machine; and a processing unit, wherein the processing unit determines the state of the industrial machine based on the operating status data acquired by the acquisition unit, the operating status data stored in the database, and the state values.

[0008] A state determination control system relating to one aspect of this disclosure comprises a state determination device and an industrial machine, wherein the state determination device includes a communication unit that transmits the operating status data and control parameters stored in the database to the industrial machine, and the industrial machine receives the operating status data and control parameters transmitted from the state determination device and operates based on the received operating status data and control parameters.

[0009] A state determination method relating to one aspect of this disclosure is a state determination method for determining the state of an industrial machine, comprising: preparing a database that stores in association the operating status data indicating the operating status of the industrial machine, control parameters for controlling the industrial machine in the operating status, and state values ​​indicating the state of the industrial machine; acquiring operating status data obtained by observing the operating status of the industrial machine; and determining the state of the industrial machine based on the acquired operating status data, the operating status data and state values ​​stored in the database.

[0010] A computer program relating to one aspect of this disclosure is a computer program that causes a computer capable of accessing a database which stores in association operating status data indicating the operating status of an industrial machine, control parameters for controlling the industrial machine in said operating status, and status values ​​indicating the state of the industrial machine, to execute a process to determine the state of the industrial machine, wherein the computer obtains operating status data obtained by observing the operating status of the industrial machine, and causes the computer to execute a process to determine the state of the industrial machine based on the obtained operating status data, the operating status data and status values ​​stored in the database. [Effects of the Invention]

[0011] According to this disclosure, the state of industrial machinery can be determined using a database for DD-PID control and sensors for operation control. [Brief explanation of the drawing]

[0012] [Figure 1] This is a schematic diagram showing an example configuration of the state determination control system according to this embodiment 1. [Figure 2] This is a block diagram showing an example configuration of the control device according to this embodiment 1. [Figure 3] This is a block diagram showing an example configuration of the state determination device according to this embodiment 1. [Figure 4] This is a conceptual diagram showing an example of a record layout for a database according to this embodiment 1. [Figure 5] This flowchart shows the processing procedure related to state determination according to Embodiment 1. [Figure 6] This is an explanatory diagram that conceptually illustrates the separation process of common data. [Figure 7] This graph shows the static characteristics of the Hammerstein model, an example of an industrial machine. [Figure 8] This graph shows the control results obtained using DD-PID control of the Hammerstein model. [Figure 9]It is a chart showing various parameters used in the state determination process in the Hammerstein model. [Figure 10] It is a chart showing the state determination result in the Hammerstein model. [Figure 11] It is a perspective view showing a stage moving device having a slider-crank mechanism which is an example of an industrial machine. [Figure 12] It is a graph showing the control result by the DD-PID control of the stage moving device. [Figure 13] It is a chart showing various parameters used in the state determination process in the slider-crank mechanism. [Figure 14] It is a chart showing the state determination result in the slider-crank mechanism. [Figure 15] It is a schematic diagram showing a configuration example of a state determination control system equipped with a molding machine as an example of an industrial machine. [Figure 16] It is a block diagram showing a configuration example of the state determination device according to Embodiment 2.

Mode for Carrying Out the Invention

[0013] Specific examples of a state determination control system and the like according to an embodiment of the present disclosure will be described below with reference to the drawings. Note that the present disclosure is not limited to these examples, and is indicated by the claims, and is intended to include all modifications within the meaning and scope equivalent to the claims. Also, at least a part of the embodiments described below may be arbitrarily combined.

[0014] (Embodiment 1) FIG. 1 is a schematic diagram showing a configuration example of the state determination control system according to Embodiment 1. The state determination control system according to Embodiment 1 includes an industrial machine 1 connected via a communication network and a state determination device 2. The state determination control system is a control system that performs DD-DID control of the industrial machine 1 and functions as a state determination system that determines the state of the industrial machine 1.

[0015] The state determination device 2 has a database 20 that is used for both DD-PID control and state determination of the industrial machine 1. The state determination device 2 reads a DD-PID control dataset corresponding to the state of the industrial machine 1 from the database 20 and transmits it to the industrial machine 1. The industrial machine 1 has a controlled machine 10 and a control device 11 that controls the operation of the controlled machine 10. The control device 11 receives the control dataset transmitted from the state determination device 2 and stores it in the control database (control DB) 1a. The control device 11 uses the control database 1a to perform DD-PID control of the operation of the controlled machine 10. On the other hand, the control device 11 transmits operating status data indicating the current operating status of the controlled machine 10 to the state determination device 2. The state determination device 2 receives the operating status data transmitted from the industrial machine 1 and performs a process to calculate a state value indicating the state of the controlled machine 10 based on the received operating status data and the information in the database 20.

[0016] <Industrial Machinery 1> The controlled machines 10 that constitute the industrial machine 1 include manufacturing equipment, machine tools, industrial robots, testing and analysis equipment, conveying equipment, power transmission equipment, ironmaking machinery, cleaning equipment, etc., and also include any other machines used in factories and businesses.

[0017] The controlled machine 10 is equipped with one or more sensors 10a that detect physical quantities necessary for controlling the operation of the industrial machine 1, which are information indicating the operating status of the industrial machine 1. These physical quantities include, for example, voltage, current, temperature, humidity, torque, pressure, speed of moving parts, acceleration, rotation angle, position, fluid flow rate, and velocity. The sensors 10a are, for example, current sensors, voltage sensors, temperature sensors, humidity sensors, torque sensors, pressure sensors, speed sensors, acceleration sensors, rotation angle sensors, positioning sensors, flow sensors, and flow meters. The sensors 10a output measurement signals indicating these physical quantities to the control device 11. These measurement signals include a device output value y(t) indicating the output of the operating controlled machine 10, and an observed quantity x(t) other than the output, which is observed using the sensors 10a. Note that the observed quantity x(t) is not essential information and does not need to be included in the measurement signal. In other words, the control device 11 may be configured to control the operation of the machine 10 to be controlled without using the observed quantity. Furthermore, the device output value y(t) and the observed quantity x(t) may be detected using the same sensor 10a, or they may be detected using different sensors 10a.

[0018] Figure 2 is a block diagram showing an example configuration of the control device 11 according to this embodiment 1. The control device 11 is a computer that controls the operation of the machine to be controlled 10, and its hardware configuration includes a control unit 11a, a storage unit 11b, an input / output unit 11c, a communication unit 11d, an operation unit 11e, and a display unit 11f. The control device 11 may also be a server device connected to a network. Furthermore, the control device 11 may be configured with multiple computers for distributed processing, or it may be implemented by multiple virtual machines provided within a single server, or it may be implemented using a cloud server.

[0019] The control unit 11a is a processor and includes arithmetic circuits such as a CPU (Central Processing Unit) and a multi-core CPU, internal storage devices such as ROM (Read Only Memory) and RAM (Random Access Memory), I / O terminals, a timing unit, etc. The control unit 11a is connected to the storage unit 11b, input / output unit 11c, communication unit 11d, operation unit 11e, and display unit 11f.

[0020] The storage unit 11b is a non-volatile memory such as a hard disk, EEPROM (Electrically Erasable Programmable ROM), or flash memory. The storage unit 11b stores the control database 1a and the control target value r(t) necessary for DD-PID control of the controlled machine 10. The control database 1a stores multiple control data sets used for DD-PID control of the controlled machine 10.

[0021] The control dataset consists of an information vector, which is the operating status data φ(i), and a control parameter θ(i), which is the PID control gain, and is expressed by the following equations (1) to (3).

[0022]

number

[0023] The control database 1a is information provided by the state determination device 2. The control unit 11a receives the control dataset transmitted from the state determination device 2 via the communication unit 11d and stores the received control dataset in the control database 1a. In other words, the control unit 11a stores the received control dataset in the storage unit 11b.

[0024] As mentioned above, the observed quantity x(t) is not always necessary, but it is useful when you want to appropriately represent the operating conditions as an information vector.

[0025] The input / output unit 11c includes an output circuit that outputs a signal and an input circuit that receives a signal. The output circuit outputs a control signal to the controlled machine 10 for controlling the operation of the controlled machine 10 according to the control of the control unit 11a. The control signal is a signal indicating the control input value u(t) that is input to the controlled machine 10. The input circuit receives the measurement signal output from the sensor 10a and provides the input measurement signal to the control unit 11a as measurement data. The measurement data includes the device output value y(t) indicating the output of the operated controlled machine 10, and an observed quantity x(t) other than the device output that can be observed using the sensor 10a.

[0026] The control unit 11a of the control device 11 calculates operating status data φ(t) which is expressed using the target values ​​r(t), r(t-1)... of the device output, the device output values ​​y(t), y(t-1)..., the control input values ​​u(t-1), u(t-2)... used in previous processing steps, and the observed quantities x(t), observed quantities x(t)... during the operation of the machine to be controlled 10, and stores it in the storage unit 11b. The control unit 11a reads the control parameter θ(t) from the control database 1a using the operating status data φ(t) as a key. Based on the read control parameter θ(t), the control unit 11a calculates the control input value u(t) and controls the operation of the machine to be controlled 10 by outputting a control signal based on the calculated control input value u(t) to the machine to be controlled 10.

[0027] The communication unit 11d is a communication circuit that sends and receives information according to a predetermined communication protocol. The communication unit 11d is connected to the status determination device 2 via a communication network, and the control unit 11a can send and receive various information to and from the status determination device 2 via the communication unit 11d. Specifically, the control unit 11a transmits multiple operating status data φ(t) stored in the storage unit 11b to the status determination device 2 via the communication unit 11d. When determining the current state of the industrial machine 1, the control unit 11a transmits multiple most recent operating status data φ(t) stored in the storage unit 11b to the status determination device 2. Furthermore, the communication unit 11d receives control data sets transmitted from the status determination device 2 at appropriate intervals and stores the received control data sets in the control database 1a.

[0028] The control unit 11e is an input device for operating the industrial machine 1. The control unit 11e is an interface for controlling the operation of the controlled machine 10 by setting, for example, the target value r(t) of the control, other operating conditions of the controlled machine 10, etc. The control unit 11e consists of, for example, operation buttons, operation keys, and a touch panel provided on the display unit 11f, and provides data indicating the operation content to the control unit 11a.

[0029] The display unit 11f is a display device such as a liquid crystal display panel or an organic EL display panel. The display unit 11f displays the operation screen necessary for operating the industrial machine 1 and also displays the status of the controlled machine 10.

[0030] <State determination device 2> Figure 3 is a block diagram showing an example configuration of the state determination device 2 according to this embodiment 1. The state determination device 2 is a computer and comprises a processing unit 21, a storage unit 22, and a communication unit (acquisition unit) 23. The storage unit 22 and the communication unit 23 are connected to the processing unit 21.

[0031] The processing unit 21 is a processor and includes arithmetic processing circuits such as a CPU, multi-core CPU, GPU (Graphics Processing Unit), GPGPU (General-purpose computing on graphics processing units), TPU (Tensor Processing Unit), ASIC, FPGA, NPU (Neural Processing Unit), internal storage devices such as ROM and RAM, and I / O terminals. The processing unit 21 functions as the state determination device 2 according to this embodiment 1 by executing the computer program P stored in the storage unit 22 described later, and implements the state determination method according to this embodiment 1. Each functional part of the state determination device 2 may be implemented in software, or some or all of them may be implemented in hardware.

[0032] The storage unit 22 is a non-volatile memory such as a hard disk, EEPROM, or flash memory. The storage unit 22 stores a computer program P for determining the state of the industrial machine 1 and a database 20 for performing DD-PID control and state determination.

[0033] The computer program P and database 20 may be recorded on the recording medium 3 in a manner that is computer-readable. The storage unit 22 stores the computer program P and database 20 read from the recording medium 3 by a reading device (not shown). The recording medium 3 is a semiconductor memory such as flash memory. The recording medium 3 may also be an optical disc such as a CD (Compact Disc)-ROM, DVD (Digital Versatile Disc)-ROM, or BD (Blu-ray® Disc). Furthermore, the recording medium 3 may be a magnetic disc such as a flexible disk or hard disk, or a magneto-optical disc. In addition, the computer program P and database 20 may be downloaded from an external server (not shown) connected to a communication network (not shown) and stored in the storage unit 22.

[0034] The communication unit 23 is a communication circuit that sends and receives information according to a predetermined communication protocol such as Ethernet (registered trademark). The communication unit 23 is connected to the control device 11 via a communication network, and the processing unit 21 can send and receive various information to and from the control device 11 via the communication unit 23. For example, the communication unit 23 receives operating status data φ(t) transmitted from the control device 11. The communication unit 23 also transmits a control data set to the control device 11 that associates operating status data φ(i) and control parameters θ(i) according to the state of the industrial machine 1.

[0035] Figure 4 is a conceptual diagram showing an example of the record layout of the database 20 according to this embodiment 1. The database 20 stores multiple datasets necessary for DD-PID control of the industrial machine 1 and for determining the state of the industrial machine 1. The dataset consists of operating status data φ(i) indicating multiple operating states of the industrial machine 1, control parameters θ(i) for controlling the industrial machine 1 in the operating state, and state values ​​z(i) indicating the state of the industrial machine 1.

[0036] The processing unit 21 reads a control dataset [φ(i),θ(i)] from the database 20 according to the status of the industrial machine 1, and transmits the read control dataset to the control device 11 of the industrial machine 1. Furthermore, the processing unit 21 uses multiple state evaluation datasets, particularly those consisting of operating status data φ(i) and state value z(i), from among the datasets stored in the database 20 to numerically determine the state of the industrial machine 1. The state evaluation dataset is represented by the following equation (4).

[0037]

number

[0038] <State determination process> The state determination device 2 according to this embodiment performs a process to numerically determine the state of the industrial machine 1 using a dataset prepared in advance, mainly for use in DD-PID control. Specifically, the processing unit 21 extracts a plurality of datasets [φ(i), z(i)] located near the operating status data φ(t) acquired from the industrial machine 1 from the database 20, and performs a process to calculate the state value z(t) based on the extracted datasets. Incidentally, since the state value calculation process is performed using a limited dataset pre-stored in database 20, the state value has upper and lower limits. When the state of industrial machine 1 is near these upper and lower limits, there is a problem in that the error in the calculated state value z(t) becomes large. Therefore, logit transformation was used to improve the accuracy of state determination near the upper and lower limits.

[0039] Figure 5 is a flowchart showing the processing procedure for state determination according to Embodiment 1. The processing unit 21 of the state determination device 2 acquires a plurality of operating status data φ(t) transmitted from the industrial machine 1 (step S11). The plurality of operating status data φ(t) is information obtained by measurement at multiple points in time. For example, in the case of an industrial machine 1 such as a manufacturing device, a plurality of operating status data φ(t) obtained by measurement during one cycle time are transmitted to the state determination device 2, and the processing unit 21 acquires the plurality of operating status data φ(t).

[0040] Next, the processing unit 21 performs a common data separation process on the dataset stored in the database 20 (step S12). Step S12 is a process to determine whether the state of the industrial machine 1 is near the upper or lower limit of the state value z(i) of the dataset stored in the database 20, and to determine the state level of the industrial machine 1.

[0041] Figure 6 is a conceptual diagram illustrating the separation process of common data. The database (z1) in the left figure represents the dataset where the state value z(i) is z1. Similarly, databases (z2) and (z3) represent the datasets where the state value z(i) is z2 and z3, respectively. The state level is determined by determining which state value dataset the operating status data φ(t) acquired in step S11 belongs to. Since the datasets stored in database 20 are mainly for DD-PID control, there are datasets where the operating status data φ(i) is similar despite the state value z(i) being different. Therefore, in this embodiment, in order to improve the accuracy of determining the state level of the industrial machine 1, a process is performed to exclude datasets from the datasets stored in database 20 that have different state values ​​z(i) and similar operating status data φ(i). In Figure 6, the database in the center shows the state after excluding datasets that may worsen the accuracy of determining the state level. The figure on the right schematically shows the datasets that have been excluded as not being used for determining the state level.

[0042] More specifically, the processing unit 21 calculates the statistical distance d between the operating status data φ1 of each dataset having a first state value and the operating status data φ2 of each dataset having a second state value. For example, the processing unit 21 calculates the statistical distance between the operating status data φ1 and φ2 using the weighted L1 norm (Manhattan distance) represented by the following equation (5). The statistical distance shown in the following equation (5) is just one example.

[0043]

number

[0044] Then, the processing unit 21 excludes from the database 20 any datasets that contain driving condition data where the statistical distance is closer than a predetermined value dc. Furthermore, the processing unit 21 may perform a process to exclude from the multiple acquired operating status data φ(t) any operating status data φ(t) whose statistical distance from the operating status data (i) of the excluded dataset is closer than a predetermined value dc. Hereafter, the dataset obtained after common data separation processing has been performed on database 20 will be referred to as the state level evaluation dataset. Furthermore, the operating status data obtained after common data separation processing has been performed on multiple operating status data φ(t) will be referred to as the operating status data to be evaluated.

[0045] Next, the processing unit 21 determines the state level of the industrial machine 1 (step S13). The state level is an indicator used to determine whether or not to perform the logit transformation described later. For example, the processing unit 21 calculates the statistical distance between the operating status data to be evaluated and the operating status data of the state level evaluation dataset for each state value, and identifies the state value of the state level evaluation dataset with the smallest statistical distance as the state level of the industrial machine 1.

[0046] More specifically, the processing unit 21 calculates the cumulative value S of the nearest neighbor distances between data sets represented by the following equation (6), and identifies the state value of the state level evaluation dataset to which the operating condition data with the smallest cumulative value S belongs as the state level of the industrial machine 1. The classification process using the following equation (6) is just one example, and the state level of the industrial machine 1 may also be identified by applying a known statistical classification process.

[0047]

number

[0048] Next, the processing unit 21 executes a process to calculate the status value of the industrial machine 1. The status value is calculated using a dataset registered in the database 20, rather than the evaluation dataset.

[0049] The processing unit 21 normalizes the state value z(i) of each dataset registered in the database 20 to a positive value less than 1 for the convenience of numerical processing (step S14). For example, the processing unit 21 normalizes the state value using the following formula (7).

[0050]

number

[0051] Next, the processing unit 21 determines whether the state of the industrial machine 1 is near the upper or lower limit of the state value z(i) in the dataset stored in the database 20 (step S15). The state of the industrial machine 1 is represented by the state level calculated in step S13. The definitions of the upper and lower limits of the state value z(i) are not particularly limited, and the upper and lower limits can be determined by any method that can ensure the accuracy of the state value calculation. For example, the upper and lower limits of the state value z(i) of the dataset stored in the database 20 may be defined as being near the upper limit and near the lower limit. Alternatively, the processing unit 21 may determine whether a value is near the upper or lower limit by using a predetermined state value from the upper limit side and a predetermined state value from the lower limit side as thresholds. Furthermore, the processing unit 21 may determine whether a value is near the upper or lower limit by using a predetermined number of values ​​smaller than the upper limit and a predetermined number of values ​​larger than the lower limit as thresholds. The predetermined number is, for example, a predetermined percentage of the difference between the upper and lower limits.

[0052] If the state of industrial machine 1 is determined to be near the upper or lower limit (step S15: YES), the processing unit 21 performs a logit transformation on the state value z normalized in step S14 (step S16). The logit transformation is expressed by the following equation (8). The logit transformation expands the values ​​near the lower limit of the normalized state value z (near 0) to -∞, and expands the values ​​near the upper limit of the normalized state value z (near 1) to +∞. The inverse logit transformation is expressed by the following equation (9).

[0053]

number

[0054]

number

[0055] Next, the processing unit 21 calculates a state value based on the acquired operating status data and the dataset in the database 20 (step S17). Specifically, the processing unit 21 calculates the statistical distance between the acquired operating status data and the dataset stored in the database 20. The statistical distance is expressed, for example, by equation (5) above. Next, the processing unit 21 uses a predetermined number of k datasets with small statistical distances as neighboring data and calculates the state value of the operating status data φ(t) at time t using the following equations (10) and (11).

[0056]

number

[0057] For simplicity of notation, the state value is denoted as z(t). However, if logit transformation is performed in step S16, the state value of the operating condition data φ(t) is determined using the logit-transformed state value. The weighting coefficient is larger the smaller the statistical distance between the data, and smaller the larger the statistical distance between the data. The processing unit 21 then determines the operating status data φ(t) obtained from the results of a certain operation, that is, the state value of each of the acquired operating status data φ(t), and calculates the average value of the state values. Hereinafter, this average value will simply be referred to as the state value.

[0058] Next, the processing unit 21 performs an inverse logit transformation on the state value (average value) calculated in step S17 (step S18).

[0059] In step S15, if it is determined that the state of the industrial equipment is not near the upper or lower limit of the state value (step S15: NO), the processing unit 21 calculates the average value of the state value based on multiple operating condition data φ(t) without performing logit conversion (step S19). The average value of the state value is obtained by formulas (10) and (11) above, as in step S17.

[0060] After completing the processing in step S18 or step S19, the processing unit 21 reverse-converts the state value back to its value before normalization (step S20). The processing unit 21 can reverse-convert the state value back to its value before normalization using the following formula (12).

[0061]

number

[0062] Next, the processing unit 21 transmits the state value, which was inversely converted in step S20, to the control device 11 of the industrial machine 1 (step S21). The control device 11 of the industrial machine 1 receives the state value transmitted from the state determination device 2 and displays it, for example, on the display unit 11f. Note that the process of transmitting the state value to the control device 11 is just one example of a method for outputting the state value, and the state determination device 2 may be configured to output the state value indicating the state of the industrial machine 1 by transmitting it to another communication terminal, etc. Next, we will verify the effectiveness of the proposed method through numerical simulation.

[0063] <First example of industrial machinery 1: Hammerstein model> As a first example of the controlled machine 10, we consider a model in which the Hammerstein model is extended so that the system changes with each operation cycle L. The Hammerstein model is expressed by the following equation (13).

[0064]

number

[0065] Figure 7 is a graph showing the static characteristics of a Hammerstein model, which is an example of industrial machine 1. The horizontal axis represents the control input value u, and the vertical axis represents the device output value y. The solid line shows the static characteristics of the model when L=0, and the dashed line shows the static characteristics of the model when L=100. Hammerstein models have similar characteristics in the low input / output range.

[0066] Figure 8 is a graph showing the control results using DD-PID control of the Hammerstein model. Figure 8A shows the time change of the device output value y(t), and Figure 8B shows the time change of the control input value u(t).

[0067] Using the control results shown in Figure 8, the state of the number of operations L, which is an example of a state value, is determined. Database 20 stores the operating status data φ(i) and control parameter θ(i) for each of the number of operations (state values) L=0,10,...100, set in increments of 10 from 0 to 100. The target value r(t) is the same value as L. Database 20 stores a sufficient amount of data to calculate the number of operations, which is the state value of industrial machine 1.

[0068] Figure 9 is a diagram showing the various parameters used in the state determination process in the Hammerstein model. L = 1, 10, 90, 100 were considered to be near the upper and lower limits of the state values, and the number of operations L was calculated using the parameters (ny, nu, nx, dc, zmin, zmax, zmm, N, k) shown in the diagram in Figure 9.

[0069] Figure 10 is a chart showing the state determination results in the Hammerstein model. The top row shows the true value of the number of operations, the middle row shows the state level, and the bottom row shows the number of operations (state value) calculated by the state determination method of this embodiment. Even for conditions not included in database 20, generally good values ​​for the number of operations (state value) are calculated.

[0070] <Second example of industrial machinery 1: Slide crank mechanism> As a second example of the controlled machine 10, consider a stage moving device having a slide crank mechanism.

[0071] Figure 11 is a perspective view showing a stage moving device having a slide crank mechanism, which is an example of industrial machine 1. The stage moving device comprises a stage 10c that moves along two guides 10b provided on a base, a drive gear 10d rotated by a motor (not shown), a driven gear 10e, a transmission belt 10f, and a link mechanism 10g. The transmission belt 10f is a belt that transmits the driving force of the motor from the drive gear 10d to the driven gear 10e. The link mechanism 10g is a mechanism that converts the rotational motion of the driven gear 10e into the reciprocating sliding motion of the stage 10c and transmits it.

[0072] Here, the rotational angular velocity of the drive shaft is denoted as the device output value y(t), the torque of the drive shaft as the control input value u(t), the absolute angle of the driven shaft as the observed quantity x(t), and the sampling time is Ts = 0.001 (seconds).

[0073] Figure 12 is a graph showing the control results of the stage movement device using DD-PID control. Figure 12A shows the time change of the device output value y(t), and Figure 12B shows the time change of the control input value u(t).

[0074] Using the control results shown in Figure 12, the state of the weight Ms(g) of stage 10c, which is an example of a state value, is determined. Database 20 stores operating status data φ(i) and control parameters θ(i) for each weight Ms (state value) Ms=0,200...1000, set in increments of 10 from 0 to 100. Database 20 stores the observed data set at x(0)=0. The target value r(t) is a predetermined torque value. Database 20 stores a sufficient amount of data set to calculate the weight Ms of stage 10c, which is the state value of industrial machine 1.

[0075] Figure 13 is a diagram showing the various parameters used in the state determination process of the slide crank mechanism. Ms = 0 to 1000 are considered the upper and lower limits of the state value, and the weight Ms was calculated using the parameters (ny, nu, nx, dc, zmin, zmax, zmm, N, k) shown in the diagram in Figure 13. The control result being evaluated is the operating condition data obtained at x(0)=1(deg).

[0076] Figure 14 is a diagram showing the state determination results for the slide crank mechanism. The top row shows the true value of the weight of stage 10c, the middle row shows the state level, and the bottom row shows the weight of stage 10c (state value) calculated by the state determination method of this embodiment. Even for conditions not included in database 20, values ​​for generally good operation counts (state values) are calculated. Note that the results for Ms=100, 200, and 800 have a large error, but it is thought that the accuracy of weight determination can be improved by storing more data sets in database 20.

[0077] <Example 3 of Industrial Machinery 1: Molding Machine> Figure 15 is a schematic diagram showing an example of a state determination control system configuration, which includes a molding machine as an example of industrial machinery 1. This section explains the configuration of the state determination control system, rather than presenting simulation results.

[0078] Industrial machine 1 is, for example, a molding machine such as an injection molding machine or an extruder. Figure 15 shows an injection molding machine as an example of industrial machine 1. The injection molding machine, which is an example of industrial machine 1, comprises a control device 11, a mold clamping device 16 for clamping the mold 15, an injection device 17 for melting and injecting the molding material, and a sensor 10a for detecting physical quantities related to the operating status of the injection molding machine.

[0079] The mold clamping device 16 comprises a fixed platen fixed to the bed and a movable platen that slides on the bed. A fixed mold and a movable mold are provided on the fixed platen and the movable platen, respectively. The mold clamping device 16 includes a mold clamping mechanism that opens and closes the mold 15 by moving the movable platen. The mold clamping mechanism is composed of, for example, a toggle mechanism. The mold clamping device 16 is equipped with a servo motor for opening and closing the mold, and the mold clamping mechanism is operated by the torque output by the servo motor for opening and closing the mold. The servo motor for opening and closing the mold is driven by power supplied from a servo amplifier.

[0080] The injection device 17 is mounted on a base. The injection device 17 comprises a heating cylinder 17a having a nozzle at its tip, and a screw 17b rotatably disposed within the heating cylinder 17a in the circumferential and axial directions. A heater for melting the molding material is provided inside or on the outer circumference of the heating cylinder 17a. The screw 17b is driven in the rotational and axial directions by a drive device 17c.

[0081] The drive unit 17c includes an injection servo motor and a ball screw mechanism for driving the screw 17b in the axial direction. The drive unit 17c also includes a screw rotation servo motor for rotationally driving the screw 17b. Both the injection servo motor and the screw rotation servo motor are driven by power supplied from a servo amplifier.

[0082] A hopper 17d into which the molding material is fed is provided near the rear end of the heating cylinder 17a.

[0083] Sensor 10a is a circuit that detects physical quantities necessary to control the operation of industrial machine 1, and provides information indicating the operating status of industrial machine 1. Examples of sensors 10a necessary to control the operation of an injection molding machine include a current sensor that detects the current flowing through a servo amplifier, a rotational speed sensor that detects the rotational speed of a motor, a current sensor that detects heater temperature or heater current, a temperature sensor, etc. Sensor 10a outputs a measurement signal indicating a physical quantity to the control device 11.

[0084] The control device 11 outputs control signals to the injection molding machine to control the operation of the injection molding machine according to the control of the control unit 11a. For example, the output circuit of the input / output unit 11c outputs control signals to the servo amplifier to control the operation of the injection servo motor, the screw rotation servo motor, and the mold opening / closing servo motor. These control signals indicate, for example, the torque value or current value of the servo motor. The output circuit also outputs a control signal that specifies the heater temperature of the heating cylinder 17a. These control signals indicate, for example, the voltage value applied to the heater. The input circuit of the input / output unit 11c receives measurement signals output from the sensor 10a, such as the servo motor current value, rotation speed, heater current, heater temperature, etc.

[0085] In injection molding machines, DD-PID control applies to components such as the servo motor and heater for the clamping device 16. The operating status data φ, which constitutes the dataset related to servo motor control, is represented by the target motor speed r, the measured motor speed (device output value y), and the servo motor current value or torque value (control input value u). The control parameter θ is a parameter for PID control of the motor speed. The state value z is the weight of the type. The operating status data φ, which constitutes the dataset related to heater temperature control, is represented by the target heater temperature r, the measured heater temperature (device output value y), and the heater current value (control input value u). The control parameter θ is a parameter for PID control of the heater temperature. The state value z is a numerical value indicating the type or characteristics of the resin. In addition, the condition of the injection molding machine can be numerically determined by using a state value z that represents the degree of abnormality of the movable parts of the injection molding machine.

[0086] Although an injection molding machine has been described here, it goes without saying that the technology of this embodiment 1 can also be applied to extruders and other molding machines. In the case of an extruder, the processing unit 21 can numerically determine the state of the extruder using a servo motor that drives the extruder's screw and a data set that controls the cylinder heater temperature using DD-PID.

[0087] According to the state determination control system of this embodiment 1, the state of the industrial machine 1 can be determined using the database 20 for DD-PID control and the sensor 10a for operation control. According to this embodiment 1, the database for DD-PID control and the database for state determination can be shared. The processing unit 21 can numerically determine the state of the industrial machine 1 using the DD-PID control dataset. Furthermore, the processing unit 21 can control the operation of the industrial machine 1 using DD-PID control by transmitting a control dataset that associates the operating status data φ(i) and the control parameter θ(i) to the industrial machine 1.

[0088] If the state value of industrial machine 1 is near its upper or lower limit, the accuracy of calculating the state value can be improved by normalizing the state value of the dataset stored in database 20 to a positive value less than 1 and performing a logit transformation.

[0089] If the state value of industrial machine 1 is not near its upper or lower limit, the accuracy of the state value calculation can be improved by normalizing the state value of the dataset stored in database 20 to a positive value less than 1 and calculating the state value without performing a logit transformation. Performing a logit transformation when the state value is not near its upper or lower limit will actually worsen the accuracy of the state value calculation.

[0090] When determining whether the state value of industrial machine 1 is near its upper or lower limit, the accuracy of classifying state levels and calculating state values ​​can be improved by performing a common data separation process.

[0091] By using a predetermined number of datasets stored in database 20 that are statistically close to the measured driving condition data, computational costs can be reduced. Furthermore, by using weights based on statistical distance to perform a weighted average of the state values ​​of these datasets, the accuracy of calculating the state values ​​can be improved.

[0092] Although an example of using database 20 to determine the state of industrial machine 1 and perform DD-PID control has been explained, database 20 may be updated as needed. For example, if the state value z of industrial machine 1 operated by DD-PID control can be observed through inspection, etc., the operating status data φ and the state value z at the time of operation may be added to database 20 in association with each other. Of course, the operating status data φ, control parameter θ, and state value z at the time of operation may also be registered in database 20 in association with each other.

[0093] In this embodiment, an example of determining the state of the industrial machine 1 using the operating status data φ(i) and state value (i) stored in the database 20 has been described, but the state of the industrial machine 1 may also be determined using the control parameter θ(i). The control device 11 of the industrial machine 1 transmits the operating status data φ(t) and the control parameter θ(t) to the state determination device 2. The processing unit 21 of the state determination device 2 receives the operating status data φ(i) and the control parameter θ(t) transmitted from the industrial machine 1.

[0094] The processing unit 21 excludes operating status data φ(i), control parameters θ(i), and state values ​​z(i) stored in the database if the state value z(i) is different and the statistical distance of the information vector consisting of the operating status data φ(i) and control parameters θ(i) is less than a predetermined value. Based on the excluded operating status data φ(i), control parameters θ(i), and state values ​​z(i), the processing unit 21 determines whether the acquired operating status data φ(t) and control parameters θ(t) of industrial machine 1 are data located near the operating status data φ(i) and control parameters θ(i) with the maximum or minimum state value.

[0095] If the acquired operating status data φ(t) and control parameter θ(t) are near the operating status data φ(i) and control parameter θ(i) with the maximum or minimum state value among the operating status data φ(i) and control parameter θ(i) stored in the database 20, the processing unit 21 performs a logit transformation on the state value z(i) stored in the database 20.

[0096] The processing unit 21 calculates the state value z of the industrial machine 1 by weighting the state values ​​z(i) associated with a predetermined number of operating status data φ(i) and control parameter θ(i) that are statistically close to the acquired operating status data φ(t) and control parameter θ(i) among the operating status data φ(i) and control parameter θ(i) stored in the database 20, using weights based on the said statistical distance.

[0097] (Embodiment 2) The state determination device according to Embodiment 2 differs from the embodiment in that it operates as a control device for controlling the operation of industrial machinery. The other components of the state determination device are the same as those of the state determination system according to Embodiment 1, so the same reference numerals are used for the same parts, and detailed descriptions are omitted.

[0098] Figure 16 is a block diagram showing an example configuration of a state determination device according to Embodiment 2. The state determination control device (state determination device) 202 according to Embodiment 2 is a computer and comprises a processing unit 221, a storage unit 222, an input / output unit 223, an operation unit 224, and a display unit 225.

[0099] The input / output unit 223, the operation unit 224, and the display unit 225 have the same or similar configuration as the input / output unit 11c, the operation unit 11e, and the display unit 11f of the control device 11 according to Embodiment 1.

[0100] The processing unit 221 and the storage unit 222 have the same configuration as the processing unit 21 and storage unit 22 of the state determination device 2 according to Embodiment 1. The storage unit 222 stores the database 20. The processing unit 221 uses the operating status data φ(i) and control parameters θ(i) stored in the database 20 to control the operation of the controlled machine 10, similar to Embodiment 1. In addition, the processing unit 221 can determine the state of the industrial machine 1 based on the operating status data obtained by measurement using the sensor 10a and the information in the database 20, similar to Embodiment 1. The industrial machine 1 is, for example, a molding machine such as an injection molding machine or extruder as described in Embodiment 1.

[0101] In the state determination control device 202 according to Embodiment 2, the state of the industrial machine 1 can be determined using the database 20 for DD-PID control and the sensor 10a for operation control, similar to Embodiment 1.

[0102] (Note 1) A condition determination device for determining the condition of industrial machinery, A database that stores, in advance, operating status data indicating the operating status of the industrial machine, control parameters for controlling the industrial machine in said operating status, and status values ​​indicating the state of the industrial machine, An acquisition unit that acquires operating status data obtained by observing the operating status of the aforementioned industrial machine, Processing section and Equipped with, The aforementioned processing unit, Based on the operating status data acquired by the acquisition unit, the operating status data stored in the database, and the status value, the state of the industrial machine is determined. State determination device.

[0103] (Note 2) The aforementioned processing unit, The state values ​​stored in the database are logit-converted, Using the logit-converted state value, the state value corresponding to the operating status data acquired by the acquisition unit is calculated. Perform an inverse logit transformation on the calculated state value. The state determination device described in Appendix 1.

[0104] (Note 3) The aforementioned processing unit, The state values ​​stored in the database are normalized to positive values ​​less than 1 and then logit-transformed. The calculated state values ​​are inversely normalized and then subjected to an inverse logit transformation. The state determination device described in Appendix 2.

[0105] (Note 4) The aforementioned processing unit, If the operating status data acquired by the acquisition unit is near the operating status data with the maximum status value, or near the operating status data with the minimum status value, the database stores the status value and performs a logit transformation. A state determination device as described in Appendix 2 or Appendix 3.

[0106] (Note 5) The aforementioned processing unit, From the driving status data and status values ​​stored in the database, the system excludes driving status data and status values ​​that have different status values ​​and whose statistical distance from the driving status data is less than a predetermined value. Based on the excluded driving status data and status values, the system determines whether the driving status data acquired by the acquisition unit is near the driving status data with the highest status value or the driving status data with the lowest status value. The state determination device described in Appendix 4.

[0107] (Note 6) The aforementioned processing unit, The state value of the industrial machine is calculated by weighting the state values ​​associated with a predetermined number of operating status data points from the operating status data stored in the database that are statistically close in distance to the operating status data acquired by the acquisition unit, using a weight based on the statistical distance. A state determination device as described in any one of the items from Appendix 1 to Appendix 5.

[0108] (Note 7) The aforementioned processing unit, The unit reads the control parameters corresponding to the operating status data acquired by the acquisition unit from the database, and controls the industrial machine based on the read control parameters. A state determination device as described in any one of the items from Appendix 1 to Appendix 6.

[0109] (Note 8) The aforementioned industrial machinery includes a molding machine. A state determination device as described in any one of the items from Appendix 1 to Appendix 7.

[0110] (Note 9) A state determination device described in any one of the items from Appendix 1 to Appendix 8, Industrial machinery and Equipped with, The state determination device is The system includes a communication unit that transmits the operating status data and control parameters stored in the database to the industrial machine. The aforementioned industrial machine is The system receives the operating status data and control parameters transmitted from the status determination device, and operates based on the received operating status data and control parameters. State determination control system.

[0111] (Note 10) A method for determining the state of industrial machinery, A database is prepared in which operating status data indicating the operating status of the industrial machine, control parameters for controlling the industrial machine in said operating status, and status values ​​indicating the state of the industrial machine are stored in association with each other. Obtain operating data obtained by observing the operating conditions of the aforementioned industrial machine, The state of the industrial machine is determined based on the acquired operating status data and the operating status data and state values ​​stored in the database. A method for determining the state.

[0112] (Note 11) A computer program that causes a computer capable of accessing a database storing operating status data indicating the operating status of an industrial machine, control parameters for controlling the industrial machine in that operating status, and status values ​​indicating the state of the industrial machine, to perform a process to determine the state of the industrial machine, Obtain operating data obtained by observing the operating conditions of the aforementioned industrial machine, The state of the industrial machine is determined based on the acquired operating status data and the operating status data and state values ​​stored in the database. A computer program that causes the aforementioned computer to perform a process. [Explanation of symbols]

[0113] 1: Industrial machinery 1a: Control database 2: State determination device 3: Recording media 10: Machine to be controlled 10a: Sensor 11: Control device 11a: Control Unit 11b: Storage section 11c: Input / output section 11d: Communications Department 11e:Operation unit 11f: Display section 15: Mold 16: Mold clamping device 17: Injection device 17a: Heating cylinder 17b: Screw 17c: Drive unit 17d: Hoppa 20: Database 21: Processing Unit 22: Storage section 23: Communications Department 202: State determination control device 221: Processing Unit 222: Storage section 223: Input / output section 224:Operation unit 225: Display section

Claims

1. A condition determination device for determining the condition of industrial machinery, A database that stores, in advance, operating status data indicating the operating status of the industrial machine, control parameters for controlling the industrial machine in said operating status, and status values ​​indicating the state of the industrial machine, An acquisition unit that acquires operating status data obtained by observing the operating status of the aforementioned industrial machine, Processing section and Equipped with, The aforementioned processing unit, Based on the operating status data acquired by the acquisition unit and the operating status data and status values ​​stored in the database, the state of the industrial machine is determined. The unit reads the control parameters corresponding to the operating status data acquired by the acquisition unit from the database, and controls the industrial machine based on the read control parameters. State determination device.

2. A condition determination device for determining the state of an industrial machine, A database that stores, in advance, operating status data indicating the operating status of the industrial machine, control parameters for controlling the industrial machine in said operating status, and status values ​​indicating the state of the industrial machine, An acquisition unit that acquires operating status data obtained by observing the operating status of the aforementioned industrial machine, Processing section and Equipped with, The aforementioned processing unit, If the operating status data acquired by the acquisition unit is near the operating status data with the maximum status value, or near the operating status data with the minimum status value, the state value stored in the database is logit converted. Using the logit-converted state value, the state value corresponding to the operating status data acquired by the acquisition unit is calculated. The state of the industrial machine is determined by performing an inverse logit transformation on the calculated state value. State determination device.

3. The aforementioned processing unit, From the driving status data and status values ​​stored in the database, the system excludes driving status data and status values ​​that have different status values ​​and whose statistical distance from the driving status data is less than a predetermined value. Based on the excluded driving status data and status values, the system determines whether the driving status data acquired by the acquisition unit is near the driving status data with the highest status value or the driving status data with the lowest status value. The state determination device according to claim 2.

4. A condition determination device for determining the condition of an industrial machine, A database that stores, in advance, operating status data indicating the operating status of the industrial machine, control parameters for controlling the industrial machine in said operating status, and status values ​​indicating the state of the industrial machine, An acquisition unit that acquires operating status data obtained by observing the operating status of the aforementioned industrial machine, Processing section and Equipped with, The aforementioned processing unit, The state value of the industrial machine is calculated by weighting the state values ​​associated with a predetermined number of operating status data points from the operating status data stored in the database that are statistically close in distance to the operating status data acquired by the acquisition unit, using a weight based on the statistical distance. State determination device.

5. The aforementioned processing unit, The state values ​​stored in the database are logit-converted, Using the logit-converted state value, the state value corresponding to the operating status data acquired by the acquisition unit is calculated. Perform an inverse logit transformation on the calculated state value. The state determination device according to claim 1.

6. The aforementioned processing unit, The state values ​​stored in the database are normalized to positive values ​​less than 1 and then logit-transformed. The calculated state values ​​are inversely normalized and then subjected to an inverse logit transformation. The state determination device according to claim 5.

7. The aforementioned industrial machinery includes a molding machine. A state determination device according to any one of claims 1 to 6.

8. A state determination device according to any one of claims 1 to 6, Industrial machinery and Equipped with, The state determination device is The system includes a communication unit that transmits the operating status data and control parameters stored in the database to the industrial machine. The aforementioned industrial machine is The system receives the operating status data and control parameters transmitted from the status determination device, and operates based on the received operating status data and control parameters. State determination control system.

9. A method for determining the state of industrial machinery, A database is prepared in which operating status data indicating the operating status of the industrial machine, control parameters for controlling the industrial machine in said operating status, and status values ​​indicating the state of the industrial machine are stored in association with each other. Obtain operating data obtained by observing the operating conditions of the aforementioned industrial machine, Based on the acquired operating status data and the operating status data and status values ​​stored in the database, the state of the industrial machine is determined. The control parameters corresponding to the acquired operating status data are read from the database, and the industrial machine is controlled based on the read control parameters. A method for determining the state.

10. A method for determining the state of an industrial machine, A database is prepared in which operating status data indicating the operating status of the industrial machine, control parameters for controlling the industrial machine in said operating status, and status values ​​indicating the state of the industrial machine are stored in association with each other. Obtain operating data obtained by observing the operating conditions of the aforementioned industrial machine, If the acquired driving status data is near the driving status data with the maximum status value, or near the driving status data with the minimum status value, the state value stored in the database is logit-converted. Using the logit-converted state values, the state values ​​corresponding to the acquired operating status data are calculated. The state of the industrial machine is determined by performing an inverse logit transformation on the calculated state value. A method for determining the state.

11. A method for determining the state of an industrial machine, A database is prepared in which operating status data indicating the operating status of the industrial machine, control parameters for controlling the industrial machine in said operating status, and status values ​​indicating the state of the industrial machine are stored in association with each other. Obtain operating data obtained by observing the operating conditions of the aforementioned industrial machine, The state value of the industrial machine is calculated by weighting the state values ​​associated with a predetermined number of operating status data points that are statistically close to the acquired operating status data points, using a weight based on the statistical distance, from the operating status data stored in the database. A method for determining the state.

12. A computer program that causes a computer capable of accessing a database storing operating status data indicating the operating status of an industrial machine, control parameters for controlling the industrial machine in that operating status, and status values ​​indicating the state of the industrial machine, to perform a process to determine the state of the industrial machine, Obtain operating data obtained by observing the operating conditions of the aforementioned industrial machine, Based on the acquired operating status data and the operating status data and status values ​​stored in the database, the state of the industrial machine is determined. The control parameters corresponding to the acquired operating status data are read from the database, and the industrial machine is controlled based on the read control parameters. A computer program that causes the aforementioned computer to perform a process.

13. A computer program for causing a computer that can access a database storing in association with operating status data indicating the operating status of an industrial machine, control parameters for controlling the industrial machine in the operating status, and status values ​​indicating the state of the industrial machine, to perform a process to determine the state of the industrial machine, Obtain operating data obtained by observing the operating conditions of the aforementioned industrial machine, If the acquired driving status data is near the driving status data with the maximum status value, or near the driving status data with the minimum status value, the state value stored in the database is logit-converted. Using the logit-converted state values, the state values ​​corresponding to the acquired operating status data are calculated. The state of the industrial machine is determined by performing an inverse logit transformation on the calculated state value. A computer program that causes the aforementioned computer to perform a process.

14. A computer program for causing a computer that can access a database storing in association with operating status data indicating the operating status of an industrial machine, control parameters for controlling the industrial machine in the operating status, and status values ​​indicating the state of the industrial machine, to execute a process to determine the state of the industrial machine, Obtain operating data obtained by observing the operating conditions of the aforementioned industrial machine, The state value of the industrial machine is calculated by weighting the state values ​​associated with a predetermined number of operating status data points that are statistically close to the acquired operating status data points, using a weight based on the statistical distance, from the operating status data stored in the database. A computer program that causes the aforementioned computer to perform a process.

Images