Method and device for forest fire early detection

Abstract

The invention relates to a method for detecting a forest fire, comprising the method steps of detecting a first sensor signal by means of a first sensor for detecting and / or analyzing a gas, detecting a second sensor signal, the second sensor signal being different from the first sensor signal, and analyzing the first sensor signal, the analysis of the first sensor signal being dependent on the second sensor signal. The invention also relates to a device for the early detection of a forest fire.

Description

[0001] METHOD AND DEVICE FOR EARLY DETECTION OF A FOREST FIRE

[0002] The invention describes a method for detecting a forest fire with the method steps of detecting a first sensor signal with a first sensor for detecting and / or analyzing a gas, detecting a second sensor signal, wherein the second sensor signal is different from the first sensor signal, analyzing the first sensor signal, wherein the analysis of the first sensor signal is dependent on the second sensor signal, and a device for the early detection of a forest fire.

[0003] State of the art

[0004] The methods approved so far by the authorities of the European Union (EU) and the American Food and Drug Administration (FDA) for determining SPF are all harmful to the volunteers involved by causing erythema, i.e.

[0005] The larger a forest fire, the more difficult it is to determine its direction and speed of spread. Weather, wind, soil conditions, and vegetation determine its path and speed of spread, which can change within a short period of time. It is therefore very important to detect a forest fire very early in order to minimize damage and keep the fire manageable, as well as to give the fire department a decisive time advantage.

[0006] During a forest fire, the complex thermal decomposition processes (distillation, pyrolysis, charring, and the oxidation of the resulting gas products during flame combustion) occur simultaneously and often in close proximity to one another. The thermal decomposition of fuels occurs in front of and along the fire line, while enclaves of intermittent open flame often persist far behind the flame front. Flame combustion generally occurs between 800°C - 1200°C. Smoldering ground fires occur between 300°C - 600°C. Combustible gases, particularly volatile organic compounds (VOCs), are formed more rapidly at temperatures above 200°C, reaching their peak at 320°C. VOC is the collective term for organic, carbon-containing substances that evaporate into the gas phase at room temperature or higher temperatures, especially terpenes. In addition, various organic compounds, e.g.Methanol, carbon dioxide, carbon monoxide, and molecular hydrogen are formed. Flaming combustion only begins at 425°C to 480°C. Flame temperatures of 700°C to 1300°C are most common. In this temperature range, carbon dioxide, nitrogen oxides, and volatile sulfur-containing compounds (VSCs), especially sulfur dioxide, are primarily formed. Smoldering fires spread slowly, approximately 3 cm / h, and can generate ground temperatures above 300°C for several hours, with peak temperatures of 600°C.

[0007] Earth observation data, particularly in the form of aerial and satellite imagery, can potentially help in detecting wildfires. The sharp increase in available earth observation data, particularly aerial and satellite imagery, enables widespread detection of wildfires. While satellite data are useful for detecting and fighting fires, they have one drawback: They usually reach emergency responders only with a delay. Geostationary satellites, due to their great distance, provide only low image resolutions, and non-geostationary satellites must orbit the Earth before they can provide new images.

[0008] Another option for detecting forest fires is to install a network of gas sensors directly in the forest. These sensors detect gases released during the onset of forest fires, thus enabling early detection of forest fires before they can be detected from a distance by optical systems. However, due to the varying vegetation in forests and the different soil composition, different gases and gas concentrations are produced, making error-free detection very difficult. Furthermore, the increasing temperatures during the different phases of forest fire development alone result in different gases and gas concentrations. The object of the present invention is therefore to provide a method for early detection of forest fires that has increased detection accuracy, is infinitely expandable, and is cost-effective to install and maintain.

[0009] It is also an object of the invention to provide a forest fire early detection system which has increased detection accuracy, is expandable as required and is cost-effective to install and maintain.

[0010] Description of the invention

[0011] This object is achieved by means of the method according to the invention for detecting a forest fire. Advantageous embodiments of the invention are also set forth in the subclaims.

[0012] The method according to the invention for detecting a forest fire comprises three method steps: In the first method step, a first sensor signal is acquired using a first sensor for detecting and / or analyzing a gas. The first sensor can be, for example, an optical sensor, gas sensor, and / or particle sensor.

[0013] In the second method step, a second sensor signal is detected, which is different from the first sensor signal. The second sensor can be, for example, a temperature sensor, a pressure sensor, and / or a humidity sensor.

[0014] In the third step, the first sensor signal is analyzed, whereby the analysis of the first sensor signal depends on the second sensor signal. The analysis of the first sensor signal is performed in such a way that the first sensor signal is corrected for factors such as air temperature, air pressure, and humidity. In addition, the baseline and long-term deviations of the measured values ​​are compensated. This analysis can be used to make predictions about the occurrence of a forest fire.

[0015] In a further development of the invention, the first sensor signal is used for gas detection by means of a resistance measurement. The gas or gas mixture to be measured influences the conductivity of a gas-sensitive sensor layer. This change in resistance serves as the measured variable. The gas sensor is, for example, an inorganic metal oxide semiconductor (MOX), organic phthalocyanine, or a conductive polymer.

[0016] In a further embodiment of the invention, the second sensor signal alone is unsuitable for gas analysis. The second sensor signal is used to detect environmental parameters such as air pressure, air temperature, and air humidity, which are then used to correct the first sensor signal for analysis.

[0017] In a further embodiment of the invention, an environmental parameter different from the gas composition of the environment is analyzed from the second sensor signal. The second sensor signal is used to detect, in particular, environmental parameters such as air pressure, air temperature, and air humidity, which are used to correct the first sensor signal during its analysis.

[0018] In a further aspect of the invention, the humidity and / or temperature are determined from the second sensor signal. These parameters influence gas detection using the first sensor signal to such an extent that the first sensor signal is corrected using the second sensor signal for analysis.

[0019] In a further embodiment of the invention, the second sensor signal is analyzed before the first sensor signal is analyzed. To correct the first sensor signal using the second sensor signal, it is necessary to analyze the second sensor signal and determine the environmental parameters, such as air pressure and air temperature, used to analyze the first sensor signal.

[0020] In a further embodiment of the invention, the first sensor signal is analyzed using comparison values. During the analysis, the type and concentration of the detected gas are compared with a comparison value. The comparison values ​​can be defined and stored for each gas detected by the first sensor signal. In a further development of the invention, the comparison values ​​are stored in a comparison matrix. The comparison matrix contains a comparison value for each value of each environmental parameter.

[0021] In a further embodiment of the invention, the comparison values ​​used for analysis are selected based on the second sensor signal and / or the environmental parameter determined from the second sensor signal. The first signal is then analyzed using the comparison value used.

[0022] In a further embodiment of the invention, the comparison values ​​are determined for different second sensor signals and / or different values ​​of the environmental parameter determined from the second sensor signal. The comparison values ​​are determined and stored under laboratory conditions and / or realistic conditions, e.g., in the forest to be monitored.

[0023] In a further aspect of the invention, the comparison values ​​are determined for a specific region. Depending on the extent of the forest to be monitored, the comparison values ​​for different regions of the forest to be monitored also vary. Using different comparison values ​​for different regions, a regionally spatially resolved correction of the baseline and long-term deviations can be performed. This also takes regional differences, e.g., in gas composition and specific environmental conditions, into account when analyzing the first sensor signal.

[0024] In a further embodiment of the invention, the comparison values ​​comprise tp comparison values ​​and / or tn comparison values. For the purposes of this document, tp comparison values ​​(true positive comparison values) are comparison values ​​recorded from a forest experiencing a forest fire. Such comparison values ​​include, for example, temperature, humidity, wind direction and strength, and / or the composition of gases. The tp measurement data (true positive measurement data) recorded with these tp comparison values ​​are accordingly measurement data measured from a forest that also experiences a forest fire. The recorded tp measurement data therefore represent measurement data from a forest that experiences a forest fire. By recording tp comparison values, the baseline and long-term deviations of the first sensor signals are compensated. In addition, design-related deviations of the sensors are recorded, averaged, and compensated.

[0025] For the purposes of this document, tn comparison values ​​(true negative comparison values) are comparison values ​​recorded from a forest that has not experienced a forest fire. Such comparison values ​​include, for example, temperature, humidity, wind direction and strength, and / or the composition of gases. The tn measurement data recorded with these tn comparison values ​​(true negative measurement data) are accordingly measurement data measured from a forest that has also not experienced a forest fire. The recorded tn measurement data therefore represent measurement data from a forest that has not experienced a forest fire. By recording tn comparison values, the baseline and long-term deviations of the initial sensor signals are compensated. In addition, design-related deviations of the sensors are recorded, averaged, and compensated.

[0026] In a further embodiment of the invention, the comparison values ​​for analyzing the first sensor signal are selected using an AI / ML function. The AI / ML function is used in this invention to improve the efficiency of the sensor device of the terminal device. The AI / ML function enables improved application-specific recording of the gases to be detected. In particular, the sensitivity of a forest fire early detection system is increased. In addition, the accuracy is increased, i.e. false positives are reduced or eliminated. Forest fire detection based on threshold values ​​is error-prone and generates many false positives. With machine learning models, especially models created on the basis of locally acquired data, the accuracy can be increased enormously, as the model adapts to reality. The AI / ML function enables improved application-specific recording of recorded measured values.The algorithm also corrects the recorded measured values ​​with regard to air humidity. Additionally, the baseline and long-term deviations of the measured values ​​are compensated. By evaluating this data, statements can be made about the current situation during forest fires. In a further embodiment of the invention, the AI / ML function uses the values ​​of the second sensor signal to analyze the first sensor signal. In a further embodiment of the invention, the AI / ML function uses the values ​​of the second sensor signal to analyze the first sensor signal. With the help of the AI / ML function, a forest fire early warning system is enabled to recognize patterns and regularities based on existing data sets and algorithms. The findings can be generalized and used for new problem solutions or for the analysis of previously unknown data. Reinforcement learning is preferred for this purpose.Through reward and punishment, the algorithm learns a tactic for how to act in potentially occurring situations to maximize the system's utility. The AI / ML function interacts with the environment and is evaluated by a cost function or reward system. With reinforcement learning, the AI / ML function is not shown which action is the right one in which situation, but rather receives positive or negative feedback through the cost function. Based on the cost function, it then assesses which action is the right one at a given time. Thus, the system learns to maximize the reward function through "reinforcement" through praise or punishment.

[0027] In a further embodiment of the invention, the comparison values ​​are selected based on the analysis of the second sensor signal. The comparison values ​​used for analysis are selected based on the ambient parameters (humidity and air temperature) analyzed and determined from the second sensor signal. Furthermore, the gas composition detected by the first sensor signal is corrected using a correction factor depending on the temperature and humidity of the ambient air.

[0028] In a further embodiment of the invention, the comparison values ​​are selected based on the analysis of the second sensor signal. The comparison values ​​used for analysis are selected based on the ambient parameters (air humidity and air temperature) analyzed and determined from the second sensor signal. Furthermore, the gas composition detected by the first sensor signal is corrected using a correction factor depending on the temperature and humidity of the ambient air. This object is also achieved by means of a device for the early detection of a forest fire. Advantageous embodiments are set forth in the following subclaims.

[0029] The device according to the invention for the early detection of a forest fire comprises a first sensor, which is intended and suitable for detecting a first sensor signal. The first sensor is preferably a gas sensor for gas detection and, optionally, for detecting the concentration of one or, optionally, a plurality of gases.

[0030] The device according to the invention further comprises a second sensor, which is intended and suitable for detecting a second sensor signal, wherein the first sensor is different from the second sensor. The second sensor serves, for example, to detect environmental parameters, such as air pressure and air temperature.

[0031] The device also includes an analysis unit designed and adapted to analyze the first sensor signal using the second sensor signal and / or an environmental parameter determined from the second sensor signal. The analysis unit includes a microcontroller, a memory, and a suitable software program for analyzing sensor signals.

[0032] In a further development of the invention, the second sensor is a humidity sensor and / or a temperature sensor with which environmental parameters can be detected.

[0033] In a further aspect of the invention, the first sensor is a gas sensor. The gas sensor is, for example, an inorganic metal oxide semiconductor (MOX), organic phthalocyanine, or a conductive polymer. In addition to heavy smoke, a forest fire produces a variety of gases, in particular carbon dioxide and carbon monoxide. The temperature of the gases is also recorded. In addition to the type and concentration of the gases produced during a forest fire, their temperature is an indicator of a forest fire. The type and concentration of these gases are characteristic of a forest fire and can be detected using a suitable first sensor. In a further embodiment of the invention, the analysis unit comprises a memory in which a comparison matrix and / or comparison values ​​are stored, which include different values ​​of the second sensor signals and / or different values ​​of the environmental parameter values ​​determined from the second sensor signals.In a further development of the invention, the comparison matrix comprises comparison values. For analysis purposes, the type and concentration of the detected gas can be compared and corrected using a comparison value. The comparison values ​​can be defined and stored for each gas detected by the first sensor. A comparison value is stored in the comparison matrix for each value of each environmental parameter.

[0034] In a further embodiment of the invention, the comparison values ​​include comparison values ​​for different regions. Depending on the extent of the forest to be monitored, the comparison values ​​for different regions of the forest to be monitored also vary. Using different comparison values ​​for different regions, a regionally spatially resolved correction of the baseline and long-term deviations can be performed. This also takes regional differences, e.g., in gas composition and specific environmental conditions, into account when analyzing the first sensor signal.

[0035] In a further embodiment of the invention, the comparison values ​​comprise tp comparison values ​​and / or tn comparison values. In a further embodiment of the invention, the comparison values ​​comprise tp comparison values ​​and / or tn comparison values. For the purposes of this document, tp comparison values ​​(true positive comparison values) are comparison values ​​recorded from a forest experiencing a forest fire. Such comparison values ​​include, for example, temperature, humidity, wind direction and strength, and / or the composition of gases. The tp measurement data (true positive measurement data) recorded with these tp comparison values ​​are accordingly measurement data measured from a forest that also experiences a forest fire. The recorded tp measurement data therefore represent measurement data from a forest that experiences a forest fire. By recording tp comparison values, the baseline and long-term deviations of the first sensor signals are compensated.In addition, design-related deviations of the sensors are recorded, averaged, and compensated. For the purposes of this document, tn comparison values ​​(true negative comparison values) are comparison values ​​recorded from a forest that has not experienced a forest fire. Such comparison values ​​include, for example, temperature, humidity, wind direction and strength, and / or the composition of gases. The tn measurement data recorded with these tn comparison values ​​(true negative measurement data) are accordingly measurement data measured from a forest that has also not experienced a forest fire. The recorded tn measurement data therefore represent measurement data from a forest that has not experienced a forest fire. By recording tn comparison values, the baseline and long-term deviations of the initial sensor signals are compensated. In addition, design-related deviations of the sensors are recorded, averaged, and compensated, in particular, the accuracy is increased, i.e.False positives are reduced or eliminated. Threshold-based forest fire detection is error-prone and generates many false positives.

[0036] In a further embodiment of the invention, the analysis unit comprises an AI / ML unit with the aid of which the analysis of the first sensor signal can be carried out. The AI / ML unit has an AI / ML function. The AI / ML function is used in this invention to improve the efficiency of the sensor device of the terminal device. The AI / ML function enables improved application-specific detection of the gases to be detected. In particular, the sensitivity of a forest fire early detection system is increased. In addition, the accuracy is increased, i.e. false positives are reduced or eliminated. Forest fire detection based on threshold values ​​is error-prone and generates many false positives. With machine learning models, especially models created on the basis of locally acquired data, the accuracy can be increased enormously because the model adapts to reality.The AI / ML function enables improved application-specific recording of recorded measured values. Furthermore, the algorithm corrects the recorded measured values ​​for humidity. In addition, the baseline and long-term deviations of the measured values ​​are compensated. By evaluating this data, statements can be made about the current situation during forest fires. Exemplary embodiments of the method according to the invention and the forest fire early detection system according to the invention are shown in simplified schematic form in the drawings and are explained in more detail in the following description.

[0037] They show:

[0038] Fig. 1 : Inventive method for detecting a forest fire

[0039] Fig. 2: Another embodiment of the method according to the invention,

[0040] Application of a comparison matrix

[0041] Fig. 3: Another embodiment of the method according to the invention,

[0042] Comparison values ​​in a comparison matrix

[0043] Fig. 4: Another embodiment of the method according to the invention, analysis of the first signal is carried out with the comparison matrix

[0044] Fig. 5: Correction factors of the ambient temperature at different

[0045] Humidity levels

[0046] Fig. 6: Correction factors for humidity at different

[0047] Ambient temperatures

[0048] Fig. 7: Example of a stationary device for the early detection of a forest fire

[0049] Fig. 8: Structure of a forest fire early detection system comprising a LoRa radio network with transmission of result data and ML data, devices for the early detection of a forest fire arranged stationary

[0050] Fig. 9: Structure of a forest fire early detection system comprising a LoRa radio network with transmission of result data and ML data, mobile devices for the early detection of a forest fire

[0051] Fig. 10: Detailed view of a forest fire early detection system comprising a LoRa radio network with transmission of result data and ML data, mobile and stationary ML units

[0052] Fig. 11 : Embodiment of a mobile device for the early detection of a forest fire with an ML unit

[0053] Fig. 1 shows an embodiment of the method according to the invention for detecting a forest fire. The method begins with the detection eS1 of a first sensor signal, wherein the detection eS1 is carried out by means of a first sensor S1, 330 (see Fig. 8, Fig. 9, Fig. 11). In all embodiments, the first sensor S1, 330 is a gas sensor that detects gas using the principle of electrical resistance measurement. The gas to be detected is carbon monoxide; optionally, other gases such as volatile organic compounds (VOCs), volatile sulfur-containing compounds (VSCs), carbon dioxide and molecular hydrogen can be detected with the first sensor S1, 330. Optionally, the concentrations of the gases can also be detected.

[0054] Simultaneously with the first detection eS1, a detection eS2 of a second sensor signal for detecting ambient parameters takes place by means of a second sensor S2, 340 (see Fig. 8, Fig. 9, Fig. 11). The second sensor S2, 340 is different from the first sensor S1, 330 and therefore detects eS2 a second sensor signal that is different from the first sensor S1, 330 and is unsuitable for gas analysis on its own. In all embodiments, the second sensor S2, 340 is a sensor for determining the ambient temperature. The detected second sensor signal eS2 is therefore suitable for determining the temperature of the ambient air.

[0055] The next method step involves analyzing aS1 the first detected eS1 sensor signal. The detected second eS2 sensor signal is used for analyzing aS1. The ambient air temperature is determined using the second sensor signal. The analysis aS1 of the first sensor signal is performed in such a way that, in this exemplary embodiment, the baseline is compensated using a correction factor CF depending on the second sensor signal. The eS1 gas composition detected using the first sensor signal is thus corrected using a correction factor CF depending on the ambient air temperature (see Fig. 5, Fig. 6).

[0056] This is followed by a threshold comparison SV. If the concentration of the eS1 and analyzed aS1 gases (here: carbon monoxide) detected by the first sensor S1, 330 exceeds an adjustable and stored threshold, a message sM is sent in the next method step. The message includes at least the content that a forest fire has been detected. Optionally, the message also contains a unique ID of the sM device for early detection of a forest fire ED, 100 sending the message and its position. Fig. 2 shows a further embodiment of the method according to the invention. At the same time, the first sensor signal eS1 (gas detection) is detected by a first sensor S1, 330, and the second sensor signal eS2 (for detecting ambient parameters) is detected by a second sensor S2, 340.

[0057] In the next method step, an analysis aS1 of the first detected eS1 sensor signal takes place. In this case, the detected second eS2 sensor signal is also used for the analysis aS1. In this exemplary embodiment, the analysis aS1 is carried out using comparison values ​​stored in a comparison matrix VM. The selection of the comparison values ​​used for the analysis aS1 is made based on the ambient parameter (air temperature) determined from the second sensor signal. In other words, different comparison values ​​are used for the analysis aS1 for different ambient parameters (see Fig. 5, Fig. 6). The comparison values ​​stored in the comparison matrix VM are determined for different second sensor signals and / or different values ​​of the ambient parameter eS2 determined from the second sensor signal. The threshold value comparison SV then also takes place.If the concentration of the eS1 and analyzed aS1 gas detected by the first sensor S1, 330 exceeds the stored threshold value, the message sM is sent in the next process step.

[0058] A further embodiment of the method according to the invention is shown in Fig. 3. Again, the detection eS1 of the first sensor signal (gas detection) by means of a first sensor S1, 330 and the detection eS2 of a second sensor signal for detecting ambient parameters (air temperature) by means of a second sensor S2, 340 take place simultaneously.

[0059] In this exemplary embodiment, the analysis aS1 of the first signal is not carried out using the detected second sensor signal eS2. Therefore, the gas composition eS1 detected using the first sensor signal is not corrected depending on the ambient air temperature during the analysis aS1. In this exemplary embodiment, the threshold value comparison SV is carried out using the comparison matrix VM. The threshold value comparison SV is performed in such a way that the threshold values ​​are compensated depending on the second sensor signal. In addition to the comparison values ​​for different ambient parameters, the comparison matrix VM also contains the threshold values ​​dependent on the ambient parameters (air temperature).

[0060] Fig. 4 shows an embodiment of the method according to the invention, wherein an analysis of the detected second signal eS2 also takes place aS2. First, the detection eS1 of the first sensor signal (gas detection) takes place simultaneously by means of a first sensor S1, 330, and the detection eS2 of a second sensor signal for detecting environmental parameters, in this embodiment, air humidity and air temperature, takes place by means of the second sensor S2, 340. The second signal is analyzed aS2, and the air humidity and air temperature are determined.

[0061] In the next process step, an analysis aS1 of the first acquired eS1 sensor signal is performed, with the analysis aS2 of the second acquired eS2 signal occurring before the analysis aS1 of the first acquired eS1 signal. Thus, the air humidity and air temperature are determined before the analysis aS1 of the first acquired eS1 sensor signal is performed.

[0062] In this exemplary embodiment, the analysis aS1 of the first signal takes place in two separate steps: In the first step aS1.1, the detected first signal eS1 is analyzed, whereby the gas composition is determined. In the second step, the threshold value comparison SV is carried out using comparison values ​​stored in a comparison matrix VM. The comparison values ​​used for the analysis aS1 are selected based on the ambient parameters (air humidity and air temperature) analyzed from aS2 and determined from the second sensor signal. In addition, the gas composition eS1 detected by means of the first sensor signal is corrected depending on the temperature and humidity of the ambient air using a correction factor CF (see Fig. 5, Fig. 6).Optionally, the comparison values ​​for analysis aS1 of the first sensor signal are selected using an AI / ML function, whereby the AI / ML function applies the values ​​of the second sensor signal to analyze the first sensor signal and the AI / ML function analyzes the values ​​of the second sensor signal. Also optionally, the AI / ML function analyzes the values ​​of the second sensor signal. The comparison values ​​include tn comparison values ​​and tp comparison values. The tn comparison values ​​are generated by the forest W to be monitored, which is not experiencing a forest fire, in particular, for example, ambient temperature, air humidity, wind direction and strength and / or the composition of gases. The tn comparison values ​​are preferably recorded and generated by the natural forest W for which the use of the forest fire early detection system 1 is intended.Depending on the extent of the forest W to be monitored, tn comparison values ​​from different regions of the natural environment are also recorded, whereby the tn comparison values ​​differ from region to region.

[0063] The tp reference values ​​are generated from true-positive events. In the simplest case, a true-positive event is a forest fire. For this purpose, forest components, such as the fauna found in forests, forest soil components, and / or loose material on the forest floor, are heated and / or burned at various temperatures in a laboratory, the resulting gases are detected, and converted into tp reference values. Additionally, the tp reference values ​​are determined for a specific region by using fauna, forest soil components, and / or loose material on the forest floor from a specific region in the laboratory experiments.

[0064] If the concentration of the eS1 and analyzed aS1 gas detected by the first sensor S1, 330 exceeds the stored threshold value, the message sM is sent in the next process step.

[0065] Fig. 5 shows an example of the curve of the correction factor CF as a function of the ambient air temperature T at different air humidity levels AHn, where: AH1>AH2>AH3>AH4>AH5>AH6>AH7. The correction factor CF is higher at higher air humidities AHn. The maximum of the correction factor CF at different air humidities AHn shifts to higher values ​​of the air temperature T.

[0066] Fig. 6 shows an exemplary embodiment of the curve of the correction factor CF as a function of the ambient air humidity AH at different values ​​of the air temperature Tn, where: T1>T2>T3>T4>T5>T6>T7 applies. The correction factor CF is higher at higher air temperatures Tn. The maximum of the correction factor CF at different air temperatures Tn shifts to higher values ​​of the ambient air humidity AH.

[0067] Fig. 7 shows an embodiment of a stationary device ED for the early detection of a forest fire, hereinafter also referred to as a terminal device ED. For the detection of a forest fire, an individual terminal device ED has a sensor unit which has sensors S1, S2 for detecting eS1, eS2 sensor signals (see Fig. 7). The terminal device ED has a first sensor S1. The first sensor S1 is a gas sensor for gas detection and recording the concentration of one or optionally a plurality of gases. The second sensor S2 is used to record environmental parameters, in particular air pressure and air temperature. A terminal device ED has a communication interface K1 and the antenna A to a gateway G for data exchange. In order to be able to install and operate the terminal device ED in inhospitable and particularly rural areas far from any power supply, the terminal device ED is equipped with a self-sufficient power supply E.In the simplest case, the power supply E is a battery, which can also be designed to be rechargeable. However, the use of capacitors, especially supercapacitors, is also possible. A somewhat more complex and cost-intensive power supply E, but one that offers a very long service life for the end device ED, is the use of solar cells.

[0068] The analysis unit C is connected to the sensors S1, S2, the power supply E, and the communication interface K1. The analysis unit C has a memory and a microcontroller that controls the aforementioned components of the terminal device ED. The memory of the analysis unit C also stores the comparison values, the comparison matrix VM, the threshold values, and optionally the tn comparison values, tp comparison values, and the AI / ML function. All of these components are housed in a housing for protection against the elements.

[0069] Fig. 8 and Fig. 9 each show an exemplary embodiment of a forest fire early detection system 1. The forest fire early detection system 1 comprises a plurality of stationary devices ED for the early detection of a forest fire (hereinafter referred to as terminal devices ED) (Fig. 8). Additionally, forest fire detection devices 100 are arranged at different positions. The forest fire detection device 100 comprises the main components of the forest fire detection station and the forest fire detection unit 300 (see Fig. 11).

[0070] In this exemplary embodiment, the forest fire early detection system 1 comprises a LoRaWAN mesh gateway network 10 with a plurality of terminal devices ED, wherein eight terminal devices ED each communicate with a gateway G via a single-hop connection FSK. The gateways G are connected to each other and to border gateways BGD. The border gateways BGD are connected to the internet network server NS, either via a wired connection WN or via a wireless connection using the internet protocol IP. Messages can thus be sent sM from a terminal device ED to the internet network server NS. Similarly, a mobile device 300 for the early detection of a forest fire is connected to a gateway G and can send sM messages to the internet network server NS.

[0071] A gateway G has a communication interface to both an end device ED for data exchange and a border gateway BGD. The connection to the border gateway BGD can be established via a meshed multi-hop network (MHF), while the connection to the end device ED is a single-hop connection (FSK). The two communication interfaces of the gateway G use different communication channels, so the sender can be identified via the communication channel used.

[0072] A border gateway BGD has a communication interface to a gateway G and to the network server NS. The border gateway BGD then sends the data from a terminal device ED, which was sent to the border gateway BGD via a single-hop and multi-hop connection, directly to the network server NS using an Internet protocol (IP). Communication between the border gateway BGD and the Internet network server NS can be wired (WN) or wireless (IP). Each communication interface of the border gateway BGD uses its own communication channel, which is different from the other communication interfaces. Further embodiments of the forest fire early detection system 1 are conceivable, e.g., a direct wireless connection from a device ED 300 for the early detection of a forest fire to a central network server without detours via gateways G.

[0073] Fig. 10 shows a detailed view of an early forest fire detection system 1 according to the invention. The early forest fire detection system 1 comprises a LoRaWAN mesh gateway network 10 with a plurality of devices ED, 300 for early detection of a forest fire, wherein the devices ED, 100 for early detection of a forest fire communicate with a gateway G via a single-hop connection FSK. The gateways G are connected to each other and to border gateways BGD. The border gateways BGD are connected to the internet network server NS, either via a wired connection WN or via a wireless connection using the internet protocol IP.

[0074] Fig. 11 shows an embodiment of a mobile device 300 for the early detection of a forest fire, which, like a stationary terminal device ED sensors, has a first sensor 330 for detecting eS1 of a first sensor signal and a second sensor 340 for detecting eS2 of a second sensor signal, as well as an ML unit 310. The mobile device 300 for the early detection of a forest fire is designed as a flight-capable drone that is autonomous, automatic, and / or remotely controllable. The mobile device 300 for the early detection of a forest fire has a drive unit 320 with a plurality of rotors 322 driven by motors 321. The motors 321 are typically electric motors and are supplied with energy by a rechargeable energy storage device (battery). The mobile device 300 for the early detection of a forest fire can be controlled by pivoting the rotors 322 and varying the speed of the individual motors 321.

[0075] The mobile device 300 for early detection of a forest fire has the first sensor 330, which is a gas sensor. Additionally, the mobile device 300 for early detection of a forest fire has a second sensor 340, which is designed as a pressure and temperature sensor. The ML unit 310 stores comparison values, the comparison matrix VM, the threshold values, and optionally the tn comparison values, tp comparison values, and the AI / ML function in a memory and is detachably connected to the forest fire detection unit 300 via the connection 312 in the receptacle 311.

[0076] The forest fire detection unit 300 according to the invention also has a navigation sensor 350 that detects objects in the surroundings of the mobile device 300 for early detection of a forest fire. The navigation sensor 350 has one or a plurality of cameras and / or time-of-flight-based sensors (e.g., radar, ultrasound, lidar) that detect obstacles during the flight of the mobile device 300 for early detection of a forest fire. The obstacles are detected, recognized, and analyzed by the control unit arranged in the mobile device 300 for early detection of a forest fire such that the mobile device 300 for early detection of a forest fire automatically avoids the obstacles during its flight.

[0077] B EZ UG S CHARACTERS LIST

[0078] 1 forest fire detection system

[0079] 10 LoRaWAN mesh gateway network

[0080] ED terminal / device for early detection of a forest fire

[0081] G Gateway

[0082] NS Internet Network Server

[0083] IP Internet Protocol

[0084] MHF multi-hop radio network

[0085] BGD Border Gateway

[0086] FSK FSK modulation

[0087] WN Wired connection

[0088] W Forest

[0089] 100 Device for early detection of a forest fire

[0090] 200 forest fire detection stations

[0091] 300 forest fire detection unit

[0092] 310 ml unit

[0093] 311 recording

[0094] 312 Detachable connection

[0095] 320 Flight propulsion / propulsion unit

[0096] 321 engine

[0097] 322 Rotor

[0098] 330 First Sensor

[0099] 340 Second Sensor

[0100] 350 navigation sensor

[0101] A antenna

[0102] C Analysis unit

[0103] E Energy supply

[0104] K1 communication interface

[0105] ML ML unit S1 First sensor / gas sensor

[0106] S2 Second sensor / temperature sensor

[0107] CF correction factor

[0108] T, Tn temperature

[0109] AH. AHn Humidity eS1 Acquisition of first sensor signal eS2 Acquisition of second sensor signal aS1, aS1.1 Analysis of first sensor signal aS2 Analysis of second sensor signal sM Sending a message

[0110] SV threshold comparison

[0111] VM comparison matrix

Claims

PATENT CLAIMS 1 . Method for detecting a forest fire with the following steps: • Acquiring (eS1) a first sensor signal with a first sensor for detecting and / or analyzing a gas, • Detecting (eS2) a second sensor signal, wherein the second sensor signal is different from the first sensor signal, • Analysis (aS1) of the first sensor signal, whereby the analysis (aS1) of the first sensor signal depends on the second sensor signal.

2. Method for detecting a forest fire according to claim 1, characterized in that the first sensor signal is used for gas detection by means of a resistance measurement.

3. A method for detecting a forest fire according to claim 1 or 2, characterized in that the second sensor signal alone is unsuitable for gas analysis.

4. Method for detecting a forest fire according to one or more of the preceding claims, characterized in that from the second sensor signal a gas composition of the Environment of various environmental parameters is analyzed (aS2).

5. Method for detecting a forest fire according to claim 4, characterized in that the humidity and / or temperature is determined from the second sensor signal (aS2).

6. Method for detecting a forest fire according to claim 4 or 5, characterized in that the analysis (aS2) of the second sensor signal takes place before the analysis (aS1) of the first sensor signal.

7. Method for detecting a forest fire according to one or more of the preceding claims, characterized in that the analysis (aS1) of the first sensor signal is carried out on the basis of comparison values.

8. Method for detecting a forest fire according to claim 7, characterized in that the comparison values ​​are stored in a comparison matrix (VM).

9. Method for detecting a forest fire according to claim 7 or 8, characterized in that the comparison values ​​used for analysis are selected on the basis of the second sensor signal and / or the environmental parameter (aS2) determined from the second sensor signal.

10. Method for detecting a forest fire according to one or more of claims 7 to 9, characterized in that the comparison values ​​are determined for different second sensor signals and / or different values ​​of the environmental parameter (aS2) determined from the second sensor signal.

11. Method for detecting a forest fire according to one or more of the Claims 7 to 10, characterized in that the comparison values ​​are determined for a specific region.

12. Method for detecting a forest fire according to one or more of the Claims 7 to 11, characterized in that the comparison values ​​comprise tp comparison values ​​and / or tn comparison values.

13. Method for detecting a forest fire according to one or more of the Claims 7 to 12, characterized in that the comparison values ​​for the analysis (aS1) of the first sensor signal are selected by means of an AI / ML function.

14. A method for detecting a forest fire according to claim 13, characterized in that the AI / ML function uses the values ​​of the second sensor signal to analyze the first sensor signal.

15. Method for detecting a forest fire according to claim 13 or 14, characterized in that the AI / ML function analyzes the values ​​of the second sensor signal (aS2).

16. Method for detecting a forest fire according to one or more of the Claims 13 to 15, characterized in that the selection of the comparison values ​​is carried out on the basis of the analysis (aS2) of the second sensor signal.

17. Device for the early detection of a forest fire with: • a first sensor (S1, 330) which is provided and suitable for detecting a first sensor signal (eS1), • a second sensor (S2, 340) which is intended and suitable for detecting a second sensor signal (eS2), wherein the first sensor (S1, 330) is different from the second sensor (S2, 340), • an analysis unit (C) which is intended and suitable for analyzing the first sensor signal using the second sensor signal and / or an environmental parameter determined from the second sensor signal (aS2).

18. Device for the early detection of a forest fire according to claim 17, characterized in that the second sensor (S2, 340) is a humidity sensor and / or a temperature sensor.

19. Device for the early detection of a forest fire according to claim 17 or 18, characterized in that the first sensor (S1, 330) is a gas sensor.

20. Device for the early detection of a forest fire according to one or more of claims 17 to 19, characterized in that the analysis unit (C) comprises a memory in which a comparison matrix (VM) and / or comparison values ​​are stored which comprise values ​​of the environmental parameter for different values ​​of the second sensor signals and / or for different values ​​of the values ​​of the environmental parameter determined from the second sensor signals (aS2).

21. Device for the early detection of a forest fire according to claim 20, characterized in that the comparison matrix (VM) contains comparison values.

22. A device for the early detection of a forest fire according to claim 20 or 21, characterized in that the comparison values ​​comprise comparison values ​​for different regions.

23. A device for the early detection of a forest fire according to one or more of claims 17 to 22, characterized in that the comparison values ​​comprise tp comparison values ​​and / or tn comparison values.

24. Device for the early detection of a forest fire according to one or more of claims 17 to 23, characterized in that the analysis unit (C) comprises an AI / ML unit with the aid of which the analysis (aS1) of the first sensor signal can be carried out.

Images