Centralized photovoltaic power station fire alarm intelligent monitoring method and system

Abstract

The invention relates to the technical field of photovoltaic power station intelligent monitoring, and discloses a centralized photovoltaic power station fire alarm intelligent monitoring method and system, and the method comprises the steps: installing a sensor on each photovoltaic module, collecting the data of each photovoltaic module in real time, and obtaining a temperature distribution diagram of the photovoltaic module; analyzing the temperature distribution diagram, identifying a hot spot area and judging whether abnormity exists or not; when an abnormity is detected, an alarm is triggered and on-site workers are notified to respond to a fire risk in time. Through multi-sensor data fusion and Gaussian smoothing processing, in combination with a dynamic threshold calculation model and a clustering algorithm, accurate identification and real-time monitoring of the hot spot area of the photovoltaic module are realized, the accuracy and response efficiency of fire early warning are significantly improved, the problems of high false alarm rate, serious missing alarm and poor environmental adaptability in the prior art are overcome, and the method is suitable for popularization and application. The safe operation of the photovoltaic power station is effectively guaranteed, and the fire risk and economic loss are reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent monitoring of photovoltaic power stations, and in particular to a centralized photovoltaic power station fire alarm intelligent monitoring method and system. Background Art

[0002] With the growing global demand for renewable energy, photovoltaic power stations have been widely used as an important source of green and clean energy. However, the large-scale development of photovoltaic power stations has also brought many safety management issues, among which fire risk is a major problem that needs to be solved in the operation of photovoltaic power stations. Common fire hazards in photovoltaic power stations mainly come from local overheating of components (i.e. hot spot phenomenon), electrical failures and external environmental factors (such as high temperature, low humidity, lightning strikes, etc.). Traditional fire alarm monitoring methods mostly rely on a single sensor (such as smoke sensor or temperature sensor) or only monitor the macroscopic temperature changes of photovoltaic power stations, which makes it difficult to timely and accurately identify abnormal characteristics of local hot spot areas. At the same time, lacking the support of data fusion and intelligent analysis technology, existing systems often cannot provide accurate warnings in the early stages of fires. This delayed response mechanism can easily lead to the spread of fire, resulting in serious consequences such as damage to photovoltaic components and even shutdown of power stations.

[0003] Although multi-sensor fusion technology and image processing algorithms have been gradually introduced into the field of photovoltaic power station monitoring in recent years, the existing technology still has many shortcomings. For example, the traditional hot spot detection method relies on the judgment rule of fixed thresholds, which is difficult to adapt to complex and changeable environmental conditions (such as sunshine intensity, wind speed, ambient temperature, etc.). In addition, the existing monitoring systems mostly use a single algorithm (such as simple threshold segmentation or edge detection), which has high false alarm and missed alarm rates in complex scenarios and cannot effectively support real-time fire analysis and dynamic emergency response. Therefore, how to build an intelligent monitoring method that can integrate multi-source data, dynamically adjust the monitoring threshold and combine advanced algorithms has become an important research direction in the field of photovoltaic power station fire alarm management. 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: the technical problems of low accuracy in hot spot recognition, poor dynamic adaptability and delayed response in fire alarm monitoring of traditional photovoltaic power stations.

[0006] In order to solve the above technical problems, the present invention provides the following technical solutions: a centralized photovoltaic power station fire alarm intelligent monitoring method, comprising:

[0007] Sensors are installed on each photovoltaic module to collect data of each photovoltaic module in real time and obtain the temperature distribution map of the photovoltaic module.

[0008] Analyze the temperature distribution map, identify the hot spot area and determine whether there is an abnormality;

[0009] When an abnormality is detected, an alarm is triggered and on-site staff are notified to respond to the fire risk in a timely manner.

[0010] As a preferred solution of the centralized photovoltaic power station fire alarm intelligent monitoring method of the present invention, wherein: the sensors include a temperature sensor, an infrared thermal imaging camera, a current sensor, an environmental sensor and a smoke sensor;

[0011] The data of photovoltaic modules include temperature data, current data, environmental data, thermal imaging data and smoke data.

[0012] As a preferred solution of the centralized photovoltaic power station fire alarm intelligent monitoring method of the present invention, wherein: obtaining the temperature distribution diagram of the photovoltaic module includes fusing the data of the photovoltaic module collected by all sensors to form a photovoltaic module status data set;

[0013] The PV module status data set is mapped to each pixel of the image through two-dimensional coordinates to form a temperature distribution map of the PV module.

[0014] As a preferred solution of the centralized photovoltaic power station fire alarm intelligent monitoring method described in the present invention, wherein: identifying the hot spot area and judging whether there is an abnormality includes identifying the hot spot area and judging whether to issue an alarm.

[0015] As a preferred solution of the centralized photovoltaic power station fire alarm intelligent monitoring method described in the present invention, the identification of hot spot areas includes performing Gaussian smoothing on the temperature data in the temperature distribution diagram, combining the output result of the clustering algorithm with the set temperature threshold, and further screening and marking the temperature abnormal area. The formula is expressed as follows:

[0016]

[0017] Wherein, H(x, y) represents an indicator function of whether it is a hot spot area, H(x, y) = 1 represents a hot spot area, and H(x, y) = 0 represents a non-hot spot area;

[0018] T(x,y) represents the temperature value at the position (x,y); μ T represents the mean of temperature data, σ T represents the standard deviation of temperature data; α represents the threshold for determining temperature anomaly; x0, y0 represent the center coordinates of the hot spot area; σ G represents the standard deviation of Gaussian smoothing, Represents the indicator function.

[0019] As a preferred solution of the centralized photovoltaic power station fire alarm intelligent monitoring method of the present invention, the determination of whether to issue an alarm includes evaluating the area of ​​the hot spot region. If the total area of ​​the hot spot region exceeds a set minimum threshold, the region is considered abnormal and an alarm signal is triggered. The formula is expressed as follows:

[0020]

[0021] Where A represents the alarm signal, A=1 represents triggering the alarm, and A=0 represents not triggering the alarm); N th represents the minimum hot spot area threshold, H(x,y) represents the indicator function of whether it is a hot spot area, N th Indicates the minimum area threshold of the hot spot area.

[0022] As a preferred solution of the centralized photovoltaic power station fire alarm intelligent monitoring method described in the present invention, wherein: when an abnormality is detected, an alarm is triggered and the on-site staff is notified, and timely response to the fire risk includes combining real-time environmental data and historical abnormal data, using normalization function and weighted analysis to calculate the dynamic threshold N of the hot spot area with real-time adaptability th , the formula is:

[0023]

[0024] Among them, N th represents the dynamic threshold of the hot spot area, S represents the sunlight intensity, T e represents the ambient temperature, σ V Indicates the smoothed value of wind speed, H d represents the area of ​​historical abnormal region, γ represents the adjustment coefficient, and λ represents the weight factor of historical data;

[0025] N th The range of is [0.1,10];

[0026] When N th <0.5, record the current hot spot data and store it in the system database, do not send an alarm to the staff, and only use it for internal monitoring of the system;

[0027] When 0.5≤N th <2, send warning information to staff, reminding them to pay attention to abnormalities in the area;

[0028] When N th When the value is ≥2, an emergency alarm signal is triggered, and an alarm message is sent to on-site and remote operation and maintenance personnel at the same time, automatically cutting off the circuit of the relevant PV panels.

[0029] As a preferred solution of the centralized photovoltaic power station fire intelligent monitoring method described in the present invention, the data module installs sensors on each photovoltaic module to collect data of each photovoltaic module in real time and obtain the temperature distribution diagram of the photovoltaic module.

[0030] The recognition module analyzes the temperature distribution map, identifies the hot spot area and determines whether there is an abnormality;

[0031] The response module, when an abnormality is detected, triggers an alarm and notifies on-site staff to respond to fire risks in a timely manner.

[0032] 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.

[0033] 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.

[0034] Beneficial effects of the invention: The centralized photovoltaic power station fire alarm intelligent monitoring method provided by the invention realizes accurate identification and real-time monitoring of hot spot areas of photovoltaic components through multi-sensor data fusion and Gaussian smoothing processing, combined with a dynamic threshold calculation model and a clustering algorithm, significantly improving the accuracy and response efficiency of fire warnings, overcoming the problems of high false alarm rate, serious missed alarms and poor environmental adaptability in the prior art, effectively ensuring the safe operation of photovoltaic power stations, and reducing fire risks and economic losses. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] 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.

[0036] Figure 1 This is an overall flow chart of a centralized photovoltaic power station fire intelligent monitoring method provided by the first embodiment of the present invention. DETAILED DESCRIPTION

[0037] 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.

[0038] Example 1, reference Figure 1 , is an embodiment of the present invention, and provides a centralized photovoltaic power station fire alarm intelligent monitoring method, comprising:

[0039] S1: Install sensors on each photovoltaic module to collect data of each photovoltaic module in real time and obtain the temperature distribution map of the photovoltaic module.

[0040] Sensors include temperature sensors, infrared thermal imaging cameras, current sensors, environmental sensors, and smoke sensors;

[0041] The data of photovoltaic modules include temperature data, current data, environmental data, thermal imaging data and smoke data.

[0042] Acquiring the temperature distribution map of the photovoltaic module includes fusing the data of the photovoltaic module collected by all sensors to form a photovoltaic module status data set;

[0043] The PV module status data set is mapped to each pixel of the image through two-dimensional coordinates to form a temperature distribution map of the PV module.

[0044] By installing multiple types of sensors (including temperature sensors, infrared thermal imaging cameras, etc.) on each photovoltaic module, multi-dimensional data (such as temperature, current, environment and smoke data) is collected in real time, and the temperature distribution map of the photovoltaic module is generated through data fusion technology to intuitively reflect the temperature change characteristics of the module surface. By mapping the data to two-dimensional coordinates to form a visual temperature distribution map, it can not only comprehensively monitor the operating status of the photovoltaic module, but also provide an accurate data basis for subsequent hot spot area identification and fire risk analysis, thereby significantly improving the accuracy and reliability of the monitoring system.

[0045] S2: Analyze the temperature distribution map, identify the hot spot area and determine whether there is an abnormality.

[0046] Identifying the hot spot area and determining whether there is an abnormality includes identifying the hot spot area and determining whether to issue an alarm.

[0047] Gaussian smoothing is performed on the temperature data in the temperature distribution map. The output results of the clustering algorithm are combined with the set temperature threshold to further screen and mark the temperature abnormality areas. The formula is expressed as follows:

[0048]

[0049] Wherein, H(x, y) represents an indicator function of whether it is a hot spot area, H(x, y) = 1 represents a hot spot area, and H(x, y) = 0 represents a non-hot spot area;

[0050] T(x,y) represents the temperature value at the position (x,y); μ T represents the mean of temperature data, σ T represents the standard deviation of temperature data; α represents the threshold for determining temperature anomaly; x0, y0 represent the center coordinates of the hot spot area; σ G represents the standard deviation of Gaussian smoothing, Represents the indicator function.

[0051] The determination of whether to issue an alarm includes evaluating the area of ​​the hot spot region. If the total area of ​​the hot spot region exceeds the set minimum threshold, the region is considered abnormal and an alarm signal is triggered. The formula is expressed as:

[0052]

[0053] Where A represents the alarm signal, A=1 represents triggering the alarm, and A=0 represents not triggering the alarm); N th represents the minimum hot spot area threshold, H(x,y) represents the indicator function of whether it is a hot spot area, N th Indicates the minimum area threshold of the hot spot area.

[0054] By analyzing the temperature distribution map, combining Gaussian smoothing and clustering algorithms to optimize the screening and marking of abnormal temperature areas, the hot spot area of ​​PV modules can be accurately identified, significantly improving the accuracy and reliability of detection. At the same time, by evaluating the area of ​​the hot spot area and comparing it with the set minimum threshold, a dynamic abnormality judgment and alarm triggering mechanism is implemented to ensure that an alarm is issued only when there is a real risk of fire, effectively reducing the false alarm and missed alarm rates, and providing accurate intelligent monitoring guarantee for the safe operation of photovoltaic power stations.

[0055] S3: When an abnormality is detected, an alarm is triggered and on-site staff are notified to respond to the fire risk in a timely manner.

[0056] When an abnormality is detected, an alarm is triggered and the on-site staff is notified to respond to the fire risk in a timely manner. This includes combining real-time environmental data and historical abnormal data, using normalization functions and weighted analysis to calculate the dynamic threshold N of the hot spot area that is adaptive in real time. th , the formula is:

[0057]

[0058] Among them, N th represents the dynamic threshold of the hot spot area, S represents the sunlight intensity, T e represents the ambient temperature, σ V Indicates the smoothed value of wind speed, H d represents the area of ​​historical abnormal region, γ represents the adjustment coefficient, and λ represents the weight factor of historical data;

[0059] N th The range of is [0.1,10];

[0060] When N th <0.5, record the current hot spot data and store it in the system database, do not send an alarm to the staff, and only use it for internal monitoring of the system;

[0061] When 0.5≤N th <2, send warning information to staff, reminding them to pay attention to abnormalities in the area;

[0062] When N th When the value is ≥2, an emergency alarm signal is triggered, and an alarm message is sent to on-site and remote operation and maintenance personnel at the same time, automatically cutting off the circuit of the relevant PV panels.

[0063] By combining real-time environmental data (such as sunshine intensity, ambient temperature, wind speed) and historical abnormal data, the adaptive threshold of the hot spot area is dynamically calculated to ensure that the monitoring system can accurately identify the severity of hot spot anomalies according to actual conditions. Through the optimization method of normalization function and weighted analysis, the system can effectively distinguish normal fluctuations from potential risks, and trigger response mechanisms such as recording, early warning and emergency alarm in a graded manner, thereby reducing false alarms and missed alarms, significantly improving the accuracy and real-time nature of fire risk warnings, and providing fast and efficient decision support for fire handling.

[0064] Embodiment 2 is an embodiment of the present invention, which provides a centralized photovoltaic power station fire alarm intelligent monitoring system, including:

[0065] The data module installs sensors on each photovoltaic module to collect data of each photovoltaic module in real time and obtain the temperature distribution map of the photovoltaic module.

[0066] The recognition module analyzes the temperature distribution map, identifies the hot spot area and determines whether there is an abnormality.

[0067] The response module, when an abnormality is detected, triggers an alarm and notifies on-site staff to respond to fire risks in a timely manner.

[0068] Embodiment 3, an embodiment of the present invention, is different from the first two embodiments in that:

[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] Embodiment 4,

[0074] The test was conducted in a 10,000 square meter photovoltaic power station, which includes 1,000 photovoltaic modules. Five sensors were installed on each photovoltaic module, including a temperature sensor (with an accuracy of ±0.1°C), an infrared thermal imaging camera (with a resolution of 320×240 pixels), a current sensor (with a range of 0-50A), an environmental sensor (to detect wind speed, sunshine intensity, temperature and other parameters), and a smoke sensor.

[0075] The data from each sensor is transmitted to the centralized control system in real time via the wireless network to build a real-time status data set of the photovoltaic components.

[0076] The control system fuses all collected data, maps the sensor data to each pixel of the image through two-dimensional coordinates, and generates a temperature distribution map of the photovoltaic module.

[0077] The temperature distribution graph is denoised using a Gaussian smoothing function to reduce the interference of environmental fluctuations and equipment noise on the data.

[0078] The K-means clustering algorithm is applied to the processed temperature distribution map to divide the temperature values ​​into normal and abnormal areas.

[0079] Combine the temperature mean, standard deviation and dynamic threshold formula to identify the hot spot area. Use the real-time hot spot area evaluation formula to calculate whether the hot spot area exceeds the set threshold.

[0080] If the hot spot area exceeds the threshold, the system automatically triggers an alarm signal and notifies the operation and maintenance personnel via text messages and the console, while cutting off the circuits of the relevant photovoltaic modules to prevent the accident from expanding.

[0081] In multiple simulation experiments, data was collected by manually creating hot spots (by blocking specific areas of the photovoltaic module or heating the surface of the module) and natural environment monitoring (exposure to different sunlight intensities for a long time) to verify the detection accuracy and response speed of the system. The experimental data are shown in Table 1.

[0082] Table 1 Experimental data table

[0083]

[0084]

[0085] Through the above experimental data, it can be seen that the centralized photovoltaic power station fire alarm intelligent monitoring method of the present invention has obvious advantages in abnormal hot spot area identification and alarm:

[0086] The data shows that when the average temperature of the PV modules is normal (such as normal modules 1, 2, and 3) and the hot spot area is lower than the set threshold, the system can accurately determine that it is not an abnormal situation.

[0087] For abnormal components (such as abnormal components 1, 2, and 3), the average temperature is obviously high, and the hot spot areas are 3.5㎡, 2.8㎡, and 2.2㎡, respectively, all exceeding the dynamic threshold NthN_{th}Nth. The system can trigger the alarm signal in time.

[0088] Gaussian smoothing significantly reduces temperature fluctuations (e.g., the standard deviation of normal components is less than 3°C) and effectively filters out noise interference. The clustering algorithm combined with the dynamic threshold formula can not only distinguish between normal and abnormal areas, but also adjust the threshold in real time according to environmental parameters (e.g., sunlight intensity, ambient temperature), enhancing system adaptability.

[0089] Traditional fire alarm monitoring methods are mostly based on a single temperature threshold, which is easily affected by environmental fluctuations and equipment noise, and has a high false alarm rate. However, the present invention achieves a higher detection accuracy through dynamic threshold adjustment and multi-parameter fusion analysis.

[0090] Experimental data show that the system can still maintain efficient detection under high sunlight intensity and high temperature environments (such as abnormal components 2 and 3), avoiding the problem of underreporting that may occur in traditional methods under extreme environments.

[0091] The system realizes the automation of the entire process from data collection to real-time alarm, significantly improving the efficiency of fire risk monitoring in photovoltaic power plants. The dynamic threshold calculation combined with multi-dimensional parameters makes the monitoring method more robust and flexible, and adapts to various photovoltaic power plant operation scenarios.

[0092] 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 centralized photovoltaic power station fire alarm intelligent monitoring method, characterized in that: include: Install sensors on each photovoltaic module to collect data of each photovoltaic module in real time and obtain the temperature distribution map of the photovoltaic module; Analyze the temperature distribution map, identify the hot spot area and determine whether there is an abnormality; When an abnormality is detected, an alarm is triggered and on-site staff are notified to respond to the fire risk in a timely manner.

2. The centralized photovoltaic power station fire alarm intelligent monitoring method according to claim 1, characterized in that: The sensors include a temperature sensor, an infrared thermal imaging camera, a current sensor, an environmental sensor and a smoke sensor; The data of photovoltaic modules include temperature data, current data, environmental data, thermal imaging data and smoke data.

3. The centralized photovoltaic power station fire alarm intelligent monitoring method according to claim 2, characterized in that: Acquiring the temperature distribution map of the photovoltaic module includes fusing the data of the photovoltaic module collected by all sensors to form a photovoltaic module status data set; The PV module status data set is mapped to each pixel of the image through two-dimensional coordinates to form a temperature distribution map of the PV module.

4. The centralized photovoltaic power station fire alarm intelligent monitoring method according to claim 3, characterized in that: Identifying the hot spot area and determining whether there is an abnormality includes identifying the hot spot area and determining whether to issue an alarm.

5. The centralized photovoltaic power station fire alarm intelligent monitoring method according to claim 4, characterized in that: The identification of hot spot areas includes performing Gaussian smoothing on the temperature data in the temperature distribution map, combining the output results of the clustering algorithm with the set temperature threshold, and further screening and marking the temperature abnormality area. The formula is expressed as: Among them, H ( x,y ) Indicator function indicating whether it is a hot spot area, H ( x,y ) =1 indicates hot spot area, H ( x,y ) =0 indicates non-hot spot area; T ( x,y ) Indicates location ( x,y ) Temperature value at T represents the mean of temperature data, σ T represents the standard deviation of temperature data; α represents the threshold for determining temperature anomaly; x0, y0 represent the center coordinates of the hot spot area; σ G represents the standard deviation of Gaussian smoothing, Represents the indicator function.

6. The centralized photovoltaic power station fire alarm intelligent monitoring method according to claim 5, characterized in that: The determination of whether to issue an alarm includes evaluating the area of ​​the hot spot region. If the total area of ​​the hot spot region exceeds a set minimum threshold, the region is considered abnormal and an alarm signal is triggered. The formula is: Where A represents the alarm signal, A=1 represents triggering the alarm, and A=0 represents not triggering the alarm); N th Indicates the minimum hot spot area threshold, H ( x,y ) Indicator function indicating whether it is a hot spot area, N th Indicates the minimum area threshold of the hot spot area.

7. The centralized photovoltaic power station fire alarm intelligent monitoring method according to claim 6, characterized in that: When an abnormality is detected, an alarm is triggered and the on-site staff is notified to respond to the fire risk in a timely manner, including combining real-time environmental data and historical abnormal data, using normalization function and weighted analysis to calculate the dynamic threshold N of the hot spot area with real-time adaptability th , the formula is: Among them, N th represents the dynamic threshold of the hot spot area, S represents the sunlight intensity, T e represents the ambient temperature, σ V Indicates the smoothed value of wind speed, H d represents the area of ​​historical abnormal region, γ represents the adjustment coefficient, and λ represents the weight factor of historical data; N th The value range of is [0.1,10]; When N th <0.5, record the current hot spot data and store it in the system database, do not send an alarm to the staff, and only use it for internal monitoring of the system; When 0.5≤N th <2, send warning information to staff, reminding them to pay attention to abnormalities in the area; When N th When the value is ≥2, an emergency alarm signal is triggered, and an alarm message is sent to on-site and remote operation and maintenance personnel at the same time, automatically cutting off the circuit of the relevant PV panels.

8. A centralized photovoltaic power station fire alarm intelligent monitoring system using the method according to any one of claims 1 to 7, characterized in that: The data module installs sensors on each photovoltaic module to collect data of each photovoltaic module in real time and obtain the temperature distribution map of the photovoltaic module; The recognition module analyzes the temperature distribution map, identifies the hot spot area and determines whether there is an abnormality; The response module, when an abnormality is detected, triggers an alarm and notifies on-site staff to respond to fire risks in a timely manner.

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 centralized photovoltaic power station fire alarm intelligent monitoring 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 centralized photovoltaic power station fire alarm intelligent monitoring method according to any one of claims 1 to 7 are implemented.

Images