Intelligent inspection robot state monitoring method and system based on big data simulation
The intelligent inspection method based on big data simulation enables flexible configuration of sensor combinations, data standardization, and anomaly coordination assessment. This solves the problems of low resource utilization and poor anomaly detection accuracy in traditional inspection systems, and improves fault location efficiency and visualization analysis capabilities.
Patent Information
- Application Number
- CN202511494439.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-10-20
- Publication Date
- 2025-12-12
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
Existing intelligent inspection systems have fixed sensor binding methods, low resource utilization, difficulty in data fusion and analysis, poor accuracy in anomaly detection, lack of multi-robot collaborative visualization, and difficulty in identifying upstream and downstream anomaly correlations in distributed scenarios.
An intelligent inspection method based on big data simulation is adopted. Through modules for task configuration and modal management, data acquisition and tag management, state matrix generation and anomaly collaborative evaluation and display, it realizes flexible configuration of sensor combinations, data frequency normalization, state matrix standardization and anomaly collaborative evaluation and visualization.
It improves the accuracy and efficiency of anomaly detection, can identify cross-regional fault chains, shorten fault location time, and enhance the emergency response capabilities of maintenance personnel.
Smart Images
Figure CN121105019A_ABST
Abstract
Description
TECHNICAL FIELD
[0001] The present application relates to the technical field of robots, in particular to a state monitoring method and system for intelligent inspection robots based on big data simulation. BACKGROUND
[0002] In the technical field of robots, intelligent inspection robots are widely used in device state monitoring in scenarios such as machine rooms and factories. However, the existing technology has the following shortcomings:
[0003] Insufficient flexibility in task configuration: In traditional inspection systems, the binding mode of sensors and inspection tasks is fixed, and it is difficult to dynamically adjust the sensor combination according to different inspection scenarios, resulting in low resource utilization and poor task adaptability.
[0004] Lack of data standardization processing: The collection frequencies of various sensors are inconsistent, the data time series are chaotic, and there is a lack of a unified label system, making it difficult to fuse and analyze multi-source data, affecting the accuracy of anomaly detection.
[0005] Strong isolation of anomaly detection: A single robot only judges the state based on its own sensor data, and cannot perceive the associated anomalies in adjacent inspection areas. In a distributed scenario, the anomaly correlation between upstream and downstream areas is difficult to mine, resulting in delayed fault location.
[0006] Weak visual analysis capability: There is a lack of visual display means for multi-robot cooperative anomaly, and operation and maintenance personnel are difficult to quickly grasp the overall abnormal distribution, affecting the emergency response efficiency. SUMMARY
[0007] The purpose of the present application is to provide a state monitoring method and system for intelligent inspection robots based on big data simulation to solve the problems raised in the background.
[0008] In order to solve the above technical problems, the present application provides the following technical solutions:
[0009] The state monitoring system for intelligent inspection robots based on big data simulation comprises a task configuration and mode management module, a data acquisition and label management module, a state matrix generation module, and an abnormality cooperative evaluation and display module.
[0010] The task configuration and mode management module is used to define the inspection task mode, configure the sensors, robots and task parameters, and generate a task mode set.
[0011] The data acquisition and label management module is used to normalize the sensor collection frequency, generate a collection time sequence, trigger the inspection task and attach task behavior labels and sequence trigger labels.
[0012] The state matrix generation module generates a standardized robot state matrix based on the task modality set and the time sequence, and converts the standardized matrix into an abnormal state matrix according to a data abnormality condition.
[0013] The abnormality cooperative evaluation and display module is configured to simulate evaluation of the abnormality cooperative degree of the robot in the inspection area, and trigger visual display of the abnormal area based on a preset threshold.
[0014] Further, the task configuration and modality management module includes a sensor task mapping unit and a resource unified configuration unit.
[0015] The sensor task mapping unit is configured to establish a mapping relationship between a sensor combination and an inspection task, and generate a task modality set.
[0016] The resource unified configuration unit is configured to encode and configure parameters for sensors, intelligent inspection robots, and inspection tasks.
[0017] Further, the data acquisition and label management module includes a time sequence initialization unit and a label generation unit.
[0018] The time sequence initialization unit is configured to define time nodes and generate a unified acquisition time sequence.
[0019] The label generation unit is configured to generate a task modality label set and a sequence trigger label set according to the number of task executions.
[0020] Further, the state matrix generation module includes a standardized matrix construction unit and an abnormality label conversion unit.
[0021] The standardized matrix construction unit is configured to generate a standardized state matrix with time nodes as rows and sensor types as columns.
[0022] The abnormality label conversion unit is configured to convert the standardized matrix into a binary-labeled abnormal state matrix according to a data abnormality condition.
[0023] Further, the abnormality cooperative evaluation and display module includes a simulation evaluation unit and a visualization unit.
[0024] The simulation evaluation unit is configured to calculate the inspection abnormality cooperative degree between robots through Boolean logic operations based on the abnormal state matrix.
[0025] The visualization unit is configured to preset a cooperative degree threshold to trigger visual display of the abnormal area on a display interface, which is distributed based on the inspection tasks, wherein one inspection task corresponds to one type of display interface.
[0026] The application discloses an intelligent inspection robot state monitoring method based on big data simulation.
[0027] Step S1. Task mode configuration and sensor overall planning: define an inspection task mode based on a sensor combination carried by an intelligent inspection robot, uniformly configure the sensor, the intelligent inspection robot and an inspection task, construct a task mode set corresponding to the inspection task, and explicitly define a required sensor combination of each inspection task.
[0028] Step S2. Data acquisition and labeling: normalize a sensor acquisition frequency, generate a uniform acquisition time sequence, drive the robot to execute the inspection task through a trigger instruction, add labels to sensor acquisition data and the time sequence according to the number of task executions, form a task mode label set and a sequence trigger label set.
[0029] Step S3. State matrix generation and abnormality marking: based on the task mode set and the sequence trigger label set, generate a standardized robot state matrix, take a time node as a row and a sensor type as a column, and according to an abnormality of sensor acquisition data, convert the standardized robot state matrix into an abnormal robot state matrix marked in binary.
[0030] Step S4. Abnormality coordination degree evaluation and visualization: for intelligent inspection robots existing an upstream and downstream relationship in an inspection area, calculate the inspection abnormality coordination degrees of the two robots based on the abnormal state matrix, and trigger abnormal area visualization display.
[0031] Further, the specific implementation process of the step S1 comprises:
[0032] Run a task mode based on a combination of sensors, wherein one task mode corresponds to one inspection task, the sensors are integrated in the intelligent inspection robot, and the intelligent inspection robot is used for inspecting the inspection area.
[0033] In a control center, uniformly configure the sensor, the intelligent inspection robot and the inspection task, mark the i-th sensor as , mark the x-th intelligent inspection robot as , mark the l-th inspection task as , and overall plan the inspection task corresponding sensors to form a task mode set , wherein I represents a type code number of the sensor.
[0034] Further, the specific implementation process of the step S2 comprises:
[0035] The control center normalizes the acquisition frequency of each sensor, ensuring that all sensors collect inspection data under a unified acquisition time series. The acquisition time series is initialized and recorded as follows: ,in, Let R represent the r-th data collection time point, and R represent the total number of data collection time points.
[0036] Set the trigger command for the inspection task, and the inspection task will be executed. The trigger command is sent to the intelligent inspection robot, and the intelligent inspection robot controls the task modal set during the inspection process. Each sensor in the system collects inspection data for the inspection area; based on the number of times the inspection task is executed, task behavior tags are attached to each inspection task, forming a task modality tag set, denoted as... ,and A sequence trigger tag is attached to the initialized acquisition time series, forming a sequence trigger tag set, denoted as . ,and f represents the number of times the inspection task is executed.
[0037] Furthermore, the specific implementation process of step S3 includes:
[0038] Based on the task modality set and the initial acquisition time series, a standardized robot state matrix is generated. The row numbers of the standardized robot state matrix are the encoding numbers of the initial acquisition time nodes, and the column numbers are the sensor type encoding numbers. The matrix element in the r-th row and i-th column of the standardized robot state matrix is denoted as... ;
[0039] On the fth inspection mission During the process, if the sensor At the time point of data collection If abnormal inspection data is collected, then the matrix elements are set to... If the sensor At the time point of data collection If no abnormalities are found in the collected inspection data, then set the matrix elements... Then, the inspection task will be performed on the fth time. During the process, the standardized robot state matrix is transformed into an abnormal robot state matrix, denoted as... .
[0040] Furthermore, the specific implementation process of step S4 includes:
[0041] On the fth inspection mission During the process, if the intelligent inspection robot Inspection areas and intelligent inspection robots , the upstream and downstream relationship exists between the inspection areas of the intelligent inspection robots, the abnormal robot state matrix is simulated and evaluated based on the abnormal robot state matrix , the intelligent inspection robot , the inspection abnormality coordination degree between the intelligent inspection robot
[0042]
[0043] , wherein represents the inspection abnormality coordination degree, represents the number of values 1 contained after the Boolean logical intersection operation between the abnormal robot state matrix and the abnormal robot state matrix , x≠y; represents the number of values 1 contained after the Boolean logical intersection operation between the abnormal robot state matrix and the abnormal robot state matrix , x≠y;
[0044] The preset inspection abnormality coordination threshold is set, and if the inspection abnormality coordination degree is greater than or equal to the inspection abnormality coordination threshold, the inspection area of the intelligent inspection robot and the inspection area of the intelligent inspection robot are displayed on the display interface.
[0045] In the above method, the inspection area refers to a machine room, in a distributed machine room scene, the upstream machine room is the starting end of the entire resource supply chain, in a distributed machine room architecture, it provides basic resources, data input or support for other machine rooms or systems, some large Internet companies will build a distributed machine room cluster, the upstream machine room is in the core position or close to the network entrance, responsible for data aggregation and forwarding, and provides network connection and communication services for downstream machine rooms; one intelligent inspection robot can be placed in an inspection area, and the intelligent inspection robot is used to realize the normalization inspection in the machine room. When the sensors of the intelligent inspection robot perceive that there is a security risk in the machine room, such as temperature, humidity and the like affecting the normal operation of the equipment in the machine room, the stability of the equipment room is affected in the upstream and downstream relationship; through the Boolean logical operation (intersection / union) to quantify the robot abnormality coordination degree, the cross-area fault chain that cannot be found by traditional isolated detection can be identified, for example, the A area cooling system failure (upstream) may cause the B area network equipment overheating and trigger traffic anomaly (downstream), the coordination degree evaluation can quickly locate the fault conduction path, and the fault positioning time is shortened from hours to minutes.
[0046] Compared with existing technologies, the beneficial effects achieved by this invention are as follows: In the intelligent inspection robot state monitoring method and system based on big data simulation provided by this invention, task modality configuration and sensor coordination are implemented. Inspection task modalities are defined based on sensor combinations, and a task modality set is constructed. Data acquisition and labeling are performed by normalizing sensor acquisition frequencies, generating a unified time series and attaching labels. State matrix generation and anomaly marking are also implemented, generating a standardized state matrix and converting it into a binary anomaly matrix. Anomaly coordination degree evaluation and visualization are achieved by calculating the anomaly coordination degree of robot inspections with upstream and downstream relationships and visually displaying abnormal areas. This invention can adapt to inspection areas with upstream and downstream relationships and perform remote synchronous inspections of various inspection areas. Simultaneously, it achieves multi-robot collaborative state monitoring through big data simulation, enabling flexible construction of task modalities among sensors to execute different inspection needs, thereby improving the accuracy and efficiency of anomaly detection. Attached Figure Description
[0047] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used together with the embodiments of the invention to explain the invention and do not constitute a limitation thereof.
[0048] Figure 1 This is a schematic diagram illustrating the steps of the intelligent inspection robot status monitoring method based on big data simulation of the present invention. Detailed Implementation
[0049] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.
[0050] In this first embodiment: a state monitoring system for intelligent inspection robots based on big data simulation is provided. The system includes: a task configuration and modality management module, a data acquisition and tag management module, a state matrix generation module, and an anomaly collaborative assessment and display module.
[0051] The task configuration and modality management module is used to define inspection task modalities, configure sensors, robots and task parameters, and generate task modality sets.
[0052] The task configuration and modal management module includes a sensor task mapping unit and a resource unified configuration unit.
[0053] The sensor task mapping unit is used to establish the mapping relationship between sensor combinations and inspection tasks, and to generate a task mode set.
[0054] The resource unified configuration unit is configured to encode and configure parameters for sensors, intelligent inspection robots, and inspection tasks.
[0055] The data acquisition and label management module is configured to normalize sensor acquisition frequency, generate an acquisition time sequence, trigger an inspection task, and attach a task behavior label and a sequence trigger label.
[0056] The data acquisition and label management module includes a time sequence initialization unit and a label generation unit.
[0057] The time sequence initialization unit is configured to define time nodes and generate a unified acquisition time sequence.
[0058] The label generation unit is configured to generate a task modality label set and a sequence trigger label set according to the number of task executions.
[0059] The state matrix generation module is configured to generate a standardized robot state matrix based on a task modality set and a time sequence, and convert the standardized robot state matrix into an abnormalized state matrix according to data abnormality.
[0060] The state matrix generation module includes a standardized matrix construction unit and an abnormality marker conversion unit.
[0061] The standardized matrix construction unit is configured to generate a standardized state matrix with time nodes as rows and sensor types as columns.
[0062] The abnormality marker conversion unit is configured to convert the standardized matrix into a binary-marked abnormalized state matrix according to data abnormality.
[0063] The abnormality coordination evaluation and display module is configured to simulate the abnormality coordination degree of a robot inspection area, and trigger the visual display of an abnormal area based on a preset threshold.
[0064] The abnormality coordination evaluation and display module includes a simulation evaluation unit and a visualization unit.
[0065] The simulation evaluation unit is configured to calculate the inspection abnormality coordination degree between robots through Boolean logic operations based on the abnormalized state matrix.
[0066] The visualization unit is configured to preset a coordination degree threshold to trigger the visual display of an abnormal area on a display interface, and the display interface is distributed based on an inspection task, wherein one inspection task corresponds to one type of display interface.
[0067] For example, based on the distributed display interface of the inspection task (such as W1 corresponding to the core room interface and W2 corresponding to the edge room interface), when the coordination degree exceeds the threshold, the interface automatically highlights the abnormal area and marks the associated robot (A_x→A_y), and the operation and maintenance personnel can intuitively master the abnormal diffusion range, preferentially check the upstream root problem, and avoid invalid checking of the downstream.
[0068] Please refer to Figure 1 In the second embodiment: an intelligent inspection robot state monitoring method based on big data simulation is provided, which is applicable to the above-mentioned first embodiment, and the method comprises the following steps:
[0069] Step S1. Task mode configuration and sensor coordination: define the inspection task mode based on the sensor combination carried by the intelligent inspection robot, uniformly configure the sensors, intelligent inspection robots and inspection tasks, build a task mode set corresponding to the inspection task, and clearly define the required sensor combination of each inspection task;
[0070] For example, the intelligent inspection robot can be equipped with two hub motors and four universal wheels, supporting 360-degree rotation in place
[0071] ; can be equipped with multiple industrial-grade cameras for image data acquisition, supporting the movement trajectory of the lifting shaft covering the upper and lower boundaries of the standard cabinet; can be equipped with a set of infrared thermal imaging and a set of visible light sensors to support the detection of the surface temperature of the equipment in the cabinet;
[0072] For example, based on the combination of sensors, a task mode is formed, wherein one task mode corresponds to one inspection task, the sensors are integrated into the intelligent inspection robot, and the intelligent inspection robot is used to inspect the inspection area;
[0073] In the control center, the sensors, intelligent inspection robots and inspection tasks are uniformly configured, the i-th sensor is recorded as , the x-th intelligent inspection robot is recorded as , the l-th inspection task is recorded as , and the inspection tasks are coordinated The corresponding sensors are recorded as , wherein I represents the type code number of the sensor.
[0074] For example, in a distributed machine room cluster, the upstream core machine room (A area) needs to monitor the server temperature and heat distribution (configure temperature sensor S1, infrared thermal imager S3), and the downstream edge machine room (B area) needs to focus on the equipment humidity and network traffic (configure humidity sensor S2, network traffic sensor S4); By defining the task mode of the sensor combination (such as core equipment inspection task W1 corresponds to {S1, S3}, and edge equipment inspection task W2 corresponds to {S2, S4}), the sensor resources can be flexibly configured according to the characteristics of the equipment in different machine rooms, avoiding resource waste caused by "one-size-fits-all" fixed configuration (such as redundant configuration of thermal imagers in edge machine rooms).
[0075] Step S2. Data collection and labeling: normalize the sensor collection frequency, generate a unified collection time sequence, drive the robot to perform the inspection task through the trigger instruction, and add labels to the sensor collection data and time sequence according to the number of task executions, forming a task mode label set and a sequence trigger label set;
[0076] For example, the control center normalizes the collection frequency of each sensor, so that each sensor collects inspection data under a unified collection time sequence, initializes the collection time sequence, and records the initialized collection time sequence as , where represents the rth collection time node, and R represents the total number of collection time nodes;
[0077] For example, the control center normalizes the sensor collection frequency of A_x (upstream) and A_y (downstream) to 1 time / minute, generates a unified sequence of 60 time nodes / hour, and adds labels to the data according to the number of task executions (f=1 / 2 / 3);
[0078] Set the trigger instruction of the inspection task, and issue the trigger instruction of the inspection task to the intelligent inspection robot. The intelligent inspection robot collects inspection data of the inspection area in the task mode set during the inspection process; based on the number of task executions, the inspection task is labeled with a task behavior label, and a task mode label set is formed, denoted as , and , the initialized collection time sequence is labeled with a sequence trigger label, and a sequence trigger label set is formed, denoted as , and , and f is the number of task executions.
[0079] Step S3. State matrix generation and anomaly marking: based on the task modality set and the sequence trigger label set, a standardized robot state matrix is generated, taking time nodes as rows and sensor types as columns, and according to the abnormality of the sensor collected data, the standardized robot state matrix is converted into a binary marked abnormal robot state matrix;
[0080] For example, based on the task modality set and the initialized collection time sequence, a standardized robot state matrix is generated, the row number of the standardized robot state matrix is the encoding number of the initialized collection time node, and the column number of the standardized robot state matrix is the category encoding number of the sensor, then the matrix element of the rth row and the ith column of the standardized robot state matrix is denoted as ;
[0081] In the fth execution of the inspection task , if the sensor collects abnormal inspection data at the collection time node , the matrix element is set, if the sensor collects no abnormal inspection data at the collection time node , the matrix element is set, then in the fth execution of the inspection task , the standardized robot state matrix is converted into an abnormal robot state matrix, denoted as .
[0082] Step S4. Abnormal coordination degree evaluation and visualization: for intelligent inspection robots with upstream and downstream relationships in the inspection area, the inspection abnormal coordination degree of the two is calculated based on the abnormal state matrix, and the abnormal area visualization display is triggered;
[0083] For example, in the fth execution of the inspection task , if the inspection area of the intelligent inspection robot has an upstream and downstream relationship with the inspection area of the intelligent inspection robot , based on the abnormal robot state matrix, the inspection abnormal coordination degree between the intelligent inspection robot and the intelligent inspection robot in the fth execution of the inspection task is simulated and evaluated:
[0084] ;
[0085] In the formula, represents the inspection abnormal coordination degree, represents the abnormal robot state matrix and the abnormal robot state matrix the number of values 1 contained after performing a Boolean logical intersection operation between the abnormal robot state matrix and the normal robot state matrix, an abnormal robot state matrix and an abnormal robot state matrix the number of values 1 contained after performing a Boolean logical intersection operation between the abnormal robot state matrix and the normal robot state matrix, x≠y;
[0086] a preset abnormality inspection synergy threshold value, if the abnormality inspection synergy degree of the intelligent inspection robot is greater than or equal to the abnormality inspection synergy threshold value, then the inspection area of the intelligent inspection robot is displayed on the display interface together with the inspection area of the intelligent inspection robot .
[0087] It should be noted that when cross-regional cross-inspection task linkage detection is required, the abnormality inspection synergy threshold value , is set as the inspection task, and h≠l, that is, the abnormality synergy degree of the core equipment inspection task W1 and the edge equipment inspection task W2 in the upstream and downstream cross-regions can be simulated and evaluated.
[0088] It should be noted that in this document, relational terms such as first and second and the like can only be used to distinguish one entity or action from another entity or action, without necessarily requiring or implying that these entities or actions are in any way mutually exclusive or in any way arranged or ordered in succession or time. In addition, the terms “comprising”, “including” or any other variant thereof are intended to cover non-exclusive inclusion, so that a process, method, article or apparatus including a list of elements does not only include those elements, but also includes other elements not explicitly listed or inherent to such a process, method, article or apparatus.
[0089] Finally, it should be noted that the above only describes the preferred embodiments of the present application, and is not intended to limit the present application, although the present application has been described in detail with reference to the foregoing embodiments, and those skilled in the art can still modify the technical solutions recorded in the foregoing embodiments, or make equivalent replacements to some technical features. Any modification, equivalent replacement, improvement, etc. made within the spirit and principles of the present application shall be included in the protection scope of the present application.
Claims
1. A method for intelligent inspection robot state monitoring based on big data simulation, characterized in that, The method comprises the following steps: Step S1. Task mode configuration and sensor coordination: defining the inspection task mode based on the sensor combination carried by the intelligent inspection robot, uniformly configuring the sensors, intelligent inspection robots and inspection tasks, constructing the task mode set corresponding to the inspection task, and determining the required sensor combination for each inspection task; Step S2. Data acquisition and labeling: normalizing the sensor acquisition frequency, generating a unified acquisition time sequence, driving the robot to execute the inspection task through a trigger instruction, adding labels to the sensor acquisition data and time sequence according to the number of task executions, and forming a task mode label set and a sequence trigger label set; Step S3. State matrix generation and abnormality marking: based on the task mode set and the sequence trigger label set, a standardized robot state matrix is generated, with time nodes as rows and sensor types as columns, and the standardized robot state matrix is converted into a binary marked abnormal robot state matrix according to the abnormality of the sensor acquisition data; Step S4. Abnormality coordination evaluation and visualization: for intelligent inspection robots with upstream and downstream relationships in the inspection area, the inspection abnormality coordination degree of the two is calculated based on the abnormal state matrix, and the abnormal area visualization display is triggered. 2.The big data simulation based intelligent inspection robot state monitoring method according to claim 1, characterized in that, The specific implementation process of step S1 comprises: A task mode is formed based on the combination of sensors, wherein one task mode corresponds to one inspection task, the sensors are integrated into an intelligent inspection robot, and the intelligent inspection robot is used to inspect the inspection area; At the control center end, the sensors, intelligent inspection robots and inspection tasks are configured uniformly, the i-th sensor is recorded as , the x-th intelligent inspection robot is recorded as , the l-th inspection task is recorded as , and the overall inspection task is The corresponding sensor is recorded as a task modal set , wherein I represents the type code number of the sensor. 3.The big data simulation based intelligent patrol robot state monitoring method according to claim 2, characterized in that, The specific implementation process of step S2 comprises: The control center normalizes the collection frequency of each sensor, so that each sensor collects the inspection data under a unified collection time sequence. The collection time sequence is initialized, and the initialized collection time sequence is recorded as wherein, r represents the rth collection time node, and R represents the total number of collection time nodes. The trigger instruction of the inspection task is set, and the trigger instruction of the inspection task is sent to the intelligent inspection robot The intelligent inspection robot controls the sensors in the task modal set to collect the inspection data of the inspection area during the inspection process; based on the execution times of the inspection task, the task behavior labels are respectively attached to the inspection task, and a task modal label set is formed, denoted as , and , the sequence trigger label set is formed by attaching the sequence trigger label to the initialized collection time sequence, denoted as , and , f is the execution times of the inspection task. 4.The big data simulation based intelligent patrol robot state monitoring method according to claim 3, characterized in that, The specific implementation process of step S3 comprises: Based on the task modality set and the initialized acquisition time sequence, a standardized robot state matrix is generated, a row number of the standardized robot state matrix is an encoding number of the initialized acquisition time node, a column number of the standardized robot state matrix is a kind encoding number of the sensor, and a matrix element of an rth row and an ith column of the standardized robot state matrix is recorded as ; On the fth inspection mission During the process, if the sensor At the time point of data collection If abnormal inspection data is collected, then the matrix elements are set to... If the sensor At the time point of data collection If no abnormalities are found in the collected inspection data, then set the matrix elements... Then, the inspection task will be performed on the fth time. During the process, the standardized robot state matrix is transformed into an abnormal robot state matrix, denoted as... .
5. The intelligent inspection robot condition monitoring method based on big data simulation according to claim 4, characterized in that, The specific implementation process of step S4 comprises: In the fth execution of the inspection task In the process, if the inspection area of the intelligent inspection robot And the inspection area of the intelligent inspection robot There is an upstream and downstream relationship, based on the abnormal robot state matrix, simulating the evaluation of the inspection abnormality coordination degree between the intelligent inspection robot In the fth execution of the inspection task And the intelligent inspection robot Between them: ； In the formula, represents the degree of cooperation of the inspection exception, represents the abnormal robot state matrix and the abnormal robot state matrix the number of values 1 contained after the Boolean logical intersection operation between represents the abnormal robot state matrix and the abnormal robot state matrix the number of values 1 contained after the Boolean logical intersection operation between A preset inspection abnormality coordination threshold is set. If the inspection abnormality coordination degree is greater than or equal to the inspection abnormality coordination threshold, the inspection area of the intelligent inspection robot is displayed on the display interface together with the inspection area of the intelligent inspection robot . 6. The intelligent inspection robot state monitoring system based on big data simulation, executes the intelligent inspection robot state monitoring method as claimed in any one of claims 1-5, characterized in that, The system comprises a task configuration and mode management module, a data acquisition and label management module, a state matrix generation module, and an abnormality coordination evaluation and display module; The task configuration and mode management module is used to define the inspection task mode, configure the sensor, robot and task parameters, and generate the task mode set; The data acquisition and label management module is used to normalize the sensor acquisition frequency, generate the acquisition time sequence, trigger the inspection task and add the task behavior label and the sequence trigger label; The state matrix generation module generates a standardized robot state matrix based on the task mode set and the time sequence, and converts it into an abnormal state matrix according to the data abnormality; The abnormality coordination evaluation and display module is used to simulate the abnormality coordination degree of the robot inspection area, and trigger the abnormal area visualization display based on the preset threshold.
7. The big data simulation based intelligent patrol robot condition monitoring system according to claim 6, characterized in that: The task configuration and mode management module comprises a sensor task mapping unit and a resource unified configuration unit; The sensor task mapping unit is used to establish the mapping relationship between the sensor combination and the inspection task, and generate the task mode set; The resource unified configuration unit is used to encode and configure the parameters of the sensors, intelligent inspection robots and inspection tasks.
8. The big data simulation based intelligent patrol robot condition monitoring system according to claim 6, characterized in that: The data acquisition and label management module comprises a time sequence initialization unit and a label generation unit; The time sequence initialization unit is used to define the time nodes and generate a unified acquisition time sequence; The label generation unit is configured to generate a task modality label set and a sequence trigger label set according to the number of task executions.
9. The big data simulation based intelligent patrol robot condition monitoring system according to claim 6, wherein: The state matrix generation module comprises a standardized matrix construction unit and an abnormality label conversion unit. The standardized matrix construction unit is configured to generate a standardized state matrix with time nodes as rows and sensor types as columns. The abnormality label conversion unit is configured to convert the standardized matrix into a binary-labeled abnormal state matrix according to data abnormality conditions.
10. The big data simulation based intelligent patrol robot condition monitoring system according to claim 6, wherein: The abnormality coordination evaluation and display module comprises a simulation evaluation unit and a visualization unit. The simulation evaluation unit is configured to calculate the inspection abnormality coordination degree among robots through Boolean logic operation based on the abnormal state matrix. The visualization unit is configured to preset a coordination degree threshold, trigger the visualization display of abnormal areas on a display interface, and distribute the display interface based on inspection tasks, wherein one inspection task corresponds to one display interface.