An industrial PLC intelligent decision-making method and system based on distributed control

Abstract

The application discloses an industrial PLC intelligent decision-making method and system based on distributed control, relates to the technical field of industrial automation control, and comprises the following steps: S1, acquiring process image records and sequential control step records, calling I / O address tables and communication variable tables; S2, establishing boundary I / O mapping relationships and determining boundary communication variables; S3, extracting boundary process image sequences; S4, identifying output image change directions and input feedback change directions, calculating asynchronous communication time lag differences, and pairing to generate boundary transition records; S5, generating boundary alignment results; S6, calculating a residence time constraint value by using a boundary transition switching control algorithm based on average residence time; and S7, triggering adjacent PLC nodes to perform cooperative control decision-making. The application can improve the cooperative decision-making stability and control continuity in the distributed control process of an industrial PLC.

Description

Technical Field

[0001] This invention relates to the field of industrial automation control technology, and in particular to an intelligent decision-making method and system for industrial PLCs based on distributed control. Background Technology

[0002] With the continuous development of intelligent manufacturing, flexible production lines, and distributed industrial control systems, application scenarios where multiple PLC nodes collaborate to complete work section handover, equipment linkage, and sequential control are increasingly common. Existing industrial PLC control systems typically rely on process images, I / O address tables, communication variable tables, and sequential control programs to achieve data exchange and control coordination between adjacent PLC nodes. However, the following problems are prevalent in practical applications: Adjacent PLC nodes typically refresh their input and output images independently according to their respective scan cycles. Differences in process image recordings on the timeline make it difficult to accurately attribute changes in output images and input feedback corresponding to handover actions within the same work segment to the same control process. Furthermore, the I / O address configurations and communication variable configurations between adjacent PLC nodes are usually maintained separately by engineers, resulting in a lack of unified dynamic correlation between boundary I / O mappings and boundary communication variables. This can easily lead to unclear correspondences between control outputs, input feedback, and communication data. During work segment handover, there is an asynchronous communication time lag between input feedback changes and output image changes. Existing sequential control typically advances directly based on local step activation and step transfer states, making it difficult to compensate for scan cycle offsets between adjacent PLC nodes. Simultaneously, boundary input feedback may experience short-term jitter, insufficient feedback retention, and frequent switching. Traditional transfer criteria are prone to premature triggering or prolonged retention, leading to unstable collaborative control decisions between adjacent PLC nodes and consequently affecting the reliability and continuity of industrial PLC distributed control.

[0003] Therefore, how to provide an intelligent decision-making method and system for industrial PLCs based on distributed control is a problem that urgently needs to be solved by those skilled in the art. Summary of the Invention

[0004] One objective of this invention is to propose an intelligent decision-making method and system for industrial PLCs based on distributed control. This invention establishes boundary I / O mapping relationships by acquiring process image records and sequential control step records, determines boundary communication variables, generates boundary transfer records and boundary alignment results, and employs a boundary transfer switching control algorithm based on average dwell time to calculate dwell time constraint values. It corrects the initial transfer criterion to form a modified transfer criterion, thus solving the problems of asynchronous process image refresh between adjacent PLC nodes, misalignment of sequential control step time axes, and instability in collaborative control decisions caused by input feedback jitter in existing technologies. This achieves intelligent decision-making for industrial PLCs with asynchronous communication time delay compensation and stable boundary transfer constraints.

[0005] An industrial PLC intelligent decision-making method based on distributed control according to an embodiment of the present invention includes the following steps: S1. Obtain the process image record and sequential control step record formed by the adjacent PLC node within the corresponding scan cycle, and retrieve the I / O address table and communication variable table of the adjacent PLC node. S2. Establish boundary I / O mapping relationships based on the I / O address table, and determine the boundary communication variables between adjacent PLC nodes according to the communication variable table and the boundary I / O mapping relationships; S3. Based on the boundary I / O mapping relationship, extract the boundary process image sequence corresponding to the boundary I / O mapping relationship from the process image record; S4. Identify the output image change edge and input feedback change edge in the boundary process image sequence, calculate the asynchronous communication time delay difference, and generate boundary transfer records by pairing them. S5. Based on the boundary transfer record, align the two sets of sequential control steps in adjacent PLC nodes that correspond to the boundary process image sequence with the asynchronous communication time delay difference as the compensation benchmark to generate the boundary alignment result. S6. Generate initial transfer criteria based on boundary alignment results; calculate the dwell time constraint value of boundary transfer records using a boundary transfer switching control algorithm based on average dwell time. S7. Correct the initial transfer criterion with the residence time constraint value to form a modified transfer criterion. Write the modified transfer criterion into the boundary communication variable to trigger adjacent PLC nodes to execute collaborative control decisions.

[0006] Optionally, S1 specifically includes: According to the scan cycle of adjacent PLC nodes, the input image formed at the beginning of the scan cycle and the output image formed at the end of the scan cycle of adjacent PLC nodes are obtained respectively. The input image, output image, PLC node identifier and scan cycle number are combined to generate process image records for adjacent PLC nodes within the corresponding scan cycle. Extract the sequential control step number, step activation status, and step transfer status corresponding to the scan cycle number from the sequential control program of adjacent PLC nodes, and generate a sequential control step record; Retrieve the I / O address table of the adjacent PLC node. The I / O address table includes the input address, output address, I / O point name, and the PLC node identifier to which the I / O point belongs. Retrieve the communication variable table of the adjacent PLC node. The communication variable table includes the communication variable name, communication variable address, communication variable data type, and the PLC node identifier to which the communication variable belongs.

[0007] Optionally, S2 specifically includes: Extract the input address, output address, and PLC node identifier of the adjacent PLC node from the I / O address table; Based on the PLC node identifier to which the I / O point belongs, the input address and output address are matched to generate the boundary I / O mapping relationship; Extract the communication variable addresses and data types corresponding to the boundary I / O mapping relationships from the communication variable table; By combining the communication variable address and the communication variable data type, the boundary communication variables between adjacent PLC nodes are determined.

[0008] Optionally, S3 specifically includes: Extract the output address, input address, and corresponding PLC node identifier from the boundary I / O mapping relationship; Based on the output address, extract the output image value under the corresponding scan cycle number from the process image record to generate the output image sequence; Based on the input address, extract the input image value under the corresponding scan cycle number from the process image record to generate an input feedback sequence. The output image sequence and input feedback sequence are arranged according to the scan cycle number to generate a boundary process image sequence corresponding to the boundary I / O mapping relationship.

[0009] Optionally, S4 specifically includes: Perform adjacent scan cycle difference processing on the output image sequence in the boundary process image sequence, identify the scan cycle number where the output image value changes, and generate the output image change edge; Perform adjacent scan cycle difference processing on the input feedback sequence in the boundary process image sequence, identify the scan cycle number where the input image value changes, and generate the input feedback change edge; According to the boundary I / O mapping relationship, the output image change edge and the corresponding input feedback change edge are paired to generate change edge pairing items; The difference between the scan cycle number corresponding to the input feedback change edge in the pairing item and the scan cycle number corresponding to the output image change edge is calculated to obtain the asynchronous communication time delay difference. The boundary transfer record is generated by combining the output image change edge, input feedback change edge, asynchronous communication time delay, output address, input address, and corresponding PLC node identifier.

[0010] Optionally, S5 specifically includes: Based on the output address, input address and corresponding PLC node identifier, extract two sets of sequential control steps corresponding to the boundary process image sequence from the sequential control step record, and generate the first sequential control step sequence and the second sequential control step sequence. Using the asynchronous communication time delay difference as a compensation benchmark, the scan cycle sequence numbers in the first and second sequential control step sequences are compensated and converted to generate the first and second compensation step sequences. Based on the output image change edge and input feedback change edge in the boundary transfer record, the first compensation step sequence and the second compensation step sequence are time-axis aligned to generate sequential control step registration terms. The sequential control step registration item, boundary transfer record, asynchronous communication time delay difference, and corresponding PLC node identifier are combined to generate the boundary alignment result.

[0011] Optionally, the step of generating the initial transition criterion based on the boundary alignment result specifically includes: The sequential control step registration items, boundary transfer records, and asynchronous communication time delay differences in the boundary alignment results are correlated to form sequential step transfer relationships; Based on the asynchronous communication time delay difference, the scan cycle number in the sequential step transfer relationship is compensated to form a compensated transfer relationship; The first and second sequential control step sequences in the compensation transfer relationship are respectively mapped to the output image change edge and the input feedback change edge in the boundary transfer record to generate the inter-step transfer relationship; The initial transfer criteria are generated by combining the inter-step transfer relationship, step activation status, step transfer status, output address, input address, and corresponding PLC node identifier.

[0012] Optionally, the step of using a boundary transfer switching control algorithm based on average dwell time to calculate the dwell time constraint value of the boundary transfer record specifically involves: Determine the corresponding scan cycle number of the output image change and the corresponding scan cycle number of the input feedback change from the boundary transfer record; According to the corresponding scan cycle number along the input feedback change, the number of scan cycles in the boundary process image sequence where the input image value remains unchanged is counted, and input feedback is generated to maintain the scan cycle. Within a preset observation window, count the number of boundary transfer records corresponding to the same boundary I / O mapping relationship, and calculate the average dwell time according to the preset observation window length and the number of boundary transfer records; The difference between the input feedback scan cycle and the preset average dwell cycle is calculated to generate a single dwell difference. The difference between the average dwell time and the preset average dwell time is calculated to generate the window dwell time difference. The dwell time constraint value of the boundary transfer record is generated based on the smaller value between the single dwell difference and the window dwell difference.

[0013] Optionally, S7 specifically includes: The dwell time constraint value is correlated with the inter-step transfer relationship in the initial transfer criterion to generate a transfer correction term; The step activation state and step transition state in the initial transition criterion are corrected to form the modified transition criterion; Based on the communication variable address and communication variable data type in the boundary communication variables, the data format of the modified transfer criterion is converted to generate communication write data; Write the communication data to the boundary communication variable; Adjacent PLC nodes obtain the corrected transfer criteria from the boundary communication variables, and generate collaborative control decisions based on the corrected transfer criteria, sequential control step records, and process image records; Update the step transition status and output image value of adjacent PLC nodes according to the collaborative control decision.

[0014] Optionally, an industrial PLC intelligent decision-making system based on distributed control includes the following modules: The record acquisition module is used to acquire the process image record and sequential control step record formed by adjacent PLC nodes within the corresponding scan cycle, and to retrieve the I / O address table and communication variable table of adjacent PLC nodes. The boundary mapping module is used to establish boundary I / O mapping relationships based on the I / O address table, and to determine the boundary communication variables between adjacent PLC nodes according to the communication variable table and the boundary I / O mapping relationships. The image extraction module is used to extract the boundary process image sequence corresponding to the boundary I / O mapping relationship from the process image record based on the boundary I / O mapping relationship; The transfer generation module is used to identify the output image change edge and input feedback change edge in the boundary process image sequence, calculate the asynchronous communication time delay difference, and generate boundary transfer records in pairs. The time alignment module is used to align the two sets of sequential control steps in adjacent PLC nodes that correspond to the boundary process image sequence based on the boundary transfer record, using the asynchronous communication time delay difference as the compensation benchmark, and generate the boundary alignment result. The criterion calculation module is used to generate initial transfer criteria based on the boundary alignment results; and to calculate the dwell time constraint value of the boundary transfer record using a boundary transfer switching control algorithm based on average dwell time. The collaborative decision-making module is used to correct the initial transfer criterion with the residence time constraint value, form a modified transfer criterion, write the modified transfer criterion into the boundary communication variable, and trigger adjacent PLC nodes to execute collaborative control decisions.

[0015] The beneficial effects of this invention are: (1) The present invention extracts the boundary process image sequence based on the boundary I / O mapping relationship, identifies the output image change edge and the input feedback change edge, calculates the asynchronous communication time delay difference and generates boundary transfer records in pairs, so that the output image change and the input feedback change can form a corresponding relationship according to the scan cycle number, thereby improving the accuracy of boundary transfer identification; (2) Based on the boundary transfer record, the two sets of sequential control steps corresponding to the boundary process image sequence in the adjacent PLC nodes are aligned on the time axis according to the asynchronous communication time delay difference to generate the boundary alignment result, reduce the misalignment of sequential control steps caused by the asynchronous scanning cycle of adjacent PLC nodes, and improve the consistency of collaborative decision-making in the distributed control process. (3) The boundary transfer switching control algorithm based on average dwell time is used to calculate the dwell time constraint value, and the initial transfer criterion is corrected with the dwell time constraint value to form the modified transfer criterion. The modified transfer criterion is written into the boundary communication variable to trigger the adjacent PLC nodes to perform collaborative control decision, suppress input feedback jitter and false triggering caused by frequent switching, and improve the stability and reliability of the intelligent decision-making results of industrial PLC. Attached Figure Description

[0016] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used in conjunction with embodiments of the invention to explain the invention and do not constitute a limitation thereof. In the drawings: Figure 1 This is a flowchart of an industrial PLC intelligent decision-making method based on distributed control proposed in this invention; Figure 2 This is a schematic diagram of data interaction between adjacent PLC nodes; Figure 3 A schematic diagram for generating boundary process image sequences; Figure 4 This is a schematic diagram illustrating the time delay difference between pairing and asynchronous communication along the changing edge. Figure 5 This is a schematic diagram illustrating the alignment of the time axis for sequential control steps. Figure 6 A schematic diagram for generating intelligent decision-making results for industrial PLCs. Detailed Implementation

[0017] The present invention will now be described in further detail with reference to the accompanying drawings. These drawings are simplified schematic diagrams, illustrating only the basic structure of the invention, and therefore only show the components relevant to the invention.

[0018] refer to Figures 1-6 An intelligent decision-making method for industrial PLCs based on distributed control includes the following steps: S1. Obtain the process image record and sequential control step record formed by the adjacent PLC node within the corresponding scan cycle, and retrieve the I / O address table and communication variable table of the adjacent PLC node. S2. Establish boundary I / O mapping relationships based on the I / O address table, and determine the boundary communication variables between adjacent PLC nodes according to the communication variable table and the boundary I / O mapping relationships; S3. Based on the boundary I / O mapping relationship, extract the boundary process image sequence corresponding to the boundary I / O mapping relationship from the process image record; S4. Identify the output image change edge and input feedback change edge in the boundary process image sequence, calculate the asynchronous communication time delay difference, and generate boundary transfer records by pairing them. S5. Based on the boundary transfer record, the two sets of sequential control steps in adjacent PLC nodes that correspond to the boundary process image sequence are aligned on the time axis using the asynchronous communication time delay difference as the compensation benchmark to generate the boundary alignment result. S6. Generate initial transfer criteria based on boundary alignment results; calculate the dwell time constraint value of boundary transfer records using a boundary transfer switching control algorithm based on average dwell time. S7. Correct the initial transfer criterion with the residence time constraint value to form a modified transfer criterion. Write the modified transfer criterion into the boundary communication variable to trigger adjacent PLC nodes to execute collaborative control decisions.

[0019] In this embodiment, S1 specifically refers to: According to the scan cycle of adjacent PLC nodes, the input image of adjacent PLC nodes at the beginning of the scan cycle is collected. The input image is formed by the bit value, word value and double word value corresponding to the input address in the input image area of ​​adjacent PLC nodes. The output images of adjacent PLC nodes are collected at the end of the same scan cycle. The output images are formed by the bit values, word values ​​and double word values ​​corresponding to the output addresses in the output image areas of the adjacent PLC nodes. The scan cycle number is generated based on the scan cycle count value of the adjacent PLC nodes. The scan cycle number increments by one as each scan cycle is completed by the adjacent PLC node. Arrange the input image, output image, PLC node identifier, and scan cycle number in the order of PLC node identifier first, scan cycle number in the middle, and input image and output image last to generate process image records for adjacent PLC nodes within the corresponding scan cycle. Extract the sequential control step number corresponding to the scan cycle number from the sequential control program of the adjacent PLC node. The sequential control step number is the step number of the sequential control program that is active in the corresponding scan cycle. Extract the step activation state corresponding to the sequential control step number from the sequential control program of the adjacent PLC node. The step activation state is formed by the step flag bit corresponding to the sequential control step number. Extract the step transition state corresponding to the sequential control step number from the sequential control program of the adjacent PLC node. The step transition state is formed by the calculation result of the transition criterion corresponding to the sequential control step number. Arrange the sequential control step number, step activation status, step transfer status, PLC node identifier, and scan cycle number in the order of PLC node identifier first, scan cycle number in the middle, sequential control step number, step activation status, and step transfer status last to generate a sequential control step record. Retrieve the I / O address table of the adjacent PLC node. The I / O address table includes the input address, output address, I / O point name and the PLC node identifier to which the I / O point belongs. The input address corresponds to the input point in the input image area and the output address corresponds to the output point in the output image area. Retrieve the communication variable table of adjacent PLC nodes. The communication variable table includes the communication variable name, communication variable address, communication variable data type, and PLC node identifier to which the communication variable belongs. The communication variable address corresponds to the communication mapping address for data exchange between adjacent PLC nodes. According to the PLC node identifier and scan cycle number, the process image record and sequential control step record are arranged accordingly to form the process image record, sequential control step record, I / O address table and communication variable table corresponding to adjacent PLC nodes.

[0020] In this embodiment, S2 specifically refers to: Filter the input address, output address, I / O point name, and PLC node identifier corresponding to the adjacent PLC node from the I / O address table; According to the PLC node identifier to which the I / O point belongs, the output addresses belonging to the output-side PLC node are assigned to the output address set, and the input addresses belonging to the input-side PLC node are assigned to the input address set. Based on the I / O point names, perform name matching between the output addresses in the output address set and the input addresses in the input address set to generate address matching entries; Combine the output address, input address, output-side PLC node identifier, input-side PLC node identifier, and I / O point name in the address matching item to generate a boundary I / O mapping relationship; Filter the communication variable name, communication variable address, communication variable data type, and PLC node identifier corresponding to adjacent PLC nodes from the communication variable table; Based on the output-side PLC node identifier and input-side PLC node identifier in the boundary I / O mapping relationship, match the corresponding communication variable address and communication variable data type in the communication variable table; Based on the name correspondence between communication variable names and I / O point names, associate the communication variable address with the address matching item; By combining the communication variable address, communication variable data type, PLC node identifier to which the communication variable belongs, and address matching item, the boundary communication variable between adjacent PLC nodes is determined.

[0021] In this embodiment, S3 specifically refers to: In the boundary I / O mapping relationship, determine the output address, input address, output-side PLC node identifier, input-side PLC node identifier, and I / O point name; According to the output side PLC node identifier and output address, retrieve the output image value under the corresponding scan cycle number in the process image record; Arrange the output image value, output address, output-side PLC node identifier, I / O point name and scan cycle number in ascending order of scan cycle number to generate an output image sequence. According to the input side PLC node identifier and input address, retrieve the input image value under the corresponding scan cycle number in the process image record; Arrange the input image value, input address, input-side PLC node identifier, I / O point name and scan cycle number in ascending order of scan cycle number to generate an input feedback sequence. Establish a correspondence between the output image sequence and the input feedback sequence according to the I / O point names; According to the scan cycle number, the output image sequence and input feedback sequence after the correspondence is established are synchronously arranged to generate the boundary process image sequence; the boundary process image sequence includes I / O point name, output address, input address, output side PLC node identifier, input side PLC node identifier, scan cycle number, output image value and input image value.

[0022] In this embodiment, S4 specifically refers to: Group the output image sequence and input feedback sequence in the boundary process image sequence according to the I / O point name; Under the same I / O point name, the output image value of the next scan cycle in the output image sequence is compared with the output image value of the previous scan cycle. The scan cycle in which the output image value changes is marked as the scan cycle in which the output image changes. The output image change edge is generated by combining the corresponding scan cycle, the value before the output image change, the value after the output image change, the output address, the output side PLC node identifier, and the I / O point name. Under the same I / O point name, the input image value of the next scan cycle in the input feedback sequence is compared with the input image value of the previous scan cycle by difference. The scan cycle in which the input image value changes is marked as the input feedback change along the corresponding scan cycle. The input feedback change edge is generated by combining the corresponding scan cycle, the value before the input image change, the value after the input image change, the input address, the input-side PLC node identifier, and the I / O point name. According to the output address, input address, and I / O point name in the boundary I / O mapping relationship, the output image change edge is paired with the input feedback change edge that follows the scan cycle number to generate change edge pairing items; When the same output image change edge corresponds to multiple input feedback change edges, select the input feedback change edge whose scan cycle number is closest to the scan cycle number corresponding to the output image change edge, and generate a change edge pairing item; Under the same I / O point name, if no input feedback edge occurs before the next output image edge occurs, the output image edge is paired with the empty feedback mark to generate an unclosed edge pairing item; The asynchronous communication time delay difference is generated by subtracting the corresponding scan cycle number of the input feedback change along the pairing item from the corresponding scan cycle number of the output image change. The asynchronous communication time delay difference along the unclosed change in the paired item is generated by subtracting the corresponding scan cycle number of the output image change from the end scan cycle number of the observation window; The boundary transfer record is generated by combining the output image change edge, input feedback change edge, asynchronous communication time delay, output address, input address, output-side PLC node identifier, input-side PLC node identifier, and I / O point name.

[0023] In this embodiment, S5 specifically refers to: Based on the output address, input address, output-side PLC node identifier, and input-side PLC node identifier in the boundary transfer record, retrieve the sequential control step record corresponding to the output-side PLC node identifier and the sequential control step record corresponding to the input-side PLC node identifier in the sequential control step record respectively. According to the scan cycle number, the sequential control step records corresponding to the output PLC node identifier are sorted in ascending order to generate the first sequential control step sequence. According to the scan cycle number, the sequential control step records corresponding to the input PLC node identifier are sorted in ascending order to generate the second sequential control step sequence. Keep the scan cycle number in the first sequential control step sequence unchanged, and generate the first compensation step sequence; Subtract the asynchronous communication time delay difference from the scan cycle number in the second sequential control step sequence to generate the second compensation step sequence; In the first compensation step sequence, the scan cycle number and the output image change are controlled in the same order as the corresponding scan cycle number, and the step number, step activation state and step transfer state are marked as output side transfer step information. In the second compensation step sequence, the compensated scan cycle number and the scan cycle number after the input feedback change are the same as the scan cycle number after subtracting the asynchronous communication time delay from the corresponding scan cycle number, control step number, step activation state and step transfer state are marked as input side transfer step information. The output-side transfer step information and the input-side transfer step information are registered according to the same boundary transfer record to generate a sequence control step registration item. The sequence control step registration item includes the output-side PLC node identifier, the input-side PLC node identifier, the output-side sequence control step number, the input-side sequence control step number, the output-side step activation status, the input-side step activation status, the output-side step transfer status, the input-side step transfer status, and the compensated scan cycle number. The sequential control step registration item, boundary transfer record, asynchronous communication time delay difference, output address, input address, output-side PLC node identifier, input-side PLC node identifier, and I / O point name are combined to generate the boundary alignment result.

[0024] In this embodiment, an initial transition criterion is generated based on the boundary alignment result, specifically as follows: In the boundary alignment results, confirm the sequential control step registration items, boundary transfer records, asynchronous communication time delay, output address, input address, output-side PLC node identifier, input-side PLC node identifier, and I / O point name; According to the same I / O point name, the sequential control step registration item is established to correspond with the boundary transfer record to form a sequential step transfer relationship; the sequential step transfer relationship includes the output side sequential control step number, the input side sequential control step number, the output side step activation state, the input side step activation state, the output side step transfer state, the input side step transfer state, the output image change edge, the input feedback change edge, and the asynchronous communication time delay difference; Based on the asynchronous communication time delay difference, the input-side scan cycle number in the sequential step transfer relationship is converted to the output-side scan cycle number to form the compensated scan cycle number. Write the compensation scan cycle number into the sequential step transfer relationship to form the compensation transfer relationship; the compensation transfer relationship includes the output scan cycle number, the input scan cycle number, the compensation scan cycle number, the output sequential control step number, the input sequential control step number, the output image change edge, and the input feedback change edge; According to the corresponding scan cycle number along the output image change, the sequential control step number in the first sequential control step sequence is matched with the sequential control step number on the output side; According to the compensation scan cycle number, match the sequential control step number in the second sequential control step sequence with the sequential control step number on the input side; The corresponding results of the first sequential control step sequence and the output image change edge, and the corresponding results of the second sequential control step sequence and the input feedback change edge are combined to generate the inter-step transfer relationship. The inter-step transfer relationship includes the sequential control step number corresponding to the first sequential control step sequence, the sequential control step number corresponding to the second sequential control step sequence, the output image change edge, the input feedback change edge, the compensation scan cycle number, and the asynchronous communication time delay difference. The initial transfer criteria are generated by combining the inter-step transfer relationship, output-side step activation state, input-side step activation state, output-side step transfer state, input-side step transfer state, output address, input address, output-side PLC node identifier, input-side PLC node identifier, and I / O point name. The initial transfer criteria include the sequential control step association content corresponding to the boundary transfer, the boundary I / O address content, the PLC node identifier content, and the scan cycle compensation content.

[0025] In this embodiment, a boundary transfer switching control algorithm based on average dwell time is used to calculate the dwell time constraint value of the boundary transfer record, specifically: In the boundary transfer record, confirm the output image change along the corresponding scan cycle number, the input feedback change along the corresponding scan cycle number, the I / O point name, the output address, the input address, the output side PLC node identifier, and the input side PLC node identifier; In the boundary process mapping sequence, confirm the value of the input feedback change along the corresponding input mapping change; Starting from the corresponding scan cycle number along the input feedback change, the number of scan cycles in which the input image value is continuously equal to the value after the input image change is counted in ascending order of the scan cycle number to generate input feedback to maintain the scan cycle; the length of the observation window is determined by the maximum allowable number of scan cycles at the boundary intersection in the PLC engineering parameter table of the adjacent PLC nodes. The PLC engineering parameter table is formed by the engineering configuration data of the adjacent PLC nodes, the scan cycle parameters, and the boundary I / O point response parameters; The average dwell time threshold is determined by the boundary transfer records of the same boundary I / O mapping relationship under normal operating conditions. The average value of the input feedback hold scan period corresponding to the most recent consecutive complete boundary transfer records is taken as the average dwell time threshold. Based on the input feedback change, take the scan period range corresponding to the length of the observation window forward along the corresponding scan period number, and count the number of boundary transfer records that appear within the scan period range for the same boundary I / O mapping relationship. Divide the observation window length by the number of boundary transfer records to generate the average dwell time. When the number of boundary transfer records is zero, the observation window length is used as the average dwell time. Subtract the average dwell time threshold from the input feedback hold scan period to generate the single dwell time difference. Subtract the average dwell time threshold from the average dwell time to generate the window dwell time difference. Compare the numerical values ​​of the single dwell time difference and the window dwell time difference, and select the smaller difference value as the dwell time constraint value of the boundary transfer record; The dwell time constraint value, input feedback hold scan cycle, average dwell cycle, average dwell cycle threshold, observation window length, I / O point name, output address, input address, output-side PLC node identifier, and input-side PLC node identifier are combined to generate a dwell time constraint record.

[0026] This implementation method, by confirming the corresponding scan cycle numbers of the output image change edge and the input feedback change edge in the boundary transfer record, can limit the control output change and input feedback change corresponding to the boundary I / O mapping relationship to the same boundary transfer process, reducing transfer misjudgments caused by different scan cycles of adjacent PLC nodes; by statistically maintaining the scan cycle of input feedback, it can reflect the degree of continuous stability after the occurrence of the input feedback change edge, avoiding the misinterpretation of short-term jitter of the input image value as valid feedback; the length of the observation window is determined by the PLC engineering parameter table and is determined by the complete boundary transfer record under normal operating conditions. The average dwell period threshold enables the dwell time constraint value to match the actual scan cycle and boundary I / O point response characteristics of adjacent PLC nodes. Furthermore, by constraining the boundary transfer record together with the average dwell period, single dwell difference, and window dwell difference, the single input feedback hold state and the boundary transfer frequency within the observation window can be evaluated simultaneously. By selecting a smaller difference value to generate the dwell time constraint value, weaker stability factors can be used as the basis for transfer correction, improving the ability of the corrected transfer criterion to suppress boundary transfer jitter, frequent switching, and feedback lag, thereby improving the stability of adjacent PLC nodes in executing collaborative control decisions.

[0027] In this embodiment, S7 specifically refers to: In the initial transfer criterion, confirm the inter-step transfer relationship, step activation status, step transfer status, output address, input address, output-side PLC node identifier, input-side PLC node identifier, and I / O point name; Write the dwell time constraint value into the step transfer relationship corresponding to the same I / O point name to generate a transfer correction item; the transfer correction item includes the step transfer relationship, dwell time constraint value, output side step activation state, input side step activation state, output side step transfer state, input side step transfer state, output address, input address, output side PLC node identifier, and input side PLC node identifier; According to the dwell time constraint value, the output side step transfer state and the input side step transfer state in the transfer correction item are corrected. When the dwell time constraint value is less than zero, the output side step transfer state and the input side step transfer state are corrected to the hold state. When the dwell time constraint value is greater than or equal to zero, the output side step transfer state and the input side step transfer state are corrected to the allow transfer state. The corrected output-side step transfer status, input-side step transfer status, inter-step transfer relationship, output address, input address, output-side PLC node identifier, input-side PLC node identifier, and I / O point name are combined to form the corrected transfer criterion; In the boundary communication variables, confirm the communication variable address, communication variable data type, PLC node identifier to which the communication variable belongs, and communication variable name; According to the data type of the communication variable, the hold state, allow transfer state, output PLC node identifier, input PLC node identifier, output address and input address in the correction transfer criterion are converted into bit data, word data and double word data consistent with the data type of the communication variable; Arrange the bit data, word data, and double word data in the address order of the communication variable address to generate the communication write data; Write the communication data into the boundary communication variable address corresponding to the output PLC node identifier and the input PLC node identifier; Adjacent PLC nodes obtain the modified transfer criteria from the boundary communication variables. Based on the hold state and allow transfer state in the modified transfer criteria, and combined with the sequential control step number, step activation state and step transfer state in the sequential control step record, a collaborative control decision is generated. The collaborative control decision includes the step transfer state of the output PLC node, the step transfer state of the input PLC node, the output image value of the output PLC node, and the output image value of the input PLC node. According to the collaborative control decision, the step transfer state corresponding to the adjacent PLC node is updated to the hold state or the transfer allowed state, and the output image value corresponding to the adjacent PLC node is updated to the output image value consistent with the modified transfer criterion.

[0028] In this embodiment, an industrial PLC intelligent decision-making system based on distributed control includes the following modules: The record acquisition module is used to acquire the process image record and sequential control step record formed by adjacent PLC nodes within the corresponding scan cycle, and to retrieve the I / O address table and communication variable table of adjacent PLC nodes. The boundary mapping module is used to establish boundary I / O mapping relationships based on the I / O address table, and to determine the boundary communication variables between adjacent PLC nodes according to the communication variable table and the boundary I / O mapping relationships. The image extraction module is used to extract the boundary process image sequence corresponding to the boundary I / O mapping relationship from the process image record based on the boundary I / O mapping relationship; The transfer generation module is used to identify the output image change edge and input feedback change edge in the boundary process image sequence, calculate the asynchronous communication time delay difference, and generate boundary transfer records in pairs. The time alignment module is used to align the two sets of sequential control steps in adjacent PLC nodes that correspond to the boundary process image sequence based on the boundary transfer record, using the asynchronous communication time delay difference as the compensation benchmark, and generate the boundary alignment result. The criterion calculation module is used to generate initial transfer criteria based on the boundary alignment results; and to calculate the dwell time constraint value of the boundary transfer record using a boundary transfer switching control algorithm based on average dwell time. The collaborative decision-making module is used to correct the initial transfer criterion with the residence time constraint value, form a modified transfer criterion, write the modified transfer criterion into the boundary communication variable, and trigger adjacent PLC nodes to execute collaborative control decisions.

[0029] Example 1: To verify the feasibility of this invention in practice, it was applied to an automated assembly line for automotive parts. This production line includes a material feeding and conveying section, a pressing and positioning section, a vision inspection section, and a material unloading and transfer section. Adjacent sections are controlled by PLC-A, PLC-B, PLC-C, and PLC-D, respectively. Each PLC node operates with a 20ms scan cycle, and the sections interact with each other via input addresses, output addresses, and communication variable addresses. Before the production line modification, adjacent PLC nodes primarily executed control based on local input images, output images, and sequential control steps. When the pressing and positioning section completed workpiece positioning and handed it over to the vision inspection section, the output image change edge and the input feedback change edge often did not occur within the same scan cycle, resulting in an asynchronous communication time lag of approximately two scan cycles. Because the original control logic did not align the two sets of sequential control steps on the time axis, the vision inspection section sometimes had not yet received stable input feedback before the upstream pressing and positioning section had already entered the next sequential control step, causing short pauses, repeated positioning, and buffering waits.

[0030] The system acquires process image records and sequential control step records formed by adjacent PLC nodes within the corresponding scan cycle, and retrieves the I / O address table and communication variable table. Based on the I / O address table, it establishes boundary I / O mapping relationships and determines boundary communication variables according to the communication variable table and boundary I / O mapping relationships. It extracts the boundary process image sequence, identifies the output image change edge and input feedback change edge, calculates the asynchronous communication time delay difference, and generates boundary transfer records. It aligns the two sets of sequential control steps corresponding to the boundary process image sequence in adjacent PLC nodes according to the asynchronous communication time delay difference, generating boundary alignment results. Based on the boundary alignment results, it generates initial transfer criteria and uses a boundary transfer switching control algorithm based on average dwell time to calculate the dwell time constraint value of the boundary transfer record. It corrects the initial transfer criteria with the dwell time constraint value, forming a modified transfer criteria, which is written into the boundary communication variables, triggering adjacent PLC nodes to execute collaborative control decisions.

[0031] Table 1. Comparison of Boundary Transfer Recognition and Sequential Control Step Alignment Effects

[0032] As shown in Table 1, the conventional method only transmits communication variables between adjacent PLC nodes, lacking scanning cycle-level pairing processing for output image transition edges and input feedback transition edges, resulting in a fluctuation in transition edge pairing accuracy between 87.8% and 91.5%. With this invention, the boundary I / O mapping relationship associates the input address, output address, and corresponding PLC node identifier. The boundary process image sequence is arranged according to the scanning cycle number, and the boundary transfer record further introduces asynchronous communication time lag, enabling a more accurate correspondence between output image transition edges and input feedback transition edges. The average transition edge pairing accuracy across the entire line increased from 89.9% to 98.4%, and the number of sequential control step misalignments decreased from 11.3 times per shift to 1.7 times, indicating that the boundary transfer record and boundary alignment results can effectively reduce control misalignment caused by asynchronous scanning cycles of adjacent PLC nodes.

[0033] Table 2 Comparison of Stability of Cooperative Control Decisions

[0034] As shown in Table 2, after adopting this invention, the number of false triggers decreased from 18 times per shift to 3 times. The main reason is that the dwell time constraint value corrected the initial transfer criterion, and short-term jitter in the input feedback will not directly trigger adjacent PLC nodes to execute collaborative control decisions. The average waiting time for section handover decreased from 186ms to 112ms, indicating that after asynchronous communication time lag compensation, the two sets of sequential control steps can be judged on a time axis closer to the same control transfer process. The cumulative time of short pause lines decreased from 22.6min per shift to 6.8min, and the number of effective output pieces increased from 2870 pieces to 3018 pieces. This indicates that after correcting the transfer criterion and writing it into the boundary communication variable, adjacent PLC nodes can update the step transfer status and output image value more stably, thereby improving the continuity and reliability of the industrial PLC distributed control process.

[0035] The above description is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any equivalent substitutions or modifications made by those skilled in the art within the scope of the technology disclosed in the present invention, based on the technical solution and inventive concept of the present invention, should be covered within the scope of protection of the present invention.

Claims

1. An intelligent decision-making method for industrial PLCs based on distributed control, characterized in that, Includes the following steps: S1. Obtain the process image record and sequential control step record formed by the adjacent PLC node within the corresponding scan cycle, and retrieve the I / O address table and communication variable table of the adjacent PLC node. S2. Establish boundary I / O mapping relationships based on the I / O address table, and determine the boundary communication variables between adjacent PLC nodes according to the communication variable table and the boundary I / O mapping relationships; S3. Based on the boundary I / O mapping relationship, extract the boundary process image sequence corresponding to the boundary I / O mapping relationship from the process image record; S4. Identify the output image change edge and input feedback change edge in the boundary process image sequence, calculate the asynchronous communication time delay difference, and generate boundary transfer records by pairing them. S5. Based on the boundary transfer record, align the two sets of sequential control steps in adjacent PLC nodes that correspond to the boundary process image sequence with the asynchronous communication time delay difference as the compensation benchmark to generate the boundary alignment result. S6. Generate the initial transition criterion based on the boundary alignment results; The dwell time constraint value of the boundary transfer record is calculated using a boundary transfer switching control algorithm based on average dwell time. S7. Correct the initial transfer criterion with the residence time constraint value to form a modified transfer criterion. Write the modified transfer criterion into the boundary communication variable to trigger adjacent PLC nodes to execute collaborative control decisions.

2. The industrial PLC intelligent decision-making method based on distributed control according to claim 1, characterized in that, Specifically, S1 is: According to the scan cycle of adjacent PLC nodes, the input image formed at the beginning of the scan cycle and the output image formed at the end of the scan cycle of adjacent PLC nodes are obtained respectively. The input image, output image, PLC node identifier and scan cycle number are combined to generate process image records for adjacent PLC nodes within the corresponding scan cycle. Extract the sequential control step number, step activation status, and step transfer status corresponding to the scan cycle number from the sequential control program of adjacent PLC nodes, and generate a sequential control step record; Retrieve the I / O address table of the adjacent PLC node. The I / O address table includes the input address, output address, I / O point name, and the PLC node identifier to which the I / O point belongs. Retrieve the communication variable table of the adjacent PLC node. The communication variable table includes the communication variable name, communication variable address, communication variable data type, and the PLC node identifier to which the communication variable belongs.

3. The industrial PLC intelligent decision-making method based on distributed control according to claim 1, characterized in that, Specifically, S2 is: Extract the input address, output address, and PLC node identifier of the adjacent PLC node from the I / O address table; Based on the PLC node identifier to which the I / O point belongs, the input address and output address are matched to generate the boundary I / O mapping relationship; Extract the communication variable addresses and data types corresponding to the boundary I / O mapping relationships from the communication variable table; By combining the communication variable address and the communication variable data type, the boundary communication variables between adjacent PLC nodes are determined.

4. The industrial PLC intelligent decision-making method based on distributed control according to claim 1, characterized in that, Specifically, S3 is: Extract the output address, input address, and corresponding PLC node identifier from the boundary I / O mapping relationship; Based on the output address, extract the output image value under the corresponding scan cycle number from the process image record to generate the output image sequence; Based on the input address, extract the input image value under the corresponding scan cycle number from the process image record to generate an input feedback sequence. The output image sequence and input feedback sequence are arranged according to the scan cycle number to generate a boundary process image sequence corresponding to the boundary I / O mapping relationship.

5. The industrial PLC intelligent decision-making method based on distributed control according to claim 1, characterized in that, Specifically, S4 is: Perform adjacent scan cycle difference processing on the output image sequence in the boundary process image sequence, identify the scan cycle number where the output image value changes, and generate the output image change edge; Perform adjacent scan cycle difference processing on the input feedback sequence in the boundary process image sequence, identify the scan cycle number where the input image value changes, and generate the input feedback change edge; According to the boundary I / O mapping relationship, the output image change edge and the corresponding input feedback change edge are paired to generate change edge pairing items; The difference between the scan cycle number corresponding to the input feedback change edge in the pairing item and the scan cycle number corresponding to the output image change edge is calculated to obtain the asynchronous communication time delay difference. The boundary transfer record is generated by combining the output image change edge, input feedback change edge, asynchronous communication time delay, output address, input address, and corresponding PLC node identifier.

6. The industrial PLC intelligent decision-making method based on distributed control according to claim 1, characterized in that, Specifically, S5 is: Based on the output address, input address and corresponding PLC node identifier, extract two sets of sequential control steps corresponding to the boundary process image sequence from the sequential control step record, and generate the first sequential control step sequence and the second sequential control step sequence. Using the asynchronous communication time delay difference as a compensation benchmark, the scan cycle sequence numbers in the first and second sequential control step sequences are compensated and converted to generate the first and second compensation step sequences. Based on the output image change edge and input feedback change edge in the boundary transfer record, the first compensation step sequence and the second compensation step sequence are time-axis aligned to generate sequential control step registration terms. The sequential control step registration item, boundary transfer record, asynchronous communication time delay difference, and corresponding PLC node identifier are combined to generate the boundary alignment result.

7. The industrial PLC intelligent decision-making method based on distributed control according to claim 1, characterized in that, The process of generating the initial transition criterion based on the boundary alignment result is as follows: The sequential control step registration items, boundary transfer records, and asynchronous communication time delay differences in the boundary alignment results are correlated to form sequential step transfer relationships; Based on the asynchronous communication time delay difference, the scan cycle number in the sequential step transfer relationship is compensated to form a compensated transfer relationship; The first and second sequential control step sequences in the compensation transfer relationship are respectively mapped to the output image change edge and the input feedback change edge in the boundary transfer record to generate the inter-step transfer relationship; The initial transfer criteria are generated by combining the inter-step transfer relationship, step activation status, step transfer status, output address, input address, and corresponding PLC node identifier.

8. The industrial PLC intelligent decision-making method based on distributed control according to claim 1, characterized in that, The boundary transfer switching control algorithm based on average dwell time is used to calculate the dwell time constraint value of the boundary transfer record, specifically as follows: Determine the corresponding scan cycle number of the output image change and the corresponding scan cycle number of the input feedback change from the boundary transfer record; According to the corresponding scan cycle number along the input feedback change, the number of scan cycles in the boundary process image sequence where the input image value remains unchanged is counted, and input feedback is generated to maintain the scan cycle. Within a preset observation window, count the number of boundary transfer records corresponding to the same boundary I / O mapping relationship, and calculate the average dwell time according to the preset observation window length and the number of boundary transfer records; The difference between the input feedback scan cycle and the preset average dwell cycle is calculated to generate a single dwell difference. The difference between the average dwell time and the preset average dwell time is calculated to generate the window dwell time difference. The dwell time constraint value of the boundary transfer record is generated based on the smaller value between the single dwell difference and the window dwell difference.

9. The industrial PLC intelligent decision-making method based on distributed control according to claim 1, characterized in that, Specifically, S7 is: The dwell time constraint value is correlated with the inter-step transfer relationship in the initial transfer criterion to generate a transfer correction term; The step activation state and step transition state in the initial transition criterion are corrected to form the modified transition criterion; Based on the communication variable address and communication variable data type in the boundary communication variables, the data format of the modified transfer criterion is converted to generate communication write data; Write the communication data to the boundary communication variable; Adjacent PLC nodes obtain the corrected transfer criteria from the boundary communication variables, and generate collaborative control decisions based on the corrected transfer criteria, sequential control step records, and process image records; Update the step transition status and output image value of adjacent PLC nodes according to the collaborative control decision.

10. An industrial PLC intelligent decision-making system based on distributed control, executing the industrial PLC intelligent decision-making method based on distributed control as described in any one of claims 1 to 9, characterized in that, Includes the following modules: The record acquisition module is used to acquire the process image record and sequential control step record formed by adjacent PLC nodes within the corresponding scan cycle, and to retrieve the I / O address table and communication variable table of adjacent PLC nodes. The boundary mapping module is used to establish boundary I / O mapping relationships based on the I / O address table, and to determine the boundary communication variables between adjacent PLC nodes according to the communication variable table and the boundary I / O mapping relationships. The image extraction module is used to extract the boundary process image sequence corresponding to the boundary I / O mapping relationship from the process image record based on the boundary I / O mapping relationship; The transfer generation module is used to identify the output image change edge and input feedback change edge in the boundary process image sequence, calculate the asynchronous communication time delay difference, and generate boundary transfer records in pairs. The time alignment module is used to align the two sets of sequential control steps in adjacent PLC nodes that correspond to the boundary process image sequence based on the boundary transfer record, using the asynchronous communication time delay difference as the compensation benchmark, and generate the boundary alignment result. The criterion calculation module is used to generate the initial transition criterion based on the boundary alignment results; The dwell time constraint value of the boundary transfer record is calculated using a boundary transfer switching control algorithm based on average dwell time. The collaborative decision-making module is used to correct the initial transfer criterion with the residence time constraint value, form a modified transfer criterion, write the modified transfer criterion into the boundary communication variable, and trigger adjacent PLC nodes to execute collaborative control decisions.

Images