Factory real-time capacity coordination method and system based on MES system

Abstract

The invention relates to a factory real-time capacity coordination method and system based on an MES system. The method comprises the steps of obtaining production data of each production link of a factory in a preset time period; acquiring equipment data of each production equipment of the factory in a preset time period; performing anomaly detection by using a machine learning model based on the production data and the equipment data; when abnormity is detected, abnormal data are extracted from the production time sequence data and / or the local equipment data, feature extraction is carried out based on the abnormal data and the production plan, and production fusion features are determined; and finally, determining a productivity coordination strategy based on the production fusion features, the production time sequence data and the local equipment data. Through comprehensive acquisition and analysis of production data, accurate anomaly detection and prediction and intelligent production capacity coordination strategy formulation, production efficiency can be greatly improved, faults and resource waste are reduced, equipment use is optimized, and overall coordination and flexibility of production are improved.

Description

Technical Field

[0001] The present application relates to the field of production management, and in particular, to a method, system, device and storage medium for real-time production capacity coordination in a factory based on an MES system. Background Art

[0002] With the rapid development of the manufacturing industry and the complexity of production models, many enterprises are facing problems such as production capacity bottlenecks, uneven production scheduling, and equipment idleness. Traditional production capacity coordination methods usually rely on manual or static production plan arrangements, and fail to effectively consider the real-time changes and dynamic adjustment requirements in the production process. Although the MES (Manufacturing Execution) system can collect real-time production data, how to dynamically adjust production tasks and coordinate the production capacity of each production unit based on these data is still a difficult point in the current technology.

[0003] In the prior art, although there are some production capacity scheduling methods based on data analysis, there is a lack of an effective real-time dynamic production capacity coordination mechanism, and accurate production scheduling cannot be achieved. Especially when facing emergencies such as equipment failures and material shortages, production plans often cannot be adjusted in time, resulting in low production efficiency or resource waste. Summary of the Invention

[0004] The present application provides a method, system and storage medium for real-time production capacity coordination in a factory based on an MES system. By collecting and analyzing production data in real time, dynamically adjusting production tasks, adjusting production equipment and allocating resources, the purpose of maximizing production efficiency can be achieved.

[0005] In a first aspect, the present application provides a method for real-time production capacity coordination in a factory based on an MES system, the method comprising:

[0006] Obtaining production data of each production link in the factory within a preset time period;

[0007] Constructing production timing data based on the production data and the production process flow;

[0008] Obtaining equipment data of each production equipment in the factory within the preset time period; the equipment data includes equipment real-time power, equipment temperature, equipment vibration, equipment current and equipment voltage;

[0009] Constructing local equipment data based on the production process flow and the equipment data;

[0010] Performing anomaly detection using a machine learning model based on the production timing data, the local equipment data and the production plan;

[0011] In response to detecting an anomaly, extracting anomaly data from the production timing data and / or the local equipment data;

[0012] Extract features based on the abnormal data and the production plan to determine production integration features;

[0013] Based on the production integration features, the production time series data, and the local equipment data, determine a production capacity coordination strategy; wherein, the production capacity coordination strategy includes production equipment maintenance, production equipment power adjustment, and production equipment working time adjustment.

[0014] Further, the construction of the production time series data based on the production data and the production process flow includes:

[0015] Divide the preset time period into time nodes to obtain a plurality of production nodes;

[0016] Obtain the production capacity data of each production equipment at each production node to obtain a production capacity vector;

[0017] Combine the production capacity vectors of each production equipment based on the production process flow to obtain the production time series data; wherein, each element of the production time series data represents the production capacity at the corresponding production node.

[0018] Further, the construction of the local equipment data based on the production process corresponding to each production equipment includes:

[0019] Obtain the equipment data corresponding to each production equipment at each production node to obtain an equipment data vector;

[0020] Group the production equipment based on the production process flow;

[0021] Combine the equipment data vectors corresponding to the production equipment in each group to obtain local equipment data.

[0022] Further, the extraction of abnormal data from the production time series data and / or the local equipment data in response to detecting an abnormality includes:

[0023] Determine the type of abnormality; wherein, the type of abnormality includes equipment abnormality and equipment production capacity abnormality;

[0024] Based on the type of abnormality, use an abnormal data location model to determine the location of the abnormal data;

[0025] Extract the abnormal data from the production time series data and / or the local equipment data based on the location of the abnormal data.

[0026] Further, the determination of the location of the abnormal data based on the type of abnormality by using an abnormal data location model includes:

[0027] In response to the abnormal type being the device abnormality, input the local device data into the abnormal data location model to determine the location of the abnormal data;

[0028] and / or,

[0029] In response to the abnormal type being the device production capacity abnormality, input the production timing data into the abnormal data location model to determine the location of the abnormal data.

[0030] Further, the feature extraction is performed based on the abnormal data and the production plan to determine the production fusion features, including:

[0031] Perform feature extraction on the abnormal data to obtain abnormal data features;

[0032] Perform feature extraction on the production plan to obtain production plan features;

[0033] Obtain a first weight corresponding to the abnormal data features and obtain a second weight corresponding to the production plan features;

[0034] Fuse the abnormal data features and the production plan features based on the first weight and the second weight to determine the production fusion features.

[0035] Further, the obtaining of the first weight corresponding to the abnormal data features and the obtaining of the second weight corresponding to the production plan features include:

[0036] Obtain a first initial weight of the abnormal data features and a second initial weight of the production plan features through the following standard variance calculation formula;

[0037]

[0038] where W represents the calculated initial weight, x is the abnormal data feature or the production plan feature, and i represents the i-th feature in the abnormal data feature or the production plan feature;

[0039] Determine a first coefficient for adjusting the first initial weight and a second coefficient for adjusting the second initial weight based on the following weight adjustment formula;

[0040]

[0041] where st is the abnormal score of the timing anomaly detection, s0 is the threshold; β is an adjustable parameter for controlling the degree of weight change, and α is the calculated dynamic adjustment value;

[0042] Adjust the first initial weight based on the first coefficient and adjust the second initial weight based on the second coefficient to determine the first weight and the second weight.

[0043] In a second aspect, the present application provides a factory real-time production capacity coordination system based on an MES system. The system includes:

[0044] A first acquisition module, configured to acquire production data of each production link in the factory within a preset time period;

[0045] A first construction module, configured to construct production timing data based on the production data and the production process flow;

[0046] A second acquisition module, configured to acquire equipment data of each production device in the factory within the preset time period; the equipment data includes equipment real-time power, equipment temperature, equipment vibration, equipment current, and equipment voltage;

[0047] A second construction module, configured to construct local equipment data based on the production process flow and the equipment data;

[0048] An anomaly detection module, configured to perform anomaly detection using a machine learning model based on the production timing data, the local equipment data, and the production plan;

[0049] An anomaly data extraction module, configured to extract anomaly data from the production timing data and / or the local equipment data in response to detecting an anomaly;

[0050] A feature extraction module, configured to perform feature extraction based on the anomaly data and the production plan to determine production fusion features;

[0051] A coordination module, configured to determine a production capacity coordination strategy based on the production fusion features, the production timing data, and the local equipment data; wherein, the production capacity coordination strategy includes production equipment maintenance, production equipment power adjustment, and production equipment working time adjustment.

[0052] In a third aspect, an electronic device is provided, including a processor, a communication interface, a memory, and a communication bus. Among them, the processor, the communication interface, and the memory complete communication with each other through the communication bus;

[0053] The memory is used to store a computer program;

[0054] The processor, when executing the program stored in the memory, implements the steps of the method for coordinating the real-time production capacity of a factory based on an MES system according to any one of the embodiments of the first aspect.

[0055] In a fourth aspect, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the steps of the method for coordinating the real-time production capacity of a factory based on an MES system according to any one of the embodiments of the first aspect are implemented.

[0056] The above technical solution provided by the embodiments of the present application has the following advantages compared with the prior art: The factory real-time production capacity coordination method based on the MES system proposed by the present invention can greatly improve production efficiency, reduce failures and resource waste, optimize equipment usage, and enhance the overall coordination and flexibility of production capacity through comprehensive collection and analysis of production data, accurate anomaly detection and prediction, and intelligent production capacity coordination strategy formulation. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] The accompanying drawings herein are incorporated into the specification and constitute a part of this specification, showing embodiments consistent with the present invention and, together with the specification, are used to explain the principles of the present invention.

[0058] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the accompanying drawings required for use in the description of the embodiments or the prior art. Obviously, for those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0059] Figure 1 It is a schematic flowchart of a factory real-time production capacity coordination method based on the MES system provided by the embodiments of the present application;

[0060] Figure 2 It is a schematic flowchart of a method for determining production fusion features provided by the embodiments of the present application;

[0061] Figure 3 It is a schematic structural diagram of an electronic device provided by the embodiments of the present application. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0062] To make the objectives, technical solutions, and advantages of the embodiments of the present application clearer, the following will clearly and completely describe the technical solutions in the embodiments of the present application with reference to the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are some, but not all, of the embodiments of the present application. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present application without creative efforts fall within the scope of protection of the present application.

[0063] Figure 1 It is a schematic flowchart of a factory real-time production capacity coordination method based on the MES system provided by the embodiments of the present application. As Figure 1 shown, process 100 can be executed by an electronic device or a factory real-time production capacity coordination system based on the MES system. In some embodiments, process 100 may include the following operations.

[0064] Step 101, obtain production data of each production link in the factory within a preset time period.

[0065] The preset time period refers to a time range, such as hours, days, weeks, etc., during which the factory collects and monitors production data. For example, it can be the time period of one hour from "XX:XX:XX, XX / XX / XXXX" to "XX:XX:XX, XX / XX / XXXX".

[0066] The production link refers to various different production processes or working stages in the factory, which can include raw material preparation, processing, assembly, inspection, etc. For example, in the process of automobile manufacturing, the production links include "body welding", "engine assembly", etc.

[0067] In some embodiments, the system can obtain and store production data within the preset time period through the real-time connection of the MES (Manufacturing Execution System) with production equipment, sensors, process flows, etc. The production data can include data in multiple dimensions such as production volume, production speed, production quality, etc. For example, the system can obtain real-time process data of each production link through sensors, such as "the number of welding times in the body welding link" or "the assembly time and quantity in the engine assembly link", etc.

[0068] Step 102, construct production time-series data based on the production data and the production process flow.

[0069] The production process flow refers to the steps and sequence of each link in the production process, which reflects the specific operation specifications of product production in the factory. For example, in the mobile phone assembly process, the production process flow can include links such as "unpacking, inspection, assembly, testing, packaging".

[0070] Production time-series data arranges the production data of each production link in chronological order to form a data set with a time dimension. For example, production data can be recorded once per minute to form time-series data. Another example is that production data can be combined with the production process flow, and the data can be organized in chronological order to construct the performance of each production link in the time dimension.

[0071] In some embodiments, the constructing of production time-series data based on the production data and the production process flow may include the following operations.

[0072] S10, divide the preset time period into time nodes to obtain multiple production nodes.

[0073] A time node refers to certain specific time points within the preset time period, which are used to divide the state of the production process or the execution time of specific tasks. For example, the production process can be monitored and data can be recorded in real time every 5 minutes, with 5 minutes as a time node, or the whole points of each day, such as 8:00, 9:00, 10:00, etc. of each day, can be used as time nodes to divide the production process.

[0074] A production node refers to a specific time point of a production link divided according to time nodes within a preset time period, representing a specific stage in the production process. For example, the nodes of production line A: start work at 8:00, conduct production volume statistics at 9:00, pause at 12:00, end at 16:00, etc.

[0075] In some embodiments, time nodes can be set according to the rhythm of the production process and monitoring requirements. For example, if precise production data is needed, a time node can be selected every minute or every 5 minutes. If only the production trend is concerned, a time node can be selected every hour or every shift.

[0076] In some embodiments, the preset time period can be divided into multiple time nodes according to the set time interval. For example, if the preset time period is 1 hour and each 5 minutes is a node, then there will be 12 production nodes within that hour. Assuming the preset time period is 1 hour and the time node is divided into one node every 5 minutes, the production nodes within this preset time period can be: 08:00 - 08:05, 08:05 - 08:10, 08:10 - 08:15, …… and so on.

[0077] S11, obtain the production capacity data of each production device at each production node to obtain a production capacity vector.

[0078] Production capacity data refers to the production capacity of a production device at a specific time point or time period, which can be expressed as the output per unit time of the device or its effective operating state. For example, the production capacity can be 100 parts per hour. The effective operating state can be the load percentage or operating efficiency of the device at a certain time node.

[0079] A production capacity vector refers to a set of production capacity data of each production device under each production node. The production capacity data can be represented by a vector, where each component represents the production capacity of a device at this production node.

[0080] For example, for a certain production node, the production capacity of device A is 100 units / hour, and the production capacity of device B is 200 units / hour, then the production capacity vector of this node is [100, 200].

[0081] In some embodiments, at each production node, the production capacity data of each production device at this node can be obtained through the MES system or the device monitoring system. The obtained production capacity data is saved according to the production node. For example, for each production node, the production capacity data of each device such as Device A, Device B, Device C, etc. is recorded. For each time node, a vector containing the production capacity of each device is formed, and the production capacity data of all production nodes can be summarized into a matrix or an array. For example, assume that at a certain production node (08:00 - 08:05), there are three devices: Device A, Device B, and Device C, and their production capacities are: Device A, 100 pieces per hour; Device B, 150 pieces per hour; Device C, 50 pieces per hour. Then the production capacity vector at this production node is: Production capacity vector = [100, 150, 50].

[0082] S12. Combine the production capacity vectors of each production device based on the production process flow to obtain the production timing data.

[0083] Among them, each element of the production timing data represents the production capacity under the corresponding production node.

[0084] Combination means integrating the production capacity data of different production links (devices) according to the production process flow to obtain the complete production capacity. For example, in the production process flow of an assembly line, Device A is responsible for welding, Device B is responsible for assembly, and Device C is responsible for testing. For a certain production node, the combination of production capacity is the merger of the production capacity vectors of these three devices. The production capacity vectors of each device at each time node are merged, and the production capacity combination data of all production nodes can be stored in the form of a matrix, an array, or a table.

[0085] Finally, summarize and generate the final production timing data. It can be a two-dimensional data structure (matrix), where each row represents a production node and each column represents the production capacity of a device.

[0086] Example: Assume that there are three devices A, B, and C in a certain production process flow, and the production capacity vectors of each device are as follows:

[0087] At the time node 08:00 - 08:05, the production capacity of Device A is 100 pieces, the production capacity of Device B is 150 pieces, and the production capacity of Device C is 50 pieces.

[0088] At the time node 08:05 - 08:10, the production capacity of Device A is 110 pieces, the production capacity of Device B is 160 pieces, and the production capacity of Device C is 55 pieces.

[0089] The production timing data is a matrix or an array obtained by combining the above production capacity quantities.

[0090] Step 103. Obtain the device data of each production device in the factory within the preset time period.

[0091] Production equipment refers to various machines and equipment involved in the production process, such as numerical control machine tools, welding equipment, assembly robots, etc.

[0092] Equipment data refers to the real-time monitoring data of each device within a preset time period, including the real-time power, temperature, vibration, current, and voltage of the device. Equipment data reflects the health status and operating efficiency of the device.

[0093] In some embodiments, by docking with the real-time data acquisition system of the device, the key operation data of each production device within a preset time period can be obtained, including information such as device power, temperature, vibration, current, and voltage, to obtain equipment data.

[0094] Step 104, based on the production process flow and the equipment data, construct local equipment data.

[0095] Local equipment data refers to the equipment data related to a specific production link. For example, the power, temperature, etc. data of the welding equipment are the local equipment data of the "body welding" link. Each production link is associated with specific equipment data.

[0096] In some embodiments, according to the production process flow, the equipment data related to a certain link can be selected and extracted to form local equipment data. The local equipment data will reflect the operating state of the equipment in a specific production link. For example, for the "body welding" link, the power, temperature, etc. data of the welding equipment in this link will be extracted to form the local equipment data of this link. Exemplarily, in the production link of "body welding", the equipment data obtained by the system may include: the power of the welding equipment is 350W, the temperature of the welding equipment is 55°C, the vibration of the welding equipment is 0.06mm, etc.

[0097] In some embodiments, the constructing of the local equipment data based on the production processes corresponding to the respective production devices may include the following operations.

[0098] S20, obtain the equipment data corresponding to each production device at each production node to obtain an equipment data vector.

[0099] The equipment data vector refers to the set of all equipment data corresponding to each production device at each production node. The equipment data vector integrates the different data of each device into a vector form for subsequent processing and analysis.

[0100] For example, for the time node 08:00 - 08:05, the device data of devices A, B, and C are production output, power consumption, and fault status respectively. Then the device data vector is: [150, 5, 0] (the production output of device A is 150 units, the power consumption of device B is 5 kW, and the fault status of device C is 0, indicating normal).

[0101] In some embodiments, at each production node, the production monitoring system, sensors, or other data acquisition means can be used to obtain the device data of each device under this node. The data can include production quantity, power, temperature, load, fault status, etc. All relevant data (such as production output, power, load, etc.) of each production device are combined into a vector to represent the status data of all devices under this node.

[0102] S21, group the production devices based on the production process flow.

[0103] The production process flow describes the sequence of each link in the production process and the equipment cooperation among them. It stipulates various tasks from raw materials to finished products and the role of each device.

[0104] For example, in automobile manufacturing, the production process flow can include body welding, painting, final assembly, etc., and each link depends on different devices.

[0105] Another example is that in mobile phone manufacturing, the production process flow can include screen assembly, motherboard welding, functional testing, etc., and each link uses different devices.

[0106] Grouping means grouping relevant production devices according to their roles, functions, or working natures in the production process flow according to the requirements of the production process flow. The devices within each group cooperate in the same process link.

[0107] For example, Group 1 (Device Group 1) includes devices A and B for welding, and they jointly complete the welding task in the production process. Group 2 (Device Group 2) includes devices C and D for testing, and they are responsible for product testing.

[0108] In some embodiments, the devices can be divided into different groups according to their roles in the production process and the sequence of process steps. For example, all devices for welding form one group, and all devices for testing form another group.

[0109] S22, combine the device data vectors corresponding to the production devices in each group to obtain local device data.

[0110] Combine the device data vectors within each group to form local device data. The device data within each group can be combined by means of vector splicing, summation, or averaging.

[0111] Exemplarily, assume that at a certain production node, device A and device B form group 1, the output of device A is 100, the output of device B is 150, and the power consumption of device C (in group 2) is 5 kW. Then the local device data of group 1 is: [100, 150], the local device data of group 2 is: [5]. If the output data of device A and device B in group 1 needs to be summed, the local device data of group 1 can be obtained as:

[250] .

[0112] Step 105, based on the production timing data, the local device data, and the production plan, use a machine learning model to perform anomaly detection.

[0113] A machine learning model refers to an algorithm model that can perform prediction, classification, or detection through data training. In anomaly detection, common machine learning models can include support vector machine (SVM), random forest (RF), deep neural network (DNN), etc.

[0114] Anomaly detection refers to identifying data that does not conform to the normal production mode. Anomaly data can reflect equipment failures, production problems, or other abnormal situations.

[0115] In some embodiments, the production timing data, the local device data, and the production plan can be input into a machine learning model. The model is obtained through training based on historical data. According to the capabilities obtained from historical training, the machine learning model can determine whether there are anomalies in the input data. For example, if the temperature of a certain device suddenly rises beyond the normal range, the machine learning model can identify the temperature anomaly based on historical data and the normal range.

[0116] Step 106, in response to detecting an anomaly, extract anomaly data from the production timing data and / or the local device data.

[0117] Anomaly data refers to data detected during the production process that is different from the normal production state and can reflect equipment failures or production problems. Anomaly data can include abnormal device power, temperature, etc.

[0118] In some embodiments, once an anomaly is detected, the system can automatically extract relevant anomaly data from the production timing data and the local device data, or can also extract relevant anomaly data from the production timing data and the local device data in the manner described in the following embodiments. For example, if the machine learning model detects that the temperature of a device exceeds a preset threshold (such as 80 °C), the system can extract the temperature data during that period from the production timing data and the local device data to further analyze the cause of the anomaly.

[0119] In some embodiments, extracting abnormal data from the production timing data and / or the local device data in response to detecting an abnormality may include the following operations.

[0120] S30. Determine the type of abnormality.

[0121] Wherein, the type of abnormality includes device abnormality and device production capacity abnormality.

[0122] In some embodiments, based on the collected data, the normal operating state and production capacity of the device can be compared. If the device experiences phenomena such as downtime, failure, etc., it is a device abnormality; if the device output is lower than the normal level, it is a device production capacity abnormality.

[0123] S31. Based on the type of abnormality, use the abnormal data location model to determine the location of the abnormal data.

[0124] The abnormal data location model is a data analysis model designed to automatically locate the specific location of abnormal data in the production timing data and local device data according to the given type of abnormality and data characteristics. This model can predict and identify the abnormal parts in the data based on historical data, device operation rules, and abnormal patterns.

[0125] The location of the abnormal data refers to the specific time point or data point location where the abnormal data appears in the production timing data and / or the local device data. For example, if the output of device A decreases during a certain period, the location of the abnormal data may be a specific time node, such as 08:15.

[0126] In some embodiments, according to the type of abnormality (device abnormality or device production capacity abnormality), a suitable abnormal data location model can be selected, and the type of abnormality and relevant data can be input into the model for analysis. The model automatically analyzes the specific location of the abnormal data based on the input type of abnormality and relevant data. If it is a device abnormality, the model can locate the moment when the device stops working; if it is a device production capacity abnormality, the model can locate the time node when the output is lower than expected.

[0127] In some embodiments, in response to the type of abnormality being the device abnormality, input the local device data into the abnormal data location model to determine the location of the abnormal data; and / or, in response to the type of abnormality being the device production capacity abnormality, input the production timing data into the abnormal data location model to determine the location of the abnormal data.

[0128] The abnormal data location model is a data analysis model used to determine the location of abnormal data according to the abnormal characteristics of device data.

[0129] S32. Extract the abnormal data from the production time-series data and / or the local device data based on the abnormal data location.

[0130] In some embodiments, the system can find and extract the corresponding abnormal data from the production time-series data and the local device data according to the abnormal data location. For example, if it is located that device A fails at 08:10, the status information of device A is extracted from the device data at 08:10.

[0131] Step 107. Perform feature extraction based on the abnormal data and the production plan to determine the production fusion features.

[0132] Feature extraction refers to extracting key features from the original data that are helpful for analysis and prediction. Feature extraction is part of data preprocessing and is used to simplify the input of the model.

[0133] The production fusion features are the features obtained by combining the abnormal data with the key information of the production plan, and the production fusion features can better reflect the potential problems in the production process.

[0134] In some embodiments, the features that can reflect the abnormal state in the production process can be extracted by fusing the abnormal data with the production plan. The production fusion features can include information such as equipment failure early warning and production line scheduling problems. For example, assuming that the abnormal device temperature (80°C) is combined with the high-load production state in the production plan, the extracted production fusion features can include "the equipment is overloaded and running, resulting in too high temperature", and this feature will help the subsequent production capacity coordination decision-making.

[0135] In some embodiments, it can also be obtained in the way as Figure 2 described to obtain the production fusion features.

[0136] Step 108. Determine the production capacity coordination strategy based on the production fusion features, the production time-series data, and the local device data.

[0137] The production capacity coordination strategy refers to a series of measures taken according to the analysis results of the data in the production process to optimize the use efficiency of production equipment, reduce production bottlenecks, and improve the overall production capacity. The production capacity coordination strategy includes but is not limited to production equipment maintenance, production equipment power adjustment, and production equipment working time adjustment, etc.

[0138] Production equipment maintenance refers to inspecting, repairing, or replacing some parts of the equipment to ensure that the equipment runs in a normal state and prevent equipment failures from affecting production.

[0139] The power adjustment of production equipment refers to adjusting the power output of the equipment to ensure that the equipment operates under appropriate working loads and avoid situations of overload or insufficient power.

[0140] The adjustment of the working time of production equipment refers to adjusting the working time of production equipment, such as by increasing the working time of the equipment or adjusting work shifts, to improve production capacity, or reducing the working time of the equipment to avoid over-fatigue and reduce the risk of failures.

[0141] In some embodiments, after detecting abnormal data in the production process and extracting production fusion features, these features are used together with production timing data and local equipment data to determine the optimal production capacity coordination strategy. According to the bottlenecks or abnormal situations in the production process, an appropriate production capacity coordination strategy can be selected to ensure smooth production. For example, if the temperature of the equipment is too high and close to the failure threshold, equipment maintenance can be recommended; if the efficiency of a production link is low, the power output or working time of the equipment can be adjusted to improve production efficiency.

[0142] Exemplarily, assume that the temperature of the welding equipment in the "body welding" link suddenly rises (beyond the normal range), and the system can identify that there may be a fault with the equipment. Based on the abnormal data (such as temperature data) and the production plan, it can be determined that the equipment needs immediate maintenance to avoid further impact on production due to the fault. Therefore, an equipment maintenance strategy can be generated, recommending immediate shutdown for repair.

[0143] Again, for example, in the "engine assembly" link, the operating power of the production equipment is too large, resulting in overloading of the equipment. By analyzing the production timing data and local equipment data, it is found that the equipment power does not match the load demand in the production plan. It can be recommended to adjust the power output of the equipment to adapt it to the current production demand and avoid equipment overload or energy waste.

[0144] Once more, for example, in some cases, the working time of the equipment on the production line may be uneven, resulting in overloading of some equipment. Assume that in the "welding" link, due to the overly long working time of a certain equipment, the efficiency has decreased. The working time arrangement of this equipment can be adjusted according to the production plan and equipment data, and appropriate rest time can be allocated, thereby improving production efficiency and reducing the risk of failures.

[0145] Figure 2 It is a schematic flowchart of the method for determining production fusion features provided by the embodiments of this application. Figure 2 The illustrated process 200 can be executed by an electronic device. As Figure 2 shown, the process 200 can include the following operations.

[0146] Step 201, perform feature extraction on the abnormal data to obtain abnormal data features.

[0147] The abnormal data feature is the feature data obtained by extracting features from abnormal data. It can be represented in the form of a feature vector.

[0148] Feature extraction refers to extracting useful features for analysis from the original data, and these features can help the model better perform pattern recognition and prediction. For example, the features extracted from abnormal data can be the time point of the occurrence of the abnormality, the temperature change rate, the vibration amplitude, etc.

[0149] Step 202: Extract features from the production plan to obtain production plan features.

[0150] The production plan feature is the feature data obtained by extracting features from the production plan. It can also be represented in the form of a feature vector.

[0151] Step 203: Obtain the first weight corresponding to the abnormal data feature and obtain the second weight corresponding to the production plan feature.

[0152] The first weight refers to the weight coefficient related to the abnormal data feature. It reflects the importance of the abnormal data feature in the overall analysis. For example, if the abnormal temperature of the equipment has a greater impact on production, the weight of the temperature abnormality may be higher.

[0153] The second weight refers to the weight coefficient related to the production plan feature. It reflects the importance of the production plan feature in the overall analysis. For example, if a certain production task is decisive for achieving the production goal, the weight of the production plan feature related to this task may be higher.

[0154] In some embodiments, the first weight and the second weight can be obtained through expert experience or in the manner described in the following embodiments.

[0155] The first initial weight of the abnormal data feature and the second initial weight of the production plan feature are obtained through the following standardized variance calculation formula (1).

[0156]

[0157] Where, W represents the calculated initial weight, x is the abnormal data feature or the production plan feature, and i represents the i-th feature in the abnormal data feature or the production plan feature.

[0158] Substituting the abnormal data feature into formula (1) can calculate the first initial weight, and substituting the production plan feature into formula (1) can calculate the second initial weight.

[0159] Moreover, a first coefficient for adjusting the first initial weight and a second coefficient for adjusting the second initial weight are determined based on the following weight adjustment formula (2).

[0160]

[0161] Wherein, st is the anomaly score of the time series anomaly detection, s0 is the threshold; β is an adjustable parameter for controlling the degree of weight change, and α is the calculated adjustment value.

[0162] The first initial weight is adjusted based on the first coefficient, and the second initial weight is adjusted based on the second coefficient to determine the first weight and the second weight.

[0163] Specifically, the calculated first coefficient can be multiplied by the first initial weight to obtain the first weight, and the second coefficient can be multiplied by the second initial weight to obtain the second weight.

[0164] Step 204, based on the first weight and the second weight, fuse the anomaly data feature and the production plan feature to determine the production fusion feature.

[0165] The production fusion feature is a comprehensive feature obtained by weighted fusion of the anomaly data feature and the production plan feature. The production fusion feature can reflect both the abnormal situation and the overall impact of the production plan, thereby helping to optimize the production process.

[0166] In some embodiments, the anomaly data feature and the production plan feature can be weighted and fused according to the first weight and the second weight. For example, the features of the anomaly data and the production plan can be combined by the weighted average method or other fusion methods to obtain the production fusion feature.

[0167] Compared with the prior art, the factory real-time production capacity coordination method based on the MES system proposed by the present invention has significant advantages in terms of real-time performance, intelligence, automation, flexibility, etc. Through comprehensive collection and analysis of production data, accurate anomaly detection and prediction, and intelligent production capacity coordination strategy formulation, it can greatly improve production efficiency, reduce failures and resource waste, optimize equipment usage, and enhance the overall coordination and flexibility of production.

[0168] As Figure 3 shown, the embodiment of the present application provides an electronic device, including a processor 111, a communication interface 112, a memory 113, and a communication bus 114. Among them, the processor 111, the communication interface 112, and the memory 113 complete mutual communication through the communication bus 114.

[0169] The memory 113 is used to store a computer program.

[0170] In one embodiment of the present application, when the processor 111 executes the program stored on the memory 113, it implements the method for real-time production capacity coordination of a factory based on the MES system provided by any of the foregoing method embodiments, including:

[0171] Obtain production data of each production link in the factory within a preset time period;

[0172] Construct production timing data based on the production data and the production process flow;

[0173] Obtain equipment data of each production device in the factory within the preset time period; the equipment data includes equipment real-time power, equipment temperature, equipment vibration, equipment current, and equipment voltage;

[0174] Construct local equipment data based on the production process flow and the equipment data;

[0175] Perform anomaly detection using a machine learning model based on the production timing data, the local equipment data, and the production plan;

[0176] In response to detecting an anomaly, extract anomaly data from the production timing data and / or the local equipment data;

[0177] Perform feature extraction based on the anomaly data and the production plan to determine production fusion features;

[0178] Determine a production capacity coordination strategy based on the production fusion features, the production timing data, and the local equipment data; wherein, the production capacity coordination strategy includes production equipment maintenance, production equipment power adjustment, and production equipment working time adjustment.

[0179] The embodiment of the present application also provides a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, it implements the steps of the method for real-time production capacity coordination of a factory based on the MES system provided by any of the foregoing method embodiments.

[0180] It should be noted that in this article, relational terms such as "first" and "second" are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the term "comprising", "including" or any other variant thereof is intended to cover non-exclusive inclusion, so that a process, method, article or device comprising a series of elements not only includes those elements, but also includes other elements not expressly listed, or further includes elements inherent to such process, method, article or device. Without further limitation, an element defined by the statement "comprising an..." does not exclude the presence of additional identical elements in the process, method, article or device comprising the said element.

[0181] The above are only specific embodiments of the present invention, enabling those skilled in the art to understand or implement the present invention. Various modifications to these embodiments will be obvious to those skilled in the art, and the general principles defined herein can be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to these embodiments shown herein, but rather will be accorded the widest scope consistent with the principles and novel features claimed herein.

Claims

1. A real-time capacity coordination method for a factory based on an MES system, characterized in that: The method comprises: Obtain production data of each production link of the factory within a preset time period; Constructing production time series data based on the production data and the production process flow; Acquire equipment data of each production equipment of the factory within the preset time period; the equipment data includes real-time power of the equipment, equipment temperature, equipment vibration, equipment current and equipment voltage; Based on the production process and the equipment data, construct local equipment data; Based on the production time series data, the local equipment data and the production plan, using a machine learning model to perform anomaly detection; In response to detecting an abnormality, extracting abnormal data from the production time series data and / or the local device data; Extract features based on the abnormal data and the production plan to determine production fusion features; Based on the production fusion characteristics, the production timing data and the local equipment data, a capacity coordination strategy is determined; wherein the capacity coordination strategy includes production equipment maintenance, production equipment power adjustment and production equipment working time adjustment.

2. The method according to claim 1, characterized in that: The constructing of production time series data based on the production data and the production process flow includes: Dividing the preset time period into time nodes to obtain multiple production nodes; Obtain the capacity data of each production equipment at each production node to obtain the capacity vector; The production capacity vectors of each production equipment are combined based on the production process flow to obtain the production time series data; wherein each element of the production time series data represents the production capacity at the corresponding production node.

3. The method according to claim 2, characterized in that The constructing of local equipment data based on the production process corresponding to each production equipment includes: Acquire device data corresponding to each production device of each production node to obtain a device data vector; Based on the production process flow, the production equipment is grouped; The equipment data vectors corresponding to the production equipment in each group are combined to obtain local equipment data.

4. The method according to claim 1, characterized in that: In response to detecting an abnormality, extracting abnormal data from the production time series data and / or the local device data includes: Determine the abnormality type; wherein the abnormality type includes equipment abnormality and equipment capacity abnormality; Based on the abnormal type, determine the abnormal data location using an abnormal data location model; Based on the abnormal data position, the abnormal data is extracted from the production timing data and / or the local device data.

5. The method according to claim 4, characterized in that The determining the abnormal data location based on the abnormal type by using an abnormal data location model includes: In response to the abnormality type being the device abnormality, inputting the local device data into an abnormal data location model to determine the abnormal data location; and / or, In response to the abnormality type being the equipment capacity abnormality, the production time series data is input into an abnormal data location model to determine the abnormal data location.

6. The method according to claim 1, characterized in that The extracting features based on the abnormal data and the production plan to determine the production fusion features includes: Extracting features from the abnormal data to obtain features of the abnormal data; Extracting features from the production plan to obtain production plan features; Obtaining a first weight corresponding to the abnormal data feature, and obtaining a second weight corresponding to the production plan feature; The abnormal data feature and the production plan feature are fused based on the first weight and the second weight to determine the production fusion feature.

7. The method according to claim 6, characterized in that The obtaining of the first weight corresponding to the abnormal data feature and the obtaining of the second weight corresponding to the production plan feature include: The first initial weight of the abnormal data feature and the second initial weight of the production plan feature are obtained by the following standardized variance calculation formula; Wherein, W represents the calculated initial weight, x represents the abnormal data feature or the production plan feature, and i represents the i-th feature in the abnormal data feature or the production plan feature; Determine a first coefficient for adjusting the first initial weight and a second coefficient for adjusting the second initial weight based on the following weight adjustment formula; Among them, st is the anomaly score of time series anomaly detection, s0 is the threshold; β is the adjustable parameter that controls the degree of weight change, and α is the calculated adjustment value; The first initial weight is adjusted based on the first coefficient, and the second initial weight is adjusted based on the second coefficient to determine the first weight and the second weight.

8. A factory real-time capacity coordination system based on MES system, characterized in that: The system comprises: The first acquisition module is used to acquire production data of each production link of the factory within a preset time period; A first construction module is used to construct production time series data based on the production data and the production process flow; A second acquisition module is used to acquire equipment data of each production equipment of the factory within the preset time period; the equipment data includes real-time power of the equipment, equipment temperature, equipment vibration, equipment current and equipment voltage; A second construction module is used to construct local equipment data based on the production process and the equipment data; An anomaly detection module, used to perform anomaly detection using a machine learning model based on the production time series data, the local equipment data and the production plan; an abnormal data extraction module, for extracting abnormal data from the production time series data and / or the local device data in response to detecting an abnormality; A feature extraction module, used to extract features based on the abnormal data and the production plan to determine production fusion features; A coordination module is used to determine a production capacity coordination strategy based on the production fusion characteristics, the production timing data and the local equipment data; wherein the production capacity coordination strategy includes production equipment maintenance, production equipment power adjustment and production equipment working time adjustment.

9. An electronic device, characterized in that: It includes a processor, a communication interface, a memory and a communication bus, wherein the processor, the communication interface and the memory communicate with each other through the communication bus; Memory, used to store computer programs; The processor is used to implement the steps of the factory real-time capacity coordination method based on the MES system as described in any one of claims 1 to 7 when executing the program stored in the memory.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the factory real-time capacity coordination method based on the MES system as described in any one of claims 1 to 7 are implemented.

Images