Intelligent fireproof detection method and system for nuclear power station

Abstract

The invention relates to the field of nuclear power plant technology fire protection fire alarm, and discloses a nuclear power plant fireproof intelligent detection method and system, and the method comprises the steps: collecting the surrounding environment data of a first object in real time, and carrying out the consistent data processing; analyzing a first feature of the first object by using the processed surrounding environment data of the first object, and performing anomaly detection; and through a positioning algorithm, positioning the abnormal area, and carrying out abnormity processing. Through real-time acquisition and processing of environment data and combination of an advanced spectrum technology and a multi-point positioning algorithm, an abnormal area can be efficiently and accurately detected and positioned, so that the response speed and precision of an abnormal event are improved, and a more reliable solution is provided especially in the aspects of fire early warning or gas leakage and the like in a complex environment.

Description

Technical Field

[0001] The present invention relates to the field of nuclear power plant technical fire protection and fire alarm, and in particular to a nuclear power plant fire protection intelligent detection method and system. Background Art

[0002] In complex and highly sensitive industrial environments such as nuclear power plants, fire is an extremely dangerous accident that may lead to serious consequences, including the leakage of radioactive materials and large-scale economic losses. Traditional fire detection systems often rely on a single type of sensor, such as temperature sensors or smoke detectors, which may have blind spots in environments such as nuclear power plants because extreme environmental conditions may cause some sensors to fail or respond slowly.

[0003] The main problem faced by existing fire detection systems in nuclear power plant environments is that they cannot provide comprehensive, real-time and accurate fire warnings under extreme conditions. Single-point sensors are easily affected by environmental factors such as radiation, high temperature and electromagnetic interference, which reduces their reliability and effectiveness. In addition, the lack of multi-dimensional data analysis capabilities makes it difficult for traditional systems to accurately identify fire hazards at an early stage, thus missing the best time for intervention. Summary of the invention

[0004] In view of the above-mentioned problems, the present invention is proposed.

[0005] Therefore, the technical problem solved by the present invention is to detect and locate abnormal areas in real time and accurately in a complex environment.

[0006] In order to solve the above technical problems, the present invention provides the following technical solutions: a nuclear power plant fire prevention intelligent detection method, comprising:

[0007] Collecting surrounding environment data of the first object in real time and performing consistency data processing;

[0008] Analyzing the first feature of the first object and performing anomaly detection using the processed surrounding environment data of the first object;

[0009] Through the positioning algorithm, the abnormal area is located and the abnormality is handled.

[0010] As a preferred solution of the nuclear power plant fire prevention intelligent detection method described in the present invention, the surrounding environment data of the first object includes data reflecting the real-time status and change trend of physical, chemical and optical characteristics related to the abnormality of the surrounding environment.

[0011] As a preferred solution of the nuclear power plant fire prevention intelligent detection method described in the present invention, the consistency data processing includes eliminating errors in the surrounding environment data of the first object and improving the consistency and quality of the surrounding environment data of the first object.

[0012] As a preferred solution of the nuclear power plant fire prevention intelligent detection method described in the present invention, the abnormality detection includes: using historical data to train the detection model, inputting the processed surrounding environment data of the first object into the trained model, analyzing the first feature of the first object, and judging whether an abnormality occurs based on the analysis results.

[0013] As a preferred solution of the nuclear power plant fire prevention intelligent detection method described in the present invention, the positioning of the abnormal area includes using spectral technology to perform spectral analysis on the first object, and the analysis results are combined with a positioning algorithm to achieve the positioning of the abnormal area.

[0014] As a preferred solution of the nuclear power plant fire prevention intelligent detection method described in the present invention, the detection model includes but is not limited to using a convolutional neural network to construct a detection model, learn and extract the first feature of the first object for analysis, and determine whether an abnormality occurs.

[0015] As a preferred solution of the nuclear power plant fire prevention intelligent detection method described in the present invention, the positioning algorithm includes but is not limited to accurately positioning the abnormal area by combining a multi-point positioning algorithm with the results of spectral analysis.

[0016] A nuclear power plant fire prevention intelligent detection system, wherein:

[0017] A data module collects the surrounding environment data of the first object in real time and performs consistency data processing;

[0018] A detection module, using the processed surrounding environment data of the first object to analyze the first feature of the first object and perform anomaly detection;

[0019] The positioning module locates the abnormal area through the positioning algorithm and performs abnormal processing.

[0020] A computer device comprises: a memory and a processor; the memory stores a computer program, wherein the processor implements the steps of any one of the methods of the present invention when executing the computer program.

[0021] A computer-readable storage medium stores a computer program, wherein the computer program, when executed by a processor, implements the steps of any one of the methods of the present invention.

[0022] Beneficial effects of the invention: The intelligent fire prevention detection method for nuclear power plants provided by the invention can efficiently and accurately detect and locate abnormal areas by real-time collection and processing of environmental data, combined with advanced spectral technology and multi-point positioning algorithms, thereby improving the response speed and accuracy of abnormal events, especially in complex environments, such as fire warnings or gas leaks, etc., providing a more reliable solution. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] 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.

[0024] Figure 1 An overall flow chart of a nuclear power plant fire prevention intelligent detection method provided for the first embodiment of the present invention. DETAILED DESCRIPTION

[0025] In order to make the above-mentioned purposes, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the drawings of the specification. Obviously, the described embodiments are part of the embodiments of the present invention, but not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary persons in the art without creative work should fall within the scope of protection of the present invention.

[0026] Example 1, reference Figure 1 , as an embodiment of the present invention, provides a nuclear power plant fire prevention intelligent detection method, comprising:

[0027] S1: Collecting surrounding environment data of the first object in real time and performing consistency data processing.

[0028] In one implementation of the present application, the first object refers to nuclear power equipment, which collects environmental data (such as temperature, radiation, gas concentration, etc.) around the equipment in real time and performs consistency processing to ensure the accuracy of the data. On this basis, a deep learning model is used to detect anomalies in the processed data to identify potential equipment failures or safety hazards. If an anomaly is detected, the area where the anomaly occurs is accurately located by combining spectral technology with a multi-point positioning algorithm. After positioning, the system will automatically take response measures, such as starting the backup system, adjusting equipment parameters, or triggering alarms to ensure the safety of equipment and personnel, thereby improving the monitoring efficiency and emergency response capabilities of nuclear power equipment.

[0029] In an optional embodiment, the first object can also be a chemical production line, which collects environmental data (such as temperature, pressure, gas concentration, vibration, etc.) around the production line in real time and performs consistency processing to ensure the reliability and accuracy of the data. Then, the processed data is subjected to anomaly detection using a deep learning model to promptly detect potential equipment failures or production anomalies. If an anomaly is detected, the specific area where the anomaly occurs is accurately located through spectral technology combined with a multi-point positioning algorithm to help identify the source of the fault or the leak. After positioning, the system can automatically adjust production parameters, start safety devices or trigger alarms to ensure the safe and stable operation of the production line and reduce possible accident risks, thereby improving the safety and response efficiency of chemical production.

[0030] The surrounding environment data of the first object includes data reflecting the real-time status and change trend of physical, chemical and optical characteristics of the surrounding environment related to the abnormality.

[0031] The consistency data processing includes eliminating errors in the surrounding environment data of the first object and improving the consistency and quality of the surrounding environment data of the first object.

[0032] Furthermore, through the real-time collection of environmental data (such as temperature, radiation, gas concentration, etc.), the operating status and potential risks around the equipment can be fully reflected. Data consistency processing ensures that noise interference and errors in the collection process are effectively eliminated, avoiding misjudgments or missed judgments due to inaccurate data. The design of this step is crucial to improving the intelligence level, accuracy and stability of the system, and is the core guarantee for subsequent abnormal detection and emergency response.

[0033] S2: Analyze the first feature of the first object and perform anomaly detection using the processed surrounding environment data of the first object.

[0034] Performing anomaly detection includes training a detection model using historical data, inputting processed surrounding environment data of the first object into the trained model, analyzing a first feature of the first object, and determining whether an anomaly occurs based on the analysis result.

[0035] In one implementation of the present application, the detection model uses a convolutional neural network to collect environmental data (such as temperature, smoke concentration, gas composition, etc.) through sensors, and pre-processes the data to ensure that it is suitable for input into the CNN model. The CNN model extracts local features in the data through multiple convolutional layers, reduces the dimension and enhances robustness through the pooling layer, and finally outputs the anomaly detection results through the fully connected layer. The model can automatically identify normal and abnormal modes, thereby accurately judging abnormal conditions in the environment.

[0036] Once an anomaly is detected, the CNN model will sound an alarm and combine the positioning algorithm to accurately locate the abnormal area and provide accurate processing information. At the same time, the model can be continuously optimized through incremental learning to improve detection accuracy and adaptability. This CNN-based anomaly detection method improves the system's ability to handle complex data, ensures the efficiency and accuracy of real-time anomaly detection, and thus enhances the security and reliability of the system.

[0037] In an optional embodiment, the detection model can also collect and pre-process environmental data (such as temperature, humidity, smoke concentration, etc.), and then input the processed data into a trained deep neural network for feature extraction and anomaly identification. The network learns normal and abnormal patterns, identifies potential abnormal behaviors, and outputs the type and severity of the anomaly. Combined with the multi-point positioning algorithm, the system further accurately locates the source of the anomaly based on the location of the sensor data to ensure rapid response and processing of abnormal events.

[0038] Furthermore, by analyzing the processed environmental data, anomaly detection is performed on the key features of nuclear power equipment (such as temperature, gas concentration, radiation level, etc.), which can monitor the equipment operating status and environmental changes in real time and identify potential faults or safety risks in advance. By using the deep learning model to analyze the characteristics of the first object, the present invention not only improves the accuracy of fault detection, but also enhances the intelligence level of the system, effectively avoiding the risk of misjudgment caused by data lag or human intervention in traditional monitoring methods. While improving equipment safety and response speed, this design optimizes the operation and management of nuclear power plants and reduces the risk of accidents caused by equipment failures or environmental anomalies.

[0039] S3: Use the positioning algorithm to locate the abnormal area and perform abnormal processing.

[0040] Positioning the abnormal area includes performing spectral analysis on the first object using spectral technology, and combining the analysis result with a positioning algorithm to achieve positioning of the abnormal area.

[0041] In one implementation of the present application, the positioning algorithm is a multi-point positioning algorithm, which uses spectral technology to perform spectral analysis on the first object, extract abnormal features, and combine with the multi-point positioning algorithm to achieve accurate positioning of the abnormal area. The specific implementation steps include: first, using a spectral sensor to collect spectral data of different areas of the first object, and then analyzing these data through a signal processing algorithm to identify abnormal features; then, using a multi-point positioning algorithm to combine sensor data at different locations with spatial positions, determine the specific location of the abnormal source through triangulation positioning technology, and improve positioning accuracy through iterative optimization; finally, confirm the location of the abnormal area and process it in combination with other sensor data, so as to achieve fast and accurate positioning of the abnormal area. This solution ensures efficient response to abnormal events in complex environments.

[0042] In an optional embodiment, the positioning algorithm may also be a two-way ranging positioning algorithm. By combining the results of spectral analysis, the two-way ranging positioning algorithm uses the distance measurement data between multiple sensors to accurately determine the location of the abnormal area. This method improves the accuracy of positioning through two-way ranging technology, effectively reduces the positioning error caused by sensor layout and environmental factors, thereby ensuring that the source of the abnormality can be accurately identified and quickly processed.

[0043] Furthermore, by introducing intelligent real-time monitoring and anomaly detection systems, the safety and operating efficiency of nuclear power equipment can be significantly improved. Compared with traditional monitoring methods, the present invention utilizes real-time environmental data collection and consistency processing mechanisms to ensure that the data is accurate and reliable, and at the same time uses deep learning algorithms to detect anomalies in the data and automatically identify potential faults or safety hazards. In addition, the present invention combines spectral technology and multi-point positioning algorithms to accurately locate abnormal areas, which can quickly respond to potential problems and reduce false alarms, greatly improving the efficiency of emergency response. When an anomaly is detected, the system automatically takes response measures, such as starting the backup system, adjusting equipment parameters, or triggering an alarm, thereby enhancing the safety, stability and emergency handling capabilities of nuclear power equipment, reducing human intervention, and improving the safety management level of nuclear power plants.

[0044] Embodiment 2 is an embodiment of the present invention, which provides a nuclear power plant fire prevention intelligent detection system, including:

[0045] The data module collects the surrounding environment data of the first object in real time and performs consistency data processing.

[0046] The detection module uses the processed surrounding environment data of the first object to analyze the first feature of the first object and perform abnormality detection.

[0047] The positioning module locates the abnormal area through the positioning algorithm and performs abnormal processing.

[0048] Embodiment 3, an embodiment of the present invention, is different from the first two embodiments in that:

[0049] Use spectral imaging sensors to obtain real-time spectral data in the nuclear power plant environment, use high-resolution spectral imaging sensors to capture multi-spectral images inside the nuclear power plant, and eliminate non-fire-related light source interference; dynamically adjust the sensitivity and resolution of the spectral imaging sensor to adapt to different ambient lighting conditions inside the nuclear power plant; convert the spectral image into a digital spectral data set;

[0050] Compare and analyze the spectral data with a preset fire characteristic spectral template, pre-establish a spectral template database containing characteristics of multiple fire types, and compare the spectral data acquired in real time with the template database; calculate the fire probability factor based on the comparison result to evaluate the possibility of fire occurrence, and trigger the subsequent multi-sensor fusion program when the probability factor exceeds a set threshold;

[0051] The multi-sensor fusion technology integrates temperature, humidity, smoke concentration and gas composition information to form a composite detection signal, integrates the data streams from temperature and humidity sensors, smoke detectors and gas analyzers, and removes irrelevant or invalid sensor data; various sensor data are fused into a unified composite detection signal, and then the weight of the composite detection signal is adjusted to reflect the importance of different sensor data in fire detection;

[0052] The composite detection signal is collected by distributed network nodes deployed in key areas of the nuclear power plant, and local preprocessing is performed, and wireless sensor network nodes are deployed in key areas of the nuclear power plant; the transmission path of the composite detection signal is optimized through a self-organizing routing protocol between wireless sensor network nodes, and edge computing technology is implemented on the nodes to perform preliminary data processing and screening; the preprocessed data is uploaded to a central processing unit for analysis;

[0053] Applying a deep learning-based anomaly detection network to perform real-time analysis on the composite detection signal to identify potential fire risks, using a convolutional neural network model to train the fire anomaly detection network, and implementing an online learning mechanism to enable the network to continuously update and optimize its own performance; analyzing the time series characteristics of the composite detection signal to identify possible fire patterns, classifying the detected abnormal events, and distinguishing between fire and other types of abnormal situations;

[0054] The fire risk index usually includes two main parts: Fire Hazard Index (FHI) and Frequency of Ignition (Fi). The calculation formula is as follows:

[0055] FRI=FHI×FiFRI=FHI×Fi

[0056] Among them, FHI is determined by evaluating potential hazards (PH) and protective measures (PM), and the calculation formula is as follows:

[0057] FHI=PHPM

[0058] Determining the Fire Hazard Index (FHI)

[0059] The Fire Hazard Index is a ratio of potential hazards to protective measures. Potential hazards are determined based on the building's risk profile, while protective measures are a total score obtained by evaluating various fire safety factors. Specific fire safety factors include:

[0060] (1) Fire prevention (including limiting the spread of fire)

[0061] (2) Use of automatic fire protection measures and systems

[0062] (3) Intervention by professional firefighters

[0063] Each fire safety factor is assigned a score ranging from 0 to 25. These scores are assigned based on the specific circumstances of the building and the applicability of the fire safety element.

[0064] Determine the ignition frequency (Fi)

[0065] The firing frequency can be determined based on different statistics or standards.

[0066] When the fire risk level output by the anomaly detection network exceeds a predetermined threshold, the early warning system is activated, and multiple levels of fire risk thresholds are set, corresponding to different early warning levels; when the detected risk level reaches a specific threshold, the early warning system of the corresponding level is immediately activated, and an alarm is sent to relevant personnel; the emergency communication link is started to ensure that the early warning information can be delivered in a timely manner in extreme cases;

[0067] Start the visualization interface of the remote monitoring center to display the location and level of fire risk, and send an alarm to the designated security personnel at the same time. Develop a remote monitoring center software interface with interactive functions to dynamically display the geographical coordinates and risk level of the fire risk location on the interface; provide historical data query function to help analyze fire trends and causes. Preferably, the automatic retrieval of the nearest fire-fighting equipment resources includes: establishing a fire-fighting equipment resource database to record the location and status of each equipment; quickly locating the nearest available fire-fighting equipment according to the fire risk level and geographical location; commanding robots or unmanned vehicles to transport fire-fighting equipment to the fire source through the automated dispatching system, and updating the allocation of fire-fighting resources in real time;

[0068] Automatically retrieve the nearest fire-fighting equipment resources, formulate the optimal fire-fighting plan based on risk level and distribution, implement the fire-fighting plan, monitor the effect feedback during the fire-fighting process, and adjust the strategy until the fire is completely extinguished or the risk is reduced to a safe level. Use drones and fixed cameras to continuously monitor the progress of the fire-fighting operation; adjust the fire-fighting strategy based on on-site feedback, and conduct on-site cleanup and safety inspections after the fire is completed.

[0069] 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.

[0070] 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.

[0071] 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.

[0072] 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-mentioned 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, it can be implemented by any one of the following technologies known in the art or their combination: 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.

[0073] Example 4: The experimental environment is set in a simulated test area of ​​a nuclear power plant. The test objects are key equipment and areas in the area, including high temperature, high humidity, smoke concentration and other environmental parameters that may affect the occurrence of fire. First, a variety of sensor devices are deployed in the test area, including highly sensitive temperature sensors, humidity sensors, smoke sensors, gas composition sensors and high-precision imaging devices for acquiring spectral data. The data of these devices are collected in real time, and consistency data processing is performed through the method of the present invention.

[0074] The raw data collected by each sensor is checked for consistency in real time to remove possible noise and redundant information. Then, the processed data is input into the deep learning model, which combines the spectral analysis results to determine whether there is a potential fire risk. In this process, the method of the present invention uses a combination of a multi-point positioning algorithm and a two-way ranging positioning algorithm to accurately locate the abnormal area. The output results of the algorithm will be calibrated in the test area, thereby providing real-time data support for emergency response. The experimental data are shown in Table 1.

[0075] Table 1 Experimental data table

[0076]

[0077] By comparing the above experimental data, it can be clearly seen that the method of the present invention exhibits obvious advantages in many aspects.

[0078] In the comparison between test object 1 and test object 3, the probability of fire anomaly detected by the traditional method was 0%, while the probability of abnormality shown in the spectral analysis results of the method of the present invention (test object 2 and test object 6) was 1% and 2%. This shows that the method of the present invention can accurately identify potential fire risks through spectral analysis, and is more sensitive and accurate than the traditional method.

[0079] The method of the present invention performs well in terms of detection time. The detection time of test object 2 and test object 6 is 10 seconds and 9 seconds respectively, while the detection time of the traditional method is generally longer, especially the detection time of test object 3 and test object 4 is 18 seconds and 19 seconds. This shows that the present invention can respond to fire risks more quickly and provide timely warnings for nuclear power plants.

[0080] Positioning accuracy is a key advantage of the method of the present invention. By combining spectral analysis and multi-point positioning algorithms, the present invention can achieve more accurate positioning of abnormal areas than traditional methods. The positioning accuracy of test subjects 2 and 6 is 5 meters and 6 meters, while the positioning accuracy of traditional methods is generally lower, with a minimum of 10 meters. This means that in the event of a fire, the location of the fire source can be determined more accurately, thereby providing more effective emergency response support.

[0081] The environmental data consistency processing effect of the method of the present invention is better, especially in the processing of important parameters such as temperature, humidity, smoke concentration and gas composition. The data consistency of test object 2 and test object 6 is higher than that of traditional methods, which can effectively reduce false positive results caused by data errors, thereby improving the accuracy of fire warning.

[0082] In practical applications, the combination of the deep learning algorithm and spectral analysis of the present invention can not only achieve efficient and accurate fire detection, but also optimize the determination of the fire location through a two-way ranging positioning algorithm. The combination of algorithms makes the detection results of the present invention more reliable and can provide more accurate fire warnings and emergency responses in complex environments such as nuclear power plants.

[0083] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A nuclear power plant fire prevention intelligent detection method, characterized in that: include: Collecting surrounding environment data of the first object in real time and performing consistency data processing; Analyzing the first feature of the first object and performing anomaly detection using the processed surrounding environment data of the first object; Through the positioning algorithm, the abnormal area is located and the abnormality is handled.

2. The nuclear power plant fire prevention intelligent detection method according to claim 1, characterized in that: The surrounding environment data of the first object includes data reflecting the real-time status and change trend of physical, chemical and optical characteristics of the surrounding environment related to the abnormality.

3. The nuclear power plant fire prevention intelligent detection method according to claim 2, characterized in that: The consistency data processing includes eliminating errors in the surrounding environment data of the first object and improving the consistency and quality of the surrounding environment data of the first object.

4. The nuclear power plant fire prevention intelligent detection method according to claim 3, characterized in that: The abnormality detection includes training a detection model using historical data, inputting processed surrounding environment data of the first object into the trained model, analyzing a first feature of the first object, and determining whether an abnormality occurs based on the analysis result.

5. The nuclear power plant fire prevention intelligent detection method according to claim 4, characterized in that: The positioning of the abnormal area includes performing spectral analysis on the first object using spectral technology, and combining the analysis result with a positioning algorithm to realize the positioning of the abnormal area.

6. The intelligent fire prevention detection method for a nuclear power plant according to claim 5, characterized in that: The detection model includes, but is not limited to, using a convolutional neural network to construct a detection model, learn and extract a first feature related to the first object for analysis, and determine whether an abnormality occurs.

7. The intelligent detection method for fire prevention in a nuclear power plant according to claim 6, characterized in that: The positioning algorithm includes, but is not limited to, using a multi-point positioning algorithm combined with the results of spectral analysis to accurately locate the abnormal area.

8. A nuclear power plant fire prevention intelligent detection system, characterized by: A data module collects the surrounding environment data of the first object in real time and performs consistency data processing; A detection module, using the processed surrounding environment data of the first object to analyze the first feature of the first object and perform anomaly detection; The positioning module locates the abnormal area through the positioning algorithm and performs abnormal processing.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the nuclear power plant fire prevention intelligent detection method 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 a processor, the steps of the nuclear power plant fire prevention intelligent detection method according to any one of claims 1 to 7 are implemented.

Images