MES management system and method for platform-based intelligent manufacturing

Abstract

The invention discloses a platform intelligent manufacturing MES management system and method, and relates to the technical field of intelligent manufacturing, and the method comprises the steps: firstly, collecting and preprocessing sensor looseness and signal stability data in real time; secondly, two types of data are comprehensively calculated and normalized to obtain a signal acquisition abnormal coefficient, and whether the signal is abnormal or not is judged according to the signal acquisition abnormal coefficient; and if yes, executing a third step and carrying out early warning prompt, and if not, returning to the first step to continue monitoring. According to the method, sensor looseness and signal data are fused, time sequence analysis is introduced to achieve early accurate early warning, a maintenance mode is upgraded to predictive maintenance, quality risks and shutdown are effectively prevented, meanwhile, a standardized and parameterized algorithm model can adapt to various sensors and scenes, the method becomes reusable assets of a platform MES, and the method is suitable for popularization and application. And the expansibility and the intelligent management level of the system are improved.

Description

Technical Field

[0001] This invention relates to the field of intelligent manufacturing technology, specifically to a platform-based intelligent manufacturing MES management system and method. Background Technology

[0002] Manufacturing Execution System (MES) management is a core area of ​​digital transformation in the manufacturing industry. It deeply integrates technologies such as artificial intelligence, the Internet of Things (IoT), and big data through an open and scalable technology platform, serving as the control core connecting the enterprise's planning layer and the shop floor equipment layer. Its core responsibility is to manage and optimize the entire production process from order placement to product completion in real time. Specifically, it utilizes AI algorithms for adaptive production scheduling and predictive maintenance, automatically collects sensor and quality data through IoT to achieve full-process transparency and traceability, and flexibly integrates ERP, WMS, and other systems with a platform architecture to break down data silos. Ultimately, it drives manufacturing enterprises to transform towards a flexible and intelligent production model, achieving the core goals of improving quality, reducing costs, and increasing efficiency.

[0003] Loose installation of IoT automatic data acquisition sensors can lead to data instability, manifested as drifting and jumping of measured values, frequent flashing of status signals, and loss of the ability to reflect the true state of the device. However, the current over-reliance on manual inspections and simple alarms lacks an intelligent early warning mechanism for this specific failure mode of loose installation, making it difficult to achieve early identification and intervention. Summary of the Invention

[0004] Technical problems to be solved To address the shortcomings of existing technologies, this invention provides a platform-based intelligent manufacturing MES management system and method, which solves the problem that loose sensor installation can directly lead to inaccurate data and misjudgment of status, while existing methods rely on manual inspections and simple threshold alarms, which cannot provide early and accurate warnings of this root cause of failure.

[0005] To achieve the above objectives, the present invention provides the following technical solution: a platform-based intelligent manufacturing MES management method, comprising the following specific steps: Step 1: Real-time acquisition of sensor looseness data and signal stability data, and preprocessing; Step 2: Comprehensive calculation and normalization of sensor looseness data and signal stability data to obtain a signal acquisition anomaly coefficient; analysis of whether the sensor signal is abnormal based on the signal acquisition anomaly coefficient; if signal acquisition anomaly is detected, proceed to Step 3; if signal acquisition is normal, return to Step 1; Step 3: Provide early warning for signal acquisition anomalies caused by sensor looseness.

[0006] Furthermore, the specific method for obtaining the signal acquisition anomaly coefficient is as follows: calculate the sensor loosening data to obtain the loosening anomaly coefficient, calculate the signal stability data to obtain the signal disorder coefficient, and perform a comprehensive calculation based on the loosening anomaly coefficient and the signal disorder coefficient to obtain the signal acquisition anomaly coefficient; ;in, Indicates the signal acquisition anomaly coefficient. Indicates the loosening anomaly coefficient. Indicates the signal disturbance coefficient. or It represents a positive real number.

[0007] Furthermore, the specific method for obtaining the loosening anomaly coefficient is as follows: a loosening threshold is preset, the value of the sensor loosening data is calculated with the loosening threshold to obtain the loosening value, and in chronological order, the subsequent loosening value is calculated with the previous loosening value to obtain the loosening severity weight, and the loosening severity weight is multiplied with the loosening value to obtain the loosening anomaly coefficient.

[0008] Furthermore, the loosening value is obtained in the following way: the difference between the sensor loosening data value and the loosening threshold is calculated to obtain the loosening value.

[0009] Furthermore, the specific method for obtaining the loosening severity weight is as follows: calculate the ratio of the subsequent loosening value to the previous loosening value to obtain the loosening severity weight.

[0010] Furthermore, the specific method for obtaining the signal disturbance coefficient is as follows: calculate the variance of the stable signal data to obtain the signal fluctuation value, set the disappearance time value, and add the signal fluctuation value and the disappearance time value to obtain the signal disturbance coefficient.

[0011] Furthermore, the specific method for obtaining the signal fluctuation value is as follows: a stable signal data sequence within a preset acquisition period is used to calculate the average value of all data in this sequence as the reference signal value; then, the difference between each stable signal data in the sequence and the reference signal value is calculated and squared; then, these squared values ​​are summed and divided by the duration of the acquisition period, and the final quotient is the signal fluctuation value.

[0012] Furthermore, the specific method for obtaining the disappearance time value is as follows: when the stable signal data disappears, it is recorded until the stable signal data reappears, at which point recording stops, and the recorded time is the disappearance time value.

[0013] Furthermore, the specific method for analyzing whether the sensor signal is abnormal based on the signal acquisition abnormality coefficient is as follows: a preset signal abnormality threshold is set, and the signal acquisition abnormality coefficient is compared with the signal abnormality threshold. If the signal acquisition abnormality coefficient is greater than the signal abnormality threshold, it indicates an abnormality; if the signal acquisition abnormality coefficient is less than or equal to the signal abnormality threshold, it indicates normality.

[0014] A platform-based intelligent manufacturing MES management system includes the following specific modules: a data acquisition module, a signal acquisition anomaly analysis module, and a loosening early warning module. The data acquisition module is used to collect sensor loosening data and signal stability data in real time and perform preprocessing. The signal acquisition anomaly analysis module is used to comprehensively calculate and normalize the sensor loosening data and signal stability data to obtain a signal acquisition anomaly coefficient. Based on the signal acquisition anomaly coefficient, it analyzes whether the sensor signal is abnormal. If an anomaly is detected, the loosening early warning module is executed; if the signal acquisition is normal, the process returns to the data acquisition module. The loosening early warning module provides early warnings for signal acquisition anomalies caused by sensor loosening.

[0015] Beneficial effects Compared with the prior art, the embodiments of the present invention have at least the following advantages or beneficial effects: 1. By fusing sensor loosening data with signal stability data and introducing time series analysis to capture the trend of loosening deterioration, early, accurate and intelligent early warning of sensor failure caused by loose installation is achieved. This method fundamentally changes the traditional passive mode that relies on manual inspection and simple threshold alarms, and upgrades the maintenance strategy to state-based predictive maintenance. This allows for intervention before the data becomes obviously inaccurate, effectively avoiding the resulting production quality risks and unplanned downtime.

[0016] 2. By constructing the early warning mechanism as a standardized and parameterized algorithm model, the scalability and intelligent management level of the platform-based MES system are enhanced. Through normalization processing and modular design, this model can adapt to different types of sensors and diverse industrial scenarios, becoming a platform asset that can be deployed and reused on a large scale. This not only improves the core capabilities of MES in equipment health management, but also lays a solid technical foundation for further integration of advanced analytics and optimization of plant-wide operation and maintenance strategies.

[0017] Of course, any product implementing this invention does not necessarily need to achieve all of the advantages described above at the same time. Attached Figure Description

[0018] Figure 1 This invention relates to a flowchart of a platform-based intelligent manufacturing MES management method.

[0019] Figure 2 This invention relates to a structural diagram of a platform-based intelligent manufacturing MES management system. Detailed Implementation

[0020] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0021] It should be noted that, in this document, relational terms such as "first" and "second" are used merely to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus.

[0022] Example 1: like Figure 1 As shown, this embodiment of the invention provides a platform-based intelligent manufacturing MES management method, including the following specific steps: Step 1: Collect sensor loosening data in real time using vibration sensors, acquire stable signal data in real time using PLC programmable logic controller, and perform data cleaning to remove redundant values ​​from sensor loosening data and stable signal data, providing a high-quality data source for subsequent calculations; Step 2: Combine the sensor loosening data and signal stability data for comprehensive calculation and normalization, such as the Min-Max normalization method, to eliminate the dimensions and transform the values ​​of different orders of magnitude into a unified numerical range to obtain the signal acquisition anomaly coefficient. Analyze whether the sensor signal is abnormal based on the signal acquisition anomaly coefficient. If the analysis shows that the signal acquisition is abnormal, proceed to Step 3. If the analysis shows that the signal acquisition is normal, return to Step 1. Step 3: When the signal acquisition anomaly coefficient exceeds the preset threshold, the system will automatically generate a structured early warning work order. This work order not only clearly indicates the specific sensor number, installation location, and equipment to which the anomaly belongs, but also includes the real-time anomaly coefficient value, historical trend charts, and preliminary diagnostic conclusions. For example, if the loosening worsens and causes signal instability, this work order will be pushed to the mobile terminal of the relevant responsible personnel or the workstation dashboard of the MES system through the integration interface in real time. It can also simultaneously trigger the on-site audible and visual alarms to issue area warnings. The early warning information includes clear handling suggestions and response time limits, thus forming a closed-loop management process from automatic system diagnosis to personnel receiving instructions and then to on-site verification and intervention, ensuring that abnormal situations can be responded to and handled in a timely and accurate manner.

[0023] Example 2 differs from Example 1 in that: The specific method for obtaining the signal acquisition anomaly coefficient is as follows: The loosening data of the sensor is calculated to obtain the loosening anomaly coefficient, and the stable signal data is calculated to obtain the signal disorder coefficient. The signal acquisition anomaly coefficient is obtained by combining the loosening anomaly coefficient and the signal disorder coefficient. ; in, This represents the signal acquisition anomaly coefficient, reflecting abnormal signal acquisition conditions of the IoT automatic acquisition sensors. This represents the loosening anomaly coefficient, reflecting the loosening status of the IoT automatic data acquisition sensors. This represents the signal disturbance coefficient, reflecting the signal disturbance level of the IoT automatic data acquisition sensor. or Represents a positive real number to avoid the fact that the signal acquisition anomaly coefficient is meaningless when the loosening anomaly coefficient or the signal disorder coefficient is zero.

[0024] The specific method for obtaining the loosening anomaly coefficient is as follows: A loosening threshold is preset. The values ​​of the sensor loosening data are calculated with the loosening threshold to obtain the loosening value. In chronological order, the next loosening value is calculated with the previous loosening value to obtain the loosening severity weight. The loosening severity weight is multiplied with the loosening value to obtain the loosening anomaly coefficient. First, the original loosening data is quantified into loosening values ​​with clear physical meaning by setting a loosening threshold to measure the current degree of deviation from the baseline. The key is to introduce time series analysis, and obtain the loosening severity weight by calculating the ratio of the loosening values ​​at different times, thereby quantifying the deterioration trend of loosening. Finally, the two are multiplied so that the assessment result includes both the current severity and the trend of change. This means that even if the current absolute value of loosening is not high, if it is in a state of rapid aggravation, the system can still generate a significant anomaly coefficient, thereby achieving a true early warning and improving fault judgment from static threshold to dynamic prediction.

[0025] The specific method for obtaining the loosening value is as follows: The difference between the sensor's loosening data and the loosening threshold is calculated to obtain the loosening value. The difference calculation not only intuitively reflects the deviation of the current loosening degree from the loosening threshold, but more importantly, it transforms the original, continuous physical signal into a scalar with clear mathematical meaning. This transformation is a prerequisite for subsequent calculation of the loosening severity weight and the loosening anomaly coefficient, enabling the system to go beyond the simple judgment of whether it exceeds the limit and achieve quantitative tracking and early warning of the loosening development trend.

[0026] The specific method for obtaining the weight of severe loosening is as follows: The ratio of the subsequent loosening value to the previous loosening value is calculated to obtain the loosening severity weight. The loosening severity weight injects dynamic time dimension perception into the early warning model. It no longer examines a single data point in isolation, but rather keenly captures the deterioration trend of the loosening state by analyzing the rate of change of the loosening value. When the subsequent loosening value is greater than the previous loosening value, the loosening severity weight is greater than 1, and the system will identify the high-risk pattern of accelerated loosening, thereby significantly improving the early warning sensitivity of the final anomaly coefficient. This method enables it to identify and warn of early signs of failure that are currently not high in absolute value but are rapidly deteriorating.

[0027] The specific method for obtaining the signal disturbance coefficient is as follows: The signal stability data is subjected to variance calculation to obtain the signal fluctuation value. The disappearance time value is set, and the signal fluctuation value and the disappearance time value are added together to obtain the signal disturbance coefficient. Not only is the noise level and stability of the signal in the signal state quantified by variance calculation, but the most serious fault state of complete signal loss, namely the disappearance time value, is also captured separately. The two are added together so that the evaluation model can respond to the complete fault spectrum from gradual degradation of signal quality to sudden interruption of communication.

[0028] The specific methods for obtaining signal fluctuation values ​​are as follows: A stable data sequence of signals within a preset acquisition period is used, and the average value of all data in this sequence is calculated as the reference signal value. Then, the difference between each stable data point in the sequence and the reference signal value is calculated and squared. Then, these squared values ​​are summed and divided by the acquisition period duration. The final quotient is the signal fluctuation value. This not only measures the degree of dispersion of the signal around its average value, but more importantly, by dividing by the acquisition period duration, the total fluctuation is normalized to the fluctuation intensity per unit time.

[0029] The specific method for obtaining the disappearance time value is as follows: When stable signal data disappears, it is recorded until stable signal data reappears, at which point recording stops. The recorded time is the disappearance time value. Unlike the fluctuation value that measures the quality of a signal, it directly determines whether a signal is present or not. This quantification enables the system to clearly distinguish between intermittent interference and continuous interruption, ensuring that the early warning mechanism maintains the highest sensitivity to the complete loss of connection of the sensor due to complete loosening, power failure, or broken circuit. It is an indispensable key dimension for comprehensive and reliable signal quality assessment.

[0030] The specific method for analyzing whether a sensor signal is abnormal based on the signal acquisition anomaly coefficient is as follows: A preset signal abnormality threshold is set. The signal acquisition abnormality coefficient is compared with the signal abnormality threshold. If the signal acquisition abnormality coefficient is greater than the signal abnormality threshold, it indicates an abnormality. If the signal acquisition abnormality coefficient is less than or equal to the signal abnormality threshold, it indicates normality. First, during the stable operation phase of the production line or equipment, historical data on long-term signal acquisition anomaly coefficients should be collected. Statistical analysis should be used to establish an initial threshold baseline. Second, manual calibration is required, taking into account the criticality of specific sensors, the tolerance range of the process, and the acceptable cost of false alarms. For critical monitoring points, the threshold should be appropriately tightened to improve early warning sensitivity; for non-critical points, it can be appropriately relaxed to reduce interference. In addition, this threshold should be designed as a configurable parameter to allow for differentiated settings based on different equipment models, installation locations, and application scenarios. Finally, a threshold optimization mechanism should be established. By continuously collecting early warning feedback and real fault cases, the threshold should be periodically reviewed and dynamically adjusted to ensure that it meets the actual operating status and maintenance needs of the production line in the long term.

[0031] Example 3: like Figure 2 As shown: A platform-based intelligent manufacturing MES management system includes the following specific modules: Data acquisition module: Used to acquire sensor loosening data and signal stability data in real time, and to perform preprocessing; Signal Acquisition Anomaly Analysis Module: This module integrates sensor looseness data with stable signal data, performs normalization processing, and obtains a signal acquisition anomaly coefficient. Based on this coefficient, it analyzes whether the sensor signal is abnormal. If an abnormal signal acquisition is detected, the looseness warning module is executed. If the signal acquisition is normal, the module returns to the data acquisition module. Loosening warning module: Provides warnings for abnormal signal acquisition caused by sensor loosening.

[0032] The preferred embodiments of the present invention disclosed above are merely illustrative of the invention. These preferred embodiments do not exhaustively describe all details, nor do they limit the invention to the specific implementations described. Clearly, many modifications and variations can be made based on the content of this specification. This specification selects and specifically describes these embodiments to better explain the principles and practical applications of the invention, thereby enabling those skilled in the art to better understand and utilize the invention. The invention is limited only by the claims and their full scope and equivalents.

Claims

1. A platform-based intelligent manufacturing MES management method, characterized in that: Comprise the following specific steps: Step one: real-time acquisition of sensor loosening data and signal stability data, and preprocessing; Step two: comprehensive calculation of sensor loosening data and signal stability data, and normalization processing, obtaining signal acquisition abnormality coefficient, analyzing whether the sensor signal is abnormal according to the signal acquisition abnormality coefficient, if the signal acquisition is abnormal, executing step three, if the signal acquisition is normal, returning to step one; Step three: warning prompt of signal acquisition abnormality caused by sensor loosening. 2.The MES management method of platformized intelligent manufacturing according to claim 1, characterized in that: The specific acquisition method of the signal acquisition abnormality coefficient is as follows: The sensor loosening data is calculated to obtain the loosening abnormality coefficient, the signal stability data is calculated to obtain the signal disorder coefficient, and the signal acquisition abnormality coefficient is obtained by comprehensive calculation according to the loosening abnormality coefficient and the signal disorder coefficient; ; wherein, represents a signal acquisition abnormality coefficient, represents a looseness abnormality coefficient, represents a signal disorder coefficient, or represents a positive real number. 3.The MES management method of platformized intelligent manufacturing according to claim 2, characterized in that: The specific acquisition method of the loosening abnormality coefficient is as follows: The value of the sensor loosening data is calculated with the loosening threshold value to obtain the loosening value, and the latter loosening value is calculated with the former loosening value in time sequence to obtain the loosening severity weight, and the loosening severity weight is multiplied with the loosening value to obtain the loosening abnormality coefficient.

4. The MES management method of platformized intelligent manufacturing according to claim 3, characterized in that: The specific acquisition method of the loosening value is as follows: The value of the sensor loosening data is calculated with the loosening threshold value to obtain the loosening value.

5. The MES management method of platformized intelligent manufacturing according to claim 3, characterized in that: The specific acquisition method of the loosening severity weight is as follows: The latter loosening value is calculated with the former loosening value to obtain the loosening severity weight.

6. The MES management method of platformized intelligent manufacturing according to claim 2, characterized in that: The specific acquisition method of the signal disorder coefficient is as follows: The signal stability data is calculated to obtain the signal fluctuation value, the signal fluctuation value is added with the disappearance time value to obtain the signal disorder coefficient.

7. The MES management method of platformized intelligent manufacturing according to claim 6, characterized in that: The specific acquisition method of the signal fluctuation value is as follows: The signal stability data sequence in the preset acquisition period is calculated, the average value of all data in the sequence is taken as the reference signal value, then each signal stability data in the sequence is subtracted from the reference signal value and squared; Then, the sum of these square values is divided by the length of the acquisition period, and the final quotient is the signal fluctuation value.

8. The MES management method of platformized intelligent manufacturing according to claim 6, characterized in that: The specific acquisition method of the disappearance time value is as follows: When the signal stability data disappears, record, until the signal stability data appears again, stop recording, and the recorded time is the disappearance time value. 9.The MES management method of platformized intelligent manufacturing of claim 1, wherein: The specific method for analyzing whether the sensor signal is abnormal according to the signal acquisition abnormality coefficient is as follows: The signal acquisition abnormality coefficient is compared with the signal abnormality threshold value, if the signal acquisition abnormality coefficient is greater than the signal abnormality threshold value, it indicates abnormality, if the signal acquisition abnormality coefficient is less than or equal to the signal abnormality threshold value, it indicates normality.

10. A platformized intelligent manufacturing MES management system for implementing the platformized intelligent manufacturing MES management method of any one of claims 1-9, characterized in that, The MES management system of the platformized intelligent manufacturing comprises a data acquisition module, a signal acquisition abnormality analysis module and a loosening warning prompt module; The data acquisition module is used for real-time acquisition of sensor loosening data and signal stability data, and preprocessing; The signal acquisition anomaly analysis module is configured to: perform comprehensive calculation on the sensor loosening data and the signal stability data, perform normalization processing, obtain a signal acquisition anomaly coefficient, analyze whether the sensor signal is abnormal according to the signal acquisition anomaly coefficient, execute the loosening early warning prompt module if the signal acquisition is abnormal, and return to the data acquisition module if the signal acquisition is normal. The loosening early warning prompt module is configured to: perform early warning prompt on the signal acquisition anomaly caused by the sensor loosening.

Images