Distributed job node state prediction method and system based on software-defined control

Abstract

The present application proposes a distributed operation node state prediction method and system based on software-defined control, which belongs to the field of industrial equipment management. Among them, the method identifies heterogeneous operation equipment by collecting the state data of the same model equipment set and performing heterogeneous state sorting; uploading the heterogeneous equipment model, state data and underground coal mine environmental parameters to the application layer; the application layer extracts the equipment control parameters, calls the state prediction model based on the heterogeneous equipment model, and performs state prediction by integrating the environmental parameters and control parameters; when the predicted data is consistent with the heterogeneous equipment state data, health identification is performed. The present application solves the technical problem that the existing technology cannot accurately identify the heterogeneous state and health assessment of the same model equipment set in the distributed operation environment, resulting in low reliability and operation and maintenance efficiency of distributed operations, and achieves the technical effect of accurately identifying the status of heterogeneous equipment and realizing predictive health management, thereby improving the reliability and operation and maintenance efficiency of distributed operations.

Description

Technical Field

[0001] The present invention relates to the field of industrial equipment management, and in particular to a distributed job node status prediction method and system based on software-defined control. Background Art

[0002] With the continuous improvement of industrial automation and intelligence, the number and complexity of equipment in distributed operating environments are increasing. Especially in environments such as underground coal mines, the status monitoring and health management of sets of equipment of the same model have become the key to ensuring operational safety and efficiency.

[0003] Currently, traditional equipment status monitoring methods rely primarily on regular inspections and single-point monitoring. These methods are unable to accurately identify heterogeneous devices of the same model, leading to the misidentification of faulty and healthy equipment. Furthermore, the lack of effective health assessment mechanisms allows only post-event repairs, without early warning. These issues lead to frequent equipment failures, high maintenance costs, and increased risk of operational interruptions in distributed operating environments, severely hampering the reliability and operational efficiency of distributed operations. Summary of the Invention

[0004] The present invention addresses the technical problem that the existing technology is unable to accurately identify the heterogeneous status and health assessment of a set of devices of the same model in a distributed operation environment, resulting in low distributed operation reliability and operation and maintenance efficiency. It provides a distributed operation node status prediction method and system based on software-defined control to solve the problem.

[0005] The technical solution of the present invention to solve the above technical problems is as follows:

[0006] In the first aspect, the present invention provides a distributed job node status prediction method based on software-defined control, which is applied to an SDC system. The SDC system includes a hardware abstraction layer, a control logic layer and an application layer. The application layer includes a visual configuration interface and a data analysis module. The execution steps include: collecting the device status data of the same model equipment set that performs the first task through the hardware abstraction layer, performing heterogeneous state sorting, and obtaining the operating heterogeneous equipment; when the number of the operating heterogeneous equipment is not equal to 0, uploading the heterogeneous equipment model, heterogeneous equipment status data and underground coal mine environmental parameters to the application layer; extracting the equipment control parameters of the operating heterogeneous equipment from the control logic layer through the application layer, calling the equipment status prediction model of the data analysis module based on the heterogeneous equipment model, processing the underground coal mine environmental parameters and the equipment control parameters to perform status prediction, and obtaining equipment status prediction data; when the equipment status prediction data is consistent with the heterogeneous equipment status data, performing a health mark on the same model equipment set in the visual configuration interface.

[0007] In a second aspect, the present invention provides a distributed job node state prediction system based on software-defined control, which is applied to an SDC system. The SDC system includes a hardware abstraction layer, a control logic layer, and an application layer. The application layer includes a visual configuration interface and a data analysis module. The distributed job node state prediction system includes: a data acquisition and sorting unit, which is used to collect device state data of a set of devices of the same model that perform a first task through the hardware abstraction layer, perform heterogeneous state sorting, and obtain heterogeneous devices for operation; a heterogeneous device uploading unit, which is used to upload the heterogeneous device model, heterogeneous device state data, and underground coal mine environmental parameters to the application layer when the number of heterogeneous devices for operation is not equal to 0; a state prediction and analysis unit, which is used to extract the device control parameters of the heterogeneous devices for operation from the control logic layer through the application layer, call the device state prediction model of the data analysis module based on the heterogeneous device model, process the underground coal mine environmental parameters and the device control parameters to perform state prediction, and obtain device state prediction data; and a health mark display unit, which is used to mark the health of the set of devices of the same model in the visual configuration interface when the device state prediction data is consistent with the heterogeneous device state data.

[0008] The beneficial effects of the present invention are:

[0009] First, the device status data of the same-model device set executing the first task is collected through the hardware abstraction layer, and heterogeneous status sorting is performed to obtain the operating heterogeneous devices, thereby realizing the accurate screening and identification of devices with abnormal status in the same-model device set; when the number of operating heterogeneous devices is not equal to 0, the heterogeneous device model, heterogeneous device status data and underground coal mine environmental parameters are uploaded to the application layer, thereby establishing a data basis for the association of heterogeneous device information and environmental factors; the device control parameters of the operating heterogeneous devices are extracted from the control logic layer through the application layer, and the device status prediction model of the data analysis module is called based on the heterogeneous device model. The underground coal mine environmental parameters and equipment control parameters are processed to perform status prediction, and the equipment status prediction data is obtained, thereby realizing intelligent status prediction based on multi-dimensional parameter fusion; when the equipment status prediction data is consistent with the heterogeneous device status data, the health mark of the same-model device set is performed in the visual configuration interface, thereby completing the verification of the prediction results and the visual management of the equipment health status.

[0010] Through the above technical solution, distributed job node status prediction based on software-defined control architecture is realized, which effectively solves the problems of inaccurate heterogeneous state identification and lack of health assessment, accurately identifies the status of heterogeneous devices and realizes predictive health management, thereby improving the reliability and operation and maintenance efficiency of distributed operations. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] Figure 1 A schematic diagram of the flow of a distributed job node state prediction method based on software-defined control provided by the present invention;

[0012] Figure 2 This is a structural diagram of the distributed job node status prediction system based on software-defined control provided by the present invention.

[0013] In the accompanying drawings, the components represented by the reference numerals are as follows:

[0014] Data collection and sorting unit 11, heterogeneous device uploading unit 12, status prediction and analysis unit 13, health mark display unit 14. DETAILED DESCRIPTION

[0015] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without making any creative efforts shall fall within the scope of protection of the present invention.

[0016] In the description of the present invention, the terms "first" and "second" are used for descriptive purposes only and should not be understood to indicate or imply relative importance or implicitly specify the number of the technical features indicated. Therefore, a feature specified as "first" or "second" may explicitly or implicitly include one or more of the specified features. In the description of the present invention, "plurality" means two or more, unless otherwise specifically defined.

[0017] In the description of the present invention, the term "for example" is used to mean "used as an example, illustration or illustration". Any embodiment of the present invention described as "for example" is not necessarily to be construed as being more preferred or advantageous than other embodiments. The following description is given to enable any person skilled in the art to implement and use the present invention. In the following description, details are listed for the purpose of explanation. It should be understood that a person of ordinary skill in the art can recognize that the present invention can be implemented without using these specific details. In other examples, well-known structures and processes are not elaborated in detail to avoid obscuring the description of the present invention with unnecessary details. Therefore, the present invention is not intended to be limited to the embodiments shown, but is consistent with the widest scope consistent with the principles and features disclosed herein.

[0018] Example 1, as Figure 1 As shown, an embodiment of the present invention provides a distributed job node status prediction method based on software-defined control, which is applied to an SDC system. The SDC system includes a hardware abstraction layer, a control logic layer and an application layer. The application layer includes a visual configuration interface and a data analysis module.

[0019] Specifically, the software-defined control-based distributed job node status prediction method provided in this application embodiment is built on a software-defined control (SDC) system. This SDC system adopts a layered, decoupled architecture design, including a hardware abstraction layer, a control logic layer, and an application layer, to achieve virtualized management of hardware resources and software-based implementation of control logic.

[0020] The hardware abstraction layer (HAL) is responsible for abstracting and encapsulating the underlying physical devices, shielding the differences between different hardware devices and providing a standardized device access interface for upper layers. This HAL enables unified management and control of various heterogeneous devices in distributed operation nodes, including but not limited to sensors, actuators, controllers, and other downhole operation equipment. The control logic layer runs core control algorithms and business logic, implementing intelligent control of the operation nodes through software. This control logic layer is decoupled from the HAL and application layers, allowing control logic to be developed, deployed, and maintained independently of the specific hardware platform, enhancing flexibility and scalability. The application layer, as the top layer, handles human-computer interaction and data analysis, including a visual configuration interface and a data analysis module. The visual configuration interface provides operators with intuitive equipment status monitoring and configuration capabilities, supporting real-time display of equipment operating status, alarm information, and health assessment results. The data analysis module integrates machine learning algorithms and data mining technologies to achieve intelligent prediction of equipment status and anomaly detection, providing a basis for preventive maintenance.

[0021] Through the three-tier architecture of the SDC system, the SDC system can achieve comprehensive monitoring, intelligent analysis and accurate prediction of distributed operation nodes, laying the foundation for improving industrial production efficiency and equipment reliability.

[0022] The distributed job node state prediction method includes:

[0023] S1. Collect device status data of a set of devices of the same model that execute the first task through the hardware abstraction layer, perform heterogeneous state sorting, and obtain heterogeneous devices for the operation.

[0024] Specifically, the hardware abstraction layer (HAL) first performs a unified data collection operation on the set of devices of the same model that are executing the first task, obtaining device status data. A set of devices of the same model refers to multiple devices with identical specifications and functional characteristics deployed in a distributed operating environment, which collaborate to execute the predefined first task. This device status data includes, but is not limited to, multi-dimensional information such as the device's operating parameters, performance indicators, and operating status indicators. This data is collected in real time and uniformly formatted through the standardized interfaces of the HAL.

[0025] After acquiring device status data, heterogeneous status sorting is performed. This operation aims to identify and filter out devices with abnormal or deviating operating states from a set of devices of the same model. Heterogeneous status sorting involves comparative analysis of the status data of devices of the same model through data analysis to identify device status patterns that differ significantly from normal operating patterns. After heterogeneous status sorting, operational heterogeneous devices are identified—devices within a set of devices of the same model that have been identified as having abnormal operating states or potential failure risks. These operational heterogeneous devices serve as key monitoring targets for subsequent status prediction and health assessment, providing a target list for precise preventive maintenance.

[0026] By performing heterogeneous state sorting, abnormal devices in distributed operating environments can be automatically identified, avoiding the tedious process of manually checking the status of devices one by one in traditional methods, and improving the efficiency and accuracy of anomaly detection.

[0027] S2. When the number of the operating heterogeneous devices is not equal to 0, the heterogeneous device models, heterogeneous device status data and underground coal mine environmental parameters are uploaded to the application layer.

[0028] Specifically, the system first counts the number of heterogeneous devices in each operation and performs conditional analysis. If the number of heterogeneous devices is not equal to 0, it indicates that there are devices with abnormal operating conditions within the same device cluster, requiring further status prediction and analysis. This conditional analysis ensures the proper allocation of resources and avoids unnecessary prediction calculations when there are no abnormal devices.

[0029] When the number of operating heterogeneous devices is not equal to 0, data upload will be performed, specifically including heterogeneous device models, heterogeneous device status data, and underground coal mine environmental parameters. The heterogeneous device model identifies the specific model and specifications of the operating heterogeneous device, providing an index for subsequent invocation of the corresponding device status prediction model. The heterogeneous device model includes identification information such as the device manufacturer, product series, and technical parameters. The heterogeneous device status data is the real-time operating data collected in step S1 for the heterogeneous devices identified as operating. The heterogeneous device status data includes multi-dimensional operating parameters of the devices, such as physical measurements such as current, voltage, temperature, vibration, and pressure, as well as status identification information such as the device's operating mode, operating time, and fault code. The underground coal mine environmental parameters reflect the external environmental conditions of the operating heterogeneous devices. These underground coal mine environmental parameters include, but are not limited to, environmental factors that affect equipment performance, such as ambient temperature, humidity, gas concentration, dust concentration, atmospheric pressure, and ventilation conditions.

[0030] These three types of data are uniformly uploaded to the application layer, providing a complete data foundation for subsequent intelligent analysis and status prediction. By uploading heterogeneous equipment models, heterogeneous equipment status data, and underground coal mine environmental parameters to the application layer, automatic screening of abnormal equipment information and accurate transmission of valid data are achieved, laying the data foundation for improving the accuracy and pertinence of status predictions.

[0031] S3. Extract the equipment control parameters of the heterogeneous equipment from the control logic layer through the application layer, call the equipment status prediction model of the data analysis module based on the heterogeneous equipment model, process the underground coal mine environmental parameters and the equipment control parameters to perform status prediction, and obtain equipment status prediction data.

[0032] Specifically, the application layer first extracts the device control parameters of the heterogeneous devices through the data interface with the control logic layer. Device control parameters refer to the control instructions and configuration parameters issued by the control logic layer to the heterogeneous devices. These include but are not limited to the device's operating mode setting, operating speed control value, output power adjustment parameters, start and stop control signals, etc.

[0033] Next, based on the uploaded heterogeneous device models, the corresponding device status prediction model is retrieved from the data analysis module. The data analysis module pre-stores specialized prediction models trained for different device models. Each model is trained and optimized with extensive historical data, enabling accurate status prediction for specific device models. Through model matching, the most appropriate device status prediction model for the heterogeneous devices currently in operation is automatically selected, ensuring targeted and accurate prediction results.

[0034] The retrieved equipment status prediction model then receives underground coal mine environmental parameters and equipment control parameters to perform status prediction. Underground coal mine environmental parameters reflect the external working environment of the heterogeneous equipment, while equipment control parameters reflect the internal control status of the heterogeneous equipment. The equipment status prediction model uses deep learning, machine learning, or other artificial intelligence algorithms to comprehensively analyze and identify patterns in these underground coal mine environmental parameters and equipment control parameters, establishing a complex mapping relationship between environmental conditions, control parameters, and equipment status. Ultimately, through computational processing by the equipment status prediction model, equipment status prediction data is obtained. Equipment status prediction data refers to the expected equipment status data of heterogeneous equipment under specific underground coal mine environmental parameters and equipment control parameters, providing a basis for subsequent equipment health assessments and maintenance decisions.

[0035] By obtaining equipment status prediction data, intelligent status prediction based on multi-source information fusion is realized, which fully considers the combined influence of the internal control status of the equipment and external environmental conditions, and improves the accuracy and practicality of status prediction.

[0036] S4. When the device status prediction data is consistent with the heterogeneous device status data, health identification is performed on the device set of the same model in the visual configuration interface.

[0037] First, a consistency comparison operation is performed between the device status prediction data and the heterogeneous device status data. The device status prediction data is the estimated value of the device status calculated by the device status prediction model in step S3, while the heterogeneous device status data is the actual operating status of the heterogeneous operating equipment actually collected in step S1. Through data comparison, the two types of data are analyzed for deviations to determine whether the prediction results and the actual status meet the preset consistency standards. The consistency judgment is evaluated based on a preset tolerance range or similarity threshold. When the deviation between the device status prediction data and the heterogeneous device status data is within an acceptable range, it is determined that the two have reached a consistent state. The consistency between the device status prediction data and the heterogeneous device status data verifies that the heterogeneous operating equipment identified in step S1 does have an abnormal state.

[0038] When consistency conditions are met, the entire set of devices of the same model is marked as healthy on the visual configuration interface. Health marking refers to the graphical display of the overall health status of the set of devices of the same model on the visual configuration interface, using visual elements such as color coding, status icons, or numerical indicators. For example, green indicates a healthy state, yellow indicates a warning state, and red indicates an abnormal state.

[0039] It's worth noting that even though only some devices are identified as heterogeneous, when the predicted device status data is verified to be consistent with the heterogeneous device status data, the entire set of devices of the same model is marked healthy. This is because the accuracy verification of the device status prediction model demonstrates the ability to reliably monitor the status of the device model, thus providing a comprehensive assessment of the health status of the entire set of devices of the same model.

[0040] By identifying a set of devices of the same model on a visual configuration interface, an intelligent health assessment based on predictive verification is achieved. This not only verifies the accuracy of anomaly detection and status prediction, but also provides operators with an intuitive display of device health status, effectively supporting equipment management and maintenance decisions.

[0041] Furthermore, the training sample equipment of the equipment status prediction model is less than or equal to the first service life; the equipment control parameters of the heterogeneous operating equipment are extracted from the control logic layer through the application layer, the equipment status prediction model of the data analysis module is called based on the heterogeneous equipment model, the underground coal mine environmental parameters and the equipment control parameters are processed to perform status prediction, and equipment status prediction data is obtained, including:

[0042] S31. Obtain the second service period of heterogeneous equipment;

[0043] S32. Using the equipment sample indexing model of the data analysis module, set the underground coal mine environment attributes and equipment control attributes as constant attributes, retrieve a first equipment status record sample that meets the heterogeneous equipment model and the first service life, and retrieve a second equipment status record sample that meets the heterogeneous equipment model and the second service life;

[0044] S33, performing homonymous mode analysis on the device state disturbance quantity record data of the first device state record sample and the second device state record sample to obtain a device state disturbance amplitude;

[0045] S34. Extracting the device control parameters of the heterogeneous operating equipment from the control logic layer through the application layer, calling the device status prediction model of the data analysis module based on the heterogeneous equipment model, processing the underground coal mine environmental parameters and the device control parameters to perform status prediction, and obtaining initial device status prediction data;

[0046] S35. Based on the device state disturbance amplitude, correct the initial device state prediction data to obtain the device state prediction data.

[0047] In a preferred embodiment, the service life of the equipment used in training the equipment status prediction model is less than or equal to a first service life. The first service life is a preset benchmark service time, typically corresponding to the equipment's designed service life or optimal performance period. By limiting the service life range of the training sample equipment, the equipment status prediction model ensures good predictive capability and stability.

[0048] First, the second service life of the current heterogeneous equipment is obtained. The second service life refers to the actual operating time of the heterogeneous equipment from the time it was put into use to the current moment, reflecting the aging degree and performance degradation of the heterogeneous equipment.

[0049] Then, historical data retrieval was performed using the equipment sample index model pre-configured in the data analysis module. To ensure comparability of the retrieval results, the underground coal mine environmental attributes and equipment control attributes were set as constant attributes. This fixed the influence of these external factors and focused on analyzing the impact of service life on equipment status. Based on this constraint, a sample of first equipment status records matching heterogeneous equipment models and first service life was retrieved, as well as a sample of second equipment status records matching heterogeneous equipment models and second service life. These two sets of sample data provided a comparative benchmark for subsequent disturbance analysis.

[0050] Subsequently, a homogeneous mode analysis is performed on the device state disturbance data from the first and second device state record samples to quantify the impact of service life. The device state disturbance refers to the deviation between the device state data corresponding to the second service life and the device state data corresponding to the first service life. Specifically, for any state parameter (such as temperature, vibration amplitude, current, or pressure) of the same device model, the difference between the value of that parameter in the second device state record sample and the corresponding parameter value in the first device state record sample is calculated. This difference is the device state disturbance for that parameter. For example, if the average operating temperature of a certain device model is 75°C during the first service life (e.g., 3 years) and the average operating temperature of the same device model during the second service life (e.g., 5 years) is 78°C, then the device state disturbance for the temperature parameter is +3°C. Positive and negative values ​​of this disturbance, respectively, indicate an upward or downward trend in the device state parameter with increasing service life. Mode analysis of the same attribute refers to performing statistical analysis on device state parameters of the same attribute type, classifying the device state parameters according to physical attributes, such as temperature parameters, vibration parameters, electrical parameters, etc., and then performing mode statistics on the disturbance quantity data set for each type of parameter. Specifically, first, for the first and second device state record samples, the disturbance quantity of each device corresponding to the state parameter is calculated to form a disturbance quantity data set. Then, for each type of state parameter, the frequency of occurrence of the device state disturbance quantity values ​​is counted, and the disturbance value with the highest frequency of occurrence is identified as the mode disturbance value of the attribute and the corresponding device state disturbance amplitude. For example, for the temperature parameter, temperature disturbance quantity data for 100 devices of the same model are collected from the first and second device state record samples. Among them, +2°C appears 35 times, +3°C appears 28 times, and +1°C appears 20 times. Other values ​​appear less frequently. Therefore, +2°C is determined to be the mode disturbance value of the temperature parameter and the device state disturbance amplitude corresponding to the temperature parameter.

[0051] The equipment state disturbance amplitude, obtained through homogeneous mode analysis, represents the most likely typical variation in equipment state parameters under the influence of age variations. This filter outs the abnormal fluctuations and accidental factors of individual equipment and reflects the general patterns and common characteristics of changes in service age for that type of equipment. This equipment state disturbance amplitude comprises a set of disturbance amplitudes for multiple parameters, such as {temperature disturbance amplitude: +2°C, vibration disturbance amplitude: +0.5mm / s, current disturbance amplitude: -1.2A, etc.}. Each disturbance amplitude quantifies the typical impact of service age variations on the corresponding equipment state parameter, providing a reliable correction benchmark for subsequent prediction and correction.

[0052] Subsequently, according to the prediction process of S3, the equipment control parameters of the heterogeneous equipment in operation are extracted from the control logic layer through the application layer. The equipment status prediction model of the data analysis module is called based on the heterogeneous equipment model. The underground coal mine environmental parameters and equipment control parameters are processed to perform status prediction and obtain initial equipment status prediction data. This initial equipment status prediction data is based on the prediction results within the service life range of the training sample and does not yet consider the impact of the actual service life of the current equipment. Next, based on the obtained equipment status disturbance amplitude, the initial equipment status prediction data is refined and corrected to obtain equipment status prediction data. The correction process applies the equipment status disturbance amplitude as a correction factor to the initial equipment status prediction data. Specifically, the corresponding equipment status disturbance amplitude is directly applied to the initial prediction data for addition and subtraction operations, thereby obtaining equipment status prediction data that has been dynamically corrected for service life.

[0053] Through the above steps, adaptive state prediction based on service life disturbance is achieved. Compared with the prediction in step S3, the impact of equipment aging laws on state changes is fully considered. By introducing the service life correction mechanism, the prediction accuracy is improved, providing more accurate and reliable state prediction services for equipment in different service stages.

[0054] Furthermore, the device status data includes a plurality of device status data; the device status data of a set of devices of the same model that execute the first task are collected through the hardware abstraction layer, and heterogeneous state sorting is performed to obtain heterogeneous devices for the operation, including:

[0055] S11, performing similarity analysis on the plurality of device status data to obtain multiple device status similarities;

[0056] S12. Based on the similarities of the multiple device states, traverse the plurality of device state data to perform local density evaluation to obtain a plurality of device state distribution densities;

[0057] S13. Using the mean value of the distribution density, traverse the distribution densities of the plurality of device states and calculate the ratio to obtain a plurality of device state heterogeneous parameters;

[0058] S14. Based on the plurality of device status heterogeneous parameters, extract device bit numbers whose device status heterogeneous parameters are greater than or equal to a predefined heterogeneous parameter threshold from the same model device set, and add them to the operation heterogeneous devices.

[0059] In a preferred embodiment, a refined heterogeneous device identification method based on similarity analysis and density assessment is proposed, which realizes accurate identification and screening of abnormal devices in a concentration of devices of the same model through multi-level data analysis.

[0060] Device status data includes multiple pieces of device status data. Specifically, each device's status information consists of multiple dimensions of state parameters, such as physical quantities like temperature, vibration, current, and pressure, as well as state identifiers like operating mode and workload. Each device's status data forms a multidimensional device state vector, providing the data foundation for subsequent similarity calculations and heterogeneous analysis.

[0061] First, similarity analysis is performed on multiple device state data to obtain multiple device state similarities. Similarity analysis involves calculating the degree of similarity between the state data of any two devices of the same model. For example, using Euclidean distance, cosine similarity, or other distance metrics, the device state vectors of any two devices are compared pairwise to generate a device similarity matrix. Multiple device state similarities reflect the interrelationships within the state distribution of the same device model, providing a similarity benchmark for subsequent density assessment.

[0062] Then, based on the similarity of multiple device states, several device state data are traversed to perform local density assessment to obtain several device state distribution densities. Among them, local density assessment refers to taking each device as the center and counting the distribution density of similar devices in its vicinity. Specifically, taking the state data of the current device as the reference point, the local neighborhood range of the device is determined according to the preset similarity threshold or the number of neighbors, and then the device state distribution density of the devices in the neighborhood is calculated. The higher the device state distribution density, the more devices with similar states there are around the device, that is, the more the state pattern of the device conforms to the group characteristics; the lower the device state distribution density, the relatively isolated or abnormal state of the device.

[0063] Subsequently, the distribution density mean is used to traverse the distribution densities of several device states and calculate the ratio to obtain several device state heterogeneity parameters. First, the arithmetic mean of all device state distribution densities is calculated to obtain the distribution density mean, which serves as a benchmark for determining device state normality. Next, the ratio of each device's device state distribution density to the distribution density mean is calculated to obtain the device state heterogeneity parameter for that device. This device state heterogeneity parameter quantitatively reflects the degree of deviation of a single device from the group average. A smaller device state heterogeneity parameter indicates that the device's state deviates more from the normal group pattern and the degree of heterogeneity is higher.

[0064] Next, based on several device state heterogeneous parameters, heterogeneous devices that meet the screening criteria are extracted from the same-model device set. Specifically, each device's device state heterogeneous parameter is compared with a predefined heterogeneous parameter threshold, and the device tag numbers whose device state heterogeneous parameters are greater than or equal to the threshold are extracted. The device tag number is a unique identifier for a device within the same-model device set, used to accurately locate and manage specific devices. Device tags that meet the criteria are added to the operational heterogeneous devices list and serve as key monitoring targets for subsequent status prediction and health assessment.

[0065] Through the above steps, automated heterogeneous device identification is achieved. Compared with the simple threshold judgment method, it can more accurately identify abnormal individuals in a group and improve the accuracy and reliability of heterogeneous state sorting.

[0066] Furthermore, based on the similarities of the multiple device states, the plurality of device state data are traversed to perform local density evaluation to obtain a plurality of device state distribution densities, including:

[0067] S121. Extracting first device status data from the plurality of device status data;

[0068] S122. Extracting a plurality of selected device state similarities based on the first device state data from the plurality of device state similarities;

[0069] S123. Select neighboring selected device state similarities of 0.5 times the number of devices of the same model from the plurality of selected device state similarities in descending order, perform mean calculation, and obtain a first device state distribution density.

[0070] S124: Add the first device state distribution density to the plurality of device state distribution densities.

[0071] In a preferred embodiment, first, first device status data is extracted from a plurality of device status data. This first device status data refers to the status information of the device currently selected as the target for density calculation during the traversal process. Following a pre-set traversal order, each device in the set of devices of the same model is selected one by one as a target for density assessment, ensuring a comprehensive density analysis of all devices.

[0072] Then, from the multiple device state similarities, multiple selected device state similarities are extracted based on the first device state data. For example, in a device set consisting of Q devices of the same model, the first device and the remaining Q-1 devices form Q-1 device state similarities, i.e., multiple selected device state similarities. These multiple selected device state similarities are the set of similarities between the first device and all other devices, reflecting the relative position and similarity distribution of the first device within the entire device population.

[0073] Subsequently, the state similarities of the multiple selected devices are sorted from largest to smallest by numerical value, and the state similarities of the neighboring selected devices that are the first 0.5 times the number of devices of the same model are selected. The average is calculated to obtain the first device state distribution density. For example, based on the state similarities of the multiple selected devices, among the remaining Q-1 devices, the devices corresponding to the first 0.5×Q similarity values ​​are sorted in descending order of the selected device state similarity, and the devices corresponding to the first 0.5×Q similarity values ​​are selected as the neighbor devices of the first device, obtaining 0.5×Q neighbor selected device state similarities. The arithmetic average of these 0.5×Q neighbor selected device state similarities is calculated, and the resulting average is the first device state distribution density, which provides a density benchmark for subsequent heterogeneous parameter calculations.

[0074] Afterwards, the first device state distribution density is added to the multiple device state distribution densities. By repeatedly executing steps S121 to S123, density calculations are performed on each device in the same device set in sequence to obtain a state distribution density of all devices, providing a comprehensive density distribution data foundation for subsequent heterogeneous parameter calculations.

[0075] Through the above steps, accurate calculation of local density based on nearest neighbor selection is achieved. By limiting the neighborhood range and statistical calculation, the density distribution characteristics of each device in the group are effectively quantified, providing reliable data support for the accurate identification of heterogeneous devices.

[0076] Furthermore, the embodiment of the present application also includes:

[0077] S51. When the number of heterogeneous devices in the operation is equal to 0, a health mark is performed on the device set of the same model in the visual configuration interface;

[0078] S52. When the device status prediction data is inconsistent with the heterogeneous device status data, on the visual configuration interface, an execution abnormality mark is performed on the operating heterogeneous devices of the same model device set, and a health mark is performed on the non-operating heterogeneous devices of the same model device set.

[0079] In one feasible implementation, when the number of heterogeneous devices in operation is equal to 0, this indicates that during the heterogeneous status sorting process in step S1, no devices in the same device set were identified as abnormally functioning, meaning that the operating status of the entire device set is within the normal range. In this case, a health indicator is assigned to the entire device set in the visual configuration interface. This health indicator, through graphical interface elements (such as a green status indicator and a normal operation icon), visually displays the overall health status of the device set to the operator, indicating that no special status prediction or abnormality monitoring is currently required and that the device set is operating normally.

[0080] If the predicted device status data is inconsistent with the heterogeneous device status data, this indicates a significant discrepancy between the prediction results of the device status prediction model in step S3 and the actual acquired heterogeneous device status, and the prediction verification has failed. In this case, the visual configuration interface first identifies the abnormal status of the heterogeneous devices in the same device set. This abnormality identification uses eye-catching visual elements (such as red warning icons and abnormal status prompts) to highlight the abnormal status of these devices, prompting operators to pay close attention and promptly address the situation. Due to the prediction verification failure, these devices are identified as having unpredictable abnormal risks and require manual intervention or further inspection. Simultaneously, the non-operating heterogeneous devices in the same device set are identified as healthy. Non-operating heterogeneous devices are normal devices that were not identified as abnormal during the heterogeneous status sorting. Since the prediction verification failure does not affect the normal status determination of these devices, they are still marked as healthy, ensuring that operators can accurately distinguish between normal and abnormal devices in the device set.

[0081] Through the above steps, intelligent equipment status management based on heterogeneous detection results and prediction verification results is realized, providing accurate status identification and classification processing for equipment monitoring in different scenarios, and effectively supporting equipment operation and maintenance decision-making.

[0082] Furthermore, the device control parameters of the heterogeneous equipment are extracted from the control logic layer through the application layer, the device status prediction model of the data analysis module is called based on the heterogeneous equipment model, the underground coal mine environmental parameters and the device control parameters are processed to perform status prediction, and the device status prediction data is obtained, including:

[0083] S361. Based on a preset device model as a constraint, collect multiple sets of training sample data whose service life is less than or equal to the first service life, wherein any set of the multiple sets of training sample data includes device control record data, deployment environment record parameters, and a label identifying device status data;

[0084] S362: Using the label identifying the device status data as supervision and the device control record data and the deployment environment record parameters as input, train a first device status predictor;

[0085] S363. Counting a first residual data set of the first device state predictor;

[0086] S364: When the number of residuals in the first residual data set is greater than or equal to a quantity threshold, train a second device state predictor using the first residual data set as supervision and the device control record data and the deployment environment record parameters as input;

[0087] S365. Until the number of residuals in the Qth residual data set is less than the quantity threshold, the outputs of the first device state predictor, the second device state predictor, and the Qth device state predictor are summed to obtain the device state prediction model, which is stored in association with the preset device model and embedded in the data analysis module.

[0088] In a preferred embodiment, first, with the preset equipment model as a constraint condition, multiple sets of training sample data with service years less than or equal to the first service years are collected. Multiple sets of training sample data constitute the basic data set for training the equipment status prediction model, and each set of training sample data contains equipment control record data, deployment environment record parameters, and labels that identify equipment status data. Among them, the equipment control record data records the control parameter setting information of the equipment during the historical operation process; the deployment environment record parameters record the external environmental condition information of the equipment; the label that identifies the equipment status data serves as the target output of supervised learning, and identifies the actual status performance of the equipment under specific control and environmental conditions. By limiting the service life range of the training samples, the consistency and representativeness of the training data are ensured.

[0089] Then, a first device state predictor is trained using the label identifying the device state data as a supervisory signal and the device control record data and deployment environment record parameters as input features. The first device state predictor uses a machine learning algorithm (such as a neural network, support vector machine, or decision tree) to establish a mapping relationship between input features and device states through supervised learning. As a basic prediction model, the first device state predictor is capable of handling most common device state prediction tasks. Subsequently, a first residual data set of the first device state predictor is statistically analyzed. The first residual data set contains the prediction error information of the first device state predictor on the training data, that is, the set of differences between the predicted values ​​and the true labels, reflecting the data patterns and prediction blind spots that the first device state predictor has not yet learned, and provides a target direction for subsequent model improvements.

[0090] When the number of residuals in the first residual dataset is greater than or equal to a preset threshold, it indicates that the prediction performance of the first device state predictor still has room for improvement, and a second round of predictor training is initiated. Using the first residual dataset as a new supervisory signal, and still using the device control record data and deployment environment record parameters as input, a second device state predictor is trained. This second device state predictor specifically learns the prediction residual pattern of the first device state predictor to correct the prediction deviation of the first device state predictor.

[0091] The system repeats steps S363 and S364 until the number of residuals in the Qth residual data set is less than the threshold, indicating that the prediction accuracy has met the preset requirements. At this point, the outputs of the first device state predictor, the second device state predictor, and finally the Qth device state predictor are summed to obtain a final device state prediction model. This device state prediction model integrates the prediction capabilities of multiple sub-predictors and achieves improved prediction performance through summation and fusion. The final device state prediction model is stored in association with the preset device model and embedded in the data analysis module, providing model support for subsequent online predictions.

[0092] Through the above steps, progressive model training based on residual learning is achieved. Compared with the single predictor method, this integrated training method can effectively improve the accuracy and generalization ability of the prediction model, providing a more reliable technical guarantee for equipment status prediction in complex industrial environments.

[0093] Furthermore, the process of determining the label identifying the device status data includes:

[0094] S3611, receiving the device control attribute rated interval from the control logic layer and the environmental attribute constraint interval from the visual configuration interface, and randomly configuring the device control record data and the environmental record parameters;

[0095] S3612. Collect an operation sample set whose service life is less than or equal to the first service life based on the preset device model, the device control record data, and the environmental record parameters, wherein the operation sample set includes a device status data set;

[0096] S3613: Extract centroid device status data from the device status data set to obtain a label that identifies the device status data.

[0097] In a preferred embodiment, the device control attribute rated interval is first received from the control logic layer, and the environmental attribute constraint interval is received from the visual configuration interface. The device control record data and environmental record parameters are randomly configured based on these constraint ranges. The device control attribute rated interval defines the normal operating range of each device control parameter, such as the minimum and maximum values ​​of control quantities such as voltage, current, and speed; the environmental attribute constraint interval defines the typical variation range of the device deployment environment, such as the range of environmental parameters such as temperature, humidity, and pressure. Random sampling is used within the device control attribute rated interval and the environmental attribute constraint interval to generate multiple different sets of device control record data and environmental record parameter combinations, ensuring that the training samples can cover all possible operating states of the device.

[0098] Subsequently, with the preset equipment model, equipment control record data, and environmental record parameters as constraints, an operation sample set with a service life less than or equal to the first service life is collected. This operation sample set refers to the actual equipment operation records retrieved from the historical operation database under specific constraints. This collection process ensures the consistency and comparability of the sample data: the equipment model constraint ensures the hardware consistency of the sample; the control and environmental parameter constraints ensure the working condition consistency of the sample; and the service life constraint ensures the time consistency of the sample. Among them, the operation sample set includes the equipment status data set, that is, the actual operation status record of the equipment under the constraints, including multi-dimensional status parameters such as temperature, vibration, and current.

[0099] Next, centroid device state data extraction is performed on the device state dataset to obtain a label identifying the device state data. Centroid device state data extraction involves using statistical methods to extract representative central state values ​​from device state data under multiple similar operating conditions. Specifically, cluster analysis or center point calculation is performed on all device state data in the operating sample set, and the centroid value (such as the mean, median, or cluster center) of each state parameter is extracted as the typical device state representation under that operating condition combination. This label identifying the device state data is the standardized state data obtained through centroid extraction. It is highly representative and stable and can effectively guide the training process of the device state prediction model.

[0100] Through the above steps, the automatic generation and standardized processing of training labels are achieved, the technical difficulties of large-scale training sample labeling are solved, a reliable data foundation is provided for the efficient training of equipment status prediction models, and the degree of automation of model training and the quality of labels are improved.

[0101] Furthermore, performing homologous mode analysis on the device state disturbance quantity record data of the first device state record sample and the second device state record sample to obtain the device state disturbance amplitude includes:

[0102] S331. Extracting positive disturbance amount record data and negative disturbance amount record data from the device state disturbance amount record data, wherein a positive disturbance amount refers to a value when the second device state record is greater than the first device state record value, and a negative disturbance amount refers to a value when the second device state record is less than the first device state record value;

[0103] S332, performing homogeneous mode analysis on the recorded data of the forward disturbance to obtain the amplitude of the forward disturbance of the device;

[0104] S333, performing homogeneous mode analysis on the negative disturbance record data to obtain the negative disturbance amplitude of the device;

[0105] S334: Add the device positive disturbance amplitude and the device negative disturbance amplitude to the device state disturbance amplitude.

[0106] In a preferred embodiment, by distinguishing between positive and negative disturbances, a bidirectional quantitative analysis of the device state change trend is achieved.

[0107] First, we extract positive and negative disturbance records from the equipment status disturbance data. A positive disturbance refers to a situation where the second equipment status record is greater than the first, reflecting an upward trend in equipment status parameters with increasing service life. A negative disturbance refers to a situation where the second equipment status record is less than the first, reflecting a downward trend in equipment status parameters with increasing service life. This directional classification allows us to accurately identify the direction of change in different equipment status parameters, laying the foundation for subsequent separate analyses.

[0108] Next, we perform homogeneous mode analysis on the recorded positive disturbance data to obtain the equipment's positive disturbance amplitude. This analysis counts the frequency of disturbance values ​​for all state parameters exhibiting a positive trend, by parameter type, and identifies the most frequently occurring positive disturbance value to obtain the equipment's positive disturbance amplitude. This equipment's positive disturbance amplitude reflects the typical magnitude of upward changes in the equipment's state parameters, providing a quantitative basis for positive adjustments during prediction and correction.

[0109] Simultaneously, a homogeneous mode analysis is performed on the recorded negative disturbance data to determine the equipment's negative disturbance amplitude. This analysis counts the frequency of disturbance values ​​for all state parameters exhibiting a negative trend, by parameter type, and identifies the most frequently occurring negative disturbance value to determine the equipment's negative disturbance amplitude. This equipment's negative disturbance amplitude reflects the typical magnitude of downward changes in the equipment's state parameters, providing a quantitative basis for negative adjustments during prediction and correction.

[0110] Subsequently, the positive and negative device disturbance amplitudes are added to the device state disturbance amplitude. This device state disturbance amplitude encompasses the positive and negative bidirectional disturbance information of each state parameter, forming a complete disturbance signature. For example, the disturbance amplitude of a device can be expressed as {temperature disturbance amplitude: +2°C, vibration disturbance amplitude: +0.5mm / s, current disturbance amplitude: -1.2A, ...}, fully reflecting the bidirectional variation of each device state parameter over service life.

[0111] Through the above steps, a classification statistical analysis based on the disturbance direction is achieved. Compared with a simple overall mode analysis, it can more accurately quantify the changing trend of the equipment status, provide more accurate and comprehensive correction parameters for subsequent prediction and correction, and improve the accuracy and applicability of service life correction.

[0112] Example 2, as Figure 2 As shown, based on the same inventive concept as the distributed job node status prediction method based on software-defined control provided in Example 1, an embodiment of the present invention also provides a distributed job node status prediction system based on software-defined control. The system is applied to an SDC system. The SDC system includes a hardware abstraction layer, a control logic layer, and an application layer. The application layer includes a visual configuration interface and a data analysis module. The distributed job node status prediction includes:

[0113] The data collection and sorting unit 11 is used to collect device status data of a set of devices of the same model that perform the first task through the hardware abstraction layer, perform heterogeneous status sorting, and obtain heterogeneous devices for the operation;

[0114] The heterogeneous device uploading unit 12 is used to upload the heterogeneous device model, heterogeneous device status data and underground coal mine environmental parameters to the application layer when the number of the heterogeneous devices in the operation is not equal to 0;

[0115] The state prediction and analysis unit 13 is configured to extract the device control parameters of the heterogeneous operating equipment from the control logic layer through the application layer, call the device state prediction model of the data analysis module based on the heterogeneous equipment model, process the underground coal mine environmental parameters and the device control parameters to perform state prediction, and obtain device state prediction data;

[0116] The health mark display unit 14 is used to mark the health of a set of devices of the same model on a visual configuration interface when the device status prediction data is consistent with the heterogeneous device status data.

[0117] Furthermore, the training sample equipment of the equipment status prediction model is less than or equal to the first service life; the status prediction analysis unit 13 includes the following execution steps:

[0118] Obtain a second service life for operational heterogeneous equipment;

[0119] Using the equipment sample indexing model of the data analysis module, the underground coal mine environment attributes and equipment control attributes are set as constant attributes, and a first equipment status record sample that meets the heterogeneous equipment model and the first service life is retrieved, and a second equipment status record sample that meets the heterogeneous equipment model and the second service life is retrieved;

[0120] Performing homonymous mode analysis on the device state disturbance quantity record data of the first device state record sample and the second device state record sample to obtain a device state disturbance amplitude;

[0121] Extracting the device control parameters of the heterogeneous equipment from the control logic layer through the application layer, calling the device status prediction model of the data analysis module based on the heterogeneous equipment model, processing the underground coal mine environmental parameters and the device control parameters to perform status prediction, and obtaining initial device status prediction data;

[0122] Based on the device state disturbance amplitude, the initial device state prediction data is corrected to obtain the device state prediction data.

[0123] Furthermore, the device status data includes a plurality of device status data; the data collection and sorting unit 11 includes the following execution steps:

[0124] Performing similarity analysis on the plurality of device status data to obtain a plurality of device status similarities;

[0125] Based on the multiple device state similarities, traverse the plurality of device state data to perform local density evaluation to obtain a plurality of device state distribution densities;

[0126] Using the distribution density mean, traverse the distribution densities of the plurality of device states and calculate the ratio to obtain a plurality of device state heterogeneous parameters;

[0127] Based on the plurality of device state heterogeneous parameters, device bit numbers whose device state heterogeneous parameters are greater than or equal to a predefined heterogeneous parameter threshold are extracted from the set of devices of the same model, and are added to the operation heterogeneous devices.

[0128] Furthermore, the data collection and sorting unit 11 further includes the following execution steps:

[0129] Extracting first device status data from the plurality of device status data;

[0130] extracting, from the plurality of device state similarities, a plurality of selected device state similarities based on the first device state data;

[0131] From the plurality of selected device state similarities, from large to small, select neighboring selected device state similarities that are 0.5 times the number of device sets of the same model, perform mean calculation, and obtain a first device state distribution density;

[0132] The first device state distribution density is added to the plurality of device state distribution densities.

[0133] Furthermore, the embodiment of the present application further includes a device status identification unit, and the execution steps of the device status identification unit include:

[0134] When the number of heterogeneous devices in the operation is equal to 0, a health mark is performed on the device set of the same model in the visual configuration interface;

[0135] When the device status prediction data is inconsistent with the heterogeneous device status data, the execution abnormality mark of the operating heterogeneous devices of the same model device set is performed on the visual configuration interface, and the health mark of the non-operating heterogeneous devices of the same model device set is performed.

[0136] Furthermore, the state prediction and analysis unit 13 includes the following execution steps:

[0137] Based on a preset device model as a constraint, multiple sets of training sample data with a service life less than or equal to the first service life are collected, wherein any set of the multiple sets of training sample data includes device control record data, deployment environment record parameters, and a label identifying device status data;

[0138] Using the label identifying the device status data as supervision and the device control record data and the deployment environment record parameters as input, training a first device status predictor;

[0139] Counting a first residual data set of the first device state predictor;

[0140] When the number of residuals of the first residual data set is greater than or equal to a quantity threshold, training a second device state predictor using the first residual data set as supervision and the device control record data and the deployment environment record parameters as input;

[0141] Until the number of residuals in the Qth residual data set is less than the quantity threshold, the outputs of the first device state predictor, the second device state predictor, and the Qth device state predictor are summed to obtain the device state prediction model, which is associated with the preset device model and stored, and embedded in the data analysis module.

[0142] Furthermore, the process of determining the label identifying the device status data includes:

[0143] Receive the rated interval of the device control attribute from the control logic layer and the environmental attribute constraint interval from the visual configuration interface, and randomly configure the device control record data and environmental record parameters;

[0144] Based on the preset device model, the device control record data, and the environmental record parameters, an operation sample set having a service life less than or equal to the first service life is collected, wherein the operation sample set includes a device status data set;

[0145] Centroidal device state data is extracted from the device state data set to obtain a label identifying the device state data.

[0146] Furthermore, the state prediction and analysis unit 13 further includes the following execution steps:

[0147] Extracting positive disturbance amount record data and negative disturbance amount record data from the device state disturbance amount record data, wherein a positive disturbance amount refers to a second device state record value being greater than a first device state record value, and a negative disturbance amount refers to a second device state record value being less than a first device state record value;

[0148] Performing homogeneous mode analysis on the recorded data of the forward disturbance to obtain the amplitude of the forward disturbance of the device;

[0149] Performing homogeneous mode analysis on the negative disturbance record data to obtain the negative disturbance amplitude of the equipment;

[0150] The positive disturbance amplitude of the device and the negative disturbance amplitude of the device are added to the device state disturbance amplitude.

[0151] It should be noted that, in the above embodiments, the description of each embodiment has its own focus. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0152] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Thus, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0153] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowcharts and / or block diagrams, as well as combinations of processes and / or blocks in the flowcharts and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded computer, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowcharts and / or block diagrams. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0154] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0155] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0156] Although preferred embodiments of the present invention have been described, additional changes and modifications to these embodiments may occur to those skilled in the art once the basic inventive concepts become known.

[0157] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the present invention and its equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. A distributed job node state prediction method based on software-defined control, characterized in that: Applied to the SDC system, the SDC system includes a hardware abstraction layer, a control logic layer, and an application layer. The application layer includes a visual configuration interface and a data analysis module. The execution steps include: The device status data of the same type of devices executing the first task are collected through the hardware abstraction layer, and the heterogeneous state data are sorted to obtain the heterogeneous devices for the operation; When the number of the heterogeneous devices in the operation is not equal to 0, uploading the heterogeneous device models, heterogeneous device status data and underground coal mine environmental parameters to the application layer; Extracting the device control parameters of the heterogeneous operating equipment from the control logic layer through the application layer, calling the device status prediction model of the data analysis module based on the heterogeneous equipment model, wherein the training sample equipment of the device status prediction model is less than or equal to the first service life, processing the underground coal mine environmental parameters and the device control parameters to perform status prediction, and obtaining device status prediction data, including: Obtain a second service life for operational heterogeneous equipment; Using the equipment sample indexing model of the data analysis module, the underground coal mine environment attributes and equipment control attributes are set as constant attributes, and a first equipment status record sample that meets the heterogeneous equipment model and the first service life is retrieved, and a second equipment status record sample that meets the heterogeneous equipment model and the second service life is retrieved; Performing homonymous mode analysis on the device state disturbance quantity record data of the first device state record sample and the second device state record sample to obtain a device state disturbance amplitude; Extracting the device control parameters of the heterogeneous equipment from the control logic layer through the application layer, calling the device status prediction model of the data analysis module based on the heterogeneous equipment model, processing the underground coal mine environmental parameters and the device control parameters to perform status prediction, and obtaining initial device status prediction data; Based on the device state disturbance amplitude, the initial device state prediction data is corrected to obtain the device state prediction data; When the device status prediction data is consistent with the heterogeneous device status data, a health mark is performed on the device set of the same model in the visual configuration interface.

2. The method according to claim 1, wherein The device status data includes a plurality of device status data. The device status data of a set of devices of the same model that execute the first task are collected through the hardware abstraction layer, and heterogeneous state sorting is performed to obtain heterogeneous devices for operation, including: Performing similarity analysis on the plurality of device status data to obtain a plurality of device status similarities; Based on the multiple device state similarities, traverse the plurality of device state data to perform local density evaluation to obtain a plurality of device state distribution densities; Using the distribution density mean, traverse the distribution densities of the plurality of device states and calculate the ratio to obtain a plurality of device state heterogeneous parameters; Based on the plurality of device state heterogeneous parameters, device bit numbers whose device state heterogeneous parameters are greater than or equal to a predefined heterogeneous parameter threshold are extracted from the set of devices of the same model, and are added to the operation heterogeneous devices.

3. The method according to claim 2, wherein Based on the multiple device state similarities, traverse the plurality of device state data to perform local density evaluation to obtain a plurality of device state distribution densities, including: Extracting first device status data from the plurality of device status data; extracting, from the plurality of device state similarities, a plurality of selected device state similarities based on the first device state data; From the plurality of selected device state similarities, from large to small, select neighboring selected device state similarities that are 0.5 times the number of device sets of the same model, perform mean calculation, and obtain a first device state distribution density; The first device state distribution density is added to the plurality of device state distribution densities.

4. The method according to claim 1, wherein Also includes: When the number of heterogeneous devices in the operation is equal to 0, a health mark is performed on the device set of the same model in the visual configuration interface; When the device status prediction data is inconsistent with the heterogeneous device status data, the execution abnormality mark of the operating heterogeneous devices of the same model device set is performed on the visual configuration interface, and the health mark of the non-operating heterogeneous devices of the same model device set is performed.

5. The method according to claim 1, wherein The device control parameters of the heterogeneous equipment are extracted from the control logic layer through the application layer, the device status prediction model of the data analysis module is called based on the heterogeneous equipment model, the underground coal mine environmental parameters and the device control parameters are processed to perform status prediction, and the device status prediction data is obtained, including: Based on a preset device model as a constraint, multiple sets of training sample data with a service life less than or equal to the first service life are collected, wherein any set of the multiple sets of training sample data includes device control record data, deployment environment record parameters, and a label identifying device status data; Using the label identifying the device status data as supervision and the device control record data and the deployment environment record parameters as input, training a first device status predictor; Counting a first residual data set of the first device state predictor; When the number of residuals of the first residual data set is greater than or equal to a quantity threshold, training a second device state predictor using the first residual data set as supervision and the device control record data and the deployment environment record parameters as input; Until the number of residuals in the Qth residual data set is less than the quantity threshold, the outputs of the first device state predictor, the second device state predictor, and the Qth device state predictor are summed to obtain the device state prediction model, which is associated with the preset device model and stored, and embedded in the data analysis module.

6. The method according to claim 5, wherein The process of determining the label that identifies the device status data includes: Receive the rated interval of the device control attribute from the control logic layer and the environmental attribute constraint interval from the visual configuration interface, and randomly configure the device control record data and environmental record parameters; Based on the preset device model, the device control record data, and the environmental record parameters, an operation sample set having a service life less than or equal to the first service life is collected, wherein the operation sample set includes a device status data set; Centroidal device state data is extracted from the device state data set to obtain a label identifying the device state data.

7. The method according to claim 1, wherein Performing homologous mode analysis on the device state disturbance quantity record data of the first device state record sample and the second device state record sample to obtain the device state disturbance amplitude includes: Extracting positive disturbance amount record data and negative disturbance amount record data from the device state disturbance amount record data, wherein a positive disturbance amount refers to a second device state record value being greater than a first device state record value, and a negative disturbance amount refers to a second device state record value being less than a first device state record value; Performing homogeneous mode analysis on the recorded data of the forward disturbance to obtain the amplitude of the forward disturbance of the device; Performing homogeneous mode analysis on the negative disturbance record data to obtain the negative disturbance amplitude of the equipment; The positive disturbance amplitude of the device and the negative disturbance amplitude of the device are added to the device state disturbance amplitude.

8. A distributed job node status prediction system based on software-defined control, characterized in that: The method according to any one of claims 1 to 7 is applied to an SDC system, wherein the SDC system includes a hardware abstraction layer, a control logic layer, and an application layer, wherein the application layer includes a visual configuration interface and a data analysis module, and the execution steps include: A data collection and sorting unit is used to collect device status data of a set of devices of the same model that execute the first task through a hardware abstraction layer, perform heterogeneous state sorting, and obtain heterogeneous devices for the operation; A heterogeneous device uploading unit, configured to upload the heterogeneous device model, heterogeneous device status data, and underground coal mine environmental parameters to the application layer when the number of heterogeneous devices in the operation is not equal to 0; A state prediction and analysis unit is configured to extract the device control parameters of the heterogeneous operating equipment from the control logic layer through the application layer, call the device state prediction model of the data analysis module based on the heterogeneous equipment model, process the underground coal mine environmental parameters and the device control parameters to perform state prediction, and obtain device state prediction data; The health identification display unit is used to identify the health of a set of devices of the same model in a visual configuration interface when the device status prediction data is consistent with the heterogeneous device status data.

Images