Fire-fighting equipment safety inspection method and system based on intelligent robot

Abstract

The invention discloses a fire-fighting facility safety inspection method and system based on an intelligent robot, and relates to the technical field of intelligent inspection, and the method comprises the steps: building a three-dimensional model, setting the path planning of the intelligent robot, collecting fire-fighting facility data through the intelligent robot, and carrying out the preprocessing of the collected fire-fighting facility data; constructing a support vector machine (SVM) model to perform fire-fighting equipment safety prediction, and automatically adjusting the inspection task based on a fire-fighting equipment safety prediction result; and storing, collecting and analyzing the generated fire-fighting facility data. The three-dimensional model is established to set the path planning of the intelligent robot, the intelligent robot collects and preprocesses the fire-fighting equipment data, and the SVM model is constructed to predict the safety of the fire-fighting equipment and automatically adjust the inspection task, so that the inspection accuracy and efficiency are improved, the reliability of the inspection result is also improved, and the inspection efficiency is improved. And the reliability and the intelligent degree of the inspection task are enhanced.

Description

Technical Field

[0001] The present invention relates to the field of intelligent inspection technology, and in particular to a firefighting facility safety inspection method and system based on an intelligent robot. Background Art

[0002] With the continuous expansion and complexity of buildings, the safety management of fire protection facilities has become an increasingly important issue. Traditional fire protection facility inspections mainly rely on manual operations, which are not only time-consuming and labor-intensive, but also easily affected by human factors, such as omissions and misjudgments. The frequency and scope of manual inspections are limited. It is difficult to fully cover all fire protection facilities in large buildings, especially those hidden or inaccessible areas. In order to improve the efficiency and accuracy of inspections, in recent years, intelligent robot technology has begun to be applied to the inspection of fire protection facilities. Intelligent robots have the ability of autonomous movement, environmental perception and data collection, which can greatly reduce the impact of human factors on inspection results. The existing inspection methods based on intelligent robots still have many limitations in path planning, data processing and exception handling, and it is difficult to meet the requirements of efficiency and accuracy in practical applications.

[0003] The application of intelligent robots in fire protection facility inspection in the existing technology mainly has the following shortcomings: the existing technology usually adopts a preset path for inspection and lacks the ability to adjust the inspection path according to real-time data, which can easily cause waste of resources or inspection blind spots in practical applications. The existing technology relies on simple statistical methods for data processing and fails to make full use of machine learning algorithms for deep data mining and anomaly detection, resulting in some potential risks being difficult to identify in a timely manner. Summary of the invention

[0004] In view of the problems existing in the above-mentioned existing fire protection facility safety inspection method and system based on intelligent robots, the present invention is proposed.

[0005] Therefore, the problem to be solved by the present invention is that the existing technology usually adopts a preset path for inspection and lacks the ability to adjust the inspection path according to real-time data, which can easily cause waste of resources or inspection blind spots in practical applications. The existing technology relies more on simple statistical methods in data processing and fails to fully utilize machine learning algorithms for deep data mining and anomaly detection, resulting in some potential risks being difficult to identify in a timely manner.

[0006] To solve the above technical problems, the present invention provides the following technical solutions: a fire facility safety inspection method based on an intelligent robot, which includes establishing a three-dimensional model to set the intelligent robot path planning, collecting fire facility data through the intelligent robot, and preprocessing the collected fire facility data; constructing a support vector machine (SVM) model to predict the safety of fire facilities, and automatically adjusting the inspection tasks based on the results of the fire facility safety prediction; storing, collecting and analyzing the generated fire facility data.

[0007] As a preferred solution of the fire protection facility safety inspection method based on intelligent robots of the present invention, wherein: the establishment of a three-dimensional model to set the intelligent robot path planning refers to obtaining the structural data of the building through laser scanning technology, including the floor plan, floor height, room layout and the location of fire protection facilities;

[0008] Use 3D modeling software to build a 3D model of the acquired building structure data, and mark the areas and key points that the intelligent robot needs to inspect in the built 3D model, where the key points are fire-fighting facilities;

[0009] A-star algorithm is used to plan the path for the key points of the 3D model.

[0010] As a preferred solution of the fire protection facility safety inspection method based on intelligent robots described in the present invention, the collection of fire protection facility data through intelligent robots refers to autonomously moving a robot with integrated temperature and pressure sensors and camera monitoring equipment inside a building according to a pre-set inspection route. When the robot reaches a key point, it identifies the fire protection facility through a camera and checks the appearance image of the fire protection facility. If the appearance image of the fire protection facility is damaged, the safety inspector is immediately notified by email to replace the fire protection facility. If the appearance image of the fire protection facility is normal, the temperature and pressure sensors are used to collect fire protection facility data, including temperature and pressure data.

[0011] As a preferred solution of the firefighting facility safety inspection method based on an intelligent robot of the present invention, wherein: the preprocessing of the collected firefighting facility data refers to cleaning the firefighting facility data to remove duplicate firefighting facility data;

[0012] Use the moving average method to smooth the cleaned firefighting facility data;

[0013] Normalize the smoothed firefighting facility data and convert them into a unified format;

[0014] Use statistical algorithms to calculate the mean, standard deviation, maximum and minimum values ​​of fire protection facility data to obtain statistical characteristics;

[0015] Calculate the difference value of fire protection facility data at adjacent time points to obtain the time characteristics. The formula is:

[0016] E=x t -x t-1 ,e=y m -y m-1 ,

[0017] Where E is the temperature difference, e is the pressure difference, and x t is the temperature data at time point t, x t-1 is the temperature data at time point t-1, y m is the pressure data at time point m, y m-1 is the air pressure data at time point m-1;

[0018] The calculated statistical features and time features are combined into a complete feature vector F, the formula is:

[0019] F=[X,B,x max ,x min ,E,M,C,y max ,y min ,e],

[0020] Where X is the mean of the temperature data, B is the standard deviation of the temperature data, and x max and x min are the maximum and minimum values ​​of the temperature data, M is the mean of the pressure data, C is the standard deviation of the pressure data, and y max and max are the maximum and minimum values ​​of the air pressure data respectively.

[0021] As a preferred solution of the firefighting facility safety inspection method based on intelligent robots of the present invention, wherein: the construction of the support vector machine SVM model for firefighting facility safety prediction refers to collecting historical firefighting facility data for preprocessing to obtain historical feature vectors F′, and generating a data training set;

[0022] Construct a support vector machine (SVM) model, including an input layer and a kernel function output layer;

[0023] Define the input format of the input layer as feature vector F;

[0024] Use RBF radial basis function as kernel function K(F i ,F j ), the formula is:

[0025] K(F i ,F j )=exp(-γ‖F i -F j ‖ 2 ),

[0026] Where γ is the parameter of the RBF kernel, F i and F j is the input feature vector;

[0027] The kernel function output layer formula is:

[0028]

[0029] Where f(F) is the result predicted by the model, y i is the eigenvector F i The category label, b is the bias term, α i is the Lagrange multiplier, n is the total number of support vectors;

[0030] Use the data training set to train the support vector machine (SVM) model, and use the loss function and Adam optimizer to iteratively optimize the model parameters;

[0031] The feature vector F is input into the trained support vector machine (SVM) model to obtain the fire protection facility safety prediction result.

[0032] As a preferred solution of the fire protection facility safety inspection method based on an intelligent robot described in the present invention, wherein: the automatic adjustment of the inspection task based on the results of the fire protection facility safety prediction refers to identifying abnormal fire protection facilities based on the fire protection facility safety prediction results, and automatically generating emergency task instructions, instructing the intelligent robot to go to the abnormal fire protection facilities first to re-collect fire protection facility data, and setting an abnormal threshold based on historical abnormal fire protection facility data. If the re-collected fire protection facility data is greater than and equal to the abnormal threshold, it is determined that the fire protection facility is abnormal, and the safety inspector is immediately notified by email to replace the fire protection facility. If the re-collected fire protection facility data is less than the abnormal threshold, it is determined to be normal, and the inspection status of the path planning is restored, and the unfinished routine inspection tasks are continued until the path planning inspection is completed and the robot returns to the starting position for charging.

[0033] As a preferred solution of the firefighting facility safety inspection method based on intelligent robots described in the present invention, the firefighting facility data collected and analyzed refers to the firefighting facilities collected and the firefighting facility safety prediction results generated by the analysis being stored in a database, and data security protection measures are implemented, the firefighting facilities collected and the firefighting facility safety prediction results generated by the analysis are backed up in the cloud, and the integrity of the firefighting facilities collected, the firefighting facility safety prediction results generated by the analysis and the backup data are regularly checked.

[0034] Another object of the present invention is to provide a fire protection facility safety inspection system based on an intelligent robot, which comprises:

[0035] The model collection module is used to establish a three-dimensional model to set the intelligent robot path planning, collect firefighting facility data through the intelligent robot, and pre-process the collected firefighting facility data;

[0036] Build an adjustment module to build a support vector machine (SVM) model for fire protection facility safety prediction and automatically adjust inspection tasks based on the results of fire protection facility safety prediction;

[0037] The storage module is used to store the fire protection facility data collected and analyzed.

[0038] A computer device comprises: a memory and a processor; the memory stores a computer program, and the processor implements the steps of the above-mentioned fire protection facility safety inspection method based on an intelligent robot when executing the computer program.

[0039] A computer-readable storage medium stores a computer program, which, when executed by a processor, implements the steps of the above-mentioned fire protection facility safety inspection method based on an intelligent robot.

[0040] The beneficial effects of the present invention are as follows: the present invention establishes a three-dimensional model to set the path planning of the intelligent robot, collects and pre-processes the fire protection facility data through the intelligent robot, constructs a support vector machine (SVM) model to predict the safety of fire protection facilities and automatically adjusts the inspection tasks, which not only improves the accuracy and efficiency of the inspection, but also improves the reliability of the inspection results, and enhances the reliability and intelligence of the inspection tasks. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.

[0042] Figure 1 The figure is a flow chart of the fire protection facility safety inspection method based on intelligent robots.

[0043] Figure 2 This is a structural diagram of the fire protection facility safety inspection system based on intelligent robots. DETAILED DESCRIPTION

[0044] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific embodiments of the present invention are described in detail below in conjunction with the accompanying drawings.

[0045] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0046] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The term "in one embodiment" that appears in different places in this specification does not necessarily refer to the same embodiment, nor is it a separate or selective embodiment that is mutually exclusive with other embodiments.

[0047] Example 1, reference Figure 1 , which is the first embodiment of the present invention, and which provides a fire protection facility safety inspection method based on an intelligent robot. The fire protection facility safety inspection method based on an intelligent robot includes:

[0048] S1. Establish a three-dimensional model to set the intelligent robot path planning, collect firefighting facility data through the intelligent robot, and pre-process the collected firefighting facility data;

[0049] Specifically, establishing a three-dimensional model to set the intelligent robot path planning means obtaining the structural data of the building through laser scanning technology, including the floor plan, floor height, room layout and location of fire protection facilities;

[0050] Use 3D modeling software to build a 3D model of the acquired building structure data, and mark the areas and key points that the intelligent robot needs to inspect in the built 3D model, where the key points are fire-fighting facilities;

[0051] A-star algorithm is used to plan the path for the key points of the 3D model.

[0052] The laser scanning technology is used to obtain the structural data of the building, including the floor plan, floor height, room layout and location of fire-fighting facilities. This process ensures the high accuracy and comprehensiveness of the basic data for the construction of the 3D model. Compared with the traditional manual measurement method, laser scanning technology can not only significantly improve the efficiency of data collection, but also obtain accurate 3D spatial data, which is especially important for large and complex buildings. Through this technology, the intelligent robot can more accurately locate its own spatial position and the location of the target fire-fighting facilities during inspection, thereby greatly improving the efficiency and accuracy of the inspection. The structural data of the building is modeled through 3D modeling software, and the areas and key points that the intelligent robot needs to inspect are marked in the model. This process realizes the digital reproduction of the actual building environment. The 3D model not only provides an intuitive environmental display, making path planning and task design more scientific and reasonable, but also It can update and maintain model data in real time to adapt to changes in the building environment. This model construction based on real data makes the path planning of the intelligent robot more accurate, effectively reducing the difficulty of inspection caused by environmental complexity. The A-star algorithm is used to plan the path of the key points of the three-dimensional model, ensuring the efficiency and accuracy of the intelligent robot in the inspection task. The advantage of the A-star algorithm is that it can dynamically adjust the path, avoid obstacles and select the optimal route according to the actual situation. This is especially important for large, complex and dynamically changing building environments. Through this algorithm, the robot can reach the designated fire-fighting facility location in the shortest time, reduce inspection time, and improve the efficiency of inspection tasks. The A-star algorithm can also adapt to the handling of emergency situations. For example, when a fire-fighting facility is abnormal, the robot can quickly adjust the path and give priority to the abnormal facility for processing to ensure the safety of the fire-fighting facility.

[0053] Furthermore, collecting firefighting facility data through intelligent robots refers to autonomously moving a robot that is integrated with temperature and pressure sensors and camera monitoring equipment inside a building along a pre-set inspection route. When the robot reaches a key point, it identifies the firefighting facility through a camera and checks the appearance image of the firefighting facility. If the appearance image of the firefighting facility is found to be damaged, the safety inspector is immediately notified by email to replace the firefighting facility. If the appearance image of the firefighting facility is normal, the temperature and pressure sensors are used to collect firefighting facility data, including temperature and pressure data.

[0054] By integrating temperature sensors, pressure sensors and camera monitoring equipment into intelligent robots, the robots can move autonomously inside the building according to pre-set inspection routes, thereby greatly reducing the reliance on manual inspections. Compared with traditional manual inspection methods, robots can conduct inspections 24 hours a day, avoiding possible omissions and subjective misjudgments in manual inspections, and enabling robots to complete inspection tasks efficiently and accurately in complex environments. When the robot reaches a key point, it uses the camera to identify the fire-fighting facilities and check their appearance images. If the camera detects damage to the appearance of the fire-fighting facilities, such as cracks, rust or missing parts, the robot will immediately notify the fire department. The robot notifies safety inspectors via email. This real-time appearance inspection function greatly improves the efficiency and accuracy of inspections, ensuring that fire-fighting facilities can be discovered and handled in a timely manner when problems occur. In contrast, manual inspections are not only prone to overlooking minor appearance damage due to fatigue or lack of experience, but also in large-scale buildings, manual inspections of the appearance of facilities one by one are far less efficient than robots. The intelligent data collection steps ensure the timeliness and accuracy of the data, providing a reliable data basis for subsequent safety analysis. By monitoring these key data in real time, the robot can detect and warn of potential failures of fire-fighting facilities early, thereby taking preventive maintenance measures to avoid catastrophic consequences.

[0055] Furthermore, preprocessing the collected firefighting facility data refers to cleaning the firefighting facility data and removing duplicate firefighting facility data;

[0056] Use the moving average method to smooth the cleaned firefighting facility data;

[0057] Normalize the smoothed firefighting facility data and convert them into a unified format;

[0058] Use statistical algorithms to calculate the mean, standard deviation, maximum and minimum values ​​of fire protection facility data to obtain statistical characteristics;

[0059] Calculate the difference value of fire protection facility data at adjacent time points to obtain the time characteristics. The formula is:

[0060] E=x t -x t-1 ,e=y m -y m-1 ,

[0061] Where E is the temperature difference, e is the pressure difference, and x t is the temperature data at time point t, x t-1 is the temperature data at time point t-1, y m is the pressure data at time point m, y m-1 is the air pressure data at time point m-1;

[0062] The calculated statistical features and time features are combined into a complete feature vector F, the formula is:

[0063] F=[X,B,x max ,x min ,E,M,C,y max ,y min ,e],

[0064] Where X is the mean of the temperature data, B is the standard deviation of the temperature data, and x max and x min are the maximum and minimum values ​​of the temperature data, M is the mean of the pressure data, C is the standard deviation of the pressure data, and y max and max are the maximum and minimum values ​​of the air pressure data respectively.

[0065] By removing duplicate firefighting facility data, the cleaning process reduces the noise in the data set, thereby improving the stability and prediction accuracy of the model. The cleaned data is more concise, which helps to reduce the computational burden and shorten the processing time. This process can significantly improve the reliability of the model in practical applications, especially on large-scale data sets, where it can better reflect its efficiency. Smoothing can effectively highlight the long-term trend of the data, making it easier for the model to capture potential regular changes. In the analysis of firefighting facility data, temperature and pressure data may fluctuate due to short-term environmental changes. The application of the moving average method can eliminate the impact of these short-term fluctuations, making the overall trend of the data more obvious, which helps to improve the prediction accuracy of the model. Through normalization, data of different dimensions are Scaling to the same range eliminates the bias caused by dimensional differences, allowing the model to treat each feature more fairly during the training process. This processing method not only improves the training efficiency of the model, but also makes the model's prediction results more reliable and consistent. By calculating the statistical characteristics of the data such as the mean, standard deviation, maximum and minimum values, we can fully understand the distribution and extreme value behavior of the data. These characteristics provide an important reference for subsequent model predictions. The introduction of time features can capture the dynamic information of data changes over time by calculating the differential values ​​of adjacent time points. This is of great significance for identifying the abnormal status of fire-fighting facilities. Combined with the analysis of these characteristics, it helps to improve the model's ability to recognize the status of fire-fighting facilities and enhance its sensitivity to abnormal situations.

[0066] S2. Construct a support vector machine (SVM) model to predict the safety of fire-fighting facilities, and automatically adjust the inspection tasks based on the results of the fire-fighting facility safety prediction;

[0067] Specifically, constructing a support vector machine (SVM) model for firefighting facility safety prediction refers to collecting historical firefighting facility data for preprocessing to obtain historical feature vectors F′ and generate a data training set;

[0068] Construct a support vector machine (SVM) model, including an input layer and a kernel function output layer;

[0069] Define the input format of the input layer as feature vector F;

[0070] Use RBF radial basis function as kernel function K(F i ,F j ), the formula is:

[0071] K(F i ,F j )=exp(-γ‖F i -F j ‖ 2 ),

[0072] Where γ is the parameter of the RBF kernel, F i and F j is the input feature vector;

[0073] The kernel function output layer formula is:

[0074]

[0075] Where f(F) is the result predicted by the model, y i is the eigenvector F i The category label, b is the bias term, α i is the Lagrange multiplier, n is the total number of support vectors;

[0076] Use the data training set to train the support vector machine (SVM) model, and use the loss function and Adam optimizer to iteratively optimize the model parameters;

[0077] The feature vector F is input into the trained support vector machine (SVM) model to obtain the fire protection facility safety prediction result.

[0078] Data preprocessing can remove noise from the data and unify the format of the data to ensure that the data input into the SVM model is consistent and high-quality. The extraction of feature vectors converts the original data into a form that the SVM can process. This process is crucial to the prediction effect of the model. Through the RBF kernel function, the SVM model can find the best classification boundary in high-dimensional space. Even in the face of complex firefighting facility data, the model can still effectively distinguish between normal and abnormal states. The advantage of the SVM model lies in its strong generalization ability for new data. It can accurately predict potential safety hazards in the future based on historical data. During the optimization process, the Adam optimizer dynamically adjusts the learning rate to accelerate the convergence of the model, while avoiding falling into the local optimal solution, greatly improving the training efficiency of the model, so that the system can obtain a high-precision prediction model in a short time. The safety prediction results can help managers identify potential safety hazards in advance and take corresponding measures to prevent them. Compared with traditional manual inspection methods, the automated prediction system based on the SVM model not only reduces the possibility of human errors, but also can detect problems earlier to ensure the safety and reliability of firefighting facilities.

[0079] Furthermore, automatically adjusting the inspection tasks based on the results of firefighting facility safety predictions means identifying abnormal firefighting facilities based on the results of firefighting facility safety predictions, and automatically generating emergency task instructions, instructing the intelligent robot to prioritize going to the abnormal firefighting facilities to re-collect firefighting facility data, and setting an abnormal threshold based on historical abnormal firefighting facility data. If the re-collected firefighting facility data is greater than and equal to the abnormal threshold, the firefighting facility is determined to be abnormal, and the safety inspector is immediately notified by email to replace the firefighting facility. If the re-collected firefighting facility data is less than the abnormal threshold, it is determined to be normal, and the inspection status of the path planning is restored, and the unfinished regular inspection tasks continue to be executed until the path planning inspection is completed and the robot returns to the starting position for charging.

[0080] Through the intelligent recognition system, inspection tasks can be carried out more accurately and efficiently, reducing unnecessary waste of resources and avoiding repeated inspections of normal facilities. Compared with traditional fixed inspection methods, the intelligent recognition system greatly improves the accuracy and efficiency of inspections, ensuring that facilities that are truly at risk are given priority within limited time and resources. The dynamic adjustment mechanism ensures that the system has extremely high flexibility and response speed when responding to emergencies. Traditional inspection systems often conduct inspections according to fixed routes and schedules and are unable to respond to sudden abnormal situations in a timely manner. The present invention, by introducing a real-time adjustment mechanism, can respond as soon as an abnormality is discovered, greatly reducing potential safety risks. Especially in large buildings, where the inspection routes are complex and there are many facilities, the application of a dynamic adjustment mechanism is particularly important, which can effectively optimize the inspection path and improve inspection efficiency. Historical data not only helps set thresholds, but also provides a reference for future abnormal situations, allowing the system to continuously learn. It becomes more intelligent and accurate in learning and adjustment, and uses historical data for model training and threshold setting, which is one of the key factors that the present invention can achieve high accuracy and low false alarm rate. When the robot confirms that a facility is abnormal, the system will immediately notify the safety inspectors by email, reminding them to deal with the abnormal facilities as soon as possible, ensuring that the abnormal situation can be known to the relevant personnel in the shortest time and take necessary actions. Compared with the traditional manual reporting method, email notification is not only fast and convenient, but also can avoid delays and omissions in information transmission, ensuring that all key data and events can be processed in time. This automated information transmission method greatly improves the overall efficiency and safety of inspection tasks, ensures the continuity and integrity of inspection tasks, and avoids interruption or omission of inspection content due to exception handling. The automatic recovery and cycle of inspection tasks also improves the autonomy of the system, enabling robots to independently complete inspection tasks in complex environments and reduce the necessity of manual intervention.

[0081] S3, storage, collection and analysis of firefighting facility data;

[0082] Specifically, storing the firefighting facility data generated by collection and analysis means storing the collected firefighting facilities and the results of the firefighting facility safety prediction generated by the analysis in a database, and implementing data security protection measures, backing up the collected firefighting facilities and the results of the firefighting facility safety prediction generated by the analysis in the cloud, and regularly performing integrity checks on the collected firefighting facilities, the results of the firefighting facility safety prediction generated by the analysis and the backup data.

[0083] The systematization of data storage not only facilitates subsequent retrieval and analysis, but also ensures the consistency and integrity of the data. The database can support complex query and statistical functions, helping managers to quickly obtain the required information and make scientific decisions. Through strict data protection measures, data leakage and tampering can be effectively prevented, ensuring that the system always complies with safety regulations during operation. Data security protection measures can also prevent data loss and improve the stability and reliability of the system. Storing data on a remote cloud server ensures that even if there is a problem with the local data, it can be quickly recovered and restored. This high availability is particularly suitable for responding to unpredictable catastrophic events such as hardware failures or natural disasters. With the support of cloud backup, the system can maintain the continuous availability and integrity of data and reduce losses caused by data loss. Integrity detection can promptly detect and repair errors or anomalies in the data to ensure that the data has not been tampered with or damaged during storage and transmission. This function further improves the reliability of the data and prevents misjudgments or decision-making errors caused by data errors. At the same time, regular data detection also helps to maintain the overall stability of the system and ensure the quality of data in long-term operation.

[0084] Example 2, reference Figure 2 , which is the second embodiment of the present invention, is different from the previous embodiment and provides a fire protection facility safety inspection system based on an intelligent robot, comprising:

[0085] The model collection module is used to establish a three-dimensional model to set the intelligent robot path planning, collect firefighting facility data through the intelligent robot, and pre-process the collected firefighting facility data;

[0086] Build an adjustment module to build a support vector machine (SVM) model for fire protection facility safety prediction and automatically adjust inspection tasks based on the results of fire protection facility safety prediction;

[0087] The storage module is used to store the fire protection facility data collected and analyzed.

[0088] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the methods described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc., which can store program codes.

[0089] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as an ordered list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by an instruction execution system, device or apparatus (such as a computer-based system, a system including a processor, or other system that can fetch instructions from an instruction execution system, device or apparatus and execute instructions), or in conjunction with such instruction execution systems, devices or apparatuses. For the purposes of this specification, "computer-readable medium" can be any device that can contain, store, communicate, propagate or transmit a program for use by an instruction execution system, device or apparatus, or in conjunction with such instruction execution systems, devices or apparatuses.

[0090] More specific examples of computer-readable media (a non-exhaustive list) include the following: an electrical connection with one or more wires (electronic device), a portable computer disk case (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disk read-only memory (CDROM). In addition, the computer-readable medium may even be a paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, deciphering or, if necessary, processing in another suitable manner, and then stored in a computer memory.

[0091] It should be understood that the various parts of the present invention can be implemented by hardware, software, firmware or a combination thereof. In the above embodiments, a plurality of steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, any one of the following technologies known in the art or a combination thereof can be used to implement: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

Claims

1. A fire protection facility safety inspection method based on an intelligent robot, characterized in that: include, Establish a three-dimensional model to set the intelligent robot path planning, collect firefighting facility data through the intelligent robot, and pre-process the collected firefighting facility data; Construct a support vector machine (SVM) model to predict the safety of fire-fighting facilities, and automatically adjust inspection tasks based on the results of the fire-fighting facility safety prediction; Store, collect and analyze firefighting facility data.

2. The firefighting facility safety inspection method based on an intelligent robot according to claim 1, characterized in that: The establishment of a three-dimensional model to set up intelligent robot path planning refers to obtaining structural data of the building through laser scanning technology, including floor plans, floor heights, room layouts and locations of fire-fighting facilities; Use 3D modeling software to build a 3D model of the acquired building structure data, and mark the areas and key points that the intelligent robot needs to inspect in the built 3D model, where the key points are fire-fighting facilities; A-star algorithm is used to plan the path for the key points of the 3D model.

3. The fire protection facility safety inspection method based on an intelligent robot as claimed in claim 2, characterized in that: The collecting of firefighting facility data by intelligent robots refers to autonomously moving a robot equipped with integrated temperature and pressure sensors and camera monitoring equipment inside a building along a pre-set inspection route; when the robot reaches a critical point, it identifies the firefighting facility through a camera and checks the appearance image of the firefighting facility; if the appearance image of the firefighting facility shows any damage, the safety inspector is immediately notified via email to replace the firefighting facility; if the appearance image of the firefighting facility is normal, the temperature and pressure sensors are used to collect the firefighting facility data, including temperature and pressure data.

4. The fire protection facility safety inspection method based on an intelligent robot as claimed in claim 3, characterized in that: The pre-processing of the collected fire-fighting facility data refers to cleaning the fire-fighting facility data and removing duplicate fire-fighting facility data; Use the moving average method to smooth the cleaned firefighting facility data; Normalize the smoothed firefighting facility data and convert them into a unified format; Use statistical algorithms to calculate the mean, standard deviation, maximum and minimum values ​​of fire protection facility data to obtain statistical characteristics; Calculate the difference value of fire protection facility data at adjacent time points to obtain the time characteristics. The formula is: E=x t -x t-1 ,eh? m -and m-1 , Where E is the temperature difference, e is the pressure difference, and x t is the temperature data at time point t, x t-1 is the temperature data at time point t-1, y m is the pressure data at time point m, y m-1 is the air pressure data at time point m-1; The calculated statistical features and time features are combined into a complete feature vector F, the formula is: F=[X,B,x max ,x min ,E,M,C,y max ,y min ,e], Where X is the mean of the temperature data, B is the standard deviation of the temperature data, and x max and x min are the maximum and minimum values ​​of the temperature data, M is the mean of the pressure data, C is the standard deviation of the pressure data, and y max and max are the maximum and minimum values ​​of the air pressure data respectively.

5. The fire protection facility safety inspection method based on an intelligent robot according to claim 4, characterized in that: The construction of the support vector machine (SVM) model for fire protection facility safety prediction refers to collecting historical fire protection facility data for preprocessing to obtain historical feature vectors F ′ , generate data training set; Construct a support vector machine (SVM) model, including an input layer and a kernel function output layer; Define the input format of the input layer as feature vector F; Use RBF radial basis function as kernel function K(F i ,F j ), the formula is: K(F i ,F j )=exp(-γ‖F i -F j ‖ 2 ), Where γ is the parameter of the RBF kernel, F i and F j is the input feature vector; The kernel function output layer formula is: Where f(F) is the result predicted by the model, y i is the eigenvector F i The category label, b is the bias term, α i is the Lagrange multiplier, n is the total number of support vectors; Use the data training set to train the support vector machine (SVM) model, and use the loss function and Adam optimizer to iteratively optimize the model parameters; The feature vector F is input into the trained support vector machine (SVM) model to obtain the fire protection facility safety prediction result.

6. The firefighting facility safety inspection method based on an intelligent robot as claimed in claim 5, characterized in that: The automatic adjustment of inspection tasks based on the results of firefighting facility safety prediction refers to identifying abnormal firefighting facilities based on the results of firefighting facility safety prediction, and automatically generating emergency task instructions, instructing the intelligent robot to go to the abnormal firefighting facilities first to re-collect firefighting facility data, and setting an abnormal threshold based on historical abnormal firefighting facility data. If the re-collected firefighting facility data is greater than and equal to the abnormal threshold, it is determined that the firefighting facility is abnormal, and the safety inspector is immediately notified by email to replace the firefighting facility. If the re-collected firefighting facility data is less than the abnormal threshold, it is determined to be normal, and the inspection status of the path planning is restored, and the unfinished regular inspection tasks are continued until the path planning inspection is completed and the robot returns to the starting position for charging.

7. The firefighting facility safety inspection method based on an intelligent robot according to claim 6, characterized in that: The storage of firefighting facility data generated by collection and analysis refers to storing the collected firefighting facilities and the results of the firefighting facility safety prediction generated by the analysis in a database, and implementing data security protection measures, backing up the collected firefighting facilities and the results of the firefighting facility safety prediction generated by the analysis in the cloud, and regularly performing integrity checks on the collected firefighting facilities, the results of the firefighting facility safety prediction generated by the analysis and the backup data.

8. A firefighting facility safety inspection system based on an intelligent robot based on the firefighting facility safety inspection method based on an intelligent robot according to any one of claims 1 to 7, characterized in that: include, The model collection module is used to establish a three-dimensional model to set the intelligent robot path planning, collect firefighting facility data through the intelligent robot, and pre-process the collected firefighting facility data; Build an adjustment module to build a support vector machine (SVM) model for fire protection facility safety prediction and automatically adjust inspection tasks based on the results of fire protection facility safety prediction; The storage module is used to store the fire protection facility data collected and analyzed.

9. A computer device comprising: A memory and a processor; the memory stores a computer program, characterized in that: when the processor executes the computer program, the steps of the fire protection facility safety inspection method based on an intelligent robot according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by the processor, the steps of the fire protection facility safety inspection method based on an intelligent robot according to any one of claims 1 to 7 are implemented.

Images