Method for monitoring driving state of mobile machine in thermal regeneration scene

Abstract

The invention discloses a method for monitoring the driving state of a mobile machine in a thermal regeneration scene. The method comprises the steps of installing a plurality of sensors on the mobile machine; the method comprises the following steps: acquiring sensing data, constructing a driving state anomaly detection model in an ROS data processing system, inputting the sensing data into the driving state anomaly detection model, and performing training and hyper-parameter optimization on the driving state anomaly detection model to obtain an optimized driving state anomaly detection model; and carrying out anomaly detection on the driving state of the mobile machine based on the optimized driving state anomaly detection model, and when an anomaly detection result is abnormal, carrying out visual display and cloud storage on abnormal data, and sending out early warning information. According to the method, the driving state anomaly detection model is constructed and optimized, the sensing data of the mobile machinery and various environmental influence data are combined, accurate detection of the driving state is realized, more deep and comprehensive driving state monitoring is provided in combination with visual display and cloud storage, and wider data sharing and remote cooperation are supported.

Description

Technical Field

[0001] The present invention relates to the technical field of mobile machinery monitoring, and in particular to a method for monitoring the driving state of a mobile machinery in a thermal regeneration scenario. Background Art

[0002] In modern industry and construction, mobile machinery such as excavators, bulldozers and cranes are increasingly used in thermal regeneration scenarios. The safety and efficiency of these machines during operation are crucial. Currently, monitoring the driving status of these heavy machines mainly relies on traditional sensor technology and simple data processing methods. Traditional systems usually include speed and angle sensors, as well as basic data recording devices to monitor and record the basic operating parameters of the machinery.

[0003] The data processing of existing technologies mainly focuses on real-time parameter display and basic alarm systems. These systems can provide some basic driving status information, such as speed, acceleration and mechanical tilt angle. However, the processing and analysis of this information still mostly depends on the operator's experience and judgment. In some cases, the operator needs to rely on multiple different interfaces and devices to obtain complete driving status information, which not only increases the complexity of operation, but also may distract the operator's attention, thereby reducing the safety and efficiency of the operation.

[0004] In addition, existing systems have obvious deficiencies in in-depth data analysis and long-term trend monitoring. They are generally unable to effectively process and analyze large amounts of historical data, and lack the ability to predict potential risks and anomalies. This means that even when there is a potential safety risk, operators may not be able to get timely warnings. Summary of the invention

[0005] The technical problem to be solved by the present invention is to overcome the above technical defects and provide a method for monitoring the driving status of a mobile machinery in a thermal regeneration scenario.

[0006] In order to solve the above problems, the technical solution of the present invention comprises the following steps:

[0007] S1. Installing multiple sensors including but not limited to three-dimensional accelerometers, gyroscopes and environmental sensors on mobile machinery;

[0008] S2, collecting sensor data provided by the above-mentioned sensor, wherein the sensor data includes driving state data of the mobile machine and external environment data;

[0009] S3. Use the ROS data processing system to establish a driving state abnormality detection model including a spatial feature extraction module, a spatiotemporal feature extraction module, a feature fusion module and a fully connected layer;

[0010] S4, inputting the collected sensor data into the driving state abnormality detection model, and obtaining an optimized driving state abnormality detection model through training and hyperparameter optimization;

[0011] S5. Use the optimized driving state anomaly detection model to perform real-time anomaly detection on the driving state of mobile machinery. Once the detection result shows that there is an anomaly, the abnormal data will be visualized and stored in the cloud, and an early warning message will be issued.

[0012] Furthermore, the external environment data includes but is not limited to data such as external temperature, humidity and light intensity.

[0013] Furthermore, the training process of the driving state abnormality detection model includes the following steps:

[0014] S4.1, dividing the sensor data into a training set, a test set and a validation set according to a certain ratio;

[0015] S4.2. Obtaining several hyperparameter combinations in the driving state abnormality detection model;

[0016] S4.3, training the model corresponding to each of the hyperparameter combinations based on the training set, and evaluating the model performance using the validation set, expressed as an evaluation score;

[0017] S4.4, using the firefly improved sparrow search algorithm to optimize and select the hyperparameter combination;

[0018] S4.5. Finally, an optimized driving state abnormality detection model is constructed based on the optimal hyperparameter combination.

[0019] Furthermore, the driving state abnormality detection model includes the following steps:

[0020] S5.1, inputting sensor data into the driving state abnormality detection model;

[0021] S5.2, extracting spatial features and temporal features of the sensor data through a spatial feature extraction module and a spatiotemporal feature extraction module respectively;

[0022] S5.3, fusing the extracted spatial features and temporal features into comprehensive features;

[0023] S5.4, input the comprehensive features into the fully connected layer for anomaly detection to obtain the final anomaly detection result;

[0024] S5.5. When the abnormality detection result exceeds a preset threshold, it is determined that an abnormality exists.

[0025] Furthermore, the warning information may be issued in a form including but not limited to sound, light or other suitable means to alert the operator.

[0026] Furthermore, the driving status data includes but is not limited to data such as speed, angle, temperature, pressure, three-dimensional acceleration and posture.

[0027] Furthermore, before the sensor data is input into the driving state abnormality detection model, it also includes preprocessing steps of data cleaning, data integration, data normalization and time series conversion of the sensor data.

[0028] Further, using the firefly improved sparrow search algorithm to optimize the hyperparameter combination includes the following steps:

[0029] S4.4.1, randomly initialize the position of the hyperparameter combination population;

[0030] S4.4.2. Update the position of the individual in the population, calculate the evaluation score of each hyperparameter combination, and update the position of the individual accordingly;

[0031] S4.4.3, further update the position of the hyperparameter combination by introducing the firefly perturbation mechanism;

[0032] S4.4.4. Repeat steps S4.4.2 and S4.4.3 until the predetermined number of iterations is reached, and output the optimal hyperparameter combination.

[0033] Furthermore, during the construction and optimization of the driving state abnormality detection model, a ROS data processing system is used for data processing and model training.

[0034] The advantages of the present invention compared with the prior art are:

[0035] 1. The present invention provides a method for monitoring the driving state of a mobile machine in a thermal regeneration scenario. By constructing and optimizing a driving state abnormality detection model and combining the sensor data of the mobile machine and various environmental impact data, the method realizes accurate detection of the driving state. At the same time, the method combines visualization and cloud storage to provide a more in-depth and comprehensive driving state monitoring, which can not only improve the safety and efficiency of the operation, but also support more extensive data sharing and remote collaboration, thereby meeting the needs of modern industrial applications.

[0036] 2. The present invention provides a method for monitoring the driving status of mobile machinery in a thermal regeneration scenario. Through real-time monitoring and early warning, the potential risk of accidents caused by abnormal driving status is reduced, and work efficiency and safety are improved. At the same time, comprehensive data support is provided, so that managers can better understand the operating status of mobile machinery and make more scientific and reasonable decisions. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] Figure 1It is an implementation flow chart of a method for monitoring the driving status of a mobile machinery in a thermal regeneration scenario according to the present invention.

[0038] Figure 2 It is a schematic diagram of component connections of a method for monitoring the driving status of a mobile machinery in a thermal regeneration scenario of the present invention. DETAILED DESCRIPTION

[0039] Exemplary embodiments will be described in detail herein, examples of which are shown in the accompanying drawings. When the following description refers to the drawings, the same numbers in different drawings represent the same or similar elements unless otherwise indicated. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present disclosure. Instead, they are merely examples of devices consistent with some aspects of the present disclosure as detailed in the appended claims.

[0040] The technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments; based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0041] like Figure 1 and Figure 2 As shown, this embodiment proposes a method for monitoring the driving state of a mobile machine in a thermal regeneration scenario, comprising the following steps:

[0042] S1. Installing multiple sensors including but not limited to three-dimensional accelerometers, gyroscopes and environmental sensors on mobile machinery;

[0043] S2, collecting sensor data provided by the above sensors, where the sensor data includes driving state data of the mobile machinery and external environment data;

[0044] S3. Use the ROS data processing system to establish a driving state abnormality detection model including a spatial feature extraction module, a spatiotemporal feature extraction module, a feature fusion module and a fully connected layer;

[0045] S4, inputting the collected sensor data into the driving state abnormality detection model, and obtaining an optimized driving state abnormality detection model through training and hyperparameter optimization;

[0046] S5. Use the optimized driving state anomaly detection model to perform real-time anomaly detection on the driving state of mobile machinery. Once the detection result shows that there is an anomaly, the abnormal data will be visualized and stored in the cloud, and an early warning message will be issued.

[0047] Furthermore, the driving status data includes but is not limited to data such as speed, angle, temperature, pressure, three-dimensional acceleration and posture.

[0048] Furthermore, the external environment data includes but is not limited to data such as external temperature, humidity and light intensity.

[0049] Furthermore, before the sensor data is input into the driving state abnormality detection model, the preprocessing steps of data cleaning, data integration, data normalization and time series conversion of the sensor data are also included.

[0050] Furthermore, the training process of the driving state abnormality detection model includes the following steps:

[0051] S4.1. Divide the sensor data into training set, test set and validation set according to a certain ratio;

[0052] S4.2, obtaining several hyperparameter combinations in the driving state abnormality detection model;

[0053] S4.3. Train the model corresponding to each hyperparameter combination based on the training set, and use the validation set to evaluate the model performance, expressed as an evaluation score;

[0054] S4.4, using the firefly improved sparrow search algorithm to optimize the selection of hyperparameter combinations;

[0055] S4.5. Finally, an optimized driving state abnormality detection model is constructed based on the optimal hyperparameter combination.

[0056] Furthermore, the optimization selection of the hyperparameter combination using the Firefly improved sparrow search algorithm includes the following steps:

[0057] S4.4.1, randomly initialize the position of the hyperparameter combination population;

[0058] S4.4.2. Update the position of the individual in the population, calculate the evaluation score of each hyperparameter combination, and update the position of the individual accordingly;

[0059] S4.4.3, further update the position of the hyperparameter combination by introducing the firefly perturbation mechanism;

[0060] S4.4.4. Repeat steps S4.4.2 and S4.4.3 until the predetermined number of iterations is reached, and output the optimal hyperparameter combination.

[0061] Furthermore, the driving state abnormality detection model includes the following steps:

[0062] S5.1. Input sensor data to the driving state abnormality detection model;

[0063] S5.2, extracting spatial features and temporal features of the sensor data through a spatial feature extraction module and a spatiotemporal feature extraction module respectively;

[0064] S5.3, fusing the extracted spatial features and temporal features into comprehensive features;

[0065] S5.4, input the comprehensive features into the fully connected layer for anomaly detection to obtain the final anomaly detection result;

[0066] S5.5. When the abnormality detection result exceeds the preset threshold, it is determined that an abnormality exists.

[0067] Furthermore, in the process of building and optimizing the driving state abnormality detection model, the ROS data processing system is used for data processing and model training.

[0068] Furthermore, the warning information may be issued in the form of sound, light or other suitable means to alert the operator.

[0069] A method for monitoring the driving state of a mobile machine in a thermal regeneration scenario, comprising the following components:

[0070] The sensor module is configured on the mobile machinery, including a three-dimensional accelerometer, a gyroscope and an environmental sensor, which is used to enhance the position and attitude monitoring capabilities of the mobile machinery and collect driving status data in real time, including speed, angle, temperature and pressure; for environmental data, including external temperature data, external humidity data and external light data, it can also increase the monitoring of meteorological conditions such as wind speed, wind direction, air pressure, and the perception of surrounding obstacles and traffic flow. By evaluating the potential impact of external environmental conditions, it provides a more comprehensive basis for driving behavior analysis and risk prediction.

[0071] The data processing unit, based on ROS, is connected to the sensor module and includes a high-performance processor for executing machine learning and artificial intelligence algorithms. It is used to receive and process the collected data, analyze driving behavior in real time, identify abnormal patterns, predict potential driving risks and store them. Using ROS (Robot Operating System) as a data processing platform improves the efficiency and real-time performance of data processing and transmission.

[0072] The display and alarm unit is connected to the data processing unit and includes a multifunctional graphical user interface (GUI) for providing real-time visual feedback to the operator based on the analysis results, and issuing alarms and providing driving suggestions when abnormal conditions are detected; the GUI is designed so that the operator can intuitively understand the driving status and system warnings, and the interface includes but is not limited to real-time charts, dynamic indicators and customized alarms. A customized GUI can be used to provide a highly customized graphical user interface, allowing users to adjust the interface layout, color theme and display content according to personal preferences and work needs. Interactive charts, dynamic simulations and three-dimensional visualization technology are used to enable operators to more intuitively understand the driving status and monitoring results.

[0073] The communication module is used to support data exchange between the system and external devices or networks, including wireless communication and wired connection, and supports multiple communication protocols, including but not limited to Wi-Fi, Bluetooth, NFC and LTE, to achieve efficient data exchange with other devices and cloud services.

[0074] Cloud platform for storing data.

[0075] A method for monitoring the driving status of mobile machinery in a thermal regeneration scenario, with a specific implementation process:

[0076] According to the structure of the mobile machine and the test requirements, sensors are arranged reasonably around the mobile machine to ensure that the entire surrounding environment of the mobile machine is covered. Sensors and equipment are started to collect information such as the movement status of the mobile machine and the surrounding environment in real time. Data is synchronized and recorded in real time to ensure the timing consistency between different data sources.

[0077] The collected sensor data is processed, including the following processing flow:

[0078] The collected raw data is cleaned to remove noise, erroneous data and redundant data to ensure the accuracy and validity of the data. The data from different sensors are integrated to form a unified data set. The data is normalized to eliminate the impact of different dimensions on the data model. The data is converted into a time series format suitable for model processing to ensure the data quality and consistency of the input model.

[0079] A deep learning model integrating CNN and LSTM, namely the driving state abnormality detection model, is constructed in the ROS data processing system. The model combines the spatial feature extraction capability of CNN and the time series modeling capability of LSTM. It includes an input layer, CNN module, LSTM module, fusion layer and prediction layer connected in sequence. The deep learning algorithm is used for pattern recognition and predictive analysis to improve the accuracy and response speed of the monitoring system.

[0080] Training and hyperparameter optimization of the driving state abnormality detection model includes the following steps:

[0081] Divide the sensor data into training set, test set and validation set according to the preset ratio;

[0082] Obtain several hyperparameter combinations in the driving state anomaly detection model;

[0083] The driving state abnormality detection model corresponding to each hyperparameter combination is trained based on the training set, and the performance of the driving state abnormality detection model corresponding to each hyperparameter combination is verified and evaluated through the validation set, and the evaluation results are used as the evaluation scores corresponding to each hyperparameter combination;

[0084] Firefly perturbation improved sparrow search algorithm: In the process of building the deep learning model, the network parameters are randomly initialized, which will cause the generation of local optimal solutions. Then the firefly improved sparrow search algorithm is used to optimize the deep learning model. FI SSA updates the positions of all sparrows and the optimal sparrow using firefly perturbation to improve the searchability, and compares the sparrows after disturbance with the sparrows before disturbance. If it is better, the sparrow position is updated.

[0085] Hyperparameters are mainly divided into two categories. One is training hyperparameters (the model is trained using a small batch random algorithm by default), which mainly include learning rate, batch size, and number of training rounds. The other is the structural hyperparameters of the model. Different models often have different structural hyperparameters.

[0086] Construct a sparrow population for sparrows using each hyperparameter combination, set the value range, population size and number of iterations of the hyperparameter combination, randomly initialize the position of the sparrow population, update the predator position, joiner position and sentinel position, calculate the evaluation score corresponding to each sparrow, update the position of each sparrow in the sparrow population based on the calculated evaluation score, update the sparrow position through firefly disturbance, calculate the evaluation score corresponding to each sparrow after the firefly disturbance, update the position of each sparrow in the sparrow population based on the calculated evaluation score, repeat steps S1 to S3 for iterative update until the iterative update reaches a preset number of iterations, output the optimal diagnostic model hyperparameter combination, and construct an optimized driving state abnormality detection model based on the optimal hyperparameter combination.

[0087] In the training process of the driving state abnormality detection model, the sensor data is input into the driving state abnormality detection model for abnormality detection, and training is performed according to the target loss function to obtain a trained driving state abnormality detection model. The target loss function adopts cross entropy loss (Cross Entropy Loss) or mean square error loss (Mean Square Error, MSE).

[0088] The processing process of the driving state abnormality detection model directly inputs the sensor data (such as acceleration, speed, direction, mobile machinery posture, etc.) collected from the mobile machinery in real time into the driving state abnormality detection model, and uses the spatial feature extraction module in the model to deeply analyze the input sensor data and extract the spatial features directly related to the driving state of the mobile machinery. These spatial features may include the driving trajectory, acceleration distribution, directional stability, etc. of the mobile machinery, which together reflect the motion state of the mobile machinery in the spatial dimension. Through the spatiotemporal feature extraction module, the sensor data is analyzed in time series to capture the trend and pattern of data changes over time, aiming to extract the time characteristics of the dynamic behavior of the mobile machinery, such as the rate of change of acceleration, speed fluctuation, etc. Reveal the evolution law of driving status in the time dimension, take the extracted spatial features and time features as input and send them to the feature fusion module for deep integration. These fused features not only contain the spatial position and motion state of the mobile machinery, but also incorporate the information of time change, which provides a strong basis for anomaly detection. The fused features are input into the fully connected layer, and the weights and biases inside the layer are used for complex nonlinear transformation and decision analysis. The fully connected layer can evaluate the driving status based on the fused features and output the anomaly detection results. When the anomaly detection result is greater than the preset threshold, it is judged that an anomaly exists. At this time, the abnormal data is visualized and stored in the cloud for subsequent analysis and tracing, and early warning information is issued to remind relevant personnel to take timely measures.

[0089] The ROS-based mobile machinery driving state monitoring system and method for thermal regeneration scenarios can provide more in-depth and comprehensive driving state monitoring by introducing advanced data processing units, machine learning-based analysis methods, intuitive user interfaces, and multi-protocol communication modules. This can not only improve the safety and efficiency of operations, but also support more extensive data sharing and remote collaboration, thereby meeting the needs of modern industrial applications.

[0090] Through real-time monitoring and early warning, the potential risk of accidents caused by abnormal driving status is reduced, and work efficiency and safety are improved. At the same time, comprehensive data support is provided, enabling managers to better understand the operating status of mobile machinery and make more scientific and reasonable decisions.

[0091] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device.

[0092] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

[0093] The present invention and its embodiments are described above, and such description is not restrictive. The drawings show only one embodiment of the present invention, and the actual structure is not limited thereto. In short, if ordinary technicians in the field are inspired by it, without departing from the purpose of the invention, they can design a structure and embodiment similar to the technical solution without creativity, which should belong to the protection scope of the present invention.

Claims

1. A method for monitoring the driving state of a mobile machine in a thermal regeneration scenario, characterized in that: The following steps are involved: S1. Install various sensors including but not limited to three-dimensional accelerometers, gyroscopes and environmental sensors on mobile machinery; S2, collecting sensor data provided by the above sensors, wherein the sensor data includes driving state data of the mobile machine and external environment data; S3. Use the ROS data processing system to establish a driving state abnormality detection model including a spatial feature extraction module, a spatiotemporal feature extraction module, a feature fusion module and a fully connected layer; S4, inputting the collected sensor data into the driving state abnormality detection model, and obtaining an optimized driving state abnormality detection model through training and hyperparameter optimization; S5. Use the optimized driving state anomaly detection model to perform real-time anomaly detection on the driving state of mobile machinery. Once the detection result shows that there is an anomaly, the abnormal data will be visualized and stored in the cloud, and an early warning message will be issued.

2. The method for monitoring the driving state of a mobile machine in a thermal regeneration scenario according to claim 1 is characterized in that: The driving status data includes but is not limited to data such as speed, angle, temperature, pressure, three-dimensional acceleration and posture.

3. The method for monitoring the driving state of a mobile machine in a thermal regeneration scenario according to claim 1 is characterized in that: The external environment data includes but is not limited to data such as external temperature, humidity and light intensity.

4. The method for monitoring the driving state of a mobile machine in a thermal regeneration scenario according to claim 1 is characterized in that: Before the sensor data is input into the driving state abnormality detection model, it also includes preprocessing steps of data cleaning, data integration, data normalization and time series conversion of the sensor data.

5. The method for monitoring the driving state of a mobile machine in a thermal regeneration scenario according to claim 1 is characterized in that: The training process of the driving state abnormality detection model includes the following steps: S4.1, dividing the sensor data into a training set, a test set and a validation set according to a certain ratio; S4.

2. Obtaining several hyperparameter combinations in the driving state abnormality detection model; S4.3, training the model corresponding to each of the hyperparameter combinations based on the training set, and evaluating the model performance using the validation set, expressed as an evaluation score; S4.4, using the firefly improved sparrow search algorithm to optimize and select the hyperparameter combination; S4.

5. Finally, an optimized driving state abnormality detection model is constructed based on the optimal hyperparameter combination.

6. A method for monitoring the driving state of a mobile machine in a thermal regeneration scenario according to claim 5, characterized in that: Using the firefly improved sparrow search algorithm to optimize the hyperparameter combination includes the following steps: S4.4.1, randomly initialize the position of the hyperparameter combination population; S4.4.

2. Update the position of the individual in the population, calculate the evaluation score of each hyperparameter combination, and update the position of the individual accordingly; S4.4.3, further update the position of the hyperparameter combination by introducing the firefly perturbation mechanism; S4.4.

4. Repeat steps S4.4.2 and S4.4.3 until the predetermined number of iterations is reached, and output the optimal hyperparameter combination.

7. The method for monitoring the driving state of a mobile machine in a thermal regeneration scenario according to claim 1 is characterized in that: The driving state abnormality detection model comprises the following steps: S5.1, inputting sensor data into the driving state abnormality detection model; S5.2, extracting spatial features and temporal features of the sensor data through a spatial feature extraction module and a spatiotemporal feature extraction module respectively; S5.3, fusing the extracted spatial features and temporal features into comprehensive features; S5.4, input the comprehensive features into the fully connected layer for anomaly detection to obtain the final anomaly detection result; S5.

5. When the abnormality detection result exceeds a preset threshold, it is determined that an abnormality exists.

8. The method for monitoring the driving state of a mobile machine in a thermal regeneration scenario according to claim 1 is characterized in that: During the construction and optimization of the driving state abnormality detection model, the ROS data processing system is used for data processing and model training.

9. The method for monitoring the driving state of a mobile machine in a thermal regeneration scenario according to claim 1 is characterized in that: The warning information is issued in the form of sound, light or other suitable means to alert the operator.

Images