Digital twin workshop hybrid flow production operation exception monitoring method and system

Abstract

This invention discloses a method and system for monitoring anomalies in mixed-flow production in a digital twin workshop, relating to the application of digital twin workshops in mixed-flow production. Targeting multi-variety, variable-batch mixed-flow production scenarios, this invention first constructs a three-layer production network topology model comprising equipment, workstation, and production line layers, introducing a product-process correlation matrix and a dynamic routing mechanism to uniformly model differentiated production tasks. Then, based on a three-dimensional perspective of "rated-actual-demand," a multi-level health assessment model is constructed. On this basis, hierarchical progressive anomaly monitoring is carried out: at the equipment layer, sudden performance deviations and trend degradation of equipment are identified based on a physical degradation model; at the workstation layer, combined with the characteristics of mixed-flow operations, the mismatch in work cycle time and buffer blockage / starvation states caused by product switching are monitored; at the production line layer, system delivery risks and load imbalance states are assessed based on full-line data.

Description

Technical Field

[0001] This invention pertains to the application of digital twin workshops in mixed-flow production workshops, specifically relating to a method and system for monitoring abnormal operations in mixed-flow production in digital twin workshops. Background Technology

[0002] With the development of intelligent manufacturing and digital twin technology, the production organization mode of manufacturing workshops has gradually evolved from the traditional single-flow production mode to a mixed-flow production mode composed of sequential flow line stations and island-type stations. This type of mixed-flow production system can complete the collaborative manufacturing of multiple varieties and variable batches of products within the same production line. While improving production flexibility and resource utilization, it also significantly increases the complexity of workshop structure topology and production processes.

[0003] In mixed-flow production scenarios, different product varieties have differentiated process paths and operating cycles. The same workstation may undertake different production tasks at different times, and equipment load and production demand exhibit frequent changes. Fluctuations in equipment performance, degradation, and sudden failures can all cause abnormal operating states at the equipment, workstation, or production line levels, thereby affecting the stability and controllability of the production process.

[0004] Existing workshop operation monitoring methods mostly focus on single-equipment status monitoring or production line-level analysis based on statistical indicators. They often use fixed thresholds or empirical rules to determine anomalies, making it difficult to adapt to the dynamic changes and structural diversity of production tasks in mixed-flow production. At the same time, although some methods have introduced the concept of multi-level monitoring, they lack a unified production network modeling framework. The representation methods between equipment operating status, workstation operation capacity, and production line operating capacity are inconsistent, making it difficult to form a unified and comparable description of operating status from multi-level monitoring results. Summary of the Invention

[0005] The technical problem this invention aims to solve is that existing mixed-flow production workshop operation monitoring methods mostly focus on single equipment or single process levels, relying on local state thresholds for anomaly judgment, which makes it difficult to comprehensively reflect the overall operation status of the mixed-flow production system under conditions of multiple varieties and variable batches. Furthermore, some methods employ different monitoring indicators and evaluation methods at the equipment, workstation, and production line levels, lacking a unified modeling framework. This makes it difficult to consistently represent multi-level operation statuses, resulting in insufficient systematicity and interpretability of anomaly monitoring results. In addition, in mixed-flow production scenarios, production cycle times change frequently and equipment loads fluctuate dynamically, making it difficult for single-level monitoring methods to promptly identify the gradually accumulating anomaly risks during production. Therefore, there is an urgent need for a multi-level operation anomaly monitoring method for mixed-flow production structures.

[0006] To address the aforementioned technical problems, this invention proposes a method and system for monitoring operational anomalies in mixed-flow production using a digital twin workshop. This method targets mixed-flow production systems comprising equipment, workstation, and production line layers. It constructs a multi-level production network topology model, introduces a product-process correlation matrix and a dynamic routing mechanism, and achieves unified modeling of multi-variety, variable-batch production tasks. Based on this, a multi-level operational status evaluation model is constructed using a three-dimensional perspective of "rated capacity - actual capacity - demand capacity," progressively characterizing equipment operating status, workstation equivalent operational capacity, and overall production line operational capacity. This invention conducts operational anomaly monitoring at the equipment, workstation, and production line layers: at the equipment layer, it identifies sudden performance anomalies and trend-based degradation anomalies; at the workstation layer, it identifies production status anomalies such as work cycle mismatch, buffer blockage, and starvation based on mixed-flow operation characteristics; at the production line layer, it identifies system delivery risks and operational imbalances based on full-line production status data and remaining order load, thereby achieving layered perception and monitoring of the operational status of the mixed-flow production system. This method can effectively enhance the systematicness and reliability of monitoring anomalies in mixed-flow production systems. It is applicable to the perception of workshop operation status under complex production structures and has good engineering application value.

[0007] The technical problem solved by this invention is achieved by the following technical solution:

[0008] A method for monitoring anomalies in mixed-flow production in a digital twin workshop includes the following steps:

[0009] Step 1: Construct a multi-level production network topology model for mixed-flow production systems; In the three-layer architecture including equipment layer, workstation layer and production line layer, combine the differences in process paths of multiple product varieties and the characteristics of variable batch orders to establish a unified mapping model that reflects the logical relationship between workshop physical resources and production tasks.

[0010] Step 2: Construct a multi-level health assessment model based on the multi-level production network topology model. Based on the three-dimensional perspective of "rated capacity - actual capacity - required capacity", quantify the physical health of the equipment layer, the equivalent operating capacity of the workstation layer, and the network achievable output rate of the production line layer.

[0011] Step 3: Conduct operational anomaly monitoring based on equipment status, compare actual equipment operating parameters with predicted values ​​from health assessment models in real time, and identify sudden performance anomalies and trend-based degradation anomalies in equipment.

[0012] Step 4: Conduct operational anomaly monitoring for workstation production status, and identify workstation-level congestion anomalies, starvation anomalies, and work cycle mismatch anomalies by combining the arrival patterns and product characteristics of the current mixed-flow task flow.

[0013] Step 5: Conduct operational anomaly monitoring for the production process. Based on the overall production status data and the remaining order queue, identify delivery risks and operational imbalances in the mixed-flow production system.

[0014] A digital twin workshop mixed-flow production operation anomaly monitoring system includes:

[0015] The mixed-flow production network modeling module is used to construct a multi-level production network topology model that includes equipment layer, workstation layer and production line layer, and maintain product process association matrix and dynamic routing information;

[0016] The multi-level health assessment module is used to build and update equipment operation status assessment models, workstation capability aggregation models, and production line network flow assessment models.

[0017] The equipment-level anomaly monitoring module is used to identify sudden failures and degradation trends at the physical level of the equipment.

[0018] The workstation-level anomaly monitoring module is used to identify workstation cycle time anomalies and logistics congestion in mixed-flow production mode.

[0019] The production line-level anomaly monitoring module is used to identify system-level order delivery risks and production line load imbalances.

[0020] An electronic device includes: one or more processors; and a memory for storing one or more programs, wherein when the one or more programs are executed by the one or more processors, the one or more processors implement the digital twin workshop mixed-flow production operation anomaly monitoring method.

[0021] A computer-readable storage medium having executable instructions stored thereon, which, when executed by a processor, cause the processor to implement the aforementioned method for monitoring abnormal operation of mixed-flow production in a digital twin workshop.

[0022] A computer program product includes a computer program that, when executed by a processor, implements the aforementioned method for monitoring abnormal operations in mixed-flow production in a digital twin workshop.

[0023] Beneficial effects:

[0024] 1. This invention constructs a multi-level mixed-flow production network topology model consisting of equipment layer, workstation layer, and production line layer, thereby achieving unified modeling of complex mixed-flow production structures and their dynamic production tasks, and improving the completeness of the description of workshop operation status.

[0025] 2. By adopting the three-dimensional perspective of "rated-actual-demand", dynamic coupling assessment of equipment physical status and production task requirements is realized.

[0026] 3. This invention can effectively handle the complex parallel redundancy relationship of island-type workstations in mixed-flow production lines, and improve the accuracy of bottleneck identification.

[0027] 4. This invention achieves effective identification of equipment status abnormalities, workstation production abnormalities, and production line operation abnormalities through a hierarchical and progressive operation anomaly monitoring mechanism, thereby improving the real-time performance and reliability of mixed-flow production system operation monitoring. Attached Figure Description

[0028] Figure 1 This is a schematic diagram of a method for monitoring abnormal operation of mixed-flow production in a digital twin workshop according to the present invention;

[0029] Figure 2 This is a detailed flowchart of the method of the present invention. Detailed Implementation

[0030] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention. Furthermore, the technical features involved in the various embodiments of this invention described below can be combined with each other as long as they do not conflict with each other. To achieve the above objectives, this invention adopts the following technical solution.

[0031] like Figure 1 As shown, this invention provides a method for monitoring abnormal operations in mixed-flow production in a digital twin workshop. The specific implementation is as follows:

[0032] Step 1: Construct a multi-level production network topology model for mixed-flow production systems to achieve unified modeling of the production structure and process of the digital twin workshop; in the three-layer architecture including equipment layer, workstation layer, and production line layer, combine the differences in process paths for multiple product varieties and the characteristics of variable batch orders to establish a unified mapping model reflecting the logical relationship between workshop physical resources and production tasks; the specific implementation method is as follows:

[0033] At the device layer, define physical device nodes and their rated design parameters;

[0034] At the workstation level, an ownership matrix of equipment and workstations is established to distinguish the internal topology of serial pipeline workstations and parallel island workstations; a dynamic operation standard matrix is ​​constructed to define the standard operation time for each workstation when processing different types of products.

[0035] At the production line level, a production line network diagram containing a dynamic routing matrix is ​​constructed. This dynamic routing matrix describes the differentiated logistics paths of different product varieties between workstation nodes, characterizing the production structure of different processes on the same line in mixed-flow production. Specifically, this includes:

[0036] To address the hybrid structural characteristics of the mixed-flow assembly workshop, which includes both rigid production lines and flexible processing islands, as well as the order characteristics of multiple product types and variable batch sizes, a three-layer topology structure comprising equipment, workstation, and production line layers is constructed. .in, A global network topology diagram representing the physical resources and logical relationships of the entire mixed-flow production workshop; For a set of network nodes, It includes device node sets and workstation node sets; This is a set of network edges, representing the physical connections or logical affiliations between nodes.

[0037] At the equipment layer, single-functional units in the physical workshop (such as robots, tightening guns, AGVs, etc.) are mapped to equipment nodes. ,in This represents the collection of all physical equipment within the workshop. For the first Individual device nodes; at the workstation level, device nodes are aggregated into workstation nodes based on their physical location and process attributes. Establish an attribution matrix for equipment and workstations, where Represents the set of all workstations. For the first Individual workstation nodes.

[0038] In particular, to characterize the process differences of "same station, different time" in mixed-flow production, this embodiment constructs a dynamic product-process correlation matrix. :

[0039] ;

[0040] in, Indicates workstation Processing the The standard rated operating time for a certain type of product; if the workstation does not involve the process for that product, then... At the production line layer, define directed edges between workstation nodes. Based on the variety sequence of the current production task, a dynamic routing matrix is ​​constructed. It is used to characterize the differentiated logistics paths of different product varieties between workstations.

[0041] Step 2: Construct a multi-level health assessment model based on a multi-level production network topology model. Using a three-dimensional perspective of "rated capacity - actual capacity - demand capacity," quantify the physical health of the equipment layer, the equivalent operational capacity of the workstation layer, and the network achievable output rate of the production line layer; achieve a comprehensive representation of equipment operating status, workstation equivalent production capacity, and production line operating capacity. This includes: at the equipment layer, using a stochastic degradation process to describe the time-varying trajectory of the equipment's actual production capacity and calculating the real-time health of the equipment; at the workstation layer, based on the serial or parallel topology relationships of the equipment within the workstation, aggregating the equipment layer health into the workstation's equivalent production capacity for a specific product at the current moment; at the production line layer, based on the network maximum flow theory, combining the equivalent production capacity of each workstation with the current dynamic routing structure, calculating the maximum theoretical output rate of the production line network in the current state. The specific implementation method is as follows:

[0042] a) Equipment-level health assessment model: A Wiener process with drift is used to describe the stochastic degradation of actual equipment production capacity. exist Performance degradation at any moment for:

[0043] ;

[0044] Defining device health based on degradation status. To characterize the long-term evolution of the equipment:

[0045] ;

[0046] in, It represents the drift rate, reflecting the average rate of equipment degradation, which is affected by load and environment. It represents the diffusion coefficient, which reflects the intensity of random fluctuations in the equipment degradation process. This is the preset failure threshold.

[0047] b) Workstation-level capacity aggregation model: Calculates the capacity aggregation of workstations based on their internal structure. equivalent production capacity .

[0048] For pipeline workstations with a serial architecture, the workstation's capacity is determined by the device with the lowest health status:

[0049] ;

[0050] For an island-type workstation with a parallel structure, the workstation's capacity is the sum of the capacities of all its internal parallel devices:

[0051] ;

[0052] in, For equipment The rated design capacity is determined by the equipment's factory parameters or historical best operating data.

[0053] c) Production Line Layer Network Health Assessment: Based on network topology and workstation capabilities, assess the overall system operational level. Equivalent capabilities of each workstation are compared. As network node weights, the maximum theoretical output rate of the production line under the current mixed-flow route is calculated based on the Max-FlowMin-Cut Theorem. :

[0054] ;

[0055] Step 3: Conduct operational anomaly monitoring based on equipment status, comparing actual equipment operating parameters with predicted values ​​from the health assessment model in real time to identify sudden performance anomalies and trend-based degradation anomalies. This includes: based on the equipment-level operational status assessment model, monitoring the matching degree between the actual equipment production capacity and the current production task requirements in real time, calculating the normalized deviation value of the actual equipment production capacity relative to the real-time load demand, and determining a sudden performance anomaly when the deviation value is lower than a preset negative anomaly deviation threshold; and updating the drift rate parameter of the equipment degradation process in real time using the maximum likelihood estimation method, determining a trend-based degradation anomaly when the slope of the drift rate change exceeds a preset acceleration factor. The specific implementation method is as follows:

[0056] a) Monitoring of sudden performance anomalies: Real-time monitoring of the actual capabilities of the equipment. In line with current production task requirements The degree of matching. An anomaly is determined when the following conditions are met:

[0057] ;

[0058] in, To allocate equipment according to the mixed-flow scheduling plan Real-time load; This is the threshold for abnormal deviation.

[0059] b) Health trend degradation monitoring: Equipment health based on Wiener process prediction Monitor its degradation rate. If the drift rate The presence of non-linear acceleration indicates that the equipment has entered a period of accelerated wear.

[0060] ;

[0061] in, The threshold for accelerated degradation.

[0062] Step 4: Conduct operational anomaly monitoring for workstation production status. Based on the characteristics of multi-product mixed-flow operations, identify workstation-level congestion, starvation, and work cycle mismatch anomalies, including: Mixed-flow cycle mismatch monitoring: Obtain the product type information currently being processed at the workstation and match it with the corresponding standard operating time; calculate the cycle deviation index between the actual operating cycle and the standard operating time. When the deviation index exceeds the mixed-flow tolerance range and there are no equipment faults, it is determined to be a work cycle mismatch anomaly; Buffer status monitoring: Monitor the work-in-process queue length of the workstation's input and output buffers in real time. When the output buffer queue length exceeds the dynamic upper limit and the downstream workstation is busy, it is determined to be a congestion anomaly. When the input buffer queue length is below the dynamic lower limit and the upstream workstation has output, it is determined to be a starvation anomaly. Specific implementation methods are as follows:

[0063] This step identifies logical anomalies caused by differences in the mixed-flow process, assuming there are no alarms at the equipment level.

[0064] First, perform mixed-flow cycle time mismatch monitoring. The system identifies workstations. Product varieties currently being processed Retrieve standard operating time Combined with the actual operation cycle Calculate the beat deviation index :

[0065] ;

[0066] like (Mixed flow tolerance threshold) is judged as an abnormal cycle mismatch, indicating insufficient personnel proficiency or material delivery delay.

[0067] Secondly, perform congestion and starvation monitoring. Monitor the length of the workstation input buffer in real time. and output buffer length .when Furthermore, when downstream workstations are busy, it is determined to be a congestion anomaly; when Furthermore, if the upstream station produces output, it is considered an abnormal starvation. This is the maximum capacity threshold for the set buffer.

[0068] Step 5: Conduct operational anomaly monitoring for the production process. Based on full-line production data, identify delivery risks and operational imbalances in the mixed-flow production system. This includes: performing cross-layer anomaly correlation analysis based on the capacity mapping and transmission relationship at the workstation level to achieve comprehensive perception of the mixed-flow production operation status. Delivery risk identification: Calculate the total load hours required to complete the remaining mixed-flow orders, and predict the order completion time based on the current maximum theoretical output rate of the production line. When the predicted completion time exceeds the safety confidence interval corresponding to the order deadline, it is determined as a system delivery risk anomaly. Operational imbalance identification: Calculate the real-time load rate of each workstation in the production line and calculate the overall load dispersion coefficient. When the dispersion coefficient exceeds a preset balance threshold, it is determined as a system operational imbalance anomaly. Specific implementation methods are as follows:

[0069] a) Delivery risk monitoring: Obtain the remaining order queue Calculate the total load hours required to complete the remaining mixed-flow orders. :

[0070] ;

[0071] in, Bottleneck processing station Working hours. Based on the current system output rate. Predicted completion time : .like (Order deadline) was deemed an abnormal delivery risk. Among them, This indicates the current time in the system.

[0072] b) Operational imbalance monitoring: Calculate the real-time load rate of each workstation in the production line. It is defined as the ratio of the actual operating time of the workstation to the available capacity. Then, the load dispersion factor for the entire line is calculated. :

[0073] ;

[0074] in, For the number of workstations, This represents the average load rate of all stations along the entire line. When... When the balance threshold is reached, it is judged as an operational imbalance anomaly, indicating that the current mixed-flow scheduling scheme has led to a polarization of resource utilization, and scheduling optimization and adjustment need to be triggered.

[0075] This embodiment provides a digital twin workshop mixed-flow production operation anomaly monitoring system based on the above method, including:

[0076] The mixed-flow production network modeling module is used to construct a multi-level production network topology model that includes equipment layer, workstation layer and production line layer, and maintain product process association matrix and dynamic routing information;

[0077] The multi-level health assessment module is used to build and update equipment operation status assessment models, workstation capability aggregation models, and production line network flow assessment models.

[0078] The equipment-level anomaly monitoring module is used to identify sudden failures and degradation trends at the physical level of the equipment.

[0079] The workstation-level anomaly monitoring module is used to identify workstation cycle time anomalies and logistics congestion in mixed-flow production mode.

[0080] The production line-level anomaly monitoring module is used to identify system-level order delivery risks and production line load imbalances.

[0081] As shown in Figure 2, the overall implementation process of the method of the present invention includes three progressive stages: production network modeling, multi-level health assessment, and multi-level anomaly monitoring.

[0082] 1. Production network modeling stage: Construct a three-layer topology structure including equipment layer, workstation layer and production line layer.

[0083] Based on equipment node mapping, physical resources are digitized; a product-process association matrix is ​​introduced to define the process logic of different products; and a dynamic routing matrix is ​​constructed to describe the production line logistics path.

[0084] 2. Multi-level health assessment stage: Stratified assessment is conducted based on the above model.

[0085] (1) At the equipment level, calculate the real-time health of the equipment based on the Wiener degradation process;

[0086] (2) At the workstation level, the equivalent production capacity of the workstation is calculated by combining the topological aggregation relationship;

[0087] (3) At the production line level, the maximum theoretical output rate of the system is calculated based on the maximum flow-minimum cut theorem.

[0088] 3. Multi-level anomaly monitoring stage: Conduct tiered monitoring based on the assessment results.

[0089] (1) Equipment status monitoring: Determine whether the drift rate is nonlinearly accelerated to identify trend degradation; determine whether the deviation value is greater than the threshold to identify sudden performance abnormalities.

[0090] (2) Production status monitoring: Determine whether the queue length exceeds the upper and lower limits to identify blockage / starvation anomalies; determine whether the takt deviation index is greater than the tolerance to identify mixed-flow takt mismatch.

[0091] (3) Production process monitoring: Determine whether the load dispersion coefficient is greater than the threshold to identify operational imbalance; determine whether the predicted completion time is greater than the deadline to identify delivery risks.

[0092] Through the collaborative work of the above three stages, a comprehensive closed-loop monitoring system was achieved, encompassing everything from physical resources to underlying logic and system performance.

[0093] An electronic device includes: one or more processors; and a memory for storing one or more programs, wherein when the one or more programs are executed by the one or more processors, the one or more processors implement the digital twin workshop mixed-flow production operation anomaly monitoring method.

[0094] A computer-readable storage medium having executable instructions stored thereon, which, when executed by a processor, cause the processor to implement the aforementioned method for monitoring abnormal operation of mixed-flow production in a digital twin workshop.

[0095] A computer program product includes a computer program that, when executed by a processor, implements the aforementioned method for monitoring abnormal operations in mixed-flow production in a digital twin workshop.

Claims

1. A method for monitoring abnormal operations in mixed-flow production in a digital twin workshop, characterized in that, Includes the following steps: Step 1: Construct a multi-level production network topology model for mixed-flow production systems; In the three-layer architecture including equipment layer, workstation layer and production line layer, combine the differences in process paths of multiple product varieties and the characteristics of variable batch orders to establish a unified mapping model that reflects the logical relationship between workshop physical resources and production tasks. Step 2: Construct a multi-level health assessment model based on the multi-level production network topology model. Based on the three-dimensional perspective of "rated capacity - actual capacity - required capacity", quantify the physical health of the equipment layer, the equivalent operating capacity of the workstation layer, and the network achievable output rate of the production line layer. Step 3: Conduct operational anomaly monitoring based on equipment status, compare actual equipment operating parameters with predicted values ​​from health assessment models in real time, and identify sudden performance anomalies and trend-based degradation anomalies in equipment. Step 4: Conduct operational anomaly monitoring for workstation production status, and identify workstation-level congestion anomalies, starvation anomalies, and work cycle mismatch anomalies by combining the arrival patterns and product characteristics of the current mixed-flow task flow. Step 5: Conduct operational anomaly monitoring for the production process. Based on the overall production status data and the remaining order queue, identify delivery risks and operational imbalances in the mixed-flow production system.

2. The method for monitoring abnormal operation of mixed-flow production in a digital twin workshop as described in claim 1, characterized in that, Step 1, which involves constructing a multi-level production network topology model, specifically includes: The digital twin workshop is structurally deconstructed, and a three-layer mixed-flow production network topology model including equipment layer, workstation layer and production line layer is constructed. At the device layer, define physical device nodes and their rated design parameters; At the workstation level, an ownership matrix of equipment and workstations is established to distinguish the internal topology of serial pipeline workstations and parallel island workstations; a dynamic operation standard matrix is ​​constructed to define the standard operation time for each workstation when processing different types of products. At the production line level, a production line network diagram containing a dynamic routing matrix is ​​constructed. The dynamic routing matrix is ​​used to describe the differentiated logistics paths of different product varieties between workstation nodes, so as to characterize the production structure of different processes on the same line in mixed-flow production.

3. The method for monitoring abnormal operation of mixed-flow production in a digital twin workshop according to claim 1, characterized in that, Step 2, constructing a multi-level health assessment model, specifically includes: At the equipment layer, a stochastic degradation process is used to describe the time-varying trajectory of the actual production capacity of the equipment, and the real-time health of the equipment is calculated. At the workstation layer, based on the serial or parallel topology relationship of the equipment within the workstation, the health of the equipment layer is aggregated into the equivalent production capacity of the workstation for a specific product at the current moment. At the production line layer, based on the network maximum flow theory, combined with the equivalent production capacity of each workstation and the current dynamic routing structure, the maximum theoretical output rate of the production line network in the current state is calculated.

4. The method for monitoring abnormal operation of mixed-flow production in a digital twin workshop according to claim 1, characterized in that, Step 3, which involves monitoring operational anomalies based on equipment status, specifically includes: Based on the equipment-level operation status assessment model, the matching degree between the actual production capacity of the equipment and the current production task requirements is monitored in real time. The normalized deviation value of the actual production capacity of the equipment relative to the real-time load requirements is calculated. When the deviation value is lower than the preset negative abnormal deviation threshold, it is determined to be a sudden performance abnormality of the equipment. The drift rate parameter of the equipment degradation process is updated in real time using the maximum likelihood estimation method. When the slope of the drift rate change exceeds the preset acceleration factor, it is determined to be a trend degradation abnormality.

5. The method for monitoring abnormal operation of mixed-flow production in a digital twin workshop according to claim 1, characterized in that, Step 4, which involves monitoring operational anomalies related to the production status of workstations, specifically includes: Mixed-flow cycle mismatch monitoring: Obtain the product type information of the current processing product in the workstation and match the corresponding standard operation time; calculate the cycle deviation index between the actual operation cycle and the standard operation time. When the deviation index exceeds the mixed-flow tolerance range and there is no fault in the equipment layer, it is determined to be an abnormal cycle mismatch. Buffer status monitoring: Real-time monitoring of the work-in-process queue lengths of the input and output buffers of the workstations. When the output buffer queue length exceeds the dynamic upper limit and the downstream workstation is busy, it is determined to be a blocking anomaly. When the input buffer queue length is below the dynamic lower limit and the upstream workstation has output, it is determined to be a starvation anomaly.

6. The method for monitoring abnormal operation of mixed-flow production in a digital twin workshop according to claim 1, characterized in that, Step 5, which involves monitoring operational anomalies in the production process, specifically includes: Based on the capability mapping and transmission relationship of the workstation layer, cross-layer anomaly correlation analysis is performed to achieve comprehensive perception of the mixed-flow production operation status. Delivery risk identification: Calculate the total workload required to complete the remaining mixed-flow orders, and predict the order completion time based on the current maximum theoretical output rate of the production line; when the predicted completion time exceeds the safety confidence interval corresponding to the order deadline, it is determined that the system delivery risk is abnormal; Operational imbalance identification: Calculate the real-time load rate of each workstation in the production line and calculate the load dispersion coefficient of the entire line; when the dispersion coefficient exceeds the preset balance threshold, it is determined that the system is operating abnormally.

7. A digital twin workshop mixed-flow production operation anomaly monitoring system, used to implement the digital twin workshop mixed-flow production operation anomaly monitoring method according to any one of claims 1 to 6, characterized in that, include: The mixed-flow production network modeling module is used to construct a multi-level production network topology model that includes equipment layer, workstation layer and production line layer, and maintain product process association matrix and dynamic routing information; The multi-level health assessment module is used to build and update equipment operation status assessment models, workstation capability aggregation models, and production line network flow assessment models. The equipment-level anomaly monitoring module is used to identify sudden failures and degradation trends at the physical level of the equipment. The workstation-level anomaly monitoring module is used to identify workstation cycle time anomalies and logistics congestion in mixed-flow production mode. The production line-level anomaly monitoring module is used to identify system-level order delivery risks and production line load imbalances.

8. An electronic device, characterized in that, include: One or more processors; A memory for storing one or more programs, wherein when the one or more programs are executed by the one or more processors, the one or more processors cause the one or more processors to implement the digital twin workshop mixed-flow production operation anomaly monitoring method according to any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that, It stores executable instructions that, when executed by a processor, cause the processor to implement the digital twin workshop mixed-flow production operation anomaly monitoring method as described in any one of claims 1 to 6.

10. A computer program product, comprising a computer program, characterized in that, When executed by a processor, the computer program implements the method for monitoring abnormal operation of mixed-flow production in a digital twin workshop as described in any one of claims 1 to 6.

Images