Fire early warning and extinguishing method and system of photovoltaic power generation unit

Abstract

The invention relates to the technical field of fire early warning and extinguishing, and discloses a fire early warning and extinguishing method and system for a photovoltaic power generation unit, and the method comprises the steps: deploying various types of sensors in the photovoltaic power generation unit and around the photovoltaic power generation unit, constructing a multi-layer sensor network, and collecting data; the potential fire risk is predicted based on the data collected by the multi-level sensor network; when a potential fire risk is detected, a fighting measure is immediately started. By constructing a multi-level sensor network, the problems that in an existing fire early warning system, monitoring data of a single sensor is insufficient, the data integration capacity of a multi-level sensor is poor, dynamic adaptability is lacked, and an automatic emergency response function is lacked are solved. The fire risk can be accurately predicted, and when the potential fire risk is detected, the fighting measures are automatically started. The fire early warning accuracy and the emergency response efficiency of the photovoltaic power generation unit are remarkably improved, and the safety and the stability of the system are guaranteed.

Description

Technical Field

[0001] The present invention relates to the technical field of fire early warning and fire fighting, and in particular to a fire early warning and fire fighting method and system for a photovoltaic power generation unit. Background Art

[0002] With the increasing global demand for renewable energy, photovoltaic power generation technology has been rapidly developed and widely used. As the core component of solar power generation, photovoltaic power generation units have been widely used in various scenarios. However, the efficient operation of photovoltaic power generation units depends on the complex electrical equipment inside them and the stability of the external environment. Any abnormal situation may cause equipment failure or even fire. In recent years, with the large-scale application of photovoltaic power generation systems, its fire hazards have gradually emerged. Traditional fire warning systems usually rely on data from a single sensor type, such as temperature or smoke sensors. However, since photovoltaic power generation units are usually in a complex environment, the data from a single sensor is often not enough to predict the risk of fire in a timely and accurate manner, which makes traditional fire warning systems unable to cope with complex and changing fire risks. In addition, photovoltaic power generation units are usually arranged in remote or difficult-to-monitor areas, which makes it more difficult to detect and extinguish fires in a timely manner.

[0003] Existing fire warning technologies face several major problems when dealing with fire risks in photovoltaic power generation units: First, existing systems are usually unable to effectively integrate data from multiple sensors and lack multi-level data analysis capabilities, resulting in insufficient accuracy and timeliness of fire warnings. Second, traditional fire warning systems mostly rely on static threshold settings and cannot dynamically adapt to changes in the environment in which the photovoltaic power generation unit is located, such as temperature, humidity, gas concentration, etc., which makes the system prone to missed or false alarms. In addition, after a fire occurs, existing technologies can usually only issue an alarm but cannot automatically perform effective emergency response or repair, thereby failing to effectively curb the spread of the fire in the early stages of the fire. These technical deficiencies not only increase the safety risks of photovoltaic power generation units, but may also cause huge economic losses and environmental damage. Summary of the invention

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

[0005] Therefore, the technical problems solved by the present invention are: insufficient monitoring data of a single sensor, lack of multi-level sensor data integration capability, lack of dynamic adaptability and lack of automated emergency response function in the photovoltaic power generation unit fire warning system.

[0006] In order to solve the above technical problems, the present invention provides the following technical solutions: a fire warning and fire fighting method for a photovoltaic power generation unit, comprising:

[0007] Deploy various types of sensors inside and around photovoltaic power generation units to build a multi-level sensor network and collect data;

[0008] Predict potential fire risks based on data collected by a multi-level sensor network;

[0009] When a potential fire risk is detected, firefighting measures are initiated immediately.

[0010] As a preferred solution of the fire warning and fire fighting method of the photovoltaic power generation unit described in the present invention, the multiple types of sensors include temperature sensors, smoke sensors, gas sensors, humidity sensors, pressure sensors, light sensors, current sensors, and vibration sensors;

[0011] The data collected by the multi-level sensor network include real-time temperature values ​​of different parts of the photovoltaic power generation unit, smoke concentration and its changing trend, concentration of specific gases in the air, ambient humidity value, gas pressure inside or around the photovoltaic unit, ambient light intensity and light changes on the surface of the photovoltaic unit, current intensity, vibration frequency and amplitude of the photovoltaic power generation equipment.

[0012] As a preferred solution of the fire warning and fire fighting method of the photovoltaic power generation unit of the present invention, wherein: the multi-level sensor network includes, in order to realize the all-round monitoring of the early signs of the fire of the photovoltaic power generation unit, a multi-level sensor network including a core layer, an auxiliary layer and an environmental layer is constructed;

[0013] The core layer includes a temperature sensor, a gas sensor and a current sensor;

[0014] The auxiliary layer includes a pressure sensor, a smoke sensor and a humidity sensor;

[0015] The environmental layer includes a light sensor and a vibration sensor;

[0016] When sensors at different levels detect data anomalies at the same time and the anomaly value exceeds the set crossover threshold, the sensitivity and operating frequency of the multi-level sensor system are adjusted in real time;

[0017] The cross threshold includes the cross threshold of the core layer sensor, the cross threshold of the auxiliary layer sensor and the cross threshold setting of the environment layer sensor;

[0018] The cross threshold of the core layer sensor includes when the temperature sensor detects that the temperature rise exceeds the temperature sensor threshold, the gas sensor simultaneously detects that the combustible gas concentration increases to exceed the gas sensor threshold, and the current sensor simultaneously detects that the current fluctuation exceeds the current sensor threshold;

[0019] The crossing threshold of the auxiliary layer sensor includes when the pressure sensor detects that the internal pressure changes exceeding the pressure sensor threshold, the smoke sensor simultaneously detects that the smoke concentration exceeds the smoke sensor threshold, and the humidity sensor simultaneously detects that the humidity changes exceeding the humidity sensor threshold;

[0020] The environmental layer sensor crossing threshold includes, when the light sensor detects that the light intensity change exceeds the light sensor threshold, and the vibration sensor detects that the vibration amplitude exceeds the vibration sensor threshold;

[0021] The real-time adjustment of the sensitivity and operating frequency of the multi-level sensor system includes performing real-time calculation based on the cross thresholds of sensors at different levels to form a comprehensive index; when the comprehensive index reaches or exceeds the comprehensive threshold, the multi-level sensor network will automatically adjust the sensitivity and operating frequency;

[0022] According to the fuzzy reasoning results of sensor data, the adjustment factor is determined, and the sensor data is corrected using the adaptive filter algorithm to ensure the stability of the adjustment process;

[0023] The formula of the comprehensive index is expressed as:

[0024]

[0025] When Λ(t)≥Λ th When , the formula for adjusting the sensitivity and operating frequency of the multi-level sensor system is expressed as:

[0026]

[0027] Among them, Λ(t) represents the comprehensive index, T(t) represents the real-time temperature data, G(t) represents the real-time gas concentration, I(t) represents the real-time current, P(t) represents the real-time pressure, S(t) represents the real-time smoke concentration, H(t) represents the real-time humidity, L(t) represents the real-time light intensity, V(t) represents the real-time vibration amplitude, S′(t) represents the adjusted sensor sensitivity, f′(t) represents the adjusted operating frequency, α represents the sensitivity coefficient, β represents the operating frequency adjustment coefficient, Λ th represents the comprehensive threshold, T th Represents the threshold of the temperature sensor, G th Indicates the threshold of the gas sensor, I th Represents the threshold of the current sensor, P th Indicates the pressure sensor threshold, S th Indicates the smoke sensor threshold, H th Indicates the humidity sensor threshold, L th Represents the light sensor threshold, V th Indicates the vibration sensor threshold;

[0028] The value range of Λ(t) is from 0 to positive infinity;

[0029] The larger the value of Λ(t), the more anomalies the system detects.

[0030] As a preferred solution of the fire warning and fire fighting method of the photovoltaic power generation unit described in the present invention, wherein: the comprehensive threshold includes a real-time comprehensive threshold and a historical prediction comprehensive threshold;

[0031] When the real-time comprehensive threshold is inconsistent with the historical prediction comprehensive threshold, and the data collected by the current multi-level sensor network fluctuates more than twice its long-term average, the real-time comprehensive threshold is used first;

[0032] When the fluctuation of data collected by the multi-level sensor network is less than twice its long-term average, the historical prediction comprehensive threshold is preferred;

[0033] Collect the data of each sensor in the past 6 months as historical data;

[0034] Analyze the historical data and calculate the long-term average value and standard deviation of each sensor;

[0035] The long-term average value includes the average value of the data collected by each sensor in the past 6 months.

[0036] As a preferred solution of the fire warning and fire fighting method of the photovoltaic power generation unit described in the present invention, the real-time comprehensive threshold value includes using the data of the multi-level sensor network, adopting the weighted comprehensive method, and applying the Kalman filter for smoothing to obtain the real-time comprehensive threshold value, which is expressed as follows:

[0037]

[0038] Among them, Λ real (t) represents the real-time comprehensive index calculated at time t, Λ core (t) represents the comprehensive index of the core layer at time t, Λ support (t) represents the comprehensive index of the auxiliary layer at time t, Λ env (t) represents the comprehensive index of the environmental layer at time t, T(t) represents real-time temperature data, G(t) represents real-time gas concentration, I(t) represents real-time current, P(t) represents real-time pressure, S(t) represents real-time smoke concentration, H(t) represents real-time humidity, L(t) represents real-time light intensity, V(t) represents real-time vibration amplitude, T th Represents the threshold of the temperature sensor, G th Indicates the threshold of the gas sensor, I th Represents the threshold of the current sensor, P th Indicates the pressure sensor threshold, Sth Indicates the smoke sensor threshold, H th Indicates the humidity sensor threshold, L th Represents the light sensor threshold, V th Indicates the vibration sensor threshold.

[0039] As a preferred solution of the fire warning and fire fighting method of the photovoltaic power generation unit described in the present invention, the historical prediction comprehensive threshold includes analyzing the data collected by the multi-level sensor network and its change rate using random forest to predict the comprehensive risk at the current moment, and obtaining the historical prediction comprehensive threshold, which is expressed as follows:

[0040]

[0041] Among them, Λ real (t) represents the real-time comprehensive index calculated at time t, Λ core (t) represents the comprehensive index of the core layer at time t, Λ support (t) represents the comprehensive index of the auxiliary layer at time t, Λ env (t) represents the comprehensive index of the environmental layer at time t, Λ hist (t) represents the comprehensive threshold of historical prediction, RF j represents the output value of the j-th decision tree, N represents the total number of decision trees contained in the random forest, and j represents the index of the j-th decision tree currently being processed.

[0042] As a preferred solution of the fire warning and fire fighting method of the photovoltaic power generation unit described in the present invention, when a potential fire risk is detected, immediately starting fire fighting measures includes:

[0043] When 0.8Λ th ≤Λ(t)<Λ th When the load is exceeded, the real-time monitoring log is started and the operator is notified to gradually reduce the operating load of the photovoltaic power generation unit;

[0044] When Λ(t)=Λ th When a fire occurs, the fire alarm is immediately sounded, the operator is notified, an alarm log is generated, the current risk parameters are recorded, and the standby status of the fire-fighting equipment is activated;

[0045] When Λ(t)>Λ th When a fire breaks out, the system immediately initiates a comprehensive emergency response mechanism, automatically isolates the affected PV modules, adjusts the current and load to prevent the fire from spreading, automatically repairs and restores some functions, and sends emergency notifications to designated safety managers and fire departments via the network;

[0046] The preparation state of the fire-fighting equipment includes turning on the standby mode of the fire-fighting water pump and preparing to start the relevant emergency power supply;

[0047] The comprehensive emergency response mechanism includes shutting down power, activating fire extinguishing systems, and implementing personnel evacuation plans;

[0048] During the recovery phase after the fire risk is eliminated, self-checks are performed on sensors at all levels and functional checks are performed on emergency equipment;

[0049] If any sensor or equipment abnormality is found during the self-check process, the system should immediately notify maintenance personnel to repair it and maintain a high alert state during this period.

[0050] A fire warning and fire fighting system for a photovoltaic power generation unit, wherein:

[0051] Multi-level sensor network module, deploys various types of sensors inside and around the photovoltaic power generation unit to build a multi-level sensor network for data collection. Sensor types may include temperature sensors, smoke sensors, gas sensors, humidity sensors, pressure sensors, light sensors, current sensors, vibration sensors;

[0052] Data acquisition and processing module collects real-time data from multi-level sensor networks and processes and analyzes the data;

[0053] The fire risk prediction module predicts the potential fire risk of the photovoltaic power generation unit based on the processed sensor data;

[0054] The alarm and emergency response module automatically triggers the alarm system and initiates corresponding emergency response measures when the fire risk reaches or exceeds the threshold;

[0055] After the self-repair and protection module detects a fire risk, the system automatically starts the self-repair mechanism of the photovoltaic unit, isolates the affected module, and attempts to restore system functions;

[0056] The remote monitoring and notification module records and monitors the development of fire risks, and sends fire risk notifications to relevant departments.

[0057] 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.

[0058] 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.

[0059] Beneficial effects of the present invention: The fire warning and fire fighting method for photovoltaic power generation units provided by the present invention solves the problems of insufficient monitoring data from a single sensor, lack of multi-level sensor data integration capabilities, lack of dynamic adaptability, and lack of automated emergency response functions in existing fire warning systems by constructing a multi-level sensor network and deploying various types of sensors inside and around the photovoltaic power generation unit. By real-time collection and analysis of data from a variety of sensors such as temperature, smoke, gas, humidity, pressure, light, current, and vibration, the system can accurately predict fire risks and automatically initiate fire fighting measures when potential fire risks are detected. This significantly improves the accuracy of fire warnings and emergency response efficiency of photovoltaic power generation units, and ensures the safety and stability of the system. BRIEF DESCRIPTION OF THE DRAWINGS

[0060] 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.

[0061] Figure 1 The present invention provides an overall flow chart of a fire warning and fire fighting method for a photovoltaic power generation unit according to the first embodiment of the present invention. DETAILED DESCRIPTION

[0062] 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.

[0063] Example 1, reference Figure 1 , as an embodiment of the present invention, provides a fire warning and fire fighting method for a photovoltaic power generation unit, comprising:

[0064] S1: Deploy various types of sensors inside and around the photovoltaic power generation unit to build a multi-level sensor network and collect data.

[0065] The various types of sensors include temperature sensors, smoke sensors, gas sensors, humidity sensors, pressure sensors, light sensors, current sensors, and vibration sensors.

[0066] The data collected by the multi-level sensor network include real-time temperature values ​​of different parts of the photovoltaic power generation unit, smoke concentration and its changing trend, concentration of specific gases in the air, ambient humidity value, gas pressure inside or around the photovoltaic unit, ambient light intensity and light changes on the surface of the photovoltaic unit, current intensity, vibration frequency and amplitude of the photovoltaic power generation equipment.

[0067] The multi-layer sensor network includes a core layer, an auxiliary layer and an environmental layer, which are constructed in order to realize all-round monitoring of early signs of fire in photovoltaic power generation units.

[0068] The core layer includes temperature sensor, gas sensor and current sensor.

[0069] The auxiliary layer includes pressure sensor, smoke sensor and humidity sensor.

[0070] The environmental layer includes light sensors and vibration sensors.

[0071] When sensors at different levels detect data anomalies at the same time and the anomaly values ​​exceed the set crossover threshold, the sensitivity and operating frequency of the multi-level sensor system are adjusted in real time.

[0072] The cross threshold includes the cross threshold of the core layer sensor, the cross threshold of the auxiliary layer sensor and the cross threshold setting of the environment layer sensor.

[0073] The cross threshold of the core layer sensor includes when the temperature sensor detects a temperature rise exceeding the temperature sensor threshold, the gas sensor simultaneously detects an increase in combustible gas concentration exceeding the gas sensor threshold, and the current sensor simultaneously detects a current fluctuation exceeding the current sensor threshold.

[0074] The crossing threshold of the auxiliary layer sensor includes when the pressure sensor detects that the internal pressure changes exceeding the pressure sensor threshold, the smoke sensor simultaneously detects that the smoke concentration exceeds the smoke sensor threshold, and the humidity sensor simultaneously detects that the humidity changes exceeding the humidity sensor threshold.

[0075] The environmental layer sensor crossing threshold includes when the light sensor detects that the light intensity change exceeds the light sensor threshold, and the vibration sensor detects that the vibration amplitude exceeds the vibration sensor threshold.

[0076] Real-time adjustment of the sensitivity and operating frequency of the multi-level sensor system includes real-time calculation based on the cross-thresholds of sensors at different levels to form a comprehensive index; when the comprehensive index reaches or exceeds the comprehensive threshold, the multi-level sensor network will automatically adjust the sensitivity and operating frequency.

[0077] According to the fuzzy reasoning results of sensor data, the adjustment factor is determined, and the sensor data is corrected using the adaptive filter algorithm to ensure the stability of the adjustment process.

[0078] The formula of the comprehensive index is expressed as:

[0079]

[0080] When Λ(t)≥Λ th When , the formula for adjusting the sensitivity and operating frequency of the multi-level sensor system is expressed as:

[0081]

[0082] Among them, Λ(t) represents the comprehensive index, T(t) represents the real-time temperature data, G(t) represents the real-time gas concentration, I(t) represents the real-time current, P(t) represents the real-time pressure, S(t) represents the real-time smoke concentration, H(t) represents the real-time humidity, L(t) represents the real-time light intensity, V(t) represents the real-time vibration amplitude, S′(t) represents the adjusted sensor sensitivity, f′(t) represents the adjusted operating frequency, α represents the sensitivity coefficient, β represents the operating frequency adjustment coefficient, Λ th represents the comprehensive threshold, T th Represents the threshold of the temperature sensor, G th Indicates the threshold of the gas sensor, I th Represents the threshold of the current sensor, P th Indicates the pressure sensor threshold, S th Indicates the smoke sensor threshold, H th Indicates the humidity sensor threshold, L th Represents the light sensor threshold, V th Indicates the vibration sensor threshold.

[0083] The value range of Λ(t) is from 0 to positive infinity.

[0084] The larger the value of Λ(t), the more anomalies the system detects.

[0085] Furthermore, a multi-level sensor network was constructed by deploying various types of sensors inside and around the photovoltaic power generation unit. This multi-level structure ensures comprehensive monitoring of the photovoltaic power generation unit and can simultaneously collect data in multiple dimensions, such as temperature, gas concentration, current, pressure, etc. This comprehensive data collection method can more accurately reflect the actual working status and environmental changes of the photovoltaic power generation unit, thereby providing a reliable data basis for subsequent fire risk prediction.

[0086] Furthermore, the construction of a multi-level sensor network not only improves the accuracy and comprehensiveness of data collection, but also provides a solid foundation for real-time monitoring and early warning of the system. This method effectively overcomes the limitations of traditional single sensor systems and ensures the timely discovery and handling of fire hazards in complex environments.

[0087] S2: Predict potential fire risks based on data collected by a multi-level sensor network.

[0088] The comprehensive threshold includes the real-time comprehensive threshold and the historical prediction comprehensive threshold.

[0089] When the real-time comprehensive threshold is inconsistent with the historical prediction comprehensive threshold, and the data collected by the current multi-level sensor network fluctuates more than twice its long-term average, the real-time comprehensive threshold is used first.

[0090] When the fluctuation of data collected by the multi-level sensor network is less than twice its long-term average, the historical prediction comprehensive threshold is preferred.

[0091] Collect the data of each sensor in the past 6 months as historical data.

[0092] The historical data is analyzed to calculate the long-term average and standard deviation of each sensor.

[0093] The long-term average value includes the average value of the data collected by each sensor in the past 6 months.

[0094] The real-time comprehensive threshold includes using the data of the multi-level sensor network, adopting the weighted comprehensive method, and applying the Kalman filter for smoothing to obtain the real-time comprehensive threshold, which is expressed as follows:

[0095]

[0096] Among them, Λ real (t) represents the real-time comprehensive index calculated at time t, Λ core (t) represents the comprehensive index of the core layer at time t, Λ support (t) represents the comprehensive index of the auxiliary layer at time t, Λ env (t) represents the comprehensive index of the environmental layer at time t, T(t) represents real-time temperature data, G(t) represents real-time gas concentration, I(t) represents real-time current, P(t) represents real-time pressure, S(t) represents real-time smoke concentration, H(t) represents real-time humidity, L(t) represents real-time light intensity, V(t) represents real-time vibration amplitude, T th Represents the threshold of the temperature sensor, G th Indicates the threshold of the gas sensor, I th Represents the threshold of the current sensor, Pth Indicates the pressure sensor threshold, S th Indicates the smoke sensor threshold, H th Indicates the humidity sensor threshold, L th Represents the light sensor threshold, V th Indicates the vibration sensor threshold.

[0097] The historical prediction comprehensive threshold includes using random forest to analyze the data collected by the multi-level sensor network and its change rate, predicting the comprehensive risk at the current moment, and obtaining the historical prediction comprehensive threshold. The formula is expressed as:

[0098]

[0099] Among them, Λ real (t) represents the real-time comprehensive index calculated at time t, Λ core (t) represents the comprehensive index of the core layer at time t, Λ support (t) represents the comprehensive index of the auxiliary layer at time t, Λ env (t) represents the comprehensive index of the environmental layer at time t, Λ hist (t) represents the comprehensive threshold of historical prediction, RF j represents the output value of the j-th decision tree, N represents the total number of decision trees contained in the random forest, and j represents the index of the j-th decision tree currently being processed.

[0100] Furthermore, using the data collected by the multi-level sensor network, the system can conduct a comprehensive analysis and predict potential fire risks. This process introduces the concepts of real-time comprehensive thresholds and historical prediction comprehensive thresholds. By combining the analysis of real-time data and historical data, dynamic prediction of fire risks is achieved. The application of this dual threshold enables the system to flexibly respond to different environmental changes and avoid false alarms and missed alarms. It can detect potential fire risks at an early stage and adapt to environmental changes by dynamically adjusting the system parameters, thereby improving the system's sensitivity and response speed.

[0101] S3: When a potential fire risk is detected, firefighting measures are initiated immediately.

[0102] When 0.8Λ th ≤Λ(t)<Λ th When the load of the photovoltaic power generation unit is reduced, the real-time monitoring log is started and the operator is notified to gradually reduce the operating load of the photovoltaic power generation unit.

[0103] When Λ(t)=Λ th When a fire occurs, a fire alarm is immediately sounded, the operator is notified, an alarm log is generated, the current risk parameters are recorded, and the readiness of the fire-fighting equipment is activated.

[0104] When Λ(t)>Λ th When a fire breaks out, the comprehensive emergency response mechanism is immediately activated to automatically isolate the affected photovoltaic modules, prevent the fire from spreading by adjusting the current and load, and automatically repair and restore some functions, and send emergency notifications to designated safety managers and fire departments through the network.

[0105] The readiness status of fire-fighting equipment includes turning on the standby mode of the fire-fighting pump and preparing to start the relevant emergency power supply.

[0106] A comprehensive emergency response mechanism includes shutting down power, activating fire extinguishing systems, and implementing evacuation plans.

[0107] During the recovery phase after the fire risk is eliminated, self-checks are performed on sensors at all levels and functional checks are performed on emergency equipment.

[0108] If any sensor or equipment abnormality is found during the self-check process, the system should immediately notify maintenance personnel to repair it and maintain a high alert state during this period.

[0109] Furthermore, after detecting a fire risk, the system can automatically initiate firefighting measures, including triggering an alarm, reducing the load of the power generation unit, and starting firefighting equipment. This automated emergency response mechanism ensures that the fire can be effectively controlled in the early stages to prevent the spread and further deterioration of the fire. The introduction of the automated emergency response system not only improves the efficiency of fire fighting, but also reduces the need for manual intervention. This preventive measure can minimize the loss of photovoltaic power generation units in a fire and ensure the safety and stability of the system.

[0110] On the other hand, this embodiment also provides a fire warning and fire fighting system for a photovoltaic power generation unit, which includes:

[0111] Multi-level sensor network module deploys various types of sensors inside and around the photovoltaic power generation unit to build a multi-level sensor network for data collection. Sensor types may include temperature sensors, smoke sensors, gas sensors, humidity sensors, pressure sensors, light sensors, current sensors, and vibration sensors.

[0112] The data acquisition and processing module collects real-time data from the multi-level sensor network and processes and analyzes the data.

[0113] The fire risk prediction module predicts the potential fire risk of the photovoltaic power generation unit based on the processed sensor data.

[0114] The alarm and emergency response module automatically triggers the alarm system and initiates corresponding emergency response measures when the fire risk reaches or exceeds the threshold.

[0115] After the self-repair and protection module detects the risk of fire, the system automatically starts the self-repair mechanism of the photovoltaic unit, isolates the affected module, and attempts to restore system functions.

[0116] The remote monitoring and notification module records and monitors the development of fire risks, and sends fire risk notifications to relevant departments.

[0117] If the above 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 method 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 and other media that can store program codes.

[0118] 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.

[0119] 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.

[0120] 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.

[0121] In Example 2, a 60-day simulation experiment was conducted in a photovoltaic power generation unit. The total area of ​​the photovoltaic power generation unit is 500 square meters, and three types of sensor networks are deployed: core layer, auxiliary layer and environmental layer. The core layer sensors include temperature sensors, gas sensors and current sensors, which are mainly used to detect key parameters inside the photovoltaic unit; the auxiliary layer sensors include pressure sensors, smoke sensors and humidity sensors, which are used to monitor the intermediate state between the environment and the photovoltaic unit; the environmental layer sensors include light sensors and vibration sensors, which are used to monitor the impact of the external environment on the photovoltaic power generation unit.

[0122] During the test, the photovoltaic power generation unit was operated for 12 hours a day to simulate fire risks under various environmental conditions, such as high temperature, gas leakage, abnormal current fluctuations, external pressure changes, increased smoke concentration, humidity surge, abnormal light reduction and ground vibration. Data under each condition is collected in real time through the sensor network, and fire risk prediction is performed through a data fusion algorithm based on a multi-level sensor network. When the system detects any potential fire risk, it immediately activates built-in firefighting measures, including triggering an alarm, starting the fire extinguishing system, turning off the power supply and notifying the emergency management center through the remote communication module.

[0123] During the specific test, the system not only recorded the data of various sensors, but also analyzed and recorded the parameter changes of each sensor before and after the fire in real time. Each simulated fire started with a slight abnormality and gradually increased the degree of abnormality to detect the system's response time and accuracy to fire risks. The sensitivity and operating frequency of each sensor were dynamically adjusted according to real-time data to ensure the stability and efficiency of the system in various environments. The experimental results are shown in Table 1.

[0124] Table 1 Experimental data table

[0125]

[0126] It is obvious from the above test data that as the environmental parameters gradually change, the fire risk inside the photovoltaic power generation unit gradually increases. Especially in the cases of test 4-environment D, test 5-environment E and test 6-environment F, key parameters such as temperature change, gas concentration, and current fluctuation all exceeded their respective thresholds, indicating that the multi-level sensor network can effectively detect early signs of fire and respond in a short time.

[0127] Through comparative analysis of tabular data, it can be found that the multi-level sensor network proposed in the present invention performs much better than the traditional single sensor system in fire warning. First, through the data fusion of core layer sensors, auxiliary layer sensors and environmental layer sensors, the system can accurately predict complex fire risks; secondly, when the data of each sensor exceeds the threshold, the system can dynamically adjust the sensitivity and operating frequency of each sensor in a timely manner to ensure that the fire signal can be accurately captured in different environments. Especially in Test 6-Environment F, as various parameters fluctuated violently, the system initiated firefighting measures within just a few seconds, avoiding the further spread of the fire. This fully proves the efficiency and stability of the present invention in dealing with sudden fires.

[0128] Compared with the existing technology, the present invention has significant innovation and advantages in fire warning and fire fighting. Traditional technology often relies on data from a single sensor type, which makes it difficult to predict fire risks in a complex environment in a timely and accurate manner; the present invention overcomes this shortcoming by building a multi-level sensor network and applying a data fusion algorithm. At the same time, the dynamic threshold adjustment mechanism adopted by the present invention is combined with the real-time data processing algorithm to further improve the response speed and accuracy of the system, and has extremely high practical application value. Through the verification of this embodiment, it can be seen that the system has a significant effect in improving the safety of photovoltaic power generation units, can effectively make up for the shortcomings of the existing technology, and shows strong adaptability in practical applications.

[0129] 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 fire warning and fire fighting method for a photovoltaic power generation unit, characterized in that: include: Deploy various types of sensors inside and around photovoltaic power generation units to build a multi-level sensor network and collect data; Predict potential fire risks based on data collected by a multi-level sensor network; When a potential fire risk is detected, firefighting measures are initiated immediately.

2. The fire warning and fire fighting method of the photovoltaic power generation unit according to claim 1, characterized in that: Various types of sensors include temperature sensors, smoke sensors, gas sensors, humidity sensors, pressure sensors, light sensors, current sensors, and vibration sensors; The data collected by the multi-level sensor network include real-time temperature values ​​of different parts of the photovoltaic power generation unit, smoke concentration and its changing trend, concentration of specific gases in the air, ambient humidity value, gas pressure inside or around the photovoltaic unit, ambient light intensity and light changes on the surface of the photovoltaic unit, current intensity, vibration frequency and amplitude of the photovoltaic power generation equipment.

3. The fire warning and fire fighting method for photovoltaic power generation unit according to claim 2, characterized in that: The multi-layer sensor network includes a core layer, an auxiliary layer and an environmental layer, which are constructed in order to realize all-round monitoring of early signs of fire in photovoltaic power generation units; The core layer includes a temperature sensor, a gas sensor and a current sensor; The auxiliary layer includes a pressure sensor, a smoke sensor and a humidity sensor; The environmental layer includes a light sensor and a vibration sensor; When sensors at different levels detect data anomalies at the same time and the anomaly value exceeds the set crossover threshold, the sensitivity and operating frequency of the multi-level sensor system are adjusted in real time; The cross threshold includes the cross threshold of the core layer sensor, the cross threshold of the auxiliary layer sensor and the cross threshold setting of the environment layer sensor; The cross threshold of the core layer sensor includes when the temperature sensor detects that the temperature rise exceeds the temperature sensor threshold, the gas sensor simultaneously detects that the combustible gas concentration increases to exceed the gas sensor threshold, and the current sensor simultaneously detects that the current fluctuation exceeds the current sensor threshold; The crossing threshold of the auxiliary layer sensor includes when the pressure sensor detects that the internal pressure changes exceeding the pressure sensor threshold, the smoke sensor simultaneously detects that the smoke concentration exceeds the smoke sensor threshold, and the humidity sensor simultaneously detects that the humidity changes exceeding the humidity sensor threshold; The environmental layer sensor crossing threshold includes, when the light sensor detects that the light intensity change exceeds the light sensor threshold, and the vibration sensor detects that the vibration amplitude exceeds the vibration sensor threshold; The real-time adjustment of the sensitivity and operating frequency of the multi-level sensor system includes performing real-time calculation based on the cross thresholds of sensors at different levels to form a comprehensive index; when the comprehensive index reaches or exceeds the comprehensive threshold, the multi-level sensor network will automatically adjust the sensitivity and operating frequency; According to the fuzzy reasoning results of sensor data, the adjustment factor is determined, and the sensor data is corrected using the adaptive filter algorithm to ensure the stability of the adjustment process; The formula of the comprehensive index is expressed as: When Λ(t)≥Λ th When , the formula for adjusting the sensitivity and operating frequency of the multi-level sensor system is expressed as: Among them, Λ(t) represents the comprehensive index, T(t) represents the real-time temperature data, G(t) represents the real-time gas concentration, I(t) represents the real-time current, P(t) represents the real-time pressure, S(t) represents the real-time smoke concentration, H(t) represents the real-time humidity, L(t) represents the real-time light intensity, V(t) represents the real-time vibration amplitude, S′(t) represents the adjusted sensor sensitivity, f′(t) represents the adjusted operating frequency, α represents the sensitivity coefficient, β represents the operating frequency adjustment coefficient, Λ th represents the comprehensive threshold, T th Represents the threshold of the temperature sensor, G th Indicates the threshold of the gas sensor, I th Represents the threshold of the current sensor, P th Indicates the pressure sensor threshold, S th Indicates the smoke sensor threshold, H th Indicates the humidity sensor threshold, L th Represents the light sensor threshold, V th Indicates the vibration sensor threshold; The value range of Λ(t) is from 0 to positive infinity; The larger the value of Λ(t), the more anomalies the system detects.

4. The fire warning and fire fighting method for a photovoltaic power generation unit as claimed in claim 3, characterized in that: The comprehensive threshold includes a real-time comprehensive threshold and a historical prediction comprehensive threshold; When the real-time comprehensive threshold is inconsistent with the historical prediction comprehensive threshold, and the data collected by the current multi-level sensor network fluctuates more than twice its long-term average, the real-time comprehensive threshold is used first; When the fluctuation of data collected by the multi-level sensor network is less than twice its long-term average, the historical prediction comprehensive threshold is preferred; Collect the data of each sensor in the past 6 months as historical data; Analyze the historical data and calculate the long-term average value and standard deviation of each sensor; The long-term average value includes the average value of the data collected by each sensor in the past 6 months.

5. The fire warning and fire fighting method for photovoltaic power generation unit according to claim 4, characterized in that: The real-time comprehensive threshold value includes using the data of the multi-level sensor network, adopting the weighted comprehensive method, and applying the Kalman filter for smoothing to obtain the real-time comprehensive threshold value, which is expressed as follows: Among them, Λ real (t) represents the real-time comprehensive index calculated at time t, Λ core (t) represents the comprehensive index of the core layer at time t, Λ support (t) represents the comprehensive index of the auxiliary layer at time t, Λ env (t) represents the comprehensive index of the environmental layer at time t, T(t) represents real-time temperature data, G(t) represents real-time gas concentration, I(t) represents real-time current, P(t) represents real-time pressure, S(t) represents real-time smoke concentration, H(t) represents real-time humidity, L(t) represents real-time light intensity, V(t) represents real-time vibration amplitude, T th Represents the threshold of the temperature sensor, G th Indicates the threshold of the gas sensor, I th Represents the threshold of the current sensor, P th Indicates the pressure sensor threshold, S th Indicates the smoke sensor threshold, H th Indicates the humidity sensor threshold, L th Represents the light sensor threshold, V th Indicates the vibration sensor threshold.

6. The fire warning and fire fighting method of the photovoltaic power generation unit according to claim 5, characterized in that: The historical prediction comprehensive threshold includes analyzing the data collected by the multi-level sensor network and its change rate using random forest, predicting the comprehensive risk at the current moment, and obtaining the historical prediction comprehensive threshold, which is expressed as follows: Among them, Λ real (t) represents the real-time comprehensive index calculated at time t, Λ core (t) represents the comprehensive index of the core layer at time t, Λ support (t) represents the comprehensive index of the auxiliary layer at time t, Λ env (t) represents the comprehensive index of the environmental layer at time t, Λ hist (t) represents the comprehensive threshold of historical prediction, RF j represents the output value of the j-th decision tree, N represents the total number of decision trees contained in the random forest, and j represents the index of the j-th decision tree currently being processed.

7. The fire warning and fire fighting method of the photovoltaic power generation unit according to claim 6, characterized in that: When a potential fire risk is detected, the firefighting measures initiated immediately include: When 0.8Λ th ≤Λ(t)<Λ th When the load is exceeded, the real-time monitoring log is started and the operator is notified to gradually reduce the operating load of the photovoltaic power generation unit; When Λ(t)=Λ th When a fire occurs, the fire alarm is immediately sounded, the operator is notified, an alarm log is generated, the current risk parameters are recorded, and the standby status of the fire-fighting equipment is activated; When Λ(t)>Λ th When a fire breaks out, the system immediately initiates a comprehensive emergency response mechanism, automatically isolates the affected PV modules, adjusts the current and load to prevent the fire from spreading, automatically repairs and restores some functions, and sends emergency notifications to designated safety managers and fire departments via the network; The preparation state of the fire-fighting equipment includes turning on the standby mode of the fire-fighting water pump and preparing to start the relevant emergency power supply; The comprehensive emergency response mechanism includes shutting down power, activating fire extinguishing systems, and implementing personnel evacuation plans; During the recovery phase after the fire risk is eliminated, self-checks are performed on sensors at all levels and functional checks are performed on emergency equipment; If any sensor or equipment abnormality is found during the self-check process, the system should immediately notify maintenance personnel to repair it and maintain a high alert state during this period.

8. A fire warning and fire fighting system for a photovoltaic power generation unit using the method according to any one of claims 1 to 7, characterized in that: Multi-level sensor network module, deploys various types of sensors inside and around the photovoltaic power generation unit to build a multi-level sensor network for data collection; Data acquisition and processing module collects real-time data from multi-level sensor networks and processes and analyzes the data; The fire risk prediction module predicts the potential fire risk of the photovoltaic power generation unit based on the processed sensor data; The alarm and emergency response module automatically triggers the alarm system and initiates corresponding emergency response measures when the fire risk reaches or exceeds the threshold; After the self-repair and protection module detects a fire risk, the system automatically starts the self-repair mechanism of the photovoltaic unit, isolates the affected module, and attempts to restore system functions; The remote monitoring and notification module records and monitors the development of fire risks, and sends fire risk notifications to relevant departments.

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 warning and fire fighting method of the photovoltaic power generation unit are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: The computer program, when executed by a processor, implements the steps of a fire warning and fire fighting method for a photovoltaic power generation unit.

Images