Fire early warning method and system for floating photovoltaic power station

Abstract

The invention discloses a fire early warning method and system for a floating photovoltaic power station, and relates to the technical field of fire early warning, and the method comprises the steps: collecting the monitoring parameters of a floating photovoltaic power station system in real time; enabling the collected monitoring data to form a real-time data matrix, carrying out the data fusion, obtaining historical fusion data, and calculating the difference between the real-time data at the current time point and the historical fusion data; calculating a dynamic threshold according to the environment variable and the sensor data, comparing the difference with the dynamic threshold to obtain an early warning judgment result, and judging whether to carry out early warning or not according to the early warning judgment result; and after the fire hazard early warning is received, fire hazard early warning is carried out according to preset emergency response measures. According to the invention, early fire risk detection, accurate alarm triggering and efficient emergency response are realized, and the safety and reliability of the system are improved. The achievement of the invention is supported by national key research and development plan projects, and the project name is offshore floating photovoltaic power generation key technology and core component research and development and engineering demonstration (project number: 2022YFB4200700).

Description

[0001] The results of this invention are supported by the National Key R&D Program Project, and the project name is: R&D and Engineering Demonstration of Key Technologies and Core Components for Offshore Floating Photovoltaic Power Generation (Project No.: 2022YFB4200700). Technical Field

[0002] The present invention relates to the technical field of fire warning, and in particular to a fire warning method and system for a floating photovoltaic power station. Background Art

[0003] With the continuous growth of global energy demand and the improvement of environmental protection awareness, renewable energy, especially photovoltaic power generation technology, has developed rapidly. In recent years, floating photovoltaic power stations, as a new form of photovoltaic power generation, have attracted widespread attention and application due to their advantages such as effective use of water surface space, reduction of land resource occupation, and improvement of power generation efficiency. Floating photovoltaic power stations not only have good heat dissipation effects, but also improve the efficiency of photovoltaic modules through the cooling effect of the water surface. However, with the expansion of the scale and increase in complexity of floating photovoltaic power stations, their operational safety and stability issues have become increasingly prominent. Fire is a major safety hazard in photovoltaic power generation systems. Once it occurs, it may cause serious economic losses and environmental pollution. Therefore, how to effectively warn of fire in the early stage has become a technical problem that needs to be solved urgently for floating photovoltaic power stations.

[0004] At present, the fire warning technology for photovoltaic power generation systems mainly focuses on the monitoring of electrical parameters and environmental parameters. Common methods include monitoring the voltage, current and temperature of photovoltaic modules and inverters, and combining certain preset thresholds to make fire judgments. However, these traditional warning methods have some shortcomings in practical applications. First, the static nature of the preset thresholds leads to poor adaptability of the system when facing different environmental conditions, and it is easy to have false alarms or missed alarms. Secondly, traditional methods mainly rely on a single or a small number of parameters, which cannot fully reflect the operating status of the system and environmental changes, and the accuracy and timeliness of the warning are low. In addition, most of the existing warning systems lack an effective emergency response mechanism. Even if they can issue warnings in time, they cannot take corresponding measures quickly, which limits their practical application effects.

[0005] These deficiencies in the existing technology have made the fire warning of floating photovoltaic power stations face severe challenges. A fire warning method for floating photovoltaic power stations invented by us innovatively proposes real-time collection of monitoring data including historical operating parameters, current operating parameters and current environmental parameters, and forms a real-time data matrix for data fusion. By calculating the difference between the real-time data at the current time point and the historical fusion data, and combining environmental variables and sensor data to calculate the dynamic threshold, a more accurate warning judgment result is obtained. Compared with the existing technology, this method not only improves the accuracy and timeliness of fire warning, but also ensures that effective response measures can be taken quickly after the fire warning through preset emergency response measures to minimize the risk and loss of fire.

[0006] In summary, the existing fire warning technology of floating photovoltaic power stations has problems such as static threshold, single parameter and insufficient emergency response mechanism. The fire warning method of floating photovoltaic power station invented by us proposes innovative solutions to these problems and belongs to the field of photovoltaic power generation system safety monitoring technology. Through the present invention, the operational safety and stability of floating photovoltaic power stations can be significantly improved, which has important application value and promotion prospects. Summary of the invention

[0007] In view of the above-mentioned problems, the present invention is proposed.

[0008] Therefore, in order to solve the above technical problems, the present invention provides the following technical solutions: A fire warning method for a floating photovoltaic power station, comprising: real-time collection of monitoring parameters of a floating photovoltaic power station system, wherein the monitoring parameters include historical operating parameters, current operating parameters and current environmental parameters;

[0009] The collected monitoring data is formed into a real-time data matrix and data fusion is performed to obtain historical fusion data, and the difference between the real-time data at the current time point and the historical fusion data is calculated;

[0010] Calculate a dynamic threshold value according to environmental variables and sensor data, compare the difference with the dynamic threshold value, obtain an early warning determination result, and determine whether to issue an early warning according to the early warning determination result;

[0011] After receiving a fire warning, a fire warning is issued according to the preset emergency response measures.

[0012] As a preferred solution of the fire warning method of the floating photovoltaic power station described in the present invention, wherein: the historical operation parameters include historical DC parameters, historical AC parameters and historical temperature parameters of each inverter;

[0013] The historical DC parameters include historical DC voltage, historical DC current and historical DC power;

[0014] The historical AC parameters include historical AC voltage and historical AC current;

[0015] The historical temperature parameters include historical inverter internal temperature;

[0016] The current operating parameters include the current DC parameters, the current AC parameters and the current temperature parameters of each inverter;

[0017] The current DC parameters include real-time DC voltage, real-time DC current and real-time DC power;

[0018] The current AC parameters include real-time AC voltage and real-time AC current;

[0019] The current temperature parameter includes the real-time inverter internal temperature;

[0020] The current environmental parameters include real-time temperature, real-time humidity and real-time wind speed.

[0021] As a preferred solution of the fire warning method of the floating photovoltaic power station described in the present invention, wherein: the real-time data matrix constituting the real-time data matrix includes:

[0022]

[0023] Among them, V dc (t) is the real-time DC voltage input to the inverter, I dc (t) is the real-time DC current input to the inverter, P dc (t) is the real-time DC power input to the inverter, V ac (t) is the real-time AC voltage output by the inverter, I ac (t) is the real-time AC current output by the inverter, T inv (t) is the real-time temperature inside the inverter, T(t) is the temperature of the environment where the inverter is currently located, H(t) is the humidity of the environment where the inverter is currently located, W(t) is the wind speed of the environment where the inverter is currently located, and S(t) is the real-time data matrix.

[0024] As a preferred solution of the fire warning method of the floating photovoltaic power station described in the present invention, wherein: the data fusion comprises:

[0025] Perform weighted moving average processing on historical data within a period of time and real-time data matrix, calculate the extended weighted moving average of historical data, and obtain historical fusion data. The calculation formula is as follows:

[0026]

[0027] Among them, H avg(t) is the weighted moving average of the extended historical data at time t. N is the size of the moving window. A larger N value means using more historical data and a smoother response; a smaller N value means using less historical data and can better reflect recent changes. i is the weight coefficient at time ti, which is used to adjust the influence of the data at each time point in calculating the weighted moving average;

[0028] The obtaining of the early warning determination result comprises:

[0029] Compare D(t) and θ(t): If D(t)>θ(t), the system triggers a fire warning signal;

[0030] If D(t)≤θ(t), the system does not trigger the warning signal.

[0031] As a preferred solution of the fire warning method of the floating photovoltaic power station described in the present invention, wherein: the difference between the real-time data at the current time point and the historical fusion data is calculated,

[0032] Calculate the standard deviation of the historical fusion data, and calculate the standard deviation normalized difference by calculating the difference between the real-time data at the current time point and the historical data;

[0033]

[0034] Where D(t) is the standard deviation normalized difference, σ S is the standard deviation of the historical fusion data.

[0035] As a preferred solution of the fire warning method of the floating photovoltaic power station described in the present invention, wherein: the calculation of the dynamic threshold value includes:

[0036] The dynamic threshold θ(t) is calculated based on environmental variables and sensor data. The dynamic threshold is composed of the weighted sum of multiple parameters:

[0037]

[0038] Among them, S i (t) is the sensor data, including DC voltage, DC current, DC power, AC voltage, AC current and inverter internal temperature, α0, α1, α2, α3, β i is the weight parameter obtained through training.

[0039] As a preferred solution of the fire warning method of the floating photovoltaic power station described in the present invention, the preset emergency response measures include:

[0040] If the fire sensor detects an increase in temperature or smoke concentration in a certain area, the area isolation device will be immediately activated and a small automatic fire-fighting unmanned boat will be activated. The severity of the fire in the area will be determined, and based on the preset threshold, it will be determined whether other fire-fighting equipment needs to be activated.

[0041] If the fire severity is lower than the first threshold, the basic fire-fighting equipment will be activated to carry out preliminary fire-fighting, and sound and light alarms will be issued to guide on-site personnel to evacuate and save themselves. During the fire-fighting process, the fire situation will be monitored in real time. If there are signs of the fire spreading, the fire severity level will be raised and relevant fire-fighting strategies will be implemented.

[0042] If the severity of the fire is between the first threshold and the second threshold, the water sprinkler system and the small automatic fire-fighting unmanned boat are activated simultaneously for joint fire fighting. The spraying intensity of the water sprinkler system and the fire-fighting path of the unmanned boat are adjusted according to the development of the fire, and it is determined whether it is necessary to cut off part or all of the power supply to prevent the fire from further spreading.

[0043] If the severity of the fire exceeds the second threshold, all power supplies will be cut off immediately and all fire-fighting equipment will be activated for comprehensive fire-fighting. Fire-fighting strategies will be dynamically adjusted based on real-time data, and the safe evacuation of personnel will be prioritized. After the fire incident is over, the system will automatically generate an accident report, recording the time, location, cause and handling process of the fire.

[0044] If equipment failure or poor fire extinguishing results occur during fire handling, the backup emergency plan will be immediately activated, including manual operation of isolation devices and fire extinguishing equipment and summoning on-site personnel for emergency handling; after the accident handling is completed, the equipment will be inspected and maintained in detail to ensure that it can operate normally the next time a fire occurs;

[0045] After the accident report is generated, the system will optimize and improve the preset emergency response measures based on the report content, including adjusting the sensitivity of fire sensors, optimizing the configuration and operation strategy of fire extinguishing equipment and updating backup emergency plans; and regularly checking and maintaining the transformer protection configuration to ensure its reliable operation and prevent fires caused by equipment failures.

[0046] Another object of the present invention is to provide a fire warning system for a floating photovoltaic power station. To solve the above technical problems, the present invention provides the following technical solutions: A fire warning system for a floating photovoltaic power station, comprising: a data acquisition module, a data processing module, a warning determination module and a fire response module;

[0047] The data acquisition module is used to collect monitoring parameters of the floating photovoltaic power station system in real time, and the monitoring parameters include historical operating parameters, current operating parameters and current environmental parameters;

[0048] The data processing module is used to form a real-time data matrix with the collected monitoring data and perform data fusion to obtain historical fusion data, and calculate the difference between the real-time data at the current time point and the historical fusion data;

[0049] The early warning determination module is used to calculate a dynamic threshold value according to environmental variables and sensor data, compare the difference with the dynamic threshold value, obtain an early warning determination result, and determine whether to issue an early warning according to the early warning determination result;

[0050] The fire response module is used to issue a fire warning according to preset emergency response measures after receiving a fire warning.

[0051] A computer device comprises a memory and a processor, wherein the memory stores a computer program, and wherein the processor implements the steps of the fire warning method for a floating photovoltaic power station as described above when executing the computer program.

[0052] 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 fire warning method for a floating photovoltaic power station as described above are implemented.

[0053] Beneficial effects of the present invention: By real-time collection of monitoring parameters of the floating photovoltaic power station system, including historical operating parameters, current operating parameters and current environmental parameters, a comprehensive monitoring foundation is established. The collected data is formed into a real-time data matrix and data fusion is performed to obtain historical fusion data, thereby improving the accuracy of abnormality detection. Dynamic thresholds are calculated based on environmental variables and sensor data, and early warning judgment results are obtained by comparing differences and dynamic thresholds. This step ensures dynamic adjustment of alarm sensitivity, reduces false alarms, and improves the accuracy of response. After receiving a fire warning, a fire warning is performed according to preset emergency response measures, so that the system can quickly take appropriate countermeasures, effectively reduce potential losses caused by fire, and give priority to personnel safety. Through these steps, early detection of fire risks, accurate alarm triggering and efficient emergency response are achieved, thereby achieving the beneficial effect of improving system safety and reliability. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0055] Figure 1 This is a schematic diagram of the structure of a terminal in the first embodiment of the present invention;

[0056] Figure 2 It is a schematic diagram of a flow chart in the first embodiment of the present invention;

[0057] Figure 3 FIG. 4 is a schematic diagram of the structure of a computer device in a third embodiment of the present invention. DETAILED DESCRIPTION

[0058] 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.

[0059] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0060] Example 1

[0061] Reference Figure 1 , as an embodiment of the present invention, provides a fire warning method for a floating photovoltaic power station.

[0062] First, the fire warning method for floating photovoltaic power station provided by the present application can be applied to Figure 1 In the terminal shown. Figure 1 As shown, the terminal may include one or two ( Figure 1 Only one is shown in the figure) processor and memory for storing data, wherein the processor may include but is not limited to a processing system such as a microprocessor MCU or a programmable logic device FPGA. The above terminal may also include a transmission device and an input and output device for communication functions. It can be understood by ordinary technicians in this field that Figure 1 The structure shown is only for illustration and does not limit the structure of the above terminal. Figure 1 More or fewer components as shown, or with Figure 1 Different configurations shown.

[0063] The memory can be used to store computer programs, such as the computer program corresponding to the fire warning method of the floating photovoltaic power station in this embodiment. The processor executes various functional applications and data processing by running the computer program stored in the memory, that is, implementing the above method. The memory may include a high-speed random access memory, and may also include a non-volatile memory, such as one or more magnetic storage systems, flash memory, or other non-volatile solid-state memory. In some examples, the memory may further include a memory remotely arranged relative to the processor, and these remote memories may be connected to the terminal via a network. Examples of the above network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0064] The transmission device is used to receive or send data via a network. The above-mentioned network includes a wireless network provided by a communication provider of the terminal. In one example, the transmission device includes a network adapter (Network Interface Controller, referred to as NIC), which can be connected to other network devices through a base station so as to communicate with the Internet. In one example, the transmission device can be a radio frequency (Radio Frequency, referred to as RF) module, which is used to communicate with the Internet wirelessly.

[0065] Before describing the embodiments of the present application in detail, some related concepts are first explained for the sake of clarity.

[0066] Floating photovoltaic power station: Floating photovoltaic power station is a new type of power generation method that installs photovoltaic cell modules on a floating platform on the surface of water. Compared with traditional land-based photovoltaic power stations, floating photovoltaic power stations have many advantages. First, it can save land resources, especially suitable for areas with tight land resources. Secondly, the surface temperature of the water body is low, which helps to improve the power generation efficiency of photovoltaic modules. In addition, floating photovoltaic power stations can also reduce water evaporation and protect water resources. This technology is suitable for a variety of water environments such as reservoirs, lakes, and aquatic aquaculture areas. It can effectively utilize the water surface and has little impact on the ecological environment. It has broad development prospects.

[0067] Support Vector Machine: Support vector machine (SVM) is a supervised learning model that is widely used in classification and regression analysis. Its basic principle is to find an optimal hyperplane to separate different categories in the data set with the largest interval. Support vector machines can handle linear inseparable problems in high-dimensional space by using kernel functions (such as linear kernels, polynomial kernels, Gaussian kernels, etc.), thereby improving the classification ability of the model. The advantage of SVM lies in its efficiency and accuracy, especially when dealing with small samples and high-dimensional data. However, choosing a suitable kernel function and adjusting parameters are key steps in its application. Support vector machines are widely used in text classification, image recognition, bioinformatics and other fields, and are one of the important tools in machine learning.

[0068] Standard Deviation Normalized Difference: Standard Deviation Normalized Difference (SDND) is a statistical method used to compare the difference between two sets of data. It first calculates the standard deviation of the two sets of data, and then normalizes the difference between them using the standard deviation as a benchmark. Specifically, SDND quantifies the differences between the two sets of data by calculating the difference in the standard deviations, and then divides this difference value by the average of the standard deviations of the two sets of data to eliminate the scale effect, making the comparison more fair and accurate. SDND is often used to analyze experimental data, quality control, and data comparison in engineering fields. It not only takes into account the absolute difference in the data, but also the variability of each set of data, so it can more objectively evaluate the statistically significant difference between the two sets of data. The application of SDND helps to understand data changes and differences in scientific research and engineering analysis, and is an effective means of data comparison and analysis.

[0069] like Figure 2 As shown, the embodiment of the present invention provides a fire warning method for a floating photovoltaic power station, and the method is applied to Figure 1 The terminal in is used as an example to illustrate, including the following steps:

[0070] S1: real-time collection of monitoring parameters of the floating photovoltaic power station system, wherein the monitoring parameters include historical operating parameters, current operating parameters and current environmental parameters;

[0071] Furthermore, the historical operating parameters include historical DC parameters, historical AC parameters and historical temperature parameters of each inverter; the historical DC parameters include historical DC voltage, historical DC current and historical DC power; the historical AC parameters include historical AC voltage and historical AC current; and the historical temperature parameters include historical inverter internal temperature.

[0072] Furthermore, the current operating parameters include the current DC parameters, current AC parameters and current temperature parameters of each inverter; the current DC parameters include real-time DC voltage, real-time DC current and real-time DC power; the current AC parameters include real-time AC voltage and real-time AC current; the current temperature parameters include real-time inverter internal temperature.

[0073] Furthermore, the current environmental parameters include real-time temperature, real-time humidity and real-time wind speed.

[0074] S2: The collected monitoring data is formed into a real-time data matrix and data fusion is performed to obtain historical fusion data, and the difference between the real-time data at the current time point and the historical fusion data is calculated;

[0075] Furthermore, the real-time data matrix includes:

[0076]

[0077] Among them, V dc (t) is the real-time DC voltage input to the inverter, I dc (t) is the real-time DC current input to the inverter, P dc (t) is the real-time DC power input to the inverter, V ac (t) is the real-time AC voltage output by the inverter, I ac (t) is the real-time AC current output by the inverter, T inv (t) is the real-time temperature inside the inverter, T(t) is the temperature of the environment where the inverter is currently located, H(t) is the humidity of the environment where the inverter is currently located, W(t) is the wind speed of the environment where the inverter is currently located, and S(t) is the real-time data matrix;

[0078] Furthermore, data fusion includes:

[0079] Perform weighted moving average processing on historical data within a period of time and real-time data matrix, calculate the extended weighted moving average of historical data, and obtain historical fusion data. The calculation formula is as follows:

[0080]

[0081] Among them, H avg (t) is the weighted moving average of the extended historical data at time t. N is the size of the moving window. A larger N value means using more historical data and a smoother response; a smaller N value means using less historical data and can better reflect recent changes. i is the weight coefficient at time ti, which is used to adjust the influence of the data at each time point in calculating the weighted moving average;

[0082] In one possible embodiment,

[0083] The weights may be set as follows:

[0084] Equal weight: All time points have the same weight, that is,

[0085] Linear weights: weights decrease over time, usually set to

[0086] Exponential weight: The weight decreases exponentially, usually set to w i =α(1-α) i , where α is a constant between 0 and 1.

[0087] Further, calculating the difference between the real-time data at the current time point and the historical fusion data includes:

[0088] Calculate the standard deviation of the historical fusion data, and calculate the standard deviation normalized difference by calculating the difference between the real-time data at the current time point and the historical data;

[0089]

[0090] Where D(t) is the standard deviation normalized difference, σ S is the standard deviation of the historical fusion data.

[0091] S3: Calculate a dynamic threshold value according to the environmental variables and the sensor data, compare the normalized difference of the standard deviation with the dynamic threshold value, obtain an early warning determination result, and determine whether to issue an early warning according to the early warning determination result;

[0092] Furthermore, the calculation of dynamic thresholds includes,

[0093] The dynamic threshold θ(t) is calculated based on environmental variables and sensor data. The dynamic threshold is composed of the weighted sum of multiple parameters:

[0094]

[0095] Among them, S i (t) is the sensor data, including DC voltage, DC current, DC power, AC voltage, AC current and inverter internal temperature, α0, α1, α2, α3, β i is the weight parameter obtained through training;

[0096] Furthermore, in a possible embodiment, the corresponding weights α0, α1, α2, α3, β i The parameter training method is implemented through the following steps:

[0097] Linear regression: Fit historical data using the least squares method to calculate the unknown parameters to minimize the error between the predicted value and the actual value.

[0098] Support vector machine: By training the model to maximize the distance from the data point to the segmentation hyperplane, the unknown parameters are obtained to achieve data classification and regression.

[0099] Neural network: The weights and biases are continuously adjusted through the back-propagation algorithm to minimize the loss function, so as to obtain the unknown parameters through training and realize complex pattern recognition.

[0100] Heuristic method: In some cases, if there is some prior knowledge about the model, the initial weights can be set directly or based on experience, and then fine-tuned based on the actual prediction results.

[0101] It should be noted that the value range of the dynamic threshold θ(t) in the present application may vary according to the selection of the actual parameters α and β, but is usually a positive interval on R;

[0102] Furthermore, when the ambient temperature T(t) increases, the threshold θ(t) may decrease, making the model more sensitive to fire warnings; when the humidity H(t) increases, the threshold θ(t) may increase, making the model less likely to make false alarms in high humidity environments;

[0103] Through this dynamic adjustment, the early warning system can optimize the warning threshold according to real-time environmental conditions, ensuring effective detection of fire risks in various environments.

[0104] Furthermore, the early warning determination results include:

[0105] Compare D(t) and θ(t):

[0106] If D(t)>θ(t), the system triggers a fire warning signal.

[0107] If D(t)≤θ(t), the system does not trigger the warning signal.

[0108] Furthermore, automatic fire alarms are installed around the floating photovoltaic substation. Once an abnormal situation is detected, the system will immediately trigger an alarm signal. Fire-fighting facilities such as small automatic fire-fighting unmanned boats, hand-push fire extinguishers and fire sand boxes are set up near the photovoltaic field and inverter to facilitate rapid fire extinguishing;

[0109] This method improves the accuracy and timeliness of early warning, enabling the early warning system to respond quickly in the early stages of a fire, thereby more effectively preventing the spread of fire.

[0110] S4: After receiving the fire warning, the system issues a fire warning according to the preset emergency response measures.

[0111] Furthermore, the pre-set emergency response measures include:

[0112] If the fire sensor detects an increase in temperature or smoke concentration in a certain area, the area isolation device will be immediately activated and a small automatic fire-fighting unmanned boat will be activated. At this time, the severity of the fire in the area will be judged and whether other fire-fighting equipment needs to be activated will be determined based on the preset threshold.

[0113] If the severity of the fire is lower than the first threshold, basic fire-fighting equipment, such as dry powder fire extinguishers, will be activated for preliminary fire-fighting, and sound and light alarms will be issued to guide on-site personnel to evacuate and self-rescue. During the fire-fighting process, the fire situation will be monitored in real time. If there are signs of the fire spreading, the fire severity level will be raised and relevant fire-fighting strategies will be implemented.

[0114] If the severity of the fire is between the first threshold and the second threshold, the water sprinkler fire extinguishing system and the small automatic fire-fighting unmanned boat will be activated at the same time for joint fire extinguishing. The spraying intensity of the water sprinkler fire extinguishing system and the fire-fighting path of the unmanned boat will be adjusted according to the development of the fire, and it will be determined whether it is necessary to cut off part or all of the power supply to prevent the fire from further expanding.

[0115] If the severity of the fire exceeds the second threshold, all power supplies will be cut off immediately and all fire-fighting equipment will be activated for comprehensive fire-fighting. During this process, the system will dynamically adjust the fire-fighting strategy based on real-time data and give priority to ensuring the safe evacuation of personnel. After the fire incident is over, the system will automatically generate an accident report, recording the time, location, cause and handling process of the fire.

[0116] If equipment failure occurs or the fire extinguishing effect is poor during the fire handling process, the backup emergency plan shall be immediately activated, including manual operation of isolation devices and fire extinguishing equipment, summoning on-site personnel for emergency handling, etc.; after the accident handling is completed, relevant personnel need to conduct detailed inspection and maintenance of the equipment to ensure that it can operate normally when the next fire occurs.

[0117] After the accident report is generated, the system will optimize and improve the preset emergency response measures based on the report content, including adjusting the sensitivity of fire sensors, optimizing the configuration and operation strategy of fire extinguishing equipment, updating backup emergency plans, etc.; relevant personnel need to regularly check and maintain the transformer protection configuration to ensure its reliable operation and prevent fires caused by equipment failure.

[0118] It should be noted that the first threshold and the second threshold in this application are determined by analyzing historical fire data to determine the typical temperature and smoke concentration ranges under different fire severity levels. These data are combined with the sensitivity and response time of the current fire sensor to preliminarily set the first threshold and the second threshold.

[0119] Example 2

[0120] According to an embodiment of the present invention, a fire warning system for a floating photovoltaic power station is provided, comprising: a data acquisition module, a data processing module, a warning determination module and a fire response module;

[0121] The data acquisition module is used to collect monitoring parameters of the floating photovoltaic power station system in real time, wherein the monitoring parameters include historical operating parameters, current operating parameters and current environmental parameters;

[0122] The data processing module is used to form a real-time data matrix with the collected monitoring data and perform data fusion to obtain historical fusion data, and calculate the difference between the real-time data at the current time point and the historical fusion data;

[0123] The early warning determination module is used to calculate the dynamic threshold value according to the environmental variables and the sensor data, compare the difference with the dynamic threshold value, obtain the early warning determination result, and determine whether to issue an early warning according to the early warning determination result;

[0124] The fire response module is used to issue a fire warning according to preset emergency response measures after receiving a fire warning.

[0125] The specific definition of the fire warning system of the floating photovoltaic power station can be found in the definition of the fire warning method of the floating photovoltaic power station mentioned above, which will not be repeated here. Each module in the fire warning system of the floating photovoltaic power station can be implemented in whole or in part by software, hardware and their combination. The above modules can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory of the computer device in the form of software, so that the processor can call and execute the operations corresponding to the above modules.

[0126] Example 3

[0127] Reference Figure 3 , which is the third embodiment of the present invention. Based on the first two embodiments, the embodiment of the present invention provides a computer device, which may be a server, and its internal structure diagram may be as shown in FIG. Figure 3 As shown. The computer device includes a processor, a memory and a network interface connected via a system bus. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium.

[0128] The database of the computer device is used to store action detection data. The network interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, the steps in any of the above-mentioned sparse tensor operation acceleration method embodiments are implemented.

[0129] Those skilled in the art will understand that Figure 3 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0130] In one embodiment, an embodiment of the present invention provides a computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, the steps in any of the above-mentioned sparse tensor operation acceleration method embodiments are implemented.

[0131] Those of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory may include read-only memory (ROM), magnetic tape, floppy disk, flash memory or optical memory, etc. Volatile memory may include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM).

[0132] Example 4

[0133] As an embodiment of the present invention, a fire warning method for a floating photovoltaic power station is provided. In order to verify the beneficial effects of the present invention, scientific demonstration is carried out through economic benefit calculation and simulation experiments.

[0134] Experimental preparation and detailed procedures

[0135] In this example, we simulated a floating photovoltaic power plant and implemented a fire warning method that includes real-time monitoring of system parameters and dynamic threshold calculation for early fire detection.

[0136] Experimental setup: We set up a simulation environment to simulate the operation of a floating PV power plant. The power plant includes multiple inverters, each of which monitors parameters such as DC voltage, DC current, DC power, AC voltage, AC current, inverter internal temperature, and ambient temperature, humidity, and wind speed.

[0137] The detailed process includes:

[0138] Data collection: The real-time monitoring system collects historical operating data (such as DC parameters, AC parameters, inverter temperature) and current environmental data (temperature, humidity, wind speed).

[0139] Data integration: Integrate the collected data into a real-time data matrix and calculate historical fusion data based on historical operating trends.

[0140] Threshold calculation: Dynamic thresholds are calculated based on environmental variables and sensor data to detect whether there are abnormal deviations from the historical normal range.

[0141] Fire warning system: When abnormal conditions exceeding the set threshold are detected, the system will trigger a fire warning signal. After the warning is triggered, the preset emergency response measures will be initiated, including isolation procedures and the deployment of automatic fire-fighting drones.

[0142] Experimental data table

[0143]

[0144]

[0145] The table above shows typical parameters of floating PV systems under normal operating conditions. Analysis shows that any deviation of parameters beyond dynamic thresholds (calculated based on historical data and environmental factors) may indicate a potential fire risk. This approach enhances safety through early detection capabilities and effectively reduces the risks faced by floating PV installations.

[0146] This example not only demonstrates the effectiveness of the fire warning method, but also highlights its innovation and technical advantages in preventing safety issues of floating PV systems.

[0147] 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 method for a floating photovoltaic power station, It is characterized in that include: Real-time collection of monitoring parameters of the floating photovoltaic power station system, wherein the monitoring parameters include historical operating parameters, current operating parameters and current environmental parameters; The collected monitoring data is formed into a real-time data matrix and data fusion is performed to obtain historical fusion data, and the difference between the real-time data at the current time point and the historical fusion data is calculated; Calculate a dynamic threshold value according to environmental variables and sensor data, compare the difference with the dynamic threshold value, obtain an early warning determination result, and determine whether to issue an early warning according to the early warning determination result; After receiving a fire warning, a fire warning is issued according to the preset emergency response measures.

2. The fire warning method for a floating photovoltaic power station according to claim 1, Features: The historical operating parameters include historical DC parameters, historical AC parameters and historical temperature parameters of each inverter; The historical DC parameters include historical DC voltage, historical DC current and historical DC power; The historical AC parameters include historical AC voltage and historical AC current; The historical temperature parameters include historical inverter internal temperature; The current operating parameters include the current DC parameters, the current AC parameters and the current temperature parameters of each inverter; The current DC parameters include real-time DC voltage, real-time DC current and real-time DC power; The current AC parameters include real-time AC voltage and real-time AC current; The current temperature parameter includes the real-time inverter internal temperature; The current environmental parameters include real-time temperature, real-time humidity and real-time wind speed.

3. The fire warning method for a floating photovoltaic power station according to claim 2, Features: The real-time data matrix constituting the real-time data matrix includes: Among them, V dc (t) is the real-time DC voltage input to the inverter, I dc (t) is the real-time DC current input to the inverter, P dc (t) is the real-time DC power input to the inverter, V ac (t) is the real-time AC voltage output by the inverter, I ac (t) is the real-time AC current output by the inverter, T inv (t) is the real-time temperature inside the inverter, T(t) is the temperature of the environment where the inverter is currently located, H(t) is the humidity of the environment where the inverter is currently located, W(t) is the wind speed of the environment where the inverter is currently located, and S(t) is the real-time data matrix.

4. The fire warning method for a floating photovoltaic power station according to claim 3, Features: The data fusion comprises: Perform weighted moving average processing on historical data within a period of time and real-time data matrix, calculate the extended weighted moving average of historical data, and obtain historical fusion data. The calculation formula is as follows: Among them, H avg (t) is the weighted moving average of the extended historical data at time t. N is the size of the moving window. A larger N value means using more historical data and a smoother response; a smaller N value means using less historical data and can better reflect recent changes. i is the weight coefficient at time ti, which is used to adjust the influence of the data at each time point in calculating the weighted moving average; The obtaining of the early warning determination result comprises: Compare D(t) and θ(t): If D(t)>θ(t), the system triggers a fire warning signal; If D(t)≤θ(t), the system does not trigger the warning signal.

5. The fire warning method for a floating photovoltaic power station according to claim 4, Features: The difference between the real-time data at the current time point and the historical fusion data is calculated as follows: Calculate the standard deviation of the historical fusion data, and calculate the standard deviation normalized difference by calculating the difference between the real-time data at the current time point and the historical data; Where D(t) is the standard deviation normalized difference, σ S is the standard deviation of the historical fusion data.

6. The fire warning method for a floating photovoltaic power station according to claim 5, Features: The calculation of the dynamic threshold comprises: The dynamic threshold θ(t) is calculated based on environmental variables and sensor data. The dynamic threshold is composed of the weighted sum of multiple parameters: Among them, S i (t) is the sensor data, including DC voltage, DC current, DC power, AC voltage, AC current and inverter internal temperature, α 0 , α 1 , α 2 , α 3 , β i is the weight parameter obtained through training.

7. The fire warning method for a floating photovoltaic power station according to claim 6, Features: The preset emergency response measures include: If the fire sensor detects an increase in temperature or smoke concentration in a certain area, the area isolation device will be immediately activated and a small automatic fire-fighting unmanned boat will be activated. The severity of the fire in the area will be determined, and based on the preset threshold, it will be determined whether other fire-fighting equipment needs to be activated. If the fire severity is lower than the first threshold, the basic fire-fighting equipment will be activated to carry out preliminary fire-fighting, and sound and light alarms will be issued to guide on-site personnel to evacuate and save themselves. During the fire-fighting process, the fire situation will be monitored in real time. If there are signs of the fire spreading, the fire severity level will be raised and relevant fire-fighting strategies will be implemented. If the severity of the fire is between the first threshold and the second threshold, the water sprinkler system and the small automatic fire-fighting unmanned boat are activated simultaneously for joint fire fighting. The spraying intensity of the water sprinkler system and the fire-fighting path of the unmanned boat are adjusted according to the development of the fire, and it is determined whether it is necessary to cut off part or all of the power supply to prevent the fire from further spreading. If the severity of the fire exceeds the second threshold, all power supplies will be cut off immediately and all fire-fighting equipment will be activated for comprehensive fire-fighting. Fire-fighting strategies will be dynamically adjusted based on real-time data, and the safe evacuation of personnel will be prioritized. After the fire incident is over, the system will automatically generate an accident report, recording the time, location, cause and handling process of the fire. If equipment failure or poor fire extinguishing results occur during fire handling, the backup emergency plan will be immediately activated, including manual operation of isolation devices and fire extinguishing equipment and summoning on-site personnel for emergency handling; after the accident handling is completed, the equipment will be inspected and maintained in detail to ensure that it can operate normally the next time a fire occurs; After the accident report is generated, the system will optimize and improve the preset emergency response measures according to the report content, including adjusting the sensitivity of fire sensors, optimizing the configuration and operation strategy of fire extinguishing equipment and updating backup emergency plans; The transformer protection configuration must also be checked and maintained regularly to ensure its reliable operation and prevent fires caused by equipment failure.

8. A system using the fire warning method of a floating photovoltaic power station as claimed in any one of claims 1 to 7, It is characterized in that include: Data acquisition module, data processing module, early warning determination module and fire response module; The data acquisition module is used to collect monitoring parameters of the floating photovoltaic power station system in real time, and the monitoring parameters include historical operating parameters, current operating parameters and current environmental parameters; The data processing module is used to form a real-time data matrix with the collected monitoring data and perform data fusion to obtain historical fusion data, and calculate the difference between the real-time data at the current time point and the historical fusion data; The early warning determination module is used to calculate a dynamic threshold value according to environmental variables and sensor data, compare the difference with the dynamic threshold value, obtain an early warning determination result, and determine whether to issue an early warning according to the early warning determination result; The fire response module is used to issue a fire warning according to preset emergency response measures after receiving a fire warning.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program. It is characterized in that When the processor executes the computer program, the steps of the fire warning method for a floating photovoltaic power station according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, It is characterized in that When the computer program is executed by a processor, the steps of the fire warning method for a floating photovoltaic power station according to any one of claims 1 to 7 are implemented.

Images