Operation state monitoring method and device, equipment and storage medium

Abstract

The invention discloses a running state monitoring method and device, equipment and a storage medium, and relates to the technical field of state monitoring. The method comprises the following steps: acquiring original operation data of multiple data types of target equipment; for the original operation data of each data type, data features of the original operation data in a preset time window are extracted, and the time window comprises a plurality of time points; and inputting the data feature combination of all the original operation data in the time window into the trained large model to obtain the operation state of the target equipment output by the large model.

Description

Technical Field

[0001] This application relates to the field of condition monitoring technology, specifically to a method, apparatus, equipment, and storage medium for monitoring operational status. Background Technology

[0002] Against the backdrop of Industry 4.0 and smart manufacturing, the requirements for efficient and reliable equipment operation are increasing. Operational status monitoring, as a key component in ensuring stable equipment operation, enabling early warning of faults, and optimizing equipment maintenance strategies, has received widespread attention.

[0003] Currently, equipment operation status monitoring mainly relies on traditional sensor data and simple threshold alarm systems. Taking a concrete mixer truck as an example, its operation status is typically monitored by collecting data from sensors and making judgments based on preset rules, such as determining that the truck is in operation when its rotational speed exceeds a certain fixed value. However, the thresholds of such methods are often set manually based on experience, making them fixed and difficult to adjust. Since the actual operating status of equipment is affected by various factors, fixed thresholds are difficult to adapt to complex and changing working environments, easily leading to false alarms or missed alarms, resulting in low accuracy in operation status identification. Summary of the Invention

[0004] The purpose of this application is to provide a method, apparatus, device, and storage medium for monitoring operational status.

[0005] To achieve the above objectives, the first aspect of this application provides a method for monitoring operational status, the method comprising: Acquire raw operational data of various data types from the target device; For each type of raw running data, extract the data features of the raw running data within a preset time window, which includes multiple time points; The data features of all raw running data within the time window are combined and input into the trained large model to obtain the running status of the target device output by the large model.

[0006] In this embodiment of the application, extracting data features of the original running data within a preset time window includes: extracting the instantaneous features of the original running data at each time point within the time window, and / or the overall features of the original running data within the time window.

[0007] In this embodiment, the original operating data includes original stirring drum rotation speed data, original vibration data, and original weight data. Instantaneous features of the original operating data at each time point within a time window are extracted, and / or overall features of the original operating data within the time window are extracted, including: extracting the instantaneous rotation speed, over / under the forward / reverse threshold of the original stirring drum rotation speed data at each time point within the time window, and the average rotation speed, rotation speed variance, and rotation speed variation of the original stirring drum rotation speed data within the time window; extracting the vibration energy, vibration peak value, vibration amplitude, and frequency domain features of the original vibration data within the time window; and extracting the instantaneous weight of the original weight data at each time point within the time window, and determining the average weight and weight variation of the original weight data within the time window based on the instantaneous weight.

[0008] In this embodiment of the application, the method further includes a large model training step, which includes: acquiring historical operating data of various data types of the target device under different operating states; extracting historical data features of each data type of historical operating data within a time window for each operating state; and training the large model based on the combination of historical data features of all historical operating data within the time window and the corresponding operating state to obtain a trained large model.

[0009] In the embodiments of this application, the operating state includes multiple steady-state operating states, as well as the switching state between different steady-state operating states.

[0010] In this embodiment of the application, the method further includes: when the operating state is a switching state between different steady-state operating states, identifying the target time point in which the target device switches between different steady-state operating states, and extracting the data features of the target device at the target time point.

[0011] In this embodiment of the application, the operating state includes an abnormal operating state, and the method further includes: when the operating state is an abnormal operating state, controlling the target device to output a prompt message.

[0012] A second aspect of this application provides an operating status monitoring device, comprising: a memory configured to store instructions; a processor configured to retrieve instructions from the memory and to implement an operating status monitoring method when executing the instructions.

[0013] A third aspect of this application provides an operation status monitoring device, comprising: an operation status monitoring apparatus.

[0014] A fourth aspect of this application provides a machine-readable storage medium storing instructions that cause a machine to perform a running state monitoring method.

[0015] The above technical solution acquires various types of raw operational data from the target equipment and extracts data features of each type of raw operational data within a time window. These features are then combined and input into a pre-trained large model to determine the operating status of the target equipment. The large model in this solution, trained on a large amount of data, can learn the complex mapping relationship between different data features and operating status, thereby more accurately determining the operating status of the mixer truck.

[0016] Other features and advantages of the embodiments of this application will be described in detail in the following detailed description section. Attached Figure Description

[0017] The accompanying drawings are provided to further illustrate the embodiments of this application and form part of the specification. They are used together with the following detailed description to explain the embodiments of this application, but do not constitute a limitation on the embodiments of this application. In the drawings: Figure 1 A schematic flowchart of a method for monitoring operational status according to an embodiment of this application is shown. Figure 2 This schematic diagram illustrates the structural block diagram of an operation status monitoring device according to an embodiment of the present application; Figure 3 The diagram illustrates the internal structure of a computer device according to an embodiment of this application. Detailed Implementation

[0018] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. It should be understood that the specific embodiments described herein are only for illustration and explanation of the embodiments of this application and are not intended to limit the embodiments of this application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application.

[0019] Figure 1 The illustration shows a flowchart of a method for monitoring operational status according to an embodiment of this application. Figure 1 As shown in the figure, this application provides a method for monitoring operational status, which may include the following steps: Step 101: Obtain raw operating data of various data types from the target device.

[0020] Among these, various data types reflect the operating status of the equipment from different dimensions, such as rotational speed, vibration, weight, electrical parameters, thermal indicators, and fluid characteristics. Raw operating data is directly acquired from the target equipment through sensors and other acquisition devices, without any processing or manipulation. This data most directly and truthfully reflects the equipment's operating status, preserving its original characteristics and details during operation.

[0021] Optionally, based on the operating characteristics of the equipment and the analysis objectives, the types of data to be collected are defined. For each defined data type, a sensor capable of reliably collecting the corresponding data is selected and installed in a suitable location. During operation, the sensor continuously collects raw operating data according to a preset acquisition sequence, and then actively sends it to the processor at a predetermined frequency, or uploads the corresponding raw operating data after receiving an acquisition command from the processor. Accordingly, the processor obtains raw operating data of various data types from the target equipment.

[0022] For example, speed sensors are installed on the mixer truck to collect the rotational speed data of the mixing drum. Speed ​​sensors are typically installed on the hydraulic motor or the slewing support of the mixing drum, and their working principle is based on electromagnetic induction or photoelectric induction technology. Taking an electromagnetic induction speed sensor as an example, when the mixing drum rotates, the sensing element inside the sensor cuts magnetic lines of force, generating an electrical signal proportional to the rotational speed. After processing by a conditioning circuit, this signal can accurately measure the rotational speed (RPM) and direction of rotation (forward / reverse). In some large mixer trucks, the accuracy of the speed sensor can reach ±0.1 RPM, providing a high-precision speed data basis for subsequent condition monitoring.

[0023] For example, vibration data is collected by installing vibration sensors on the mixer truck. Vibration sensors are typically installed near the chassis or the bearing housing of the mixing drum, and often employ piezoelectric or accelerometer-type structures. Piezoelectric vibration sensors utilize the property that piezoelectric materials generate an electrical charge when subjected to vibration-induced pressure, converting the vibration signal into an electrical signal. Accelerometer-type vibration sensors, on the other hand, reflect vibration conditions by measuring vibration acceleration. These sensors can measure acceleration values ​​in three axes (X, Y, Z), comprehensively capturing vehicle motion, engine vibration, road impacts, and the impact of concrete sliding inside the drum. In practical applications, the frequency response range of vibration sensors typically covers 0.1Hz-10kHz, effectively detecting vibration data at various frequencies and providing rich vibration data for feature extraction and state assessment.

[0024] For example, weight data is collected by installing weight sensors on the mixer truck. These weight sensors can be directly mounted at the support points of the mixing drum and include both resistance strain gauge and capacitive types. Resistance strain gauge load cells change resistance under pressure, and the weight can be calculated by measuring this change in resistance. Capacitive load cells measure weight by utilizing the relationship between capacitance change and pressure. These sensors can accurately measure the weight of materials inside the mixing drum with an accuracy of ±0.5%FS (full scale), providing crucial weight data support for determining the loading and unloading status of the mixer truck.

[0025] In addition to the data mentioned above, other suitable data types, including but not limited to vehicle speed, can be introduced to further improve the accuracy of the results, depending on the specific application scenario and objectives.

[0026] Step 102: For each type of raw running data, extract the data features of the raw running data within a preset time window. The time window includes multiple time points.

[0027] In this context, a time window refers to a continuous period of time selected on the timeline, containing multiple time points, used for segmented feature extraction of the original operational data. Data features refer to information that can represent the essential characteristics of the data.

[0028] Optionally, the raw operating data can be divided into multiple consecutive time windows according to preset time window parameters. For example, if the sensor continuously collects weight data for 30 minutes at a frequency of once per minute, and the preset time window is 5 minutes, it can be divided into 6 consecutive 5-minute windows, each containing 5 weight data points.

[0029] Furthermore, within each time window, statistical characteristics, time-domain characteristics, and frequency-domain characteristics can be calculated. By extracting and organizing the data characteristics of various types of raw operational data within each time window, the corresponding combination of data characteristics for that window can be obtained.

[0030] In one feasible implementation, the instantaneous features of the original running data at each time point within the time window are extracted, and / or the overall features of the original running data within the time window are extracted.

[0031] Among them, instantaneous features refer to the features calculated for each independent time point within the time window, reflecting the operating status of the equipment at that time point; while overall features are features extracted based on the data within the entire time window, used to describe the overall performance of the equipment during that period.

[0032] For each time point within a time window, features can be extracted using appropriate methods based on the determined instantaneous feature type. For example, for numerical features, the data value at that time point can be directly read; for statistical features, the mean, variance, etc., of the data within a certain range near that time point can be calculated. Simultaneously, based on the determined overall feature type, appropriate calculation methods are used to process the data within each time window to extract overall features. For example, the mean, standard deviation, and other statistical features of the data within the time window can be calculated; trend characteristics of the data can be analyzed using curve fitting methods.

[0033] For example, the instantaneous rotational speed and over / under the forward / reverse speed thresholds at each time point within a time window are extracted from the original mixing drum rotational speed data, along with the average rotational speed, speed variance, and speed variation within the time window. Instantaneous rotational speed reflects the current rotational speed of the mixing drum in real time and is a crucial indicator for judging the immediate operating status of the mixer truck. For instance, during mixing operations, the instantaneous rotational speed is usually stable within a set range; however, its value changes rapidly during start-up or shutdown. Over / under the forward / reverse speed thresholds refers to whether the rotational speed exceeds a preset forward or reverse speed threshold. If the speed exceeds the corresponding threshold, it can be determined that the mixer truck is in an abnormal operating state; for example, reverse rotation of the mixing drum may be caused by a control system malfunction or misoperation. The average rotational speed is calculated by averaging the instantaneous rotational speed over a period of time. It smooths out fluctuations in instantaneous rotational speed and reflects the average operating speed of the mixing drum within that time period, helping to assess the stability of the mixer truck over a longer period. If the average rotational speed continuously deviates from the normal rotational speed range, it may indicate a equipment malfunction or abnormal operating condition. The speed variance measures the dispersion of the rotational speed data, i.e., the stability of the rotational speed. A smaller variance indicates a more stable rotational speed; a larger variance indicates significant fluctuations in rotational speed. During normal operation, the rotational speed variance of a mixer truck typically remains within a small range. A sudden increase in variance may be caused by factors such as hydraulic system malfunctions or changes in mixing resistance due to uneven material distribution. The rotational speed variation, i.e., the rotational speed trend within a time window, can be determined through linear regression analysis of the instantaneous rotational speed within the window to determine whether the speed is steadily increasing, decreasing, or remaining stable. During the loading process, as the amount of material increases, the mixing resistance increases, and the rotational speed may gradually decrease; conversely, during unloading, the mixing resistance decreases, and the rotational speed may increase. Analyzing the rotational speed trend within an event window allows for a more accurate assessment of the mixer truck's operating stage.

[0034] For example, the vibration energy, peak value, amplitude, and frequency domain characteristics of the original vibration data within a time window are extracted. Vibration energy is calculated by the root mean square (RMS) value of the vibration data within the time window, reflecting the average vibration energy level during that period. During the mixer truck's operation, uneven road surfaces increase vibration energy, and the RMS value also increases accordingly; while when the mixing drum is stably mixing materials, the vibration energy is relatively stable, and the RMS value remains within a relatively fixed range. Therefore, the RMS value can serve as an important basis for judging the mixer truck's road conditions and mixing operation status. The peak value reflects the maximum amplitude of the vibration data. During mixer truck operation, instantaneous impacts such as engine starting and material input generate vibration peaks. By monitoring the magnitude and frequency of these peaks, it is possible to determine whether the equipment has been subjected to abnormal impacts; for example, excessively high peak values ​​may indicate foreign object collisions within the mixing drum or loosening of mechanical parts. Vibration amplitude refers to the range of variation in vibration data, which is related to the intensity of the vibration. The vibration amplitude of the mixer truck will vary under different operating conditions. For example, the vibration amplitude is relatively small when the truck is empty; while it increases when fully loaded or during high-intensity mixing operations. Analyzing changes in vibration amplitude helps to understand the load and operating status of the mixer truck. The frequency domain characteristics are obtained by converting time-domain data to the frequency domain using a Fast Fourier Transform. For example, engine operation produces vibrations at specific frequencies, tire rotation during vehicle movement exhibits another set of frequency characteristics, and the natural frequencies of an empty and full mixing drum also change due to their different overall masses.

[0035] For example, the instantaneous weight of the original weight data at each time point within a time window is extracted, and the average weight and weight change of the original weight data within the time window are determined based on the instantaneous weight. The instantaneous weight reflects the current weight of the material in the mixing drum in real time and is a direct indicator of the mixer truck's loading status. During loading, the instantaneous weight gradually increases; during unloading, the instantaneous weight gradually decreases. By monitoring the changes in instantaneous weight in real time, the material loading and unloading status of the mixer truck can be intuitively understood. The average weight is calculated by averaging the instantaneous weight over a period of time, eliminating the influence of instantaneous weight fluctuations and reflecting the average loading amount of the mixing drum during that period. For example, if the average value increases slowly or stops increasing during continuous loading over a period of time, it may indicate an abnormality in the loading process, such as hopper blockage. The weight change is mainly reflected in two aspects: the rate of change and the direction of change. By calculating the rate of change of the instantaneous weight within the time window (e.g., the weight is increasing linearly at a rate of ~X kg / s) and the direction of change (positive / negative), the dynamic characteristics of material loading and unloading can be effectively identified. During the loading process, the weight change rate is positive and relatively stable, directly corresponding to the "loading" state; while during the unloading process, the change rate turns negative. By analyzing the weight change rate and its direction, we can more accurately determine the material loading and unloading status of the mixer truck and its changing trend, providing important data support for intelligent scheduling and production management.

[0036] Step 103: Input the combined data features of all the original running data within the time window into the trained large model to obtain the operating status of the target device output by the large model.

[0037] Larger models can employ state machine-based time-series classification models, such as Long Short-Term Memory (LSTM), Gated Recurrent Unit (GRU), and Transformer models. These models possess a "memory" function, enabling them to understand the temporal dependencies of data. Taking the LSTM model as an example, it contains input gates, forget gates, and output gates. These gate structures allow for the selective memorization and forgetting of historical information, thus better handling long-term dependencies in time series data. In the monitoring of mixer truck operation status, the current state of the equipment is often closely related to its previous operating state. The LSTM model can effectively capture this time-series characteristic, improving the accuracy of state identification.

[0038] During the model training phase, historical operating data of various data types from the target device under different operating states are acquired, and historical data features are extracted within a set time window. The historical data features extracted from each data type within each time window are combined according to a specific format to form a historical data feature set. These historical data feature sets are used as input data, with the corresponding operating states as labels, and are divided into training, validation, and test sets. To improve the model's generalization ability, data augmentation and regularization techniques can be used to expand and process the data to avoid overfitting. After selecting a large model and determining its training parameters, the training set is input into the large model, and training is performed according to the set training parameters. Supervised learning algorithms, such as cross-entropy loss function combined with stochastic gradient descent optimization algorithm, can be used to continuously adjust the model's parameters so that the model's prediction results are as close as possible to the labels, ultimately obtaining a well-trained large model.

[0039] During the model application phase, various data features extracted within each time window are combined according to a specific format to form a data feature combination, which is then input into a pre-trained large model. The large model analyzes and judges the input data feature combination based on its internally learned patterns and rules. When the model finds an operating state that highly matches the data feature combination, it outputs that operating state and its probability, such as "98% probability of loading state." This judgment process is completed by calculating the similarity between the input features and each state, combined with a preset judgment threshold. Furthermore, to improve the accuracy of state judgment, the outputs of multiple large models can be integrated for comprehensive decision-making, or a voting mechanism can be used to reduce the possibility of misjudgment.

[0040] It's important to note that the operating states here include not only various steady-state operating states but also the transition states between different steady-state operating states. A steady-state operating state refers to a state where various data of the equipment remain relatively stable over a period of time, such as the loading state. Its corresponding characteristic combination is that the rotational speed stabilizes within a specific range, the weight shows a continuous increasing trend, and the vibration exhibits specific frequency characteristics. A transition state refers to the transitional state experienced by the target equipment when changing from one steady-state operating state to another. In this state, the various parameters of the equipment will change dynamically, no longer remaining stable but gradually adjusting according to certain rules to adapt to the new steady-state operating conditions. For example, when switching from the "loading" state to the "transporting" state, the corresponding data characteristic combination is that the weight stops increasing and stabilizes at a high value, the rotational speed gradually stabilizes, and the vibration characteristics become dominated by the vehicle's travel frequency.

[0041] In addition to the large model determining that the device is in a switching state, more detailed analysis can be performed to determine the target time point for the switching. For example, the large model can directly capture the target time point for the state switching; alternatively, the rate of change of data features can be calculated, and when the rate of change exceeds a certain threshold, that time point is considered the target time point; or a sliding window method can be used to find the time point within the window where the data features change most significantly as the target time point. After determining the target time point, corresponding data features can be extracted before and after it. For example, using this point as the center, data within a certain time range can be selected, and the average value, standard deviation, and other data features of each monitoring parameter within that period can be calculated, thus providing complete and accurate data records to support subsequent analysis and decision-making.

[0042] In addition to the various normal operating states mentioned above, this solution can also cover the identification of abnormal operating states. The large-scale model establishes an abnormal operating state database through learning from a large number of abnormal data samples. When a detected combination of data features matches an item in the database, the current state is determined to be abnormal. For example, if the extracted data feature combination is that the mixing drum is rotating at a forward speed of 3 RPM, but the weight data shows zero and is accompanied by severe shaking, inputting this data feature combination into the large-scale model will determine it as an "idling" abnormality, rather than a normal operating state. In this case, a timely warning is issued, along with preliminary fault cause analysis and handling suggestions. For example, for an "idling" abnormality, the system may prompt the user to check whether the material supply system or the connecting parts of the mixing drum are loose.

[0043] In addition, at the business application level, the following solutions can also be deployed: Full-process visual dashboard: The control and dispatch center's large screen displays the real-time status of all vehicles (green for driving, yellow for loading, and red for unloading).

[0044] Accurate Volumetric Automated Calculation: Weight sensors monitor weight changes in real time during the unloading process. The initial weight is recorded at the start of unloading, and the final weight is recorded at the end. Based on preset parameters such as material density, the weight difference is converted into the actual unloading volume and automatically compared with order requirements. If a discrepancy is detected, an immediate alert is issued. This function not only improves the accuracy and efficiency of volumetric calculation but also prevents "material theft" from an information technology perspective.

[0045] Intelligent dispatching: When a vehicle is detected to be in the "unloading" state, a preparation notification can be automatically sent to the next vehicle in the queue, indicating that unloading is about to begin, and simultaneously displaying information such as the estimated waiting time. At the same time, the dispatch center can coordinate unloading sites and arrange subsequent loading tasks in advance based on the system's prompts. This intelligent dispatching mechanism effectively reduces vehicle waiting time, improves the overall operational efficiency of the fleet, and lowers overall operating costs.

[0046] In this application, raw operational data of various data types from the target device are acquired, and data features within a preset time window are extracted. These features are then combined and input into a pre-trained large model to obtain the operational status. The large model learns from a large amount of data and can automatically mine complex patterns and relationships in the data, no longer limited to simple fixed threshold judgments. This avoids false alarms and false negatives caused by unreasonable fixed threshold settings.

[0047] Figure 1 This is a flowchart illustrating a runtime status monitoring method in one embodiment. It should be understood that, although... Figure 1 The steps in the flowchart are shown sequentially as indicated by the arrows, but these steps are not necessarily executed in the order indicated by the arrows. Unless otherwise explicitly stated herein, there is no strict order in which these steps are executed, and they can be performed in other orders. Figure 1 At least some of the steps in the process may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these sub-steps or stages is not necessarily sequential, but can be executed in turn or alternately with other steps or at least some of the sub-steps or stages of other steps.

[0048] In one embodiment, such as Figure 2 As shown, an operation status monitoring device 200 is provided, including a raw operation data acquisition module, a data feature extraction module, and an operation status determination module, wherein: The raw operating data acquisition module 201 is used to acquire raw operating data of various data types from the target device.

[0049] The data feature extraction module 202 is used to extract data features of the raw running data within a preset time window for each type of raw running data. The time window includes multiple time points.

[0050] The running status determination module 203 is used to input the combined data features of all raw running data within the time window into the trained large model to obtain the running status of the target device output by the large model.

[0051] The operation status monitoring device includes a processor and a memory. The above-mentioned raw operation data acquisition module, data feature extraction module, and operation status determination module are all stored as program units in the memory. The processor executes the above-mentioned program modules stored in the memory to realize the corresponding functions.

[0052] The processor contains a kernel, which retrieves the corresponding program units from memory. One or more kernels can be configured, and methods for monitoring the processor's runtime status can be implemented by adjusting kernel parameters.

[0053] The memory may include non-permanent memory in computer-readable media, such as random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM, and the memory includes at least one memory chip.

[0054] This application provides a storage medium on which a program is stored, which, when executed by a processor, implements the above-described operation status monitoring method.

[0055] This application provides a processor for running a program, wherein the program executes the above-described running status monitoring method during runtime.

[0056] In one embodiment, a computer device is provided, which may be a terminal, and its internal structure diagram may be as follows: Figure 3 As shown in the figure, the computer device includes a processor A01, a network interface A02, a display screen A04, an input device A05, and a memory (not shown) connected via a system bus. The processor A01 provides computing and control capabilities. The memory includes internal memory A03 and a non-volatile storage medium A06. The non-volatile storage medium A06 stores an operating system B01 and a computer program B02. The internal memory A03 provides an environment for the operation of the operating system B01 and the computer program B02 stored in the non-volatile storage medium A06. The network interface A02 is used for communication with external terminals via a network connection. When the computer program is executed by the processor A01, it implements a method for monitoring operating status. The display screen A04 can be a liquid crystal display (LCD) or an e-ink display. The input device A05 can be a touch layer covering the display screen, buttons, a trackball, or a touchpad mounted on the computer device casing, or an external keyboard, touchpad, or mouse.

[0057] Those skilled in the art will understand that Figure 3 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0058] This application provides a computer (electronic) device, which includes a processor, a memory, and a program stored in the memory and executable on the processor. When the processor executes the program, it implements the steps of any of the above-mentioned operation status monitoring methods.

[0059] This application also provides a computer program product that, when executed on a data processing device, is suitable for executing a program with initialization steps for monitoring operational status.

[0060] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0061] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations 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, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0062] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0063] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0064] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0065] Memory may include non-persistent memory in computer-readable media, such as random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM. Memory is an example of computer-readable media.

[0066] Computer-readable media includes both permanent and non-permanent, removable and non-removable media that can store information using any method or technology. Information can be computer-readable instructions, data structures, modules of programs, or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, CD-ROM, digital versatile optical disc (DVD) or other optical storage, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other non-transferable medium that can be used to store information accessible by a computing device. As defined herein, computer-readable media does not include transient computer-readable media, such as modulated data signals and carrier waves.

[0067] It should also be noted that 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 process, method, article, or apparatus. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element.

[0068] The above are merely embodiments of this application and are not intended to limit the scope of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of the claims of this application.

Claims

1. A method for monitoring operational status, characterized in that, The method includes: Acquire raw operational data of various data types from the target device; For each type of raw running data, extract the data features of the raw running data within a preset time window, where the time window includes multiple time points; The data features of all the original running data within the time window are combined and input into the trained large model to obtain the running status of the target device output by the large model.

2. The operation status monitoring method according to claim 1, characterized in that, Extracting data features from the raw operational data within a preset time window includes: Extract the instantaneous features of the original running data at each time point within the time window, and / or the overall features of the original running data within the time window.

3. The operation status monitoring method according to claim 2, characterized in that, The raw operating data includes raw stirring drum rotation speed data, raw vibration data, and raw weight data. The instantaneous features of the raw operating data at each time point within the time window are extracted, and / or the overall features of the raw operating data within the time window are extracted, including: Extract the instantaneous speed, over-reverse / forward / reverse threshold, and average speed, speed variance, and speed variation of the original stirring drum speed data at each time point within the time window; Extract the vibration energy, peak vibration value, vibration amplitude, and frequency domain characteristics of the original vibration data within the time window; Extract the instantaneous weight of the original weight data at each time point within the time window, and determine the average weight and weight change of the original weight data within the time window based on the instantaneous weight.

4. The operation status monitoring method according to claim 1, characterized in that, The method further includes a training step for the large model, the training step comprising: Acquire historical operating data of the target device in various operating states, including multiple data types. For each operating state, extract the historical data features of each data type within the time window; Based on the historical data feature combinations of all historical running data within the time window and the corresponding running states, the large model is trained to obtain the trained large model.

5. The operation status monitoring method according to any one of claims 1-4, characterized in that, The operating states include multiple steady-state operating states, as well as the switching states between different steady-state operating states.

6. The operation status monitoring method according to claim 5, characterized in that, The method further includes: When the operating state is a switching state between different steady-state operating states, the target time point in which the target device switches between different steady-state operating states is identified, and the data features of the target device at the target time point are extracted.

7. The operation status monitoring method according to claim 1, characterized in that, The operating state includes an abnormal operating state, and the method further includes: When the operating state is the abnormal operating state, control the target device to output a prompt message.

8. A device for monitoring operational status, characterized in that, include: The memory is configured to store instructions; A processor is configured to retrieve the instructions from the memory and, when executing the instructions, to implement the operational status monitoring method according to any one of claims 1 to 7.

9. An operational status monitoring device, characterized in that, include: The operation status monitoring device according to claim 8.

10. A machine-readable storage medium storing instructions thereon, characterized in that, When executed by the processor, the instruction causes the processor to be configured to perform the operation status monitoring method according to any one of claims 1 to 7.

Images