Operation support device, operation management system, and operation support method

Abstract

This enables the operation of controlled objects while ensuring safe driving. [Solution] The operation support device 100 includes a control command value calculation unit 112 that calculates a control command value for the controlled object 820 based on a target value of the state quantity that is the target of operation of the controlled object 820. The operation support device 100 includes a determination unit 114 that determines whether the control command value deviates from the operating conditions 121 of the controlled object. If the determination unit 114 determines that there is a deviation, the operation support device 100 includes a correction unit 115 that calculates a corrected target value by correcting the target value so that the operating conditions 121 are satisfied. The operation support device 100 includes a display control unit 116 that outputs an operation management screen including the target value and the corrected target value.

Description

[Technical Field]

[0001] The present invention relates to an operation support device, an operation management system, and an operation support method for assisting in the operation of a controlled object. [Background technology]

[0002] When an operator is running or operating a system, recommended actions may be provided as guidance. However, if people become accustomed to following guidance, they may lose the opportunity to develop the ability to make their own judgments.

[0003] Patent Document 1 describes a technology for reviewing operations performed by an operator. This publication states that "the advice generation unit generates advice regarding driving operations using the relationship between the actual operation information acquired by the actual operation information acquisition unit and the ideal operation information calculated by the ideal operation information calculation unit. Specifically, if the discrepancy between the actual operation information and the ideal operation information remains relatively large for a long period, the advice generation unit generates advice regarding driving operations that will reduce that discrepancy." [Prior art documents] [Patent Documents]

[0004] [Patent Document 1] Japanese Patent Publication No. 2017-220171 [Overview of the project] [Problems that the invention aims to solve]

[0005] The safety driving support system described in Patent Document 1 compares the actual operator's actions with ideal actions, making it difficult to use when ideal actions cannot be defined. For example, in situations where it is difficult to determine the driver's intentions, the safety driving support system may present incorrect ideal actions. Even when applied to controlled objects such as plants, it is difficult to apply in new plants where the ideal state and operation of the controlled object are unknown.

[0006] To enable operators to acquire the ability to make their own judgments and improve their proficiency, it is desirable to allow operators to perform operations that reflect their intentions within a safe range that does not lead to accidents or dangerous situations, rather than providing guidance from the outset. In a situation where safety is guaranteed, operators can acquire judgment abilities and improve their proficiency by trial and error in operations based on their own judgment. Furthermore, this approach can be applied to controlled objects whose ideal state or operation is unknown. The present invention has been made in view of the above, and aims to provide an operation support device, an operation management system, and an operation support method that enable operation while ensuring safe driving. [Means for solving the problem]

[0007] To solve the above-mentioned problems, the operation support device according to the present invention comprises: a control command value calculation unit that calculates a control command value for the controlled object based on a target value of a state quantity that is the target of operation of the controlled object; a determination unit that determines whether the control command value deviates from the operating conditions of the controlled object; a correction unit that, if the determination unit determines that it deviates, calculates a corrected target value by correcting the target value so that it satisfies the operating conditions; and a display control unit that outputs an operation management screen including the target value and the corrected target value.

[0008] The operation support device also includes a control command value calculation unit that calculates a control command value for the controlled object based on a target value of a state quantity that is the target of operation of the controlled object; a state quantity calculation unit that calculates a state quantity of the controlled object based on the control command value; a determination unit that determines whether the control command value deviates from the operating conditions of the controlled object; and a modification unit that, if the determination unit determines that it deviates, calculates a modified target value by modifying the target value so that it satisfies the operating conditions. The control command value calculation unit calculates the control command value as a first modified control command value based on the modified target value; the state quantity calculation unit calculates the state quantity of the controlled object based on the first modified control command value; the determination unit determines whether the state quantity of the controlled object deviates from the operating conditions; and if the determination unit determines that it deviates, the modification unit calculates a second modified control command value by modifying the first modified control command value so that it satisfies the operating conditions. [Effects of the Invention]

[0009] According to the present invention, it is possible to provide an operation support device, an operation management system, and an operation support method that enable operation while ensuring safe driving. Problems, configurations, and effects other than those described above will be clarified by the following description of embodiments. [Brief explanation of the drawing]

[0010] [Figure 1] This is an overall configuration diagram of the operation management system according to the first embodiment. [Figure 2] This figure illustrates the dynamic characteristics of the control system, including the operation support device and the controlled object, according to the first embodiment. [Figure 3] This figure shows the dynamic characteristics of the control system, including the operation support device and the controlled object according to the first embodiment, using an equation. [Figure 4] This diagram illustrates the process by which the correction unit according to the first embodiment corrects the target value. [Figure 5] This diagram illustrates the process by which the modification unit according to the first embodiment modifies the control command value. [Figure 6]This is a screen configuration diagram of the operation management screen according to the first embodiment. [Figure 7] This is a screen configuration diagram of the operation management screen according to the first embodiment. [Figure 8] This is a screen configuration diagram of the operation management screen according to the first embodiment. [Figure 9] This is a flowchart of the operation support process according to the first embodiment. [Figure 10] This is a flowchart of the operation support process according to the first embodiment. [Figure 11] This figure illustrates the deviation of state variables when the control period is long according to the first embodiment. [Figure 12] This figure illustrates how to avoid deviations in state variables when modifying control command values ​​at short intervals according to the first embodiment. [Figure 13] This is a functional block diagram of the operation support device according to the second embodiment. [Figure 14] This is a functional block diagram of the operation support device according to the third embodiment. [Figure 15] This is a screen configuration diagram of the operation management screen according to the third embodiment. [Figure 16] This is a functional block diagram of the operation support device according to the fourth embodiment. [Figure 17] This is a screen configuration diagram of the operation management screen according to the fourth embodiment. [Figure 18] This is a functional block diagram of the operation support device according to the fifth embodiment. [Figure 19] This is a hardware configuration diagram showing an example of a computer that implements the functions of the operation support device according to the above embodiment. [Modes for carrying out the invention]

[0011] <<Overview of Operation Support Device>> The following describes the outline of an operation support device in an embodiment for carrying out the present invention. The operation support device calculates a control command value corresponding to the target value of the operation of the controlled object instructed by the operator, and outputs it to the controlled object, provided that the operating conditions are met for a predetermined period of time. If the target value of the operation instructed by the operator does not meet the operating conditions, the operation support device modifies the target value to meet the operating conditions.

[0012] A target value is the target value for the state variable of the controlled object. Examples of target values ​​include the operating temperature and operating pressure in a chemical plant. Control command values ​​are the parameters of the control commands to the actuators equipped on the controlled object. Examples of control command values ​​include the degree of valve opening and closing and the output of a pump. Operating conditions are the conditions under which the controlled object can operate safely. Examples of operating conditions include conditions for the control command values ​​(upper and lower limits, operating range) and conditions for the state variables.

[0013] Such an operation support device ensures the safety of the controlled object even if the operator inputs (sets) an incorrect target value. As long as the controlled object is safe, the operation support device will control (operate) the controlled object according to the target value instructed by the operator. Therefore, the operator can know what state the controlled object will be in depending on the target value. Consequently, the operator can acquire the judgment ability regarding the operation of the controlled object and improve their proficiency. In addition, by outputting (displaying) both the instructed target value and the corrected target value, the operator can understand how to operate the controlled object in different states. Hereafter, the corrected target value will be referred to as the modified target value.

[0014] <Configuration of the operation support device> Figure 1 is an overall configuration diagram of the operation management system 10 according to the first embodiment. The operation management system 10 is composed of an operation support device 100 and a controlled object 820. The controlled object 820 is the object operated / controlled by the operation support device 100, and is, for example, equipment / devices installed in a chemical plant or a power plant.

[0015] The operation support device 100 is a computer and comprises a control unit 110, a storage unit 120, and an input / output unit 180. User interface devices 810 such as a display, keyboard, and mouse are connected to the input / output unit 180. The input / output unit 180 also includes a communication device and is capable of sending and receiving data with the controlled object 820. A media drive may also be connected to the input / output unit 180, enabling data exchange using a recording medium.

[0016] ≪Operation support device: storage unit≫ The memory unit 120 is composed of memory devices such as ROM (Read Only Memory), RAM (Random Access Memory), and SSD (Solid State Drive). The memory unit 120 stores operating conditions 121, control parameters 122, controlled target parameters 123, and a program 128. The program 128 includes a description of the processing of the functional units provided in the control unit 110, which will be described later. The various contents of the memory unit 120 may be stored in an external storage device such as a cloud server and read as needed.

[0017] The operating conditions 121 are the conditions under which the controlled object 820 can be operated safely. The operating conditions 121 include, for example, conditions regarding the control command value (upper and lower limits) and conditions for state variables. The control parameters 122 are parameters that the operation support device 100 refers to when controlling the controlled object 820. The control parameters 122 include parameters that are referenced, for example, when a control command value is calculated based on a target value. The controlled parameter 123 is a parameter that indicates the dynamic characteristics of the controlled object 820.

[0018] <<Operation support device: control processing>> Before describing the control unit 110, the control processing of the operation support device 100 for the controlled object 820 will be described. In the following, the dynamics of the controlled object 820 will be given by equation (1) with respect to the state variable x and the control command value u.

[0019]

number

[0020] Here, the left-hand side is the time derivative of the state variable x. In other words, equation (1) shows how the state variable x of the controlled object 820 changes when a control command value u is given. Equation (1) expresses the dynamics in a continuous-time system, but it can also be expressed in a discrete-time system (difference equation) as in equation (2).

[0021]

number

[0022] In equation (2), the subscripts k and k+1 indicate time. Note that dynamics are not expressed as equations like equations of motion or thermodynamic equations, but rather as a state variable x at the current time. k and control command value u k According to this, the state variable x at the next time step k+1 It may also be a machine learning model that calculates . Equation (2) can also represent such a machine learning model.

[0023] Figure 2 is a diagram illustrating the dynamic characteristics of a control system including the operation support device 100 and the controlled object 820 according to the first embodiment. C represents the transfer function of the operation support device 100, and P represents the transfer function of the controlled object 820. In Figure 2, a continuous-time system (see equation (1)) is assumed for simplicity of explanation. The transfer function C can be determined from the control algorithm of the control command value calculation unit 112, which will be described later. For example, if the control algorithm is proportional-integral control (PI control), its transfer function can be given by equation (3).

[0024]

number

[0025] Here, K P This is a proportional gain, and KI is the integral gain. Also, s is the Laplace transform operator. On the other hand, the transfer function P is obtained by performing the Laplace transform on Equation (1). Here, for simplicity, it is assumed that the transfer function P of the controlled object 820 is given by Equation (4).

[0026]

Number

[0027] Here, ω is the natural angular frequency and ζ is the damping coefficient, which are parameters indicating the dynamic characteristics of the controlled object 820 (see the controlled object parameter 123). Then, using Equations (5) and (6), the control command value u (see the symbol 731) and the output of the controlled object 820 (state quantity x (see the symbol 741)) can be calculated from the target value r (see the symbol 711). Note that the target value r is the value targeted for the state quantity x of the controlled object 820 input / set by the operator, who is the user of the operation support device 100.

[0028]

Number

[0029] By using these equations, for any target value r and control gains K P , K I , it is possible to calculate how the control command value u and the output of the controlled object 820 (state quantity x) change. When the upper limit value u max and the lower limit value u min of the control command value are given under the operating condition 121, it is possible to predict whether the control command value u calculated using Equation (5) conflicts with the upper limit value u max or the lower limit value u min . Similarly, when the upper limit value x max and the lower limit value x min of the output (state quantity x) are given, Equation (6) can be used. ​​​​​​Changing these parameters alters the rate of change, maximum value, and steady-state value of the control command, which may cause the control to deviate from the operating range (operating conditions 121). In this invention, parameters that can change the dynamic characteristics of the control loop in this way are called dynamic characteristic parameters (control parameters 122).

[0031] The same processing can be used when dealing with more complex controlled objects or control algorithms. Equation (7) is considered as a general representation of a complex control algorithm that includes nonlinear elements. Figure 3 shows the dynamic characteristics of a control system including the operation support device 100 and the controlled object 820 according to the first embodiment, using equation (7).

[0032]

number

[0033] p in equation (7) k This is the control parameter 122 at the time k input / set by the operator. Control parameter p k For example, the proportional gain K in proportional-integral control. P or integral gain K I This is a vector representing the combined values ​​of the control parameter p. k and target value r k This is determined by the operator's input / settings.

[0034] By performing a recursive calculation using equations (2) and (7), the control command value u at any time k can be obtained. k (See symbol 731) and state variable x k+1 (See symbol 741) can be calculated. Therefore, it is easy to determine whether these deviate from the operating range (see operating conditions 121). Note that the state quantity x k+1 This is the control command value u at time k+1. k+1 This is referenced when calculating (see equation (7)). For the sake of simplicity, unless otherwise specified, the discrete-time dynamics of equation (2) will be used below.

[0035] As explained above, in order to determine a deviation from the operating conditions 121 using the state variable x, the dynamics of the controlled object 820 must include the state variable x related to the operating conditions 121. For example, if the operating conditions 121 relate to temperature, then the state variable x must include temperature.

[0036] ≪Operation support device: Control unit≫ Returning to Figure 1, the configuration of the control unit 110 will be explained. The control unit 110 includes a CPU (Central Processing Unit) and comprises a reception unit 111, a control command value calculation unit 112, a state quantity calculation unit 113, a determination unit 114, a correction unit 115, and a display control unit 116. The control unit 110 may also include an FPGA (Field Programmable Gate Array), an ASIC (Application Specific Integrated Circuit), etc.

[0037] ≪Control Unit: Reception Department≫ The reception unit 111 receives the target value r input / set by the operator, who is a user of the operation support device 100. k (See reference numeral 711 in Figure 2) and the control parameter 122, p k (See reference numeral 721 in Figure 3)

[0038] <Control Unit: Control Command Value Calculation Unit> The control command value calculation unit 112 calculates the state variable x k , target value r k , control parameter p k Based on the control command value u k The value is calculated (see equation (7)) and output to the controlled object 820. For this reason, in Figure 3, the operation support device 100 may be read as the control command value calculation unit 112. Note that the state quantity x k These are the measured values ​​of the state variables of the controlled object 820 at time k, or the predicted values ​​of the state variables calculated by the state variable calculation unit 113, which will be described later.

[0039] ≪Control Unit: State Quantity Calculation Unit≫ The state variable calculation unit 113 calculates the state variable x k , control command value uk Based on this, the state variable x at the next time step k+1 k+1 The state variable calculation unit 113 calculates (predicts) the state variable x measured at the current time k. k and control command value u k Based on this, the state variable x at time k+1 is used as the predicted value. k+1 This calculates the value. Note that in Figure 3, the controlled object 820 may be read as the state variable calculation unit 113.

[0040] ≪Control Unit: Determination Unit≫ The determination unit 114 uses equations (5) and (6) to determine whether the control command value u and state variable x calculated from the target value r deviate from the operating conditions 121. The determination unit 114 uses equations (2) and (7) to determine whether the control command value u and state variable x calculated from the target value r k The control command value u calculated from k or state variable x k+1 The system may also determine whether or not the operating conditions 121 are being deviated from. The determination unit 114 may also determine whether or not the deviation occurs within a predetermined time period (from time k to k+N).

[0041] As explained above, the operation support device 100 controls the state quantity x of the controlled object 820. k Target value r k Based on this, the control command value u for the controlled object 820 is determined. k It includes a control command value calculation unit 112 that calculates the value. The operation support device 100 controls the control command value u k Based on this, the state variable x of the controlled object 820 k+1 It includes a state quantity calculation unit 113 that calculates the following:

[0042] ≪Control Unit: Correction Unit (Correction of Target Value)≫ The correction unit 115 corrects the target value r when the control command value u or state variable x deviates from the operating conditions 121. The correction unit 115 does not correct the target value r when the control command value u or state variable x does not deviate from the operating conditions 121.

[0043] Figure 4 shows that the modification unit 115 according to the first embodiment has set the target value r tThis is a diagram illustrating the process of correcting the target value r. t The control command value u calculated from t or state variable x t+1 If the operating conditions 121 are deviated from, the correction unit 115 adjusts the target value r t Correct the target value r t fixed Let's assume the target value is r. t It is assumed that the time up to N points in the future (t=k,...,k+N) is known. The modification unit 115 adjusts the control command value u at the time up to N points in the future. k ,…,u k+N or state variable x k+1 ,…,x k+N+1 To prevent deviation, the corrected target value r k fixed ,…,r k+N fixed The correction unit 115 calculates the correction target value r using the following equations (8) to (13). t fixed Calculate (t=k,…,k+N).

[0044]

number

[0045] The modification unit 115 solves the constraint-conditional optimization problem shown in equations (8) to (13) to obtain the modified target value r t fixed Equation (8) calculates the evaluation function J1(r) of equation (9) with equations (10) to (13) as constraints. fixed By minimizing the corrected target value r t fixed The following is calculated: Equation (9) is the original target value r t and the corrected target value r t fixed This is an evaluation function that evaluates the sum of the deviations. In other words, the correction unit 115 evaluates the original target value r t The amount of change from is as small as possible, and the modified target value r satisfies the operating conditions 121 related to the control command value u and state variable x. t fixed Calculate.

[0046] Equation (10) corresponds to the control algorithm (see Equation (7)), and Equation (11) corresponds to the dynamic characteristics of the control object 820 (see Equation (2)). Equation (12) is the operating condition 121 related to the control command value u, and Equation (13) is the operating condition 121 related to the state quantity x. By solving the optimization problem incorporating Equations (12) and (13) as constraint conditions, the control command value u and the state quantity x can be obtained as the corrected target value r that does not deviate from the operating condition 121. t fixed If the lower limit value or the upper limit value is not set, the constraint can be invalidated by using -∞ / ∞.

[0047] As described above, the operation support device 100 includes a determination unit 114 that determines whether the control command value u k deviates from the operating condition 121 of the control object 820. When the determination unit 114 of the operation support device 100 determines that there is a deviation, the operation support device 100 includes a correction unit 115 that calculates the corrected target value r t so as to satisfy the operating condition 121. t fixed

[0048] ≪Control unit: Correction unit (correction of control command value)≫ When the state quantity x deviates, the correction unit 115 may correct the control command value u t instead of the target value r. When the state quantity x does not deviate, the correction unit 115 does not correct the control command value u t . FIG. 5 is a diagram for explaining the process in which the correction unit 115 according to the first embodiment corrects the control command value u t . When the control command value u t does not deviate from the operating condition 121 and the state quantity x t deviates from the operating condition 121, the correction unit 115 corrects the control command value u t+1 to obtain the corrected control command value u t t fixed . The correction unit 115 adjusts the corrected control command value u t so that the control command value u t+1 and the state quantity x (t = k,..., k + N) do not deviate. t fixedCalculate (t = k, …, k + N). The correction unit 115 calculates the correction control command value u using the following formulas (14) to (18). t fixed to calculate.

[0049] [Number]

[0050] The correction unit 115 calculates the correction control command value u by solving the optimization problem with constraints shown in formulas (14) to (18). t fixed to calculate. Formula (14) minimizes the evaluation function J2(u fixed ) of formula (15) with formulas (16) to (18) as constraints to calculate the correction control command value u t fixed . Formula (15) is an evaluation function that evaluates the total deviation between the target value r t and the state quantity x t . In other words, the correction unit 115 calculates the correction control command value u t t such that the deviation between the target value r t and the state quantity x fixed is minimized and the driving condition 121 related to the control command value u and the state quantity x is satisfied. This is consistent with the concept of general control algorithms.

[0051] Formula (16) corresponds to the dynamic characteristics of the controlled object 820 (see formula (2)). Formula (17) is the driving condition 121 related to the control command value u, and formula (18) is the driving condition 121 related to the state quantity x. By solving the optimization problem incorporating formulas (17) and (18) as constraints, the correction control command value u t fixed can be calculated such that the control command value u and the state quantity x do not deviate from the driving condition 121.

[0052] As described so far, the correction of the target value r t and the correction of the control command value u t are almost the same process. However, in the correction of the control command value u t [[ID=四十九]] t ​​​​​​​​​​​​Since is the optimization variable, the equation corresponding to equation (10) is the control command value u t It is not included in equations (14) to (18) that show the correction.

[0053] The target value r t or control command value u t The correction method does not necessarily have to be the one described above; other methods may be used. For example, a reference governor (see reference (1) below) may be used to correct the target value, or a safety filter (see reference (2) below) may be used to correct the control command value.

[0054] References (1) A. Bemporad, "Reference governor for constrained nonlinear systems," IEEE Transactions on Automatic Control, vol.43, Issue 3, pp. 415-419, 1998. Reference (2) Ames, Aaron D., et al., "Control barrier functions: Theory and applications," 2019 18th European control conference (ECC), IEEE, 2019.

[0055] Up to this point, the target value r t Instead, control command value u t The revised target value r was explained. t fixed Based on the control command value u t This may be modified. More specifically, the control command value calculation unit 112 calculates the modified target value r t fixed Based on the control command value u t The (first modified control command value) is calculated. The state quantity calculation unit 113 calculates the state quantity based on the first modified control command value. If the determination unit 114 determines that the state quantity deviates from the operating conditions 121, the modification unit 115 adjusts the control command value u t fixed Calculate the (second modified control command value).

[0056] As explained above, the control command value calculation unit 112 calculates the corrected target value r t fixed Based on this, the control command value is recalculated as the first modified control command value. The state quantity calculation unit 113 calculates the state quantities of the controlled object 820 based on the first modified control command value. The determination unit 114 determines whether the state variables of the controlled object 820 deviate from the operating conditions 121. If the determination unit 114 determines that there is a deviation, the correction unit 115 corrects the first correction control command value to satisfy the operating conditions 121 and generates a second correction control command value u t fixed Calculate.

[0057] ≪Control Unit: Display Control Unit≫ The display control unit 116 outputs the operation management screen 610 (see Figure 6 below) to the user interface device 810 (see Figure 1), which is a display. Figure 6 is a screen configuration diagram of the operation management screen 610 according to the first embodiment. Area 611 of the operation management screen 610 displays a graph of the target value r. The horizontal axis of this graph is time, and the vertical axis is the target value. The dotted line is the target value input / set by the operator, and the solid line is the corrected target value corrected by the operation support device 100 (correction unit 115).

[0058] Area 612 displays a graph of the control command value u. The horizontal axis of this graph represents time, and the vertical axis represents the control command value. The dotted line represents the control command value based on the target value entered / set by the operator, and the solid line represents the control command value based on the modified target value. The dashed line indicates the upper limit of the control command value.

[0059] The target value entered / set by the operator is a step input that rises rapidly. Therefore, the control command value calculated by the control command value calculation unit 112 based on the target value also rises sharply, exceeding the upper limit and deviating from the operating conditions 121. To avoid this situation, the correction unit 115 corrects the target value r so that the target value r changes smoothly. fixed Calculate the corrected target value r. fixedFollowing this, as shown in the solid line graph, the control command value will not deviate from the upper limit and will satisfy operating condition 121.

[0060] In this way, the operation support device 100 displays on the operation management screen 610 what kind of situation occurs and what kind of corrections are made based on the target value instructed (input / set) by the operator. The operator can get hints on what operations to perform next time. Furthermore, inexperienced operators may not be able to determine what improvements should be made even after viewing the operation management screen 610. For such operators, the operation support device 100 may output an operation management screen 620 (see Figure 7 below) that also displays advice.

[0061] Figure 7 is a screen configuration diagram of the operation management screen 620 according to the first embodiment. The graphs of target values ​​and control command values ​​displayed in areas 622 and 624 are the same as those of the operation management screen 610 (see Figure 6). The display control unit 116 may display "Please reduce the range of change in the target value" (see advice 621) to encourage the operator to bring the target value closer to the modified target value.

[0062] Furthermore, the display control unit 116 controls the control gain (K P , K I To resolve a situation where the control gain is too high, the message "Control gain is too high" (see Advice 623) may be displayed. Lowering the control gain will prevent the control command from exceeding the upper limit. Displaying Advice 623 regarding the adjustment of control parameter 122 in this way will support not only the operator but also the technician who adjusts the control parameter.

[0063] Up to this point, we have shown the operation management screens 610 and 620 when the control command value u deviates from the operating conditions 121. If the state variable x of the controlled object 820 deviates from the operating conditions 121, the state variable x may also be displayed.

[0064] Figure 8 is a screen configuration diagram of the operation management screen 630 according to the first embodiment. The graphs of target values ​​and control command values ​​displayed in areas 631 and 634 are the same as those of the operation management screen 610 (see Figure 6). The display control unit 116 may also display graphs of the state quantity x in the case where there is no correction to the target value and in the case where there is a correction, when it is expected that the state quantity x will deviate from the operating conditions 121.

[0065] Area 633 displays a graph showing the predicted value of the state quantity if the target value is not modified, and the measured value of the state quantity after modification. The horizontal axis of this graph represents time, and the vertical axis represents the state quantity. The dotted line represents the predicted value of the state quantity based on the target value entered / set by the operator, and the solid line represents the measured value of the state quantity after the target value has been modified. The dashed line indicates the upper limit of the state quantity. Warning 632 indicates that the state quantity has reached the upper limit of operating condition 121.

[0066] In this way, the operation support device 100 displays on the operation management screen 630 what kind of situation occurs based on the target value given by the operator, and what the situation is after the correction. The operator can get hints on what operations to perform in the future.

[0067] Note that the operation management screens 610 and 620 do not include the actual state variables of the controlled device 820, so all graphs can be displayed once the operator has entered / set the parameters and the operation support device 100 has completed processing. On the other hand, the operation management screen 630 includes the measured value of the state variable x, so it can only display waveforms up to the current time. The operation management screen 630 shown in Figure 8 is an example of a screen that is displayed after the device has actually been operated. If there is no deviation from the operating conditions 121 for the control command value or state variable, the correction unit 115 does not correct the target value or control command value, and therefore the corrected target value and the control command value corresponding to the corrected target value are not displayed.

[0068] As described above, the operation support device 100 includes a display control unit 116 that outputs an operation management screen 610 including the target value and the modified target value. The operation management screen 620 displays the details of the operation to change the target value to a modified target value (see advice 621).

[0069] <<Operation support processing>> Figures 9 and 10 are flowcharts of the operation support process according to the first embodiment. The operation support process is a process that is repeated with each control cycle. Referring to Figures 9 and 10, the operator sets a target value r from time k to k+N. k ,…,r k+N The processing of the operation support device 100 after input / setting will be explained.

[0070] In step S11, the determination unit 114 starts the process of repeating steps S12 to S13 sequentially from time t k to k+N. In step S12, the control command value calculation unit 112 calculates the state variable x t , target value r t , control parameter p t Based on the control command value u t Calculate (see formula (7)). In step S13, the state quantity calculation unit 113 calculates the state quantity x t , control command value u t Based on this, the state variable x at the next time t+1 t+1 Calculate (see formula (2)).

[0071] In step S14, the determination unit 114 determines the control command value u k ,…,u k+N If the operating conditions 121 are satisfied and not deviated (step S14 → NO), the process proceeds to step S19 (see Figure 10). The determination unit 114 determines the control command value u k ,…,u k+N If the driving conditions 121 are not met and the condition deviates (step S14 → YES), the process proceeds to step S15. In step S15, the correction unit 115 sets the corrected target value r k fixed ,…,r k+N fixed Calculate (see formulas (8) to (13)). Steps S16 to S18 are the same as steps S11 to S13.

[0072] Moving on to Figure 10, we will continue the explanation of the operation support process. In step S19, the determination unit 114 determines the state quantity x k+1 ,…,x k+N+1 If the operating conditions 121 are satisfied and not deviated (step S19 → NO), the process proceeds to step S21. The determination unit 114 determines the control command value x k+1 ,…,x k+N+1 If the driving conditions 121 are not met and the condition deviates (step S19 → YES), the process proceeds to step S20. In step S20, the modification unit 115 modifies the control command value u k fixed ,…,u k+N fixed Calculate (see formulas (14) to (18)).

[0073] In step S21, the control command value calculation unit 112 outputs the control command value to the controlled object 820. In step S20, the control command value calculation unit 112 corrects the control command value u k fixed If the corrected control command value u is calculated, k fixed The control command value calculation unit 112 outputs the control command value u in step S17. k If this has been calculated, then this control command value u k The control command value calculation unit 112 outputs the control command value u calculated in step S12 if steps S17 and S20 were skipped. k Outputs. In step S22, the display control unit 116 outputs the operation management screens 610, 620, and 630.

[0074] Features of the Operation Support Device The operation support device 100 determines, based on the target value input / set by the operator, whether the control command value and state variable will deviate from the operating conditions 121 up to a predetermined time in advance (t=k,...,k+N). If a deviation occurs, the operation support device 100 corrects the target value and control command value (see step S15 in Figure 9 and step S20 in Figure 10).

[0075] With such an operation support device 100, even if the operator makes an incorrect operation, it is guaranteed that the state variables and control command values ​​of the controlled object 820 will not deviate from the operating conditions 121. In other words, the controlled object 820 can be operated safely within a range that does not cause accidents or dangerous situations. For this reason, the operator can try various operations through trial and error according to their own judgment. By repeating such trial and error, the operator will be able to understand the characteristics of the operation support device 100 and the characteristics of the operation of the controlled object 820. In turn, it is expected that the operator will acquire judgment skills and improve their proficiency by trying operations through trial and error based on their own judgment.

[0076] Furthermore, if the operator performs an inappropriate operation, intervention by the operation support device 100 will be displayed on the operation management screens 610, 620, and 630. This allows the operator to efficiently learn which operations should not be performed.

[0077] Furthermore, as the operator repeatedly performs various operations through trial and error, there is a possibility that they may discover appropriate operating methods that even experienced operators had not noticed in the past. Another advantage of the present invention is that this kind of search for optimal operation cannot be achieved by operators who follow guidance functions that reproduce the operations of experienced operators, such as those described in Patent Document 1.

[0078] Some existing control systems provide operators with recommended actions as guidance, and the operators operate the control system according to the given guidance. Such systems can only be built for control systems where the know-how of skilled operators can be utilized. In contrast, the operation support device 100 can be applied even when the ideal state or operation of a new control target 820 is unknown, by setting the operating conditions 121.

[0079] ≪Variation: Prediction Period≫ In the embodiment described above, deviations are determined by calculating control command values ​​and state variables for time up to N steps ahead. The longer the prediction time for deviation determination, the earlier the deviation can be determined, allowing the correction unit 115 to make smoother corrections. On the other hand, making predictions over a long period requires many calculations. However, if the sole purpose is to avoid deviations from the operating conditions 121, a prediction period of one control cycle step is sufficient.

[0080] ≪Modified Version: Control Command Value Calculation Unit≫ The control command value calculation unit 112 calculates the control command value u by referring to the target value r and the control parameter p. The control algorithm implemented in the control command value calculation unit 112 can be anything. For example, as mentioned above, it may be proportional-integral control or Model Predictive Control (MPC). Since MPC is a control algorithm that can handle constraint conditions, it can calculate a control command value that does not deviate from the operating conditions 121 regardless of the target value r given. For this reason, the correction unit 115 for correcting the target value is not essential. When using MPC, the control command value u can be calculated using equations (19) to (23). Here, Q and R in equation (20) are weight parameters and are treated as control parameters 122.

[0081]

number

[0082] Because the MPC has a high computational load, the calculation period for control command values ​​(control period) is often extended. In such cases, as shown in Figure 11 below, there is a possibility of deviating from the operating conditions 121 between control periods. Figure 11 is a diagram illustrating the deviation of state variables when the control period is long according to the first embodiment. The upper part of Figure 11 is a graph of state variables, and the lower part is a graph of control command values. The control period, which is the period for calculating the control command values, is Δt.

[0083] At time t3, the state variable is x A Assume that a control command is given to keep the state variable at x until time t4. If no particular changes occur in this situation, the state variable will remain x A It is maintained at the upper limit x. However, in reality, at time ta, due to the influence of disturbances, the state variable is affected by the upper limit x. max It can sometimes exceed this limit. Since the control command value maintains the control command value at time t3 during the control period Δt, the state variable at time tb will have an upper limit x max This will exceed the limit.

[0084] The correction unit 115 can resolve this situation. The calculation period of the correction unit 115 is set to a period δt that is shorter than the control period Δt, and the correction unit 115 corrects the control command value u with a period δt. For example, let δt = tb - ta.

[0085] Figure 12 is a diagram illustrating the avoidance of state variable deviation when the control command value is modified with a short period δt according to the first embodiment. At time tb, the modification unit 115 modifies the control command value, thereby preventing the state variable from exceeding the upper limit x max This prevents deviations beyond a certain limit. Since the safety filter (see reference (2)) has a lower computational load than the MPC, the period δt can be made significantly shorter compared to the control period Δt of the MPC.

[0086] ≪Variation: Operation Management Screen≫ The state variables displayed in area 633 of the operation management screen 630 (see Figure 8) are the measured values ​​of the state variables of the controlled object 820. Alternatively, the predicted values ​​of the state variables calculated by the state variable calculation unit 113 may be displayed instead of the measured values. By referring to the predicted values, the operator can understand how the future state variables will change based on the current target values, and thus accumulate operational know-how.

[0087] ≪Second Embodiment≫ In the first embodiment, the operating conditions 121, including the control command value and the upper and lower limits of the state variables, were fixed. However, depending on the controlled object 820, the upper and lower limits of other state variables may change depending on a certain state variable. For example, in a plant that controls thermal fluid, the limit value of the allowable flow rate may be changed depending on the temperature of the fluid. The operation support device 100A of the second embodiment (see Figure 13 below) can also handle cases where the operating conditions 121 change according to the state variable x.

[0088] Figure 13 is a functional block diagram of the operation support device 100A according to the second embodiment. Compared to the operation support device 100 according to the first embodiment (see Figure 1), the determination unit 114A and the operating conditions 121A are different. The operating conditions 121A do not indicate upper or lower limits for fixed control command values ​​or state variables, but rather upper and lower limits for a certain state variable. For example, the operating conditions 121A include a table that shows the upper limit of the flow rate for each fluid temperature. Such a table can be considered as equations (24) and (25).

[0089]

number

[0090] The determination unit 114A determines the control command value u calculated using equation (2) or equation (7). t or state variable x t+1 Whether operating condition 121A is satisfied is determined by whether (t=k+1,…,k+N) satisfies equations (24) and (25).

[0091] As explained above, the operating condition 121A includes a condition for a second state variable that changes according to the first state variable (see equation (25)), or a condition for a control command value that changes according to the first state variable (see equation (24)).

[0092] ≪Third Embodiment≫ As explained above, the operation support device 100 allows the operator to safely operate the controlled object 820 while allowing for trial-and-error operation. On the other hand, if the degree of freedom of operation is too high, the operator may have difficulty deciding what to try. The operation support device 100B of the third embodiment (see Figure 14 below) supports the operator's decision-making by notifying the operator of the operating status of the controlled object 820.

[0093] Figure 14 is a functional block diagram of the operation support device 100B according to the third embodiment. Compared to the operation support device 100 according to the first embodiment (see Figure 1), the control unit 110 is equipped with an operation evaluation unit 117, and the display control unit 116B is different. The operation evaluation unit 117 calculates key performance indicators (KPIs) related to the operation of the controlled object 820 based on the corrected target value, control command value, corrected control command value, and measured values ​​of the state quantities of the controlled object 820. Examples of key performance indicators (KPIs) include energy consumption and CO2 emissions. The display control unit 116B outputs the operation management screen 640 (see Figure 15, described later).

[0094] Figure 15 is a screen configuration diagram of the operation management screen 640 according to the third embodiment. The graphs of target values, state variables, and control command values ​​displayed in areas 641, 642, and 644 are the same as those of the operation management screen 630 (see Figure 8). In addition to target values, state variables, and control command values, the display control unit 116B displays KPIs 643 and 645 calculated by the operation evaluation unit 117.

[0095] By referring to KPIs 643 and 645, operators can consider what actions they need to take to increase or decrease the KPIs. Providing such additional information is expected to allow operators to gain a deeper understanding of the characteristics of the controlled system 820.

[0096] As described above, the operation support device 100B includes an operation evaluation unit 117 that calculates an evaluation index (KPI) based on at least one of the control command value, the second modified control command value, and the state variable. The operation management screen 640 includes evaluation indicators (see KPIs 643 and 645).

[0097] ≪Fourth Embodiment≫ As explained above, the operation support device 100 allows the operator to safely operate the controlled object 820 while allowing for trial-and-error operation. On the other hand, depending on the controlled object 820, even a slight change in the operating state can cause a large change in throughput. In the case of such controlled objects, it is important to minimize trial and error by the operator and quickly approach the desired operating state. The operation support device 100C according to the fourth embodiment (see Figure 16 below) supports the operation of such controlled objects 820.

[0098] Figure 16 is a functional block diagram of the operation support device 100C according to the fourth embodiment. Compared to the operation support device 100 according to the first embodiment (see Figure 1), the control unit 110 is equipped with an operation intent estimation unit 118, and the modification unit 115C and display control unit 116C are different. The operation intent estimation unit 118 estimates the operator's operation intent based on the operator's input / settings and their history. For estimating the operator's operation intent, Preference Learning, as shown in reference (3), can be used, for example. An example of operation intent is the improvement of KPIs (e.g., energy consumption).

[0099] Reference (3) W. Chu, and Z. Ghahramani, "Preference learning with Gaussian processes," Proceedings of the 22nd international conference on Machine learning, pp.137-144, 2005.

[0100] The display control unit 116C outputs the operation management screen 650 (see Figure 17 below). Figure 17 is a screen configuration diagram of the operation management screen 650 according to the fourth embodiment. The graphs of target values, state variables, and control command values ​​displayed in regions 653, 654, and 655 are the same as those of the operation management screen 630 (see Figure 8). In addition to the target values, state variables, and control command values, the display control unit 116C displays a confirmation message 651 based on the operator's operation intent calculated by the operation intent estimation unit 118. If the operator's own intention matches the operation intent calculated by the operation intent estimation unit 118, the operator presses the "YES" button 652. Then the correction unit 115C solves the constraint-bound optimization problem shown in equations (26) to (31) to correct the target value r fixed Calculate.

[0101]

number

[0102] The evaluation function in equation (27) is a function that includes either a state variable, a control command value, or a target value, and its form changes depending on the estimated operational intent. For example, if the goal is to reduce energy consumption, the evaluation function evaluates the cumulative value of energy consumption; if the goal is to reduce CO2 emissions, the evaluation function evaluates the cumulative value of CO2 emissions. In other words, the modification unit 115C calculates a modified target value that maximizes / minimizes the intended evaluation indicator (e.g., energy consumption).

[0103] As described above, the operation support device 100C includes an operation intent estimation unit 118 that estimates the intent of the instructed operation. The modification unit 115C modifies the target value to maximize or minimize an evaluation index calculated based on at least one of the intended control command value and state variable.

[0104] ≪Fifth Embodiment≫ The operation support device 100 according to the first embodiment provides advice 623 (see Figure 7) such as "The control gain is too high." Experienced operators can determine how much to lower the control gain (control parameter 122), but novice operators may not be able to determine what control gain setting is appropriate. For this reason, it is desirable to have a function that assists in setting an appropriate control gain. The operation support device 100D according to the fifth embodiment (see Figure 18 described later) assists in setting such control parameter 122.

[0105] Figure 18 is a functional block diagram of the operation support device 100D according to the fifth embodiment. Compared to the operation support device 100 according to the first embodiment (see Figure 1), the modification unit 115D of the control unit 110 is different. The modification unit 115D, like the first embodiment, modifies the target value r fixed In addition to calculating the state variable x of the controlled object 820, the corrected target value r fixed Modify control parameter 122 to approach the desired result.

[0106] The modification unit 115D modifies the control parameters 122 using a method such as Bayesian optimization. Bayesian optimization requires the selection of parameter candidates and their evaluation. The control command value calculation unit 112 calculates the control command value u by referring to the parameter candidates selected by the modification unit 115D. The state variable calculation unit 113 calculates the state variable x using the control command value u. The modification unit 115D then calculates the state variable x and the modified target value r fixed The difference is evaluated, and if the difference is sufficiently small, the parameter candidate is used to adjust the state quantity to the corrected target value r fixed It can be determined that the operation can be performed in this manner. Furthermore, by displaying the corrected control parameter 122 value on the operation management screen, the operator can understand what control gain to set.

[0107] As explained above, when the control command value calculation unit 112 calculates the control command value, the control parameter 122 (control gain K P ,K I See (reference). The correction unit 115D sets the correction target value r fixed Then, the control parameter 122 is modified so that the difference with the state variable x of the controlled object 820 becomes small.

[0108] <<Other variations>> Although several embodiments of the present invention have been described above, these embodiments are merely illustrative and do not limit the technical scope of the present invention. In the embodiments described above, the controlled object 820 is assumed to be a plant, but it is not limited to this. The controlled object 820 may be an automobile or construction machinery, and can also be used to support the operator (driver). In the embodiments described above, the operation support devices 100, 100A, 100B, 100C, and 100D output control command values / modified control command values ​​to the controlled object 820. The operation support devices 100, 100A, 100B, 100C, and 100D do not output to the actual controlled object 820, and can also be used as simulation devices for training operators.

[0109] The determination units 114 and 114A use the control command values ​​calculated by the control command value calculation unit 112 and the state variables calculated by the state variable calculation unit 113 to determine whether the control command values ​​and state variables deviate from the operating conditions 121 and 121A. The determination units 114 and 114A may also calculate the control command values ​​and state variables themselves by performing the processing of the control command value calculation unit 112 and the state variable calculation unit 113, and then determine the deviation.

[0110] The correction units 115, 115C, and 115D calculate the corrected target value and corrected control command value using the control command value calculated by the control command value calculation unit 112 and the state quantity calculated by the state quantity calculation unit 113 (see equations (10) and (11)). The correction units 115, 115C, and 115D may also calculate and correct the control command value and state quantity by themselves performing the processing of the control command value calculation unit 112 and the state quantity calculation unit 113.

[0111] The present invention can take on various other embodiments, and furthermore, various modifications such as omissions and substitutions can be made without departing from the spirit of the invention. These embodiments and their variations are included in the scope and spirit of the invention as described herein, and are also included in the scope of the invention and its equivalents as described in the claims.

[0112] Hardware Configuration The operation support devices 100, 100A, 100B, 100C, and 100D according to the above embodiment are implemented by a computer 900 having a configuration such as that shown in Figure 19. Figure 19 is a hardware configuration diagram showing an example of a computer 900 that implements the functions of the operation support devices 100, 100A, 100B, 100C, and 100D according to the above embodiment. The computer 900 includes a CPU 901, ROM 902, RAM 903, SSD 904, and an input / output interface 905 (labeled as input / output I / F (Interface) in Figure 19). Furthermore, the computer 900 includes a communication interface 906 (labeled as communication I / F in Figure 19) and a media interface 907 (labeled as media I / F in Figure 19). The computer 900 may be equipped with an HDD (Hard Disc Drive) instead of the SSD 904, or it may be equipped with an HDD in addition to the SSD 904.

[0113] The CPU 901 operates based on programs stored in the ROM 902 or SSD 904 and is controlled by the control unit 110 in Figure 1. The ROM 902 stores boot programs executed by the CPU 901 when the computer 900 starts up, as well as programs related to the computer 900's hardware.

[0114] The CPU 901 controls input devices 910, such as a mouse and keyboard, and output devices 911, such as a display and printer, via the input / output interface 905. The CPU 901 acquires data from the input devices 910 and outputs the generated data to the output devices 911 via the input / output interface 905.

[0115] SSD904 stores programs executed by CPU901 and data used by those programs. Communication interface906 receives data from other devices (e.g., controlled object 820) not shown via the communication network and outputs it to CPU901, and also transmits data generated by CPU901 to other devices via the communication network.

[0116] The media interface 907 reads a program or data stored in the recording medium 912 and outputs it to the CPU 901 via the RAM 903. The CPU 901 loads the program from the recording medium 912 onto the RAM 903 via the media interface 907 and executes the loaded program. The recording medium 912 can be an optical recording medium such as a DVD (Digital Versatile Disk), a magneto-optical recording medium such as an MO (Magneto Optical Disk), a magnetic recording medium, a conductive memory tape medium, or a semiconductor memory.

[0117] For example, when computer 900 functions as an operation support device 100, 100A, 100B, 100C, or 100D according to the above embodiment, the CPU 901 of computer 900 realizes the functions of the operation support devices 100, 100A, 100B, 100C, or 100D by executing a program 128 (see Figure 1) loaded onto RAM 903. The CPU 901 reads the program from the recording medium 912 and executes it. Alternatively, the CPU 901 may read the program from another device via a communication network, or it may install the program 128 from the recording medium 912 onto the SSD 904 and execute it. [Explanation of Symbols]

[0118] 10. Operation Management System 100,100A,100B,100C,100D Operation support device 111 Reception Department 112 Control command value calculation unit 113 State Variable Calculation Unit 114,114A Judgment section 115,115C,115D Modification section 116, 116B, 116C Display Control Unit 117 Operation Evaluation Department 118 Operation Intent Estimation Unit 121,121A Operating Conditions 122 Control Parameters 123 Controlled Parameters 610, 620, 630 Operation Management Screen

Claims

1. A control command value calculation unit calculates a control command value for the controlled object based on the target value of the state quantity that is the target of the operation of the controlled object, A determination unit that determines whether the control command value deviates from the operating conditions of the controlled object, If the determination unit determines that there is a deviation, the correction unit calculates a corrected target value by modifying the target value so that the operating conditions are met, The system includes a display control unit that outputs an operation management screen including the target value and the modified target value. Operation support equipment.

2. A control command value calculation unit calculates a control command value for the controlled object based on the target value of the state quantity that is the target of the operation of the controlled object, A state quantity calculation unit calculates the state quantity of the controlled object based on the control command value, A determination unit that determines whether the control command value deviates from the operating conditions of the controlled object, The determination unit determines that there is a deviation, and the correction unit calculates a corrected target value obtained by modifying the target value to satisfy the operating conditions, The control command value calculation unit is: Based on the aforementioned corrected target value, the control command value is calculated as the first corrected control command value. The state quantity calculation unit, Based on the first modified control command value, the state variables of the controlled object are calculated, The determination unit, Determine whether the state quantity of the controlled object deviates from the operating conditions. The aforementioned modification section is, If the determination unit determines that a deviation has occurred, it calculates a second modified control command value obtained by modifying the first modified control command value to satisfy the operating conditions. Operation support equipment.

3. The system further includes a display control unit that outputs an operation management screen including the target value and the modified target value. The operation support device according to claim 2.

4. The aforementioned operation management screen is: The details of the operation to change the target value to the modified target value are further displayed. The operation support device according to claim 3.

5. The system further includes an operation evaluation unit that calculates an evaluation index based on at least one of the control command value, the second modified control command value, and the state variable, The aforementioned operation management screen is: Includes the aforementioned evaluation indicators The operation support device according to claim 3.

6. The control command value calculation unit is: When calculating the aforementioned control command value, the control parameters are referenced. The aforementioned modification section is, The control parameters are modified so that the difference between the target value and the state variable of the controlled object becomes small. The operation support device according to claim 2.

7. The aforementioned operating conditions are: The conditions for the second state quantity that change according to the first state quantity, or Includes conditions for the control command value that change according to the first state quantity. The operation support device according to claim 2.

8. The system further includes an operation intent estimation unit that estimates the intent of the instructed operation, The aforementioned modification section is, The target value is modified to maximize or minimize the evaluation index calculated based on at least one of the control command value and the state variable, as intended. The operation support device according to claim 2.

9. The controlled object and, An operation management system comprising the operation support device described in claim 1.

10. The controlled object and, An operation management system comprising the operation support device described in claim 2.

11. The operation support device, A step of calculating a control command value for the controlled object based on the target value of the state quantity that is the target of the operation of the controlled object, The steps include determining whether the control command value deviates from the operating conditions of the controlled object, If it is determined that the control command value deviates from the operating conditions of the controlled object, the steps include calculating a modified target value by modifying the target value to satisfy the operating conditions, The steps include: outputting an operation management screen that includes the target value and the modified target value; and executing the following steps. Operation support method.

Images