JP2007058447A - Modeling method of control target system based on extended petri net and controller based on extended petri net - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a controller based on an extended Petri net useful for a design or improvement of a physical distribution system. <P>SOLUTION: This controller 101 based on the extended Petri net comprises a computer system having input means 105, 106, a display means 104, a Petri net model design means 111, and a Petri net execution means 112, connectable with external control target equipment 108. Thereby, control of a system, monitoring of the system, and performance evaluation of the system can be performed by one Petri net model. <P>COPYRIGHT: (C)2007,JPO&INPIT

Description

  The present invention relates to a modeling method based on an extended Petri net and a control device based on an extended Petri net using the same for controlling a control target system such as a physical distribution, FA system or production line in the physical distribution field or the manufacturing field. It is about.

  Sequence control is used in warehouse logistics, FA systems, production lines, etc. in the logistics and manufacturing fields. Recently, there is an increasing need for added value that can express the state of the system in detail, such as not only control but also system operation status and progress grasp, system monitoring from a remote location, display of the system error source, etc. . In addition, by evaluating the performance of production lines and logistics systems in a simulated manner, it is possible to discover the bottleneck processes and system configuration problems that are reducing productivity, and obtain hints for improving the system. The same applies to value.

  Logistics and production systems are mainly controlled by a PLC (programmable logic controller) that performs sequence control. In PLC, sequence control processing is mainly performed using a description method such as LD (ladder) or SFC (sequential function chart). Since LD arranges operating conditions for each output, the entire control flow is difficult to see, and the program is difficult for a third party other than the creator to understand. For this reason, problems such as poor maintainability have been pointed out. SFC can improve the problem that it is difficult to see the flow of LD control. However, since both are description methods that perform sequence control processing, the system itself cannot be expressed. Therefore, it is not possible to provide the added value such as system state monitoring and system performance evaluation.

  The state of products in warehouse logistics such as the logistics field and production field, and the state that parts and materials flowing through the production line become products through several work processes are generally expressed as discrete event systems. One technique for modeling discrete event systems is Petri nets. Petri nets are tools that can be modeled with graphics that can be applied to many systems.

  Petri Net was proposed by Carl Adam Petri of Germany in 1962 and represents production systems, information systems, logistics systems, etc. characterized by parallel, asynchronous, parallel, distributed, non-deterministic and established behaviors. It is a powerful means to do.

  The Petri net definition itself is as simple as an automaton, and its understanding does not require special mathematical knowledge. In particular, a major feature of Petri nets is that they can be represented as diagrams.

  A Petri net is one of directed graphs and has an initial state called initial marking. The basic graph for creating a Petri net is a weighted directed bipartite graph, and consists of two types of nodes: places and transitions. An arc is either connected from a place to a transition or connected from a transition to a place.

  In modeling using the concept of conditions and events, places represent conditions and transitions represent events. A transition has several input places and output places as preconditions and postconditions for each event.

  The Petri net diagram will be described below with reference to FIG. 2 and the like related to the present invention. The input place 201 and the input place 204, which are prerequisites for the transition 208, and the output place 209, which is a post-condition, are connected by respective arcs. The token 202 and token 203 in the input place 201 and the token 205 in the input place 204 represent initial marking.

  The condition that the transition 208 can be ignited is that the token having a weight 211 or more of the arc 206 connected to the transition 208 from the input place 201 is in the input place 201 and the arc connected to the transition 208 from the input place 204 This is a case where tokens having a weight of 207 or more exist in the input place 204. In this case, the arc weight 211 connected to the input place 201 is 2, and there are also two tokens in the input place 201. The weight of the arc connected to the input place 204 is 1 when no numerical value is described in the arc, and the number of tokens in the input place 204 is 1. Therefore, since the precondition is satisfied, the transition 208 is ignited.

  FIG. 3 shows a state where the transition 208 is ignited. As a precondition after firing, a token having a weight of 309 arc connected from the input place 301 is removed. Similarly, the input place 302 removes tokens corresponding to the weight of the arc 304 to be connected.

  Under the postponement condition after firing, a token for the weight of the arc 306 connected to the output place 307 is marked in the output place 307. In this case, since the arc weight is 1, one token is marked in the output place 307.

  The firing delay 212 of the transition 208 in FIG. 2 will be described. Temporal Petri nets are a combination of the above Petri net theory and temporal concepts. This ignition delay 212 has a function of delaying the ignition time for five steps from a state where ignition is possible. If one step time is 1 second, the ignitable transition 208 will fire after 5 seconds. In this way, using a time Petri net makes it possible to express the behavior of the system in terms of time.

In color petri nets, it is possible to give attributes to tokens. In transition, a firing rule can be expressed by a logical expression or the like using a token attribute.
JP 2003-150887 A

  There are the following four problems in system control, monitoring, and system performance evaluation using conventional Petri net theory.

  First, in the conventional Petri net theory, it has been difficult to simply model a system that preferentially handles interrupts in any situation of the system. When there were multiple types of interrupt processing, modeling was further complicated. Complicating the Petri net model affects the effect of visual representation, which is also an advantage of Petri nets, and hinders system monitoring. For example, when the transport system of FIG. 4 is controlled by making a Petri net model, in order to cope with an interrupt-like control such as an emergency stop, the control contents are transferred to all the transitions in the program that is made a Petri net model. When an emergency stop is pressed, a precondition that an emergency stop process is performed and a post-condition must be added. Adding preconditions and post-conditions to the transition not only increases the design effort of the Petri net model, but also complicates the Petri net model. In addition, since the number of transitions increases as the scale of the system increases, the system is more affected. Furthermore, if there are a plurality of types of interrupt processing, a plurality of preconditions and postponement conditions to be added to the transition are also added, so that the influence on the increase in design man-hours and the complexity of the Petri net model is further increased.

  Secondly, when the system is modeled by the conventional Petri net theory and the Petri net model is executed, it is determined at every step whether or not all transitions in the model can be ignited, If it was possible to ignite, it was necessary to perform an ignition process. If the number of transitions increases as the system size increases, performing the firing determination for each transition for all transitions imposes a heavy load on the Petri net execution process.

  Third, when a product, part, or product is expressed by giving attributes to the token using color Petri net theory, many definitions and settings are required, such as the attribute definition of the token and the attribute definition of the arc where the token transitions. For this reason, the design and attribute setting of the Petri net model are complicated, and the design man-hours of the Petri net model have increased.

  Fourth, the Petri net model for the purpose of control could not perform all of the system monitoring and system performance evaluation, so it was necessary to redesign the Petri net model according to the application.

  The present invention has been made in view of such circumstances, and an object of the present invention is to provide a modeling method capable of performing system control, system monitoring, and system performance evaluation with a single Petri net model. And It is another object of the present invention to provide a control device based on an extended Petri net using this modeling method.

As a result of intensive studies to solve this problem and repeated trial and error, the present inventors have completed the present invention as follows.
<Modeling method of controlled system based on extended Petri net>
That is, the control target system modeling method based on the extended Petri net of the present invention is a control target system modeling method based on an extended Petri net that models a system to be controlled using an extended Petri net,
Place a place representing a possible state of the system and a hierarchical structure transition representing the processing necessary for transition from a place representing a certain state of the system to a place representing the next state in a chronological order. A system state model creation step of creating a Petri net diagram and modeling the overall state of the system, and a start indicating the start of the process, from a place on the upper Petri net diagram to the next place A subdivision required for transition between a processing state place and a plurality of processing state places showing a transition, an end transition indicating the end of the processing, a state between the start transition and the end transition. Lower-order petrines created by arranging process transitions representing processes in chronological order A system processing execution model creation step that models the processing contents of the hierarchization destination included in the hierarchical transition, and a Petri net model that can transition from normal processing to interrupt processing of the system, Reflects the result of the system interrupt processing execution model creation step that enables the interrupt processing to be executed preferentially from any state during execution of the normal processing of the system and the logical input / output conditions of the external I / O device And creating a specific Petri net diagram that blocks the input / output conditions using a shared place having a function that can be shared among a plurality of Petri net diagrams. And an input / output condition blocking model creation step for modeling to enable reuse.
(Function and effect)
According to the control target system modeling method based on the extended Petri net of the present invention, the following operational effects can be obtained.
(1) In the system state model creation process and the system process execution model creation process, the system to be controlled can be modeled in order according to the system state flow and process flow. Design is possible without knowledge or design experience.
(2) When the Petri net diagram modeled in this way is executed by the Petri net execution means, the system state can be easily grasped from the Petri net diagram created in the system state model creation process or the system processing execution model creation process. it can.
(3) The Petri net diagram created in the system interrupt processing execution model creation step can be simplified as compared with a conventional Petri net model capable of interrupt processing. Therefore, the design efficiency of modeling is improved.
(4) The Petri net diagram created in the input / output condition blocking model creation step can be reflected in one shared place. For this reason, the Petri net model can be simplified by reusing the shared place in another Petri net diagram.
<Control device based on extended Petri net>
Further, the present invention may use a control target system modeling method based on the extended Petri net for a control device.
That is, the control device based on the extended Petri net of the present invention includes an input means including a pointing device and a keyboard, a display means for displaying the created and edited contents, and a Petri net model design means for designing a Petri net model by the extended Petri net. And a Petri net execution means for executing the Petri net model, and a control device based on an extended Petri net comprising a computer system connectable to an external control target device,
The Petri net model design means includes an element arrangement editing means for creating a Petri net diagram in which places, transitions, arcs connecting the places and the transitions are arranged, and various nodes for setting the nodes arranged by the element arrangement editing means. Node setting means, sheet setting means for setting the Petri net diagram in a state according to the purpose, hierarchical structure setting means for setting a hierarchical structure in the Petri net diagram, an external device to be controlled and an external device having a specific function I / O setting means for setting an association with an input / output place, and the Petri net executing means includes a hierarchical structure executing means for controlling execution of the layered Petri net diagram, and the Petri net model. Control execution means for controlling an external control target device by executing A Petri net model recording means for recording the created Petri net model, and a Petri net for displaying the creation, editing and execution of the Petri net model on the display means. And a model display means.
(Function and effect)
According to the control device based on the extended Petri net of the present invention, the following operational effects can be obtained.
(1) The hierarchical structure executing means of the control apparatus of the present invention is a petri that performs interrupt processing on interrupt processing instructions generated during normal execution processing of a Petri net model having a hierarchical structure created by the modeling method described above. Execution processing can be arbitrarily transferred to the net diagram.
(2) The Petri net execution means executes only the necessary Petri net diagram of the hierarchization destination when the system state transitions. For this reason, the transition firing determination process can be performed only for the necessary Petri net diagrams, so the load of the execution process can be reduced.
(3) Since the Petri net execution means expresses the system behavior in real time with the system state model sheet and the control processing model sheet while controlling the controlled object, the system state can be easily grasped. .
(4) By executing the I / O setting sheet execution process and initialization function of the control execution means, the Petri net diagram in which external input / output conditions are blocked is simplified, and the execution process is performed by reducing transitions in the Petri net model. Can reduce the load.
<Others>
It is more preferable that the control target system modeling method based on the extended Petri net and the control device based on the extended Petri net of the present invention further include the following configuration.
1. That is, the present invention described above is an attribute setting means for setting an attribute that enables storage, reference, and substitution of data according to a predetermined setting for all tokens in the Petri net diagram in the Petri net diagram unit. A function setting unit that assigns a function that can set external information to the attribute value assigned to the token or that can arbitrarily set the attribute value to a place or transition, and a firing conditional expression that indicates a firing rule of the transition. A transition firing condition formula setting means that can be arbitrarily set, a transition output setting means that can arbitrarily set a change of the attribute value with respect to the attribute value given to the token that is transitioned after the transition is fired, and a function setting means Function execution means for changing the attribute value of the token according to the setting, and the transition firing condition Performs ignition determination in consideration of the ignition condition set in the setting means, it is preferable to have an attribute setting firing process means for changing the attribute value of the token after firing in accordance with the setting of the transition output setting means.
(Function and effect)
According to the present invention, the following effects can be obtained.
(1) By using the attribute setting means, it is possible to define attributes only for necessary sheets with respect to a Petri net model composed of a plurality of sheets. For this reason, when the execution process is performed by the Petri net execution means, it is possible to reduce the memory consumption related to the attribute and the execution process related to the attribute to reduce the processing load.
(2) The attribute setting means can easily give an attribute to the token simply by specifying the data length that can be stored in the token by defining several kinds of attributes in the token in advance.
(3) When the data length is designated by the attribute setting means, an address is automatically given to each attribute value. Depending on the address, the transition firing condition expression setting means, the transition output setting means, or the function setting means can be easily set such as reference to individual attribute values and substitution of calculation results.
(4) By using the attribute setting firing processing means, it is possible to control the transition firing from the token attribute value, or to change the token attribute value by setting values such as the transition and function place functions.
(5) By using all the means described above, the system can be expressed in detail, and complex system control and detailed system status monitoring can be realized.
2. Furthermore, the present invention sets performance evaluation execution means that can simulate performance evaluation that enables identification of a factor that causes a decrease in productivity of the system, and specific parameters that the performance evaluation execution means uses at the time of execution. It is more preferable to have performance evaluation setting means.
(Function and effect)
According to the present invention, the following effects can be obtained.
(1) The performance evaluation execution means can evaluate the performance of the system in a simulated manner using a Petri net diagram modeled in the extended Petri net format for the original control. Thereby, depending on the setting of the performance evaluation setting procedure, it is possible to grasp the system production efficiency, the work plan, and the malfunction of the system configuration.
(2) By using the performance evaluation setting means and the performance evaluation effective means, the Petri net diagram for control can be reused for performance evaluation. As a result, the Petri net model necessary for performance evaluation can be designed more efficiently.

Next, an embodiment is given and this invention is demonstrated in detail. The following description appropriately applies not only to the control device based on the extended Petri net of the present invention but also to the modeling method of the controlled system based on the extended Petri net. It should be noted that which embodiment is optimal depends on the control target, required performance, and the like.
FIG. 1 shows a hardware configuration of a control device based on an extended Petri net. FIG. 1 describes the devices and functions required for carrying out the present invention.

In FIG. 1, a computer main body 101 is a general-purpose computer having a central processing unit 102, a main storage device 103, and the like.
The design execution means 110 based on the extended Petri net designs a Petri net model using the data input device 105 (eg, keyboard) and the pointing input device 106 (eg, mouse), and controls the control target device 108. Performs execution processing of the Petri net model to be controlled, system state display processing, system performance evaluation processing, and the like.
The display device 104 (for example, a liquid crystal display) displays the processing contents of the design execution unit 110 based on the extended Petri net. The auxiliary storage device 107 (eg, magnetic disk) records data such as the design execution means 110 based on the extended Petri net as necessary. The database 131 can be connected to the design execution means 110 based on the extended Petri net as necessary, and exchanges data with the Petri net execution means 112.
The computer main body includes a general-purpose PCI board I / O interface 109 capable of handling external input / output signals. By being processed by the central processing unit 102, an input signal from the external control target device 108 is received, reflected as data in the design execution means 110 based on the extended Petri net recorded on the main storage device 103, and extended. The design execution means 110 based on the Petri net transmits an output signal from the I / O interface 109 to the external control target device 108 as an external output signal. The control target device 108 can be controlled by the mechanism of these devices.

The Petri net model design means 111 creates a Petri net model that can control the system and evaluate the performance of the system by performing the following several design-related means.
The sheet setting unit 113 selects the type of sheet on which the Petri net diagram is drawn, and sets the execution processing interval time of the Petri net execution unit 112. The element arrangement editing unit 114 creates a Petri net diagram by inputting “place”, “transition”, and “arc” into the selected Petri net diagram sheet. The attribute setting means 115 determines whether or not there is an attribute definition setting (hereinafter referred to as attribute definition) by an extended Petri net obtained by extending the attribute definition of the color Petri net token for each sheet of the Petri net diagram, and defines the attribute for the initial marking token. Set the value by.
The firing condition formula setting means 116 that can be used only when the attribute setting is permitted in the Petri net diagram is arbitrarily ignited by utilizing the attribute of the token for the transition separately from the petri net firing condition shown in FIG. Set the firing conditional expression that represents the rule. The transition output setting means 117, which can be used only when the attribute setting is permitted in the Petri net diagram, arbitrarily sets a value for the token that transitions due to the firing of the transition. The various node setting means 118 sets an ignition delay time, a transition name, an initial marking number of place tokens, a place name, and an arc weight.
In the hierarchical structure setting 119, the upper and lower relationships of the hierarchical structure are constructed, and the sheet name and the firing order of the transition are set. The function setting means 120 assigns and sets special functions necessary for control and function functions such as operations to transitions and places. The I / O setting unit 121 assigns an external input / output signal to the control target device 108 to a place. The performance evaluation setting 122 sets parameters used for performance evaluation in a model of the control target system in the form of an extended Petri net for control purposes.

The Petri net execution means 112 can perform system control, system monitoring, and system performance evaluation by performing the following several execution-related means. The hierarchical structure execution means 123 executes the hierarchical structure of the Petri net model designed by the Petri net model design means 111 according to the extended Petri net format. The attribute setting firing processing unit 124 determines whether or not the firing condition formula that is the firing rule set by the transition firing condition formula setting unit 116 is satisfied, and performs the firing processing if all the firing conditions are satisfied. Further, in the firing process, a process that affects the attribute value of the transitioning token is performed by performing processing such as calculation of the firing output value of the token set by the transition output setting unit 117 and substitution of the reference value. The function execution unit 125 performs an execution process of each node assigned a function such as a place to which the function setting unit 120 is assigned a function for reflecting external input / output data to a token.
The control execution processing unit 126 expands information such as input signals and input data from the control target device 108 from the I / O interface 109 to the Petri net model designed by the Petri net model design unit 111 and forms an extended Petri net. The control target device 108 is controlled by performing the execution process based on the above. The performance evaluation execution unit 127 performs special execution processing based on various parameters set for performing system performance evaluation using a Petri net model capable of system control and system monitoring, and performs processing representing system behavior. . The normal execution processing unit 128 performs an execution process on the Petri net model designed by the Petri net model design unit 111 based on the format of the extended Petri net, and performs a process representing the behavior of the Petri net model.

  The Petri net model designed by the Petri net model design unit 111 is recorded on a disk such as the auxiliary storage device 107 by the Petri net model storage unit 129. A design process in which modeling is performed by the Petri net model design unit 111 and an execution process representing the behavior of the system by the Petri net execution unit 112 are displayed on the display unit 104 by the Petri net model display unit 130.

Next, an Example is given and this invention is demonstrated more concretely.
FIG. 4 shows a part of a transport system including an automatic sorting line used in a distribution center or the like. FIG. 4 also shows contacts for the transfer system to exchange input / output signals with the computer main body 101 shown in FIG. Specifically, a contact point at which the I / O interface 109 receives an input signal from each control device of the system is represented by X and a number. In addition, the I / O interface 109 represents a contact for outputting an output signal to each control device of the system by Y and a number. With these configurations, the transport system is controlled and the transport system is monitored, and the performance of the transport system is evaluated.

  First, the transport system shown in FIG. 4 will be described. This transport system is a part of a system that automatically transports goods carried by a truck or the like to an area such as a storage rack or a flat place in a distribution center. The incoming goods are input to the incoming line 401 in a box (bucket) in which a plurality of parts can be put. In the bucket, up to five types of different products can be collected as long as they are transported to the same storage rack area. Information on the product number and the number of products in the bucket can be obtained by a barcode attached to the side of the bucket. The table below shows the product information in the bucket.

  In the table above, two numbers from the left indicate storage rack areas, and two numbers from the right indicate storage rack shelf numbers. When the contents of the bucket that has been input to the arrival line 401 are in a state where the products to be transported to different areas are mixedly loaded, they are automatically sorted to the mixed loading check line 403. And the lamp (Y15) which calls the worker who checks mixed loading is turned on.

  In addition, when the total number of products in the bucket is 100 or more, the product is not transported to the storage rack area but transported to the flat field area, which is a temporary storage place for large quantities of products.

Next, for example, a method for sorting buckets in line 2 (405) will be described.
The sorting line 402 communicates with the pull-in lines 404 to 407 in each area. The bucket that has flowed from the conveyor of the arrival line 401 is detected by the position sensor 410 of the sorting line 402. In response to the detection, the conveyor of the sorting line 402 starts moving and the bucket flows on the sorting line 402. The buckets flowing through the sorting line 402 sequentially pass in front of five position sensors (X4 to X7) and five pushing cylinders (mechanisms that expand and contract by air pressure: Y10 to Y14) attached to the sorting line 402.
When the sorting destination is line 2 (405), when the position sensor 411 detects the passage of the bucket, the conveyor of the sorting line 402 stops after a certain time. At that time, the bucket is in front of the pushing cylinder 412. Air is supplied to the extrusion cylinder 412, and the cylinder is extended by the pressure. When the push-out completion sensor X32 that senses a state in which the cylinder attached to the push-out cylinder 412 is fully extended detects the completion, the supply of air into the cylinder is stopped. At this time, the bucket is pushed out to the line 2 (405) by the extended cylinder.

  After the detection of the extrusion completion sensor X32, the extrusion cylinder 412 releases the air in the cylinder and contracts the cylinder to the initial position. When the pushing cylinder 412 returns to the initial position, the initial position sensor X37 attached to the pushing cylinder 412 detects it. In response to the detection, if there is a bucket on the sorting line 402, the conveyor of the sorting line 402 is moved to continue sorting the next bucket. The position sensor 413 of the line 2 (405) detects the pushed bucket and moves the conveyor of the line 2. As a result, the bucket is transported to the storage rack in the two areas. With regard to the specification of the sorting line 402, it is possible to place a bucket with a certain interval on a maximum of five conveyors, so that sorting can be performed continuously.

  In order to control, monitor, and evaluate the performance of these transport systems with the control device based on the extended Petri net of the present invention, it is necessary to model the transport system in the extended Petri net format. Hereinafter, taking this transport system as an example, a system modeling method in the extended Petri net format, and features of the extended Petri net of the present invention and a processing method at the time of execution will be described.

FIG. 5 shows a screen configuration of the design execution means 110 based on the extended Petri net that is output to the display device 104 when the transport system is modeled in the extended Petri net format.
The main window 501 of the application can be freely moved and expanded / contracted. Mainly, the main window 501 includes a menu bar 502 for performing various operations and nodes necessary for creating a Petri net model (for places, transitions, and arcs, normal arc, suppression arc, and reciprocating arc 3) A tool bar 503 for selecting and switching arrangements, a window 506 (sheet) for creating a Petri net model, and a grid line in the 507 to arrange the nodes and a Petri net for the created net window A toolbar 504 for displaying a model in a scaled size, a toolbar 505 for executing a Petri net model such as execution of control and performance evaluation, a window 507 for modeling a hierarchical destination with an extended Petri net model, a project in the current application ( Modeled peto And it is configured to unity) net diagram in project bar 508 which is represented hierarchically.

  The Petri net diagram created when modeling the system in the extended Petri net format can be created by the design execution means 110 based on the extended Petri net as follows.

  To place each node (place, transition, arc, etc.), select the node you want to place on the toolbar 503 and left-click the pointing input device 106 (mouse) at the place where you want to place the windows 506 and 507 for creating the Petri net model. Then, it can arrange. The arc can be connected by selecting the arc to be set with the tool bar 501 and selecting the place or transition to be connected with the pointing input device 106. At this time, arcs are connected in the direction from the node selected first with the mouse to the next selected node. A plurality of windows (sheets) can be generated for the windows 506 and 507 for creating the Petri net model.

  The method of modeling the transport system of FIG. 4 in the form of an extended Petri net can be roughly divided into four steps. In these four steps, it is necessary to draw a Petri net diagram on a sheet (window) suitable for each purpose.

  The first system state model creation step is a step of creating a Petri net diagram representing a state that can occur in the system or a state in which the system is to be monitored. The creator can draw a Petri net diagram representing the state of the system on the system state model sheet.

  The second system processing execution model creation step is a Petri network that expresses the contents of processing required when transitioning from one state of the Petri net diagram created in the system state model creation step to the next state in the hierarchy destination. This is a process of creating a net diagram. The creator can draw a Petri net diagram expressing the contents of the necessary processing on the control processing model sheet.

  The third system interrupt processing execution model creation process is a Petri net diagram that represents control branch processing at a higher level of the hierarchical structure when interrupt processing is to be executed preferentially from normal processing to emergency stop. It is a process to create. The creator can draw a Petri net diagram expressing the branch process of the control on the system state model sheet.

  The fourth input / output condition blocking model creation step is a step of creating a Petri net diagram for the purpose of blocking input / output conditions using a shared place and reusing them in other sheets. The creator can create a Petri net diagram for the purpose of making the block by using an I / O setting sheet.

  FIG. 7 shows a Petri net diagram in which the transport system shown in FIG. 4 is created in the system state model creation process. The Petri net model of FIG. 7 represents the appearance of FIG. 4 with a Petri net. Places 702, 704, etc. indicate lines or portions on the lines. If the product (bucket) flowing on the line is represented by a token, it is possible to grasp what product (bucket) is flowing in which line of the transport system. In this way, in the system state model creation step, a possible state of the system or a state in which the creator wants to monitor the system is represented by a Petri net diagram. Then, the processing when transitioning from one state to the next state is expressed using hierarchical transitions 701, 705, and the like.

  In the system process execution model creation process, which is the next process, the contents of the process expressed by hierarchical transitions are modeled. FIG. 10 is a Petri net diagram of the hierarchization destination of the hierarchized transition 701 shown in FIG. A hierarchized transition 701 represents an input process for an incoming product. In this Petri net diagram of the hierarchization destination, input / output signal processing with the control device at the time of input of incoming goods is performed.

  The Petri net diagram of the hierarchization destination has a specific structure that starts with a start transition (START node) and ends with an end transition (END node), as shown in FIG. The specification of hierarchical transitions such as start transition and end transition will be described later.

  In the Petri net diagram of the hierarchization destination, as shown in FIG. 10, the contents of the incoming goods input process are first subdivided into a straight line connecting the start transition and the end transition. In FIG. 10, there are three processes: a place 1002 that indicates the state of confirmation of the incoming goods, a place 1008 that indicates the state of confirmation of the operating state of the incoming conveyor, and a place 1011 that indicates the completion of control of the incoming conveyor. It is subdivided into status places so that the progress of processing can be grasped.

  Next, process transitions indicating the subdivided processes generated by the transition of these process state places are arranged in order to create a flow of the entire process. In FIG. 10, the processing is subdivided into a processing transition 1005 indicating the confirmation processing of the received product, a processing transition 1009 for performing the operation processing of the conveyor, and a processing transition 1010 for performing the operation continuation processing on the operating conveyor.

Further, the processing transition input condition necessary for transition from the processing state place and the processing content to be performed after satisfying the input condition are modeled as a processing transition output condition.
In FIG. 10, as the input conditions of the processing transition 1005, the processing state place 1002, the external input place 1003 corresponding to the input signal of the sensor attached to the left end of the incoming line indicating the presence / absence of the incoming goods, and the external input place 1003 Modeling is performed under three input conditions of a shared place 1004 that suppresses the firing of a processing transition 1005 due to an overlapping input signal.
As the processing after the firing of the processing transition 1005, the marking indicating the state transition to the state place 1008, the marking for adding one count number of the shared place 1007 for counting the number of buckets on the arrival line, and the same as the shared place 1004 Three processes for modeling a token on a shared place 1006 sharing a place and preventing firing due to a duplicate signal in the external input place 1003 are modeled. Similarly, all processing transitions are modeled.

  Similarly, FIG. 11 shows a layered destination Petri net diagram of the layered transition 706 in FIG. 7 and FIG. 12 shows a layered destination Petri net diagram of the layered transition 707 as a model for the remaining layered transitions in FIG. To do.

  The system interrupt processing execution model creation step is a step of creating a Petri net model capable of executing interrupt processing such as emergency stop from normal processing. The normal process in FIG. 4 is to execute the process by executing the Petri net diagram created in the system state model creation process or the system process execution model creation process. When interrupt processing is performed for normal processing in an extended Petri net, modeling is performed to branch normal processing and interrupt processing in a Petri net diagram that has a hierarchical structure higher than the Petri net diagram that performs interrupt processing like normal processing. There is a need. In the case of the transfer system of FIG. 4, the upper Petri net diagram shown in FIG. 7 is hierarchized by a hierarchized place so that the hierarchical structure becomes higher than that of FIG. Modeling is in progress.

When the execution process of the Petri net diagram shown in FIG. 6 is started, a token is marked on the hierarchized place 604 that performs the processing of the transport system shown in FIG. In this way, modeling is performed so as to always mark the hierarchical place where the normal processing is performed.
The hierarchized transition 605 includes a Petri net diagram of the hierarchization destination as shown in FIG. 8, and is modeled to fire when an emergency stop button is pressed. Therefore, when the emergency stop button is pressed, the hierarchized transition 605 is fired, the token of the hierarchized place 604 that performs normal processing is removed, and the token is shifted to the hierarchized place 606 that performs processing during emergency stop. As a result, the normal process is stopped and the emergency stop process is executed. In this way, a hierarchical place that performs interrupt processing is prepared, and modeling is performed so that tokens can be transitioned from the hierarchical place that performs normal processing.

The input / output condition blocking model creation process aims to simplify modeling by blocking external input / output conditions and reusing them in other Petri net diagrams. In the case of the transport system shown in FIG. 4, a condition is set as shown in FIG. 9 such that the signal is turned on when any one of the seven emergency stop buttons (X22 to X28) is pressed.
At the time of modeling, the result of the input / output condition is reflected on an I / O shared place having a specific function that can share information such as tokens in other Petri net diagrams as in the same place. In the case of FIG. 9, the result of the input condition is reflected by the I / O shared place 901. Then, the input condition of the transition 804 is simplified by sharing the I / O shared place 901 with the I / O shared place 803 in another Petri net diagram (FIG. 8).

  The I / O setting sheet, which is a sheet of the Petri net diagram shown in FIG. 9, can be modeled as shown in FIG. 9 by a specific execution process of the extended Petri net. This will be described in the I / O setting sheet execution processing method.

  In this way, modeling in four steps facilitates modeling, and the system state can be easily grasped when the Petri net model is executed.

Next, the execution process of the Petri net model in the extended Petri net format of the control device based on the extended Petri net according to the present invention will be described.
The execution process of the Petri net model in the extended Petri net format is performed by the design execution means 110 based on the extended Petri net. The execution processing includes execution forms of the control execution means 126, the performance evaluation execution means 127, and the normal execution means 128, and a hierarchical structure execution means 123 that controls execution processing between a plurality of nets in the hierarchical structure. These will be described in order.

FIG. 13 is a flowchart schematically showing the order of execution processing of the control execution means 126 in a sheet depicting a single Petri net diagram. The basis of the execution process is the time Petri net theory that incorporates the time concept into the Petri net theory of FIGS. In the extended Petri net format, execution processing is performed in units of windows (sheets) 506 and 507 of the design execution means 501 based on the extended Petri net of FIG.
The execution process is to perform a firing process on the transition in the sheet to change the token. The transition firing process differs depending on whether or not there is an attribute setting that can be selected by the attribute setting means 115. Here, an ignition process for a sheet of a Petri net diagram for which attribute setting is not performed by the attribute setting unit 115 will be described.

  There are mainly two types of ignition processes: an ignition process 1304 without an ignition delay and an ignition process 1306 with an ignition delay.

  The ignition process without an ignition delay is to perform the ignition process only for a transition in which no delay time is set for the transition. In this case, after the recorded ignitable transition is ignited, the igniting process is performed again for the transition for which the ignition delay time is not set. Repeats the ignition process until a transition for which no ignition delay time is set cannot be ignited.

  As for the ignition process with an ignition delay, the ignition process is performed only for the transition in which the ignition delay set time is set. In this case, the recorded ignitable transition includes an igniting counter, and each time the igniting process is performed, the igniting counter is incremented by 1 to delay the igniting of the transition by a delay time, and the time is delayed. For the transition in which the values of the ignition counter and the transition delay time of the transition are recorded are the same, the token is transitioned by releasing the standby state and performing the ignition.

  Based on this ignition process, the control execution process of FIG. 13 will be described. The input refresh processes 1301 and 1310 are processes for reflecting the input signal of the control target device 108 in the modeled in the extended Petri net format using the following external input place.

An external input place X such as a place 902 or a place 1003 indicates an input connection from the control target device 108 to the control device 101 based on the extended Petri net. Regarding the reference numerals, the control device 101, the control target device 108, and the input contact number It corresponds to. Therefore, the external input place of X1 represents that the input signal from the control device connected to the input contact 1 is reflected. The input signal sent from the control device can determine the presence or absence of the input signal by marking the token of the external input place. Since this place is a functional place, there is always one token that can be marked in the place, and the specification always reflects the state of the external input signal.

  Regarding the ignition processing without I / O setting sheet ignition delay, the processing without ignition delay is the same except that the types of sheets are different. The I / O setting sheet blocks other complicated conditions applying the presence / absence of signals in the external input place and the external output place corresponding to the contact indicating the connection with the control target device in the I / O shared place. This is a sheet used for simplifying sheet modeling. Details of the I / O setting sheet will be described later.

  The ignition process from steps 1302 to 1307 is the same as that shown above.

  The output refresh process 1308 outputs, as an output signal, the state represented by the presence or absence of a token in the external output place indicating Y and the output contact number for the control target device 108 connected to the contact. Process to reflect. Specifically, for example, when the external output place Y1 is marked, an output signal is output to the control target device 108 connected to the output contact Y1.

  The time taken for the ignition process or the like is subtracted from the one step time set by the sheet setting means 113, and the remaining time is waited at 1311. In this way, it is possible to control the control target device 108 by repeatedly performing the execution process at regular intervals.

  FIG. 14 is a flowchart of the performance evaluation execution unit 127 and the normal execution unit 128. In both of these, the order of execution processes, the ignition processes 1401 and 1403 without ignition delay, and the ignition process 1402 with ignition delay are the same as the above-described ignition processes except that the parameters used for the execution process are different.

  The feature of the control execution means 126 is that the input information reflected in the external input / output place changes with time, so that the execution process is repeatedly repeated even if all transitions cannot be fired. In this respect, the specifications are different from the performance evaluation execution means 127 and the normal execution means 128.

  The hierarchical structure execution unit 123 will be described. A place 604 and a place 606 in FIG. 6 are hierarchized places, and a transition 605 and a transition 607 are hierarchized transitions. Hereinafter, each hierarchical node of the extended Petri net will be described.

A hierarchical place can have a hierarchical structure by relating it to one sheet. The Petri net diagram related to the hierarchical place in the hierarchical structure is called a Petri net diagram (sheet) of the hierarchical destination.
The layered place and the Petri net diagram of the layered destination have the following specifications.
1. When a token is marked in the hierarchized place, the Petri net diagram of the hierarchization destination is activated and an execution process is performed.
2. As long as tokens continue to exist in the hierarchized place, the petri net diagram of the hierarchized destination performs execution processing.
3. When a token in the hierarchized place transitions and goes out of the place, the Petri net diagram of the hierarchization destination is inactivated and the execution process is stopped. However, when there are still one or more tokens in the hierarchical place, the execution process is performed again according to the following setting of the sheet setting means 113.
(i) When the check box 1501 in FIG. 15 is not checked, the Petri net diagram is activated and continuously executed from the stopped state (continuous execution process).
(ii) When the check box 1501 in FIG. 15 is checked, the Petri net diagram is initialized from the state where the Petri net diagram is activated and stopped, and the execution processing is performed again (execution processing after initialization).
4). If there is no token in the hierarchical place, the execution process of the Petri net diagram of the hierarchical destination is stopped until a token is marked again in the hierarchical place.

Similarly to the hierarchized place, the hierarchized transition can have a hierarchical structure by associating one Petri net diagram. The Petri net diagram related to the hierarchized transition in a hierarchical structure is called the Petri net diagram (sheet) of the hierarchization destination. The layered transition and layered Petri net diagram have the following specifications.
1. The hierarchized transition activates the hierarchized Petri net diagram when it becomes ready to fire, and performs execution processing. However, the hierarchized transition does not reserve the firing of the input place token.
2. The Petri net diagram of the hierarchization destination is always composed of specific transitions of the start transition and end transition.
3. In the Petri net diagram of the hierarchization destination that has been activated and started execution, the start transition is ignited only once (specification of the start transition).
4). Until the end transition firing condition is met and the end transition is fired, the hierarchized transition of the hierarchizing source does not fire (end transition specification).
5. If the transition that conflicts with the hierarchized transition fires first, and the hierarchized transition cannot be fired, the Petri net diagram of the hierarchized destination that is active and executing the process immediately executes the process Stop and initialize the sheet as inactive. At this time, only the normal place is initialized.
6). When the end transition is ignited, the Petri net diagram of the hierarchization destination is made inactive, the execution process is terminated, and initialization is performed. Then, the hierarchized transition of the hierarchization source is fired and token transition is performed as usual.

  The flowchart of FIG. 16 shows the flow of execution processing of the Petri net model having a hierarchical structure based on the specifications of the hierarchized place, the hierarchized transition, and the Petri net diagram of the hierarchization destination. The ignition process 1602 shows the ignition process without an ignition delay (one step unit) and the ignition process with an ignition delay (one step unit) shown in FIGS. 13 and 14. Therefore, the process of the flowchart of FIG. 16 is performed before the firing process of the flowcharts shown in FIGS.

  The activation determination 1601 in FIG. 16, the sheet activation determination 1603 by the hierarchized node, and the sheet inactivity determination 1604 by the hierarchical node can be determined from the state of the hierarchical node.

  In the extended Petri net format of the present invention, three types of sheets for drawing Petri net diagrams are prepared so that system control and monitoring and system performance evaluation can be performed smoothly. Each type and role will be described.

The first sheet is a system state model sheet. There are three main roles for this sheet.
1. Displays the system status (system monitoring is possible).
2. When the system state transitions, it has a trigger role that executes only the necessary layered Petri net diagram.
3. At the time of system performance evaluation execution, the control processing model sheet, which is a lower Petri net diagram of the hierarchical structure, and the system state model sheet are separated, and only the system state model sheet necessary for system performance evaluation is executed.

  The second sheet is a control processing model sheet. The role of this window is a sheet that receives an instruction of execution processing from the system state model sheet and performs processing, control processing, data processing, and the like necessary for changing the state of the system. It is a sheet that actually arranges external input places, external output places, and the like.

The third sheet is an I / O setting sheet. The feature of this sheet is that it is always executed after signal refresh of a control target device (I / O device) corresponding to input / output. The role of this sheet is as follows.
1. The conditions for applying the presence / absence of signals of the I / O device are blocked to simplify the conditions of the control processing model sheet.
2. The I / O setting sheet always checks the condition of the I / O device signal, so that the condition that always reflects the signal of the I / O device can be used even if the hierarchized sheet is not always executed. It becomes like this.
3. In the I / O setting sheet, a sheet in which I / O device conditions are blocked using an I / O shared place (a functional place that can be used by sharing one place among a plurality of sheets), etc. The initialization process 1309 (same meaning as the I / O refresh process. Execution of the initialization process can be selected) is executed for each step on the sheet. This simplifies the input / output condition sheet.

There are rules for the three types of seats for efficient system control, monitoring and performance evaluation.
1. The system model net is always arranged at a position having a higher hierarchical structure than the control processing model sheet.
2. A node related to an I / O device such as an external input / output place cannot be arranged in the system state model sheet.
3. Hierarchy cannot be set on the I / O setting sheet.

  Next, taking the Petri net model of the transport system modeled in the extended Petri net form of the present invention as an example, the execution process of the hierarchical structure in the extended Petri net form and the modeling of the conventional Petri net theory become a complicated model. A modeling method that can simplify interrupt processing such as emergency stop will be described.

  FIG. 6 is a Petri net diagram at the top of the hierarchical structure serving as the root. Since this Petri net diagram is a sheet of a system state model, it is modeled to represent a change in state of the transport system. The transfer system net in FIG. 6 can express whether the transfer system is under control or emergency stop, and whether it is in a stage of transition to them.

  The place 601 indicating the initial state of the system has an initial marking. In the modeling of the transport system of FIG. 6, the token indicates the current state of the transport system. When the execution button of the toolbar 505 is selected and the execution process of the Petri net diagram modeling this transport system is started, the transition 603 is fired and the token 602 transitions to the hierarchical place 604.

  When a token is marked on the hierarchized place 604, FIG. 7 of the Petri net diagram indicating that the line control is being performed, which is the Petri net diagram of the hierarchization destination, is activated and the execution process is started. When a token is marked on the stratified place 604, the hierarchized transition 605 is ready to be fired, so that the emergency stop processing sheet, which is the Petri net diagram of the stratified destination, is activated and starts executing.

  The Petri net diagram showing the line control in FIG. 7 activated by the hierarchical place 604 is a system state model sheet. The state indicated by each place on the sheet is the state of the transport system itself, and is modeled in the same manner as the appearance of the transport system viewed from above. Each place represents each line of the transportation system, and the transition token can represent a bucket that is an incoming product and a product in the bucket. For this reason, with respect to FIG. 7, a setting for giving an attribute to the token is performed. This point will be described later in connection with the attribute setting method.

  When the sheet of the system state model indicating that the line control is being performed in FIG. 7 is executed, the hierarchized transition 701 can be fired. Then, the Petri net diagram showing the input of the received goods in FIG. 10 which is the Petri net diagram of the hierarchization destination is activated and the execution process is performed.

  The incoming product input sheet of FIG. 10 is a control processing model sheet. This sheet performs a control process when an incoming product is input. Therefore, an external input / output place is arranged in this Petri net model, and handles an input signal from a control target such as a sensor or an output signal to a control target such as a conveyor.

  On the input sheet of the incoming goods, the start transition 1001 is ignited only once (specification of the start transition), thereby marking the status place 1002 indicating the confirmation state of the incoming goods.

In order for the next transition 1005 to ignite, the left end sensor 414 (X1) of the incoming line 401 must be turned ON. The left end sensor 414 is turned on when an operator of the arrival line 401 puts a received item in front of the left end sensor 414. Therefore, a token is marked on the external input place 1003 corresponding to the left end sensor 414 after the arrival of goods.
The shared place 1004 is a functional place that allows one place to be shared by a plurality of sheets, and shares the same place as the shared place 1006. Since the shared place 1004 is connected to the transition 1005 by a suppression arc, the firing condition of the transition 1005 is satisfied if there is no token in the shared place. As the transition 1005 is fired, the token transits to the shared place 1006 and the shared place 1007 state place 1008 which are three output places.

  At this time, since the external input place 1003 is a functional place that always reflects the state of the input signal of the left end sensor 414 in the place token, the token of 1003 is not lost by the transition of the token due to the firing of the transition 1005.

  A shared place 1005 represents the number of buckets flowing in the arrival line. This shared place 1005 is used in the Petri net diagram of the hierarchized destination of the hierarchized transition 705.

  When a token is marked on the shared place 1006 after firing, the shared place 1004 that shares the same place is also marked. The role of the shared place 1004 is to make it impossible to ignite the transition 1005 until the bucket passes from the front of the left end sensor 414 and the input signal is turned OFF.

  The processing after the status place 1008 performs processing for operating the conveyor according to the status of the conveyor on the arrival line. After that process, the end transition 1012 is fired. By this firing, the hierarchized transition 701 of the hierarchization source sheet is fired and the token is changed to the function place 702.

  The token that has transitioned to the function place 702 represents an incoming product (bucket) that flows through the transport system. The function place 702 is a place having a function of reflecting the two-dimensional barcode data read by the barcode reader 409 on the attribute of the token. The function place 702 can be implemented with a function by inputting a function code and setting the function with the function setting means 120 shown in FIG. This function will be described later in connection with attribute definition.

  As described above, the execution processing of the sheet under line control shown in FIG. 7 indicates the state of the conveyance system, and the processing necessary when transitioning from the state to the next state is performed as the hierarchization destination of the hierarchized transition. The process is executed by executing the Petri net diagram. As long as there is a token in the hierarchized place 604 in FIG. 6, the sheet is hierarchized by the hierarchized nodes in FIG. 7 and the sheet, and the conveyance system is controlled by executing the execution process.

  When the hierarchized transition 605 of FIG. 6 can be fired, the emergency stop processing sheet (FIG. 8), which is the Petri net diagram of the hierarchization destination, is activated and executed.

  When the emergency stop processing sheet of FIG. 8 is executed, the start transition 801 fires only once and marks the token in the state place 802. Since the transition 804 cannot satisfy the firing condition if there is no token in the I / O shared place 803, it does not fire. This I / O shared place 803 is shared with the I / O shared place 901 in which input conditions are blocked in the I / O setting sheet of FIG. This blocking has already been described in relation to the input / output condition blocking model creation step.

  Timings of the I / O setting sheet firing process and execution process are shown in steps 1302, 1304, 1306, and 1309 in the flowchart of FIG.

In the I / O setting sheet of FIG. 9, when any of the seven emergency stop buttons on each line of the transport system is pressed, a transition to be connected is fired and a token is marked on the I / O shared place 901. .
The I / O shared place 901 has a limited capacity set to 1 in order to handle ON / OFF control signals by tokens. Therefore, only one token of the I / O shared place 901 can be marked. As long as the token of this I / O shared place 901 is not removed, the transition will not fire continuously.

The I / O setting sheet can be set by the sheet setting means 113 (FIG. 1) whether or not initialization is performed in units of steps with the setting 1501 shown in FIG. Since this I / O setting sheet is set to be initialized at each step, the token of the I / O shared place 901 is initialized at the next step, and the token is removed.
When the emergency stop button is returned, the transition to be connected is not ignited, so the token of the I / O shared place 901 is also lost.

  Since the I / O shared place 901 is shared with the I / O shared place 803 of the emergency stop processing net in FIG. 8, the input signals of the seven emergency stop buttons are blocked to simplify the firing condition of the transition 804. Yes. If even one emergency stop is pressed, the token is marked on the I / O shared place 803, and the transition 804 is fired.

  In the conventional Petri net, when a token once marked in a place is discharged, it is necessary to construct a Petri net diagram for discharging. However, the initialization setting function of the I / O setting sheet by the extended Petri net of the present invention can eliminate the need to construct a Petri net diagram for discharging the token when the emergency stop button is returned. Further, by blocking the input / output conditions of the I / O device and using them in other sheets, the system modeling can be simplified and the system design efficiency can be improved.

  After the transition 804 fires, the token transitions to the processing state place 805. Then, the end transition 806 can be ignited and ignites.

  When the emergency stop button is pressed as described above, the layered transition 605 in FIG. 6 is fired, the token of the layered place 604 is removed, and the token transits to the layered place 606. When the hierarchized transition 605 is fired, the emergency stop processing sheet FIG. 8, which is the Petri net diagram of the hierarchization destination, is inactivated and initialization is performed. At this time, only a normal place is initialized.

  Further, after the hierarchized transition 605 is fired, the line control sheet (FIG. 7) that is the Petri net diagram of the hierarchization place 604 and the hierarchization destination executed by the execution of the line control sheet. All Petri net diagrams are inactive and the execution process stops.

  The sheet under line control in FIG. 7 is not initialized because the sheet setting unit 123 (FIG. 1) does not set 1501 in FIG. In this case, when a token is marked again in the hierarchical place 604, the execution process is continued from the current sheet state.

  The layered place 606 marks the token by firing the layered transition 605. As a result, the Petri net diagram of the hierarchized destination of the hierarchized place 606 is activated and starts execution processing. In the Petri net diagram of the hierarchization destination where the execution process is started, the transport system is brought into an emergency stop state by performing the process of removing the output signal in the Petri net model of the system.

  As described above, when the execution process of the Petri net diagram shown in FIG. 7 and the Petri net diagram whose hierarchical structure is lower is stopped, the actual transport system is also stopped. At this time, the system state of the Petri net model shown in FIG. 7 is the same as the state of the actually stopped transport system.

  Since the token is marked on the hierarchized place 606, the hierarchized transition 607 can be ignited, and the Petri net diagram of the hierarchized destination is activated to start execution.

  If these hierarchical Petri net format execution methods are used, interrupt processing such as emergency stop can be easily performed regardless of the state of the system. Furthermore, it is possible to easily cope with an increase in the types of interrupt processing.

  As an example, a case where another interrupt process is desired to be added to the transport system of FIG. 4 will be described. When interrupt processing is to be performed preferentially, a token is guided to a Petri net model that performs interrupt processing in the upper route of the hierarchical structure (branch processing). This makes it possible to add and execute interrupt processing.

  In the case of FIG. 6, a layered transition different from the layered transition 605 is connected from the layered place 604. Then, a condition for generating an interrupt is modeled in the Petri net diagram of the hierarchization destination so that the hierarchized transition can be fired when the condition is satisfied. Next, a hierarchical place is connected to this hierarchical transition. If the contents of the interrupt process are modeled in the Petri net diagram of the hierarchized destination of the connected hierarchical place, the interrupt process is completed.

  In this way, if the interrupt processing generation condition is satisfied, the newly added layered transition is fired, the token of the layered place 604 is removed, and the layered place that performs the newly added interrupt processing is changed. The token is marked. Thereby, interrupt processing can be executed.

  In the conventional Petri net theory, when the transport system shown in FIG. 4 is controlled by a Petri net model, in order to cope with an interrupt-like control such as an emergency stop, the control contents are included in the Petri net model program. All the transitions in this section had to be added with the precondition and postcondition that an emergency stop process would be performed when an emergency stop was pressed. However, according to the method of executing the hierarchical structure in the extended Petri net format according to the present invention, the token is guided to the Petri net model that performs the interrupt processing on the route as shown in FIG. Processing can be easily realized. Further, even if the number of types of interrupt processing increases, it can be easily handled by adding branch processing to the route.

  Further, the method for executing a hierarchical structure in the extended Petri net format according to the present invention is characterized in that execution processing is performed by activating the Petri net diagram of the hierarchization destination by a hierarchical node at a higher level of the hierarchical structure. For this reason, only the transition of the active sheet in the Petri net model is performed by modeling the system in the extended Petri net format and executing the sheet modeled by the hierarchical structure execution method in this extended Petri net format. Execution is possible simply by making a fire determination. As a result, it is not necessary to perform firing determination for all transitions in the Petri net model, and the load of execution processing can be reduced.

  Next, attribute definition of the extended Petri net of the control device based on the extended Petri net according to the present invention and application to control will be described. When a bucket is inserted into the transport system of FIG. 4, the layered transition 701 of the Petri net diagram of FIG. 7 is fired by the execution process of the Petri net diagram of FIG. At this time, the token represents the input bucket. Furthermore, different products in the bucket can be expressed by the attribute value of the token by the attribute definition of the extended Petri net.

The extended Petri net attribute definition has two main characteristics.
The first defines three types of data types that can handle integer values, real values, character strings, or binary values (bits) that are commonly used in the control field: numeric, character, or flag types. Thus, these three types are used as token attribute definitions.
Second, three types of data length are set for each sheet on which a Petri net diagram is drawn, and data corresponding to the setting can be stored in all tokens in the Petri net diagram. The address can be specified so that it can be referenced or assigned. These attribute definitions can be set by the attribute setting means 115.
With these features, it is possible to allow attribute setting only for sheets that require attribute definition within a Petri net model consisting of multiple Petri net diagrams, and memory consumption related to attributes when execution processing is performed. And the load of the execution process related to the attribute can be suppressed.
Also, by determining the type of attribute in advance, it is possible to easily define the attribute simply by specifying the data length that can store the token.

  This attribute setting will be described using the conveyance system of FIG. 4 as an example. In the Petri net diagram shown in FIG. 7, attributes are set as shown in FIG. 15 for tokens representing buckets. The number of data can be set by 1502 for selecting the number of numerical data, 1503 for selecting the number of character data, and 1504 for selecting the number of flag data. In FIG. 7, six numerical types and six character types are designated, respectively, and a check box 1507 for allowing attribute definition to the selected sheet is checked. This makes it possible to store data of 6 numerical types and 6 character types for all tokens in FIG.

  An address is automatically set to the data stored in this token. Addresses from D1 to D6 are assigned to six numerical data. Addresses S1 to S6 are assigned to the six character-type data. As a result, when defining an attribute, by specifying an address that is automatically set for each attribute value, it is possible to easily set the reference to each attribute value or the assignment of the operation result. is there.

  A method of applying the extended Petri net attribute definition to system control and system status monitoring will be described. In the transport system of FIG. 4, buckets containing the products shown in Tables 1 to 5 are sequentially inserted. When the first bucket is inserted, the hierarchized transition 701 in the Petri net diagram of FIG. 7 is executed, and the token is changed to the function place 702 by firing. At this time, the function place 703 reserves a token until the specific function of the place is completed, and the token cannot be changed. In this function place 702, the two-dimensional barcode of the bucket flowing in the arrival line 401 is read after marking the token, and the product number in Table 1 is assigned to the token from S1 to S5 of the character type. Further, the number for the product number is substituted from D1 to D5. After the substitution is completed, the token reservation by the function place 702 is canceled, and the transition 703 can be fired.

An execution process for performing the transition setting for the sheet for which the attribute setting is permitted and the control of the conveyance system will be described. In a sheet that permits the attribute setting, the transition firing condition expression setting unit 116 and the transition output setting unit 117 can easily perform settings for performing complex control using the attribute value of the token. In the execution process of the sheet for which the attribute setting is permitted, the firing determination process is performed in consideration of the firing rule including the transition firing condition expression set by the transition firing condition expression setting unit 116 in addition to the normal transition firing determination.
In addition, the token attribute value is changed in accordance with the setting of the transition output setting unit 117 for the transitioning token after firing. The firing condition formula set by the transition firing condition formula setting means 116 and the output setting set by the transition output setting means 117 are set for each related arc, and the setting value is incorporated into the related arc. Complex firing processing using attribute values in system control is performed by the attribute setting firing processing means 124 in the execution processing.

  In the transition 703, since the setting by the transition firing condition formula setting unit 116 is not performed, the fire is fired by the normal firing determination. The output setting after firing is set by the transition output setting means 117. Therefore, the storage rack area is determined based on the attribute value of the token transitioned from the function place 703, and data indicating the storage rack area is set in D6 to be stored in the token transitioned from the function place 703.

  A setting screen of the transition output setting means 117 of the transition 703 is shown in FIG. The transition output settings are based on the attribute value of the token that transitions from the input place, and the attribute value of the token that is output from the transition to the output place. This can be done arbitrarily, such as substituting the calculated result.

The description content of 1: 1: D1 of the address D1 in the output token setting table 1702 will be described. The first number indicates the number of the arc connected to the input place where the token transitions due to the firing of the transition. The arc number is a number displayed in the number column of the input arc list 1701.
The second number is the number of the transition token. When the arc specified by the arc number has a weight of 2 or more, there are two or more tokens that transition from the input place connected to the arc. For this reason, the second number indicates the number of tokens that transition when there are multiple tokens that transition from the same arc when the transition is fired. By these, one transition token can be determined.
The third address value specifies the address of the confirmed token.

  The address D1 of the output token setting table 1702 means that the value of the token D1 that transitions from the function place 702, that is, the value of the number is substituted as it is. D2 to D5 are also assigned the value of the designated address. For D6, if the upper two-digit storage rack area (hereinafter referred to as “transport destination area”) of the product number assigned to S1 to S5 is the same for all five products, the numerical value of the transport destination area is substituted and the product is transported. If even one destination area is different, the value 6 for guiding to the mixed loading check line 403 is substituted. In this way, it is possible to easily set the attribute data of the token to be output.

  The control in which the pushing cylinder sorts the buckets according to the transfer destination area of the buckets in the sorting line 402 of FIG. 4 will be described. The token storing the data of the transport destination area in D6 in the transition 703 is sequentially transferred to a place 707 indicating a bucket on the sorting line 402 through execution processing. When a token is marked in this place 707, a hierarchized transition 706 that performs sorting processing by an extrusion cylinder and a hierarchized transition 708 that performs processing of transporting a bucket to the next extrusion cylinder can be ignited in a normal firing determination. It becomes.

In the hierarchized transition 706, the firing condition formula using the attribute value of the token representing the bucket on the line is set by the transition firing condition formula setting unit 116. A screen for setting the hierarchized transition 706 by the transition firing condition setting means 116 is shown in FIG.
The input arc list 1801 shows the input arcs. In the firing conditional expression setting 1802, the firing conditional expression can be arbitrarily set using the token attribute value in the input place connected to the input arc. There are two firing condition formulas shown in the firing condition formula setting 1802 as the conditions for the transition 706 to fire. If there is a state that can be ignited by normal firing determination, and both of the two firing condition expressions are satisfied, the transition 706 is ignited. The first firing condition formula is that the token transport destination area is 2, and the second firing condition formula is that the number of five products is 100 or less.

  In the hierarchized transition 708, the firing condition formula is set by the transition firing condition formula setting means 116. A screen for setting the hierarchized transition 708 by the transition firing condition setting means 116 is shown in FIG. The hierarchized transition 708 is configured to ignite if it can be ignited by normal firing determination and satisfies either of the two firing condition expressions shown in the firing condition expression setting 1902.

  Since the value of D6 indicating the transfer destination area of the token that has transitioned to the place 707 is 2, and the total of D1 to D5 indicating the number of products is also 100 or less, the hierarchical transition 706 satisfies the firing condition. Therefore, it is possible to ignite by the firing determination by the firing condition formula, and the execution process of the layered Petri net diagram shown in FIG. 11 is performed.

  FIG. 11 shows the processing contents controlled by the extrusion cylinder that sorts the buckets into the line 2. The bucket shown in Table 1 is sorted into line 2 by performing the execution process of the Petri net diagram of FIG.

  Among a plurality of sheets obtained by modeling the transport system of FIG. 4 in the extended Petri net format, only the sheets shown in FIG. Attributes are not defined for other sheets. In this way, the token attributes can be defined on a sheet basis as needed. As a result, the firing determination process based on the attribute need only be performed on the necessary sheets, and the load on the execution process can be reduced.

  In this way, the system is detailed by changing the attribute value of the token by the attribute definition by the extended Petri net, the firing control of the transition using the attribute value of the token by the extended Petri net, the function of the transition and function place, etc. It is possible to express in Further, complex system control and detailed system status monitoring can be realized.

  Next, a Petri net diagram that models controlled systems such as production lines and distribution systems in the form of extended Petri nets for control purposes, and system performance is simulated to reduce productivity. A performance evaluation execution means 127 that performs performance evaluation of a system that can find a process that becomes a bottleneck or a problem portion of the system configuration and obtain a hint to improve the system, and a specification that the performance evaluation execution means 127 uses at the time of execution The performance evaluation setting means 122 for setting the parameters will be described.

  In the Petri net diagram obtained by modeling the transport system shown in FIG. 4 in the extended Petri net format, a function place corresponding to a signal with the control target device is used in the control processing model sheet. For this reason, even if a Petri net model modeled with an extended Petri net is executed, the bottleneck that lowers the transfer efficiency, the tact time for introducing a bucket that does not overflow the bucket on the line, or the system The problem part of the configuration cannot be found in a simulated manner (not controlled with the transport system). This is because the execution process does not proceed unless the function place of the control process model sheet communicates with the control target device.

  The performance evaluation execution means 127 executes only the system state model sheets of a plurality of Petri net diagrams obtained by modeling the transport system in the extended Petri net format. Thereby, the performance evaluation of a conveyance system can be performed. This execution form is called performance evaluation execution.

  The parameters set by the performance evaluation setting unit 122 are used during the execution process by the performance evaluation execution unit 127. This parameter is different from the parameters set by the various node setting means 118 and the function setting means 120 used during the execution process by the control execution means 126 and the normal execution means 128.

  A method for setting the parameters used in the performance evaluation execution by the performance evaluation setting means 122 will be described. When performing performance evaluation, it is necessary to make settings for the hierarchized node and function place whose hierarchization destination sheet is the control processing model sheet. The setting method will be described using a Petri net diagram obtained by modeling the above-described transport system in the form of an extended Petri net.

First, it is necessary to decide what kind of performance evaluation is to be performed in the transport system. Here, the performance evaluation when tomorrow's work order is performed by the transport system of FIG. 4 will be performed.
The time unit, which is the time for the one-step execution process, is defined as the actual work time. In this case, the performance evaluation setting means 122 is set so that the performance evaluation execution is completed in 960 steps in which one time unit is 30 seconds and 8 hours of work per day is converted into time units.

Among the Petri net diagrams of the transport system, the Petri net diagrams of the system state model sheet are shown in FIGS. FIG. 6 controls branching of normal processing for controlling the transport system and emergency stop processing for stopping the transport system when the emergency stop button is pressed. Based on the work history, the ignition delay time of the hierarchized transition 605 is calculated and set from the number of emergency stops that occur in one day of the transport system. Functions such as statistics and probabilities can be set for the ignition delay time. In the hierarchized transition 607, the ignition delay time is calculated and set from the time of one emergency stop of a similar system.
In the performance evaluation execution, when the hierarchized destination sheet of the hierarchized transition is a control processing model sheet, the hierarchized destination sheet does not execute the execution process, and the ignition process is performed according to the ignition delay time set by the performance evaluation setting unit 122. It is a mechanism to do. For this reason, when the performance evaluation of FIG. 6 is executed, the execution process of FIG. 7 by the hierarchized place 604 is stopped by the hierarchized transition 605 and the hierarchized transition 607 for the emergency stop occurrence time of the transport system calculated from the work history. Thereby, the performance evaluation of a conveyance system can be reproduced faithfully.

  When the input tact time of the incoming goods in the transport system is 1 time unit of 30 seconds, the ignition delay time of the hierarchized transition 701 in FIG. Therefore, the performance evaluation setting unit 122 sets the ignition delay time of the hierarchical transition 701 to 1.

  The function place 702 needs to set a function by the performance evaluation setting unit 122. The function set in the performance evaluation execution may be different from the function set in the control execution. In the performance evaluation execution, for example, tomorrow's work order in the database 131 connected to the control apparatus of the present invention is read, and a function that reflects the product number and quantity is set in the attribute value of the token in the function place 702.

  Since the transition 703 is a normal transition, parameters such as attribute settings used during control execution are used.

  The hierarchized transition 705 performs control processing when sending buckets that flow on the arrival line 401 with the hierarchization destination sheet to the sorting line 402 during control execution. Therefore, the ignition delay of the hierarchized transition 705 at the time of performance evaluation set by the performance evaluation setting unit 122 is set by calculating the transport time of the bucket flowing from the left end to the right end of the arrival line in FIG.

  The hierarchized transition 706 performs a control process of the extrusion cylinder that sorts the buckets by the hierarchization destination sheet when the control is executed. Therefore, the ignition delay of the hierarchized transition 706 at the time of performance evaluation execution set by the performance evaluation setting means 122 is set by calculating the time required for the pushing cylinders to sort the buckets. For the firing condition expression and output setting by the attribute setting of the token of the hierarchical transition 706, the parameters used in the control execution process are used as they are.

  Similarly to the transition 706 for other layered transitions, the time for processing the control target in the control processing model sheet of the layered destination is set to the firing delay of the layered transition by the performance evaluation setting unit 122. Thereby, the performance evaluation execution of FIG. 7 can express the state of the system in a time series when, for example, buckets for tomorrow's work orders are flowed to the transport system of FIG.

  The process of performance evaluation execution is represented by tokens that transition Petri net diagrams. The speed of execution of performance evaluation can be adjusted by arbitrarily setting the processing time for one step.

As described above, by performing the performance evaluation, it is possible to evaluate the performance by flowing the bucket of tomorrow's work order in a simulated manner to the transport system of FIG. As a result, it is possible to check whether the tact time of the bucket to be inserted is appropriate or whether the working time of the day is appropriate. In addition, it is possible to ascertain whether the performance of the transfer system is suitable for work.
This is because the performance evaluation execution means 127 can grasp the bucket flow condition from the Petri net diagram representing the transport system of FIG.
Conversely, the speed and sorting speed for transporting the bucket of the transport system are determined by performance evaluation execution, and the design of the control processing is reviewed so that processing is performed at that time, or the processing is performed by replacing the transport device with one with higher performance. It is also possible to assist improvement activities, such as speeding up.

It is a figure which shows the system configuration example of the control apparatus based on the extended Petri net by this invention. It is a figure which shows before the ignition of the transition used for description of Petri net theory. It is a figure which shows after the ignition of the transition used for description of Petri net theory. It is a figure which shows the contact number corresponding to the structure of a conveyance system, and a control object apparatus. It is a figure which shows the window of the design execution means based on the extended Petri net by this invention. It is a figure which shows the highest net | network (route net) which modeled the conveyance system in the extended Petri net format of this invention. It is a figure which shows the net in line control which modeled the conveyance system in the extended Petri net format of this invention. It is a figure which shows the emergency stop processing net which modeled the conveyance system in the extended Petri net format of this invention. It is a figure which shows the I / O setting net which modeled the conveyance system in the extended Petri net format of this invention. It is a figure which shows the input net | network of the arrival goods which modeled the conveyance system in the extended Petri net format of this invention. It is a figure which shows the line 2 branch processing net which modeled the conveyance system in the extended Petri net format of this invention. It is a figure which shows the cylinder 2 passage net which modeled the conveyance system in the extended Petri net form of this invention. It is a flowchart figure which shows the execution process of the control execution means by the extended Petri net of this invention. It is a flowchart figure which shows the execution process of the normal execution means and performance evaluation execution means by the extended Petri net of this invention. It is a figure which shows the setting screen of the network setting means by the extended Petri net of this invention. It is a flowchart figure which shows the execution process of the hierarchical structure execution means by the extended Petri net of this invention. It is a figure which shows the setting screen of the output setting means of the transition by the extended Petri net of this invention. It is a figure which shows the setting screen of the firing condition formula setting means of the transition by the extended Petri net of this invention. It is a figure which shows the setting screen of the firing condition formula setting means of the transition by the extended Petri net of this invention.

Explanation of symbols

101 Control Device Based on Extended Petri Net 111 Petri Net Model Design Means 112 Petri Net Execution Means 113 Net Setting Means 114 Element Placement Editing Means 115 Attribute Setting Means 116 Transition Firing Condition Expression Setting Means 117 Transition Output Setting Means 118 I / O Setting Means 120 Hierarchical structure setting means

Claims (4)

A method for modeling a control target system based on an extended Petri net that models a system to be controlled using an extended Petri net,
Place a place representing a possible state of the system and a hierarchical structure transition representing the processing necessary for transition from a place representing a certain state of the system to a place representing the next state in a chronological order. Creating a Petri net diagram and modeling the overall state of the system;
Transition from one place to the next place on the upper Petri net diagram shows the state between the start transition indicating the start of the process, the end transition indicating the end of the process, the start transition, and the end transition. The hierarchical structure is represented in a subordinate Petri net diagram created by arranging a plurality of processing state places that are subdivided and processing transitions that indicate subdivided processing necessary for transition between the processing state places in time series. System process execution model creation process that models the processing contents of the hierarchization destination included in the transition,
Create a Petri net model that can transition from normal processing of the system to interrupt processing, and create a system interrupt processing execution model that enables priority execution of interrupt processing from any state during execution of normal processing of the system Process,
A specific Petri net diagram that reflects a result of a logical input / output condition of an external I / O device and blocks the input / output condition using a shared place having a function that can be shared among a plurality of Petri net diagrams. Creating and modeling an input / output condition blocking model that performs modeling that enables reuse of the shared place between the plurality of Petri net diagrams;
A method for modeling a control target system based on an extended Petri net, comprising: An input means including a pointing device and a keyboard, a display means for displaying the created and edited contents, a Petri net model design means for designing a Petri net model by an extended Petri net, and a Petri net execution means for executing the Petri net model With
A control device based on an extended Petri net consisting of a computer system connectable with an external control target device,
The Petri net model design means is:
Element arrangement editing means for creating a Petri net diagram in which a place, a transition, and an arc connecting the place and the transition are arranged;
Various node setting means for setting the nodes arranged by the element arrangement editing means;
Sheet setting means for setting the Petri net diagram in a state according to the purpose;
Hierarchical structure setting means for setting a hierarchical structure in the Petri net diagram;
Means for setting an association between an external control target device and an external input / output place having a specific function,
The Petri net execution means is:
Hierarchical structure execution means for controlling execution of the layered Petri net diagram;
Control execution means for controlling the external control target device by executing the Petri net model;
Normal execution means for executing the Petri net model and expressing the behavior of the system in Petri net,
Furthermore, Petri net model recording means for recording the created Petri net model,
A Petri net model display means for displaying the creation, editing and execution of the Petri net model on the display means;
A control device based on an extended Petri net, comprising: Attribute setting means for setting an attribute that enables storage, reference, and substitution of data according to a predetermined setting for all tokens in the Petri net diagram in the Petri net diagram unit;
A function setting means for assigning a function that can set external information to the attribute value given to the token or can arbitrarily set the change of the attribute value to a place or a transition;
A transition firing condition formula setting means capable of arbitrarily setting a firing condition formula indicating a firing rule of the transition;
Transition output setting means capable of arbitrarily setting a change of the attribute value with respect to the attribute value given to the token that transitions after the transition is ignited;
Function execution means for changing the attribute value of the token in accordance with the setting of the function setting means;
Attribute determination firing processing means for performing an ignition determination in consideration of the firing condition expression set by the transition firing condition expression setting means, and changing the attribute value of the token after firing according to the setting of the transition output setting means;
The control device based on the extended Petri net according to claim 2, comprising: Furthermore, a performance evaluation execution means that can simulate performance evaluation that enables specification of a factor part of productivity reduction of the system,
Performance evaluation setting means for setting specific parameters used by the performance evaluation execution means during execution;
The control apparatus based on the extended Petri net according to claim 2 or 3.

Images