Method for evaluating the cause of a forest fire and forest fire cause evaluation device

Abstract

The invention relates to a method for evaluating the cause of a forest fire with the method steps of detecting a measurement signal of a forest fire, analyzing the measurement signal, evaluating the analysis data obtained from the analysis of the measurement signal with an evaluation unit and determining the cause of the fire from the analysis data, as well as a forest fire evaluation device.

Description

[0001] The invention relates to a method for evaluating the cause of a forest fire with the method steps of detecting a measurement signal of a forest fire, analyzing the measurement signal, evaluating the analysis data obtained from the analysis of the measurement signal with an evaluation unit and determining the cause of the fire from the analysis data, as well as a forest fire evaluation device. State of the art

[0002] Systems for the early detection of forest fires are well known. The area to be monitored is monitored using optical sensors that can detect smoke columns resulting from a forest fire. These sensors, for example, take the form of rotatable cameras, but they have the disadvantage of being less effective at night and prone to false detections, e.g., in the case of dust clouds caused by agricultural activities. Furthermore, optical systems can usually only detect forest fires once the forest fire is already advanced and the smoke columns are visible over great distances. Monitoring using an IR camera installed in a satellite from a high orbit has the disadvantage that the resolution of the cameras over great distances prevents detection of forest fires in the early stages. A satellite is also expensive to purchase and maintain, particularly during launch.Surveillance using mini-satellites in low orbit has the disadvantage that the satellites are not geostationary, meaning they require a certain amount of time to orbit the area during which the area is not monitored. Close-meshed surveillance requires a large number of satellites, which are also costly to launch. Surveillance using satellites also results in high carbon dioxide emissions during launch.

[0003] It makes more sense to monitor the area using a number of inexpensive, mass-produced sensors that operate using optical smoke and / or gas detection. The sensors are distributed throughout the area and transmit data to a base station via radio link.

[0004] Such a system for the early detection of forest fires is presented in document US 2008 / 0309502 A1. In the event of a fire alarm, a sensor transmits information to a nearby control terminal, which then triggers an alarm using a long-range radio frequency signal.

[0005] This system has the disadvantage that the control terminal triggers the alarm and requires a powerful RF unit. The sensors require a GPS unit that continuously transmits a signal to the control terminal. Therefore, the sensors consume a lot of power, and the lifespan of the sensors' power sources (batteries) is limited. Task

[0006] It is therefore an object of the present invention to provide a method for evaluating the cause of a forest fire that operates reliably, is expandable as required, and is cost-effective to install and maintain.

[0007] It is also an object of the present invention to provide a forest fire cause evaluation device that operates reliably, is expandable as required, and is cost-effective to install and maintain.

[0008] The object is achieved by means of the method for evaluating the cause of a forest fire according to claim 1. Advantageous embodiments of the invention are set out in the subclaims.

[0009] The inventive method for evaluating the cause of a forest fire comprises four method steps: In the first method step, a measurement signal of a forest fire is recorded. The measurement signal is recorded using a suitable sensor of an evaluation unit, e.g. a temperature and / or a gas sensor. In the second method step, a measurement signal is analyzed. The analysis of the measurement signal optionally includes converting the measurement signal into analysis data, which is also optionally stored and / or sent. In the third method step, the analysis data obtained from the analysis of the measurement signal is evaluated using an evaluation unit. In the fourth method step, the cause of the fire is determined from the analysis data. To determine the cause of the fire, the measurement signal and the analysis data are optionally evaluated for the gases released during the forest fire.

[0010] In an advantageous embodiment of the invention, the measurement signal is acquired during the forest fire. Optionally, the analysis, evaluation, and determination of the cause of a forest fire also take place during the forest fire. The method according to the invention can therefore provide valuable information for containing and fighting the forest fire in the short term.

[0011] In a further development of the invention, the analysis data includes data on the substances burned by the forest fire. The data on the burned substances provide clues to the cause of the forest fire.

[0012] In a further embodiment of the invention, the cause of the fire is determined from the data on the substances burned in the forest fire. In a further aspect of the invention, the cause of the fire is determined from the data on the substances burned in the forest fire that are not part of the forest. This makes it possible to determine whether the forest fire was caused by a non-natural event, e.g., arson.

[0013] In a further embodiment of the invention, those substances that are not part of the forest are determined by comparing the analysis data with library data of burned substances from forest components. The library data of burned substances are determined in an environment that allows for reproducible conditions, e.g., in a laboratory. For this purpose, a gas analysis is performed on those burned substances from forest components from which an evaluation of the cause of a forest fire is to be determined.

[0014] In a further embodiment of the invention, the library data is ML / AI data. The library data is preferably created using a machine learning algorithm.

[0015] In a further embodiment of the invention, the evaluation of the measurement signal is initiated event-driven. If no event, in particular a forest fire, is detected, no evaluation of the measurement signal is performed. This minimizes the computational effort.

[0016] In a further embodiment of the invention, the event that triggers the evaluation is based on the acquired measurement signals and / or their analysis. The measurement signals are analyzed and checked to determine whether an event—a fire—is detected. If so, the measurement signals are evaluated.

[0017] In a further embodiment of the invention, the event that initiates the evaluation is the deviation of the recorded measurement signals and / or their analysis data from previously recorded measurement signals and / or their analysis data. The recorded measurement signals optionally include a gas analysis. A forest fire generates characteristic gases, which are determined by analysis. Optionally, the concentration of the characteristic gas is determined and analyzed. If this gas is detected, it is analyzed whether a specific concentration (threshold value) to be specified by the user is exceeded. As an alternative to exceeding the threshold value, the detection of a pattern or other anomalies in the recorded and / or analyzed measurement signals can also be used as a triggering event.

[0018] In a further aspect of the invention, the acquisition time of the measurement signal is recorded. The measurement signal receives a timestamp that is uniquely linked to the measurement signal. The timestamp contains, for example, the time of acquisition of the measurement signal, the time of commencement of acquisition of the measurement signal, and the duration of acquisition of the measurement signal. In a further embodiment of the invention, the analysis data includes the acquisition time and / or time data generated from the acquisition time.

[0019] In a further development of the invention, the evaluation includes a temporal assessment of the analysis data. In particular, the evaluation includes when which gas was detected and in which concentration. This evaluation of the correlation between time and gas concentration enables a reliable determination of the cause of a forest fire.

[0020] In an advantageous embodiment of the invention, the evaluation comprises analysis data that precedes the event that initiated the evaluation. The analysis data includes a gas analysis that precedes the event that initiated the evaluation. By comparing this analysis data with analysis data that precedes the event that initiated the evaluation, it can be determined whether the forest fire was caused by a non-natural event, e.g., arson.

[0021] In a further embodiment of the invention, the measurement signals are recorded and / or analyzed by an evaluation unit. The evaluation unit has suitable sensors for this purpose, e.g., for gas analysis and / or for detecting a forest fire. Additionally, the evaluation unit comprises an analysis unit that is suitable and intended for analyzing the recorded signals.

[0022] In a further embodiment of the invention, the analysis data is evaluated by the evaluation unit and / or the cause of the fire is determined by the evaluation unit. The analysis data is thus evaluated by the evaluation unit, which provides the analysis data and records the measurement signals. This enables a timely and decentralized determination of the cause of the fire.

[0023] In a further embodiment of the invention, the measurement signals and / or the analysis data are sent to the network server, and / or the network server evaluates the analysis data and / or determines the cause of the fire. This allows for a centralized determination of the cause of the forest fire.

[0024] In a further development of the invention, the measurement signals are analyzed by multiple evaluation units. In a further aspect of the invention, the analysis data from multiple evaluation units are used to evaluate and / or determine the cause of the fire. Wide-area monitoring of an area and determination of the cause of the fire in the monitored area are thus possible. At the same time, a redundant determination of the cause of the fire is achieved.

[0025] In an advantageous embodiment of the invention, the sensor unit has a sensor that can be operated in a first signal detection mode and a second signal detection mode. The two signal detection modes can differ due to the duration in which the detection takes place. For example, the first signal detection mode can have a shorter detection time than the second signal detection mode, or vice versa. A physical distinction between the two signal detection modes is also possible: In the first signal detection mode, for example, smoke can be detected, while in the second signal detection mode heat can be detected by means of infrared detection. Important within the meaning of the invention is the difference in the detection of the signal using different signal detection modes. This increases the reliability and accuracy of the method according to the invention because two independent detection modes are used for the early detection of a forest fire.

[0026] In a further embodiment of the invention, the signals acquired during the first signal acquisition are analyzed. The signals acquired using the first signal acquisition mode are analyzed and checked to determine whether an event—a fire—has been detected.

[0027] In a further development of the invention, the first signal acquisition is repeated in a first signal acquisition mode of the sensor unit at a time interval. The time interval can be set by a user. For example, one signal acquisition can be performed every second; other time intervals are possible. This allows the temporal resolution of the signal acquisition to be increased or decreased.

[0028] In an advantageous embodiment of the invention, the execution of a second signal acquisition in a second signal acquisition mode of the sensor unit is initiated in an event-driven manner. Such an event can be the failure of a system component of the evaluation unit, which prevents signal acquisition in the first mode. Furthermore, signal acquisition can take place in the second signal acquisition mode if an event (fire) was detected by the first signal acquisition. The signals from the first signal acquisition are thus checked, and the method operates more reliably than with signal acquisition in only one signal acquisition mode.

[0029] In a further development of the invention, the event that triggers the second signal acquisition in a second signal acquisition mode is based on the signals acquired in the first signal acquisition and / or their analysis. The signals acquired using the first signal acquisition mode are analyzed and checked to determine whether an event—a fire—has been detected. If so, a signal acquisition occurs in the second signal acquisition mode. The signals from the first signal acquisition are thus checked, making the method more reliable than with signal acquisition in only one signal acquisition mode.

[0030] In a further embodiment of the invention, the event that starts the second signal acquisition in a second signal acquisition mode is the exceeding of a threshold value of the analyzed data from the signals acquired in the first signal acquisition. The threshold value can be, for example, the concentration of a gas that naturally occurs in the area being monitored by the method according to the invention. If this gas is detected by means of a first signal acquisition, it is useful to check whether a certain concentration (threshold value) to be specified by the user is exceeded. This check is carried out with the second signal acquisition in a second signal acquisition mode. As an alternative to exceeding the threshold value, the detection of a pattern or other anomalies in the acquired and / or analyzed data can also be used as a triggering event.

[0031] In a further development of the invention, the second signals acquired by the second signal acquisition mode are evaluated. The signals acquired using the second signal acquisition mode are analyzed and checked to determine whether an event—a fire—has been detected.

[0032] In a further embodiment of the invention, a message is sent from the evaluation unit to a first gateway. The message is sent, in particular, when the evaluation of the first signal acquisition and the second signal acquisition has detected a fire. The message is sent as a data packet, either wirelessly or wired, to a first gateway (node).

[0033] In a further embodiment of the invention, the notification signal is generated in the sensor unit when the data from the detected second signals and / or analyzed from these exceeds a second threshold. The threshold can, for example, be the concentration of a gas that is of natural origin or occurs from other sources in the area being monitored by the method according to the invention. If this gas is detected by a second signal detection, the notification is generated and forwarded to a first gateway.

[0034] In a further development of the invention, the second signal acquisition mode comprises a gas analysis. The gas analysis can be performed using suitable gas detectors, e.g., catalytic gas detectors, semiconductor gas detectors, or electrochemical gas detectors. A quantitative analysis can also be performed.

[0035] In a further embodiment of the invention, the first signal acquisition mode is based on a different physical principle than the second signal acquisition mode. The first signal acquisition mode is typically designed for preliminary or rough analysis. A simple, proven, and inexpensive principle can be used for this purpose, e.g., smoke detection using an infrared LED. The second signal acquisition mode serves to increase accuracy and, if necessary, quantitative analysis. Other principles are typically used for this purpose, e.g., electrical ones.

[0036] In a further embodiment of the invention, the first signal detection mode is based on optical smoke detection. The optical chamber of the smoke detector contains an infrared LED and a light-sensitive sensor, a so-called photodiode. The infrared LED continuously emits a test light beam that is invisible to the human eye and is aligned so that it does not directly hit the light-sensitive photodiode. If smoke particles are present in the air and penetrate the optical chamber of the evaluation unit, the emitted infrared light beam from the LED is scattered and reflected by the smoke particles. Part of this scattered infrared light also falls on the light-sensitive photodiode.

[0037] The object is further achieved by means of the forest fire cause evaluation device according to the invention with an evaluation unit.

[0038] The forest fire cause evaluation device according to the invention with an evaluation unit has a sensor unit, a control unit, a communication unit, and a self-sufficient power supply unit. The sensor unit has a sensor that can detect a forest fire, e.g., a gas sensor. The communication unit enables the evaluation unit to be connected to, e.g., a network server. The control unit controls the acquisition of measurement signals by the sensor unit. In order to be able to install and operate the evaluation unit even in inhospitable and, in particular, rural areas far from any power supply, the evaluation unit is equipped with a self-sufficient power supply. The power supply can be provided, e.g., by energy storage devices - including rechargeable ones - and / or solar cells.

[0039] In a further development of the invention, the control unit is intended and suitable for controlling the signal acquisition. Signal acquisition is optionally carried out using two different signal acquisition modes, each of which is activated and controlled by the control unit. At the same time, the control unit controls the duration and timing of both signal acquisition modes.

[0040] In a further embodiment of the invention, the control unit is provided and suitable for capturing and / or storing the acquired signals in a time-resolved manner. The acquired signals are provided with a time stamp and stored in a memory of the control unit.

[0041] In a further embodiment of the invention, the evaluation unit comprises an evaluation unit that is suitable and intended to evaluate the detected signals. The signals detected using the first signal detection mode and the second signal detection mode are analyzed and checked to determine whether an event—a fire—has been detected. In a further embodiment of the invention, the evaluation unit is intended and suitable to evaluate and / or store the detected signals in a time-resolved manner. The detected signals are provided with a time stamp and stored in a memory of the evaluation unit. In a further development of the invention, the evaluation unit is part of the control unit.

[0042] In a further embodiment of the invention, the sensor unit comprises a first sensor element. The first signal acquisition mode is typically designed as a preliminary or rough analysis. A simple, proven, and inexpensive principle of the first sensor element can be applied for this purpose.

[0043] In a further development of the invention, the first sensor is a gas sensor. A gas sensor detects the presence of a defined gas, whereby the concentration of the gas must exceed a threshold value to be detected. Alternatively, instead of the threshold value, patterns or other anomalies of the detected signals can also be recorded.

[0044] In a further embodiment of the invention, the measuring principle of the first sensor element is based on electrical signal detection or optical signal detection. The second sensor element optionally serves to increase the accuracy and, if necessary, quantitatively analyze a gas. Other principles, such as electrical ones, are usually used for this purpose. The signals from the first sensor element are thus verified, and the method operates more reliably than with signal detection in only one sensor element.

[0045] In an advantageous embodiment of the invention, the evaluation unit is part of a network of evaluation units. This enables wide-area monitoring of an area and determination of the cause of a fire in the monitored area. At the same time, a redundant determination of the cause of the fire is achieved.

[0046] In a further embodiment, the network is a mesh gateway network comprising evaluation units, a network server, and gateways. In a further embodiment of the invention, the gateways of the mesh network have different functionalities and / or designs.

[0047] In the mesh gateway network, a gateway communicates directly exclusively with other gateways and evaluation units of the mesh gateway network. In a further embodiment of the invention, the mesh gateway network has first gateways that do not have a single-hop connection to a network server. In particular, communication between evaluation units and a first gateway is direct, i.e., without any further intermediate stations (single-hop connection). Communication between the gateways can take place via a direct single-hop connection; a multi-hop connection is also possible.

[0048] This simultaneously extends the range of the mesh gateway network because the first gateway is connected to the border gateway via a mesh multi-hop network, allowing it to forward data from the evaluation units to the internet network server. The connection between the border gateway and the network server is established wirelessly or wired.

[0049] In a further embodiment of the invention, the mesh gateway network comprises an LPWAN. LPWAN describes a class of network protocols for connecting low-power devices, such as battery-operated sensors, to a network server. The protocol is designed to achieve a long range and low energy consumption of the sensors at low operating costs.

[0050] In a further development of the invention, the mesh gateway network of the forest fire early detection system is a LoRaWAN mesh gateway network. LoRaWAN is particularly energy-efficient. LoRaWAN networks implement a star-shaped architecture using gateway message packets between the evaluation units and the central network server. The gateways are connected to the network server, while the evaluation units communicate wirelessly with the respective gateway via LoRa.

[0051] In a further embodiment of the invention, the evaluation units and / or the gateways have a self-sufficient power supply. To enable installation and operation of the evaluation units and the gateways connected to them even in inhospitable and, in particular, rural areas far from power supplies, the evaluation units and the gateways are equipped with a self-sufficient power supply. The power supply can be provided, for example, by energy storage devices – including rechargeable ones.

[0052] In a further development of the invention, the self-sufficient energy supply comprises an energy storage device and / or an energy conversion device. In particular, the energy supply via solar cells, which convert light energy into electrical energy, is worthy of mention. The electrical energy is usually stored in an energy storage device to ensure the energy supply even during times of low solar radiation (e.g., at night).

[0053] In a further embodiment of the invention, the sensor unit has a second sensor element. The second sensor element serves to increase accuracy and, if necessary, quantitative analysis. Other principles, such as electrical ones, are typically used for this purpose. The signals from the first sensor element are thus checked, and the method is more reliable than with signal acquisition in only one sensor element.

[0054] In a further embodiment of the invention, the sensor unit has a sensor that can be operated in a first signal detection mode and a second signal detection mode. The two signal detection modes can differ due to the duration in which the detection takes place. For example, the first signal detection mode can have a shorter detection time than the second signal detection mode, or vice versa. A physical distinction between the two signal detection modes is also possible: In the first signal detection mode, for example, smoke can be detected, while in the second signal detection mode heat can be detected by means of infrared detection. Important within the meaning of the invention is the difference in the detection of the signal using different signal detection modes. This increases the reliability and accuracy of the method according to the invention because two independent detection modes are used for the early detection of a forest fire.

[0055] In a further development of the invention, the first signal acquisition mode is different from the second signal acquisition mode. The first signal acquisition mode is typically designed for preliminary or rough analysis. A simple, proven, and inexpensive principle can be used for this purpose, e.g., smoke detection using an infrared LED. The second signal acquisition mode serves to increase accuracy and, if necessary, quantitative analysis. Other principles are typically used for this purpose, e.g., electrical.

[0056] In a further embodiment of the invention, the first signal detection mode and the second signal detection mode can be controlled separately. If an event (fire) is detected by the first signal detection mode, the signal is detected in the second signal detection mode. The signals from the first signal detection mode are thus verified, and the method operates more reliably than with signal detection in only one signal detection mode.

[0057] In a further embodiment of the invention, the first signal acquisition mode can be periodically repeated and / or the second signal acquisition mode can be activated. The signals acquired using the first signal acquisition mode are analyzed and checked to determine whether an event—a fire—has been detected. If this is the case, the signal is acquired in the second signal acquisition mode. The signals from the first signal acquisition are thus checked; the method operates more reliably than with signal acquisition in only one signal acquisition mode. Repeating the first signal acquisition mode increases the accuracy of the acquisition.

[0058] Embodiments of the method according to the invention for evaluating the cause of a forest fire and of the device according to the invention are shown in simplified schematic form in the drawings and are explained in more detail in the following description.

[0059] They show: Fig. 1: Basic principle of the procedure for evaluating the cause of a forest fire Fig. 2 a: Further development of the procedure for evaluating the cause of a forest fire Fig. 2 b: Further development of the procedure for evaluating the cause of a forest fire Fig. 3: Further development of the procedure for evaluating the cause of a forest fire, evaluation and determination of the cause of the fire outside the evaluation unit Fig. 4: Further development of the procedure for evaluating the cause of a forest fire, three evaluation units, evaluation and determination of the cause of the fire outside the evaluation units Fig. 5 ac: Examples of evaluation units Fig. 6: Embodiment of the invention in the LoRaWAN network Fig. 7: Detailed view of an embodiment of the invention in the LoRaWAN network

[0060] Fig. Figure 1 shows an embodiment of the method according to the invention for evaluating the cause of a forest fire. In the first method step, a measurement signal of the forest fire is recorded SE by an evaluation unit EV during the forest fire. The evaluation unit EV has a sensor unit S for this purpose (see Figure 1). Fig. 5). The measurement signal is then analyzed by the evaluation unit AE located in the evaluation unit EV and converted into analysis data. The analysis data obtained from the analysis A of the measurement signal are then also stored and evaluated by the evaluation unit AE EVA. In the next process step, the evaluation unit AE determines the cause of the fire from the analysis data B.

[0061] Further embodiments of the method according to the invention are shown Fig. 2. In this embodiment, SE detects the sensor unit S of the evaluation unit EV (see Fig. 5) Measurement signals in two signal detection modes. The sensor unit S initially operates in the first signal detection mode. In the first signal detection mode, the sensor unit S detects the scattered light of an infrared LED scattered by smoke using a photodiode and operates in the first sensor mode like a conventional smoke detector ( Fig. 2 a). The acquisition in the first signal acquisition mode is continuous and is repeated at adjustable time intervals t ( Fig. 2 b), in this and the other embodiments, t = 1 min. The recorded measurement signal is analyzed A by the evaluation unit AE, converted into analysis data and stored in the memory of the evaluation unit AE. If the SE and analyzed A measurement signals recorded using the first signal acquisition mode indicate a forest fire - i.e. smoke is detected and a threshold value SW1 for smoke development is reached or exceeded - the control unit C connected to the evaluation unit AE controls the sensor unit S such that the sensor unit S records signals in a second signal acquisition mode. For this purpose, the sensor unit S has two sensor elements S1, S2 whose measuring principle differs. While the first sensor element S1 is an optical smoke detector, the second sensor element S2 is a detector that detects gas using electrical and / or electrochemical methods, e.g. a semiconductor gas detector.This also increases the accuracy and reliability of the SE detection. SE detection using the second signal acquisition mode also occurs at adjustable time intervals t of 1 minute.

[0062] This SE measurement signal acquired using the second signal acquisition mode is also analyzed by the evaluation unit AE, converted into analysis data, and stored in the memory of the evaluation unit AE. The SE measurement signal acquired using the second signal acquisition mode also receives a timestamp; the time of acquisition SE of the measurement signal acquired using the second signal acquisition mode is also recorded and stored together with the analysis data.

[0063] The analysis data of the SE measurement signals acquired using the first signal acquisition mode and in particular the SE measurement signals acquired using the second signal acquisition mode contain data on the substances burned by the forest fire. The analysis A of the analysis data is performed by comparing substances burned by the forest fire with library data of burned substances from common forest components. The library data is advantageously ML / AI data acquired from the forest, preferably in laboratory experiments. If the analysis A shows that substances that are not common forest components were detected in the forest fire by the SE measurement signals, the evaluation unit AE starts and performs an evaluation E of the analysis data.

[0064] The evaluation E includes the temporal evaluation of the analysis data, each of which is provided with a time stamp. The evaluation E also includes analysis data that predates the event that triggered the evaluation. In particular, the determination B of the cause of the fire following the evaluation E determines B whether the forest fire was caused by a non-natural event, e.g., arson.

[0065] Fig. 3 shows a further embodiment of the method according to the invention. Here, the first and second sensor modes are also not operated in parallel; the sensor unit S initially operates in the first sensor mode. The sensor unit S also continuously detects the scattered light of an infrared LED scattered by smoke using a photodiode and records sensor data in the first sensor mode. If a first event SW1 is detected using the first sensor mode—i.e., smoke is detected—the control unit C controls the sensor unit S such that the sensor unit S records signals in the second sensor mode.

[0066] This SE measurement signal, acquired using the second signal acquisition mode, is also analyzed by the evaluation unit AE, provided with a time stamp, converted into analysis data, and stored in the memory of the evaluation unit AE. In contrast to the previous embodiment (see Fig. 2) sent by the control unit C of the evaluation unit EV to the Internet network server NS. The evaluation EVA and subsequent determination B of the cause of the fire are carried out with the Internet network server NS.

[0067] Fig. 4 shows a variant of the above embodiment (see Fig. 3), in which in this embodiment three different evaluation units EV are used in a forest fire cause evaluation device 1 (see Fig. 6, Fig. 7). Each evaluation unit EV records measurement signals SE1, SE2, SE3 separately from one another. Also separately from one another, A1, A2, A3 are analyzed by the evaluation unit AE located in each individual evaluation unit EV, assigned a time stamp, converted into analysis data, and stored in the memory of the evaluation unit AE that performed the analysis A1, A2, A3. The analysis data is sent from the respective control unit C of each evaluation unit EV to the Internet network server NS TR1, TR2, TR3. The evaluation EVA and subsequent determination B of the cause of the fire also takes place using the Internet network server NS.

[0068] Fig. Figure 5 shows three variants of an exemplary embodiment of an evaluation unit EV. In order to be able to install and operate the evaluation unit EV even in inhospitable and particularly rural areas far from any power supply, the evaluation unit EV 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 ( Fig. 5 a). However, the use of capacitors, especially supercapacitors, is also possible. A somewhat more complex and cost-intensive energy supply E, but one that offers a very long lifetime for the evaluation unit EV, is the use of solar cells ( Fig. 5 b). In addition to the energy conversion EK by the solar cell, an energy storage ES and power electronics are also arranged in the evaluation unit EV. Furthermore, the evaluation unit EV includes the actual sensor unit S ( Fig. 5 a, b), which records the measurement signals using optical and / or electronic methods.

[0069] The sensor unit S is preferably designed in two stages ( Fig. 5 c). The evaluation unit EV additionally has the communication interface K1. Via the communication interface K1, messages from the evaluation unit EV, in particular measurement data, are sent wirelessly as a data packet to a first gateway G via a single-hop FSK connection via LoRa (chirp frequency spread spectrum modulation) or frequency modulation.

[0070] The control unit C is connected to the sensor unit S, the power supply E, the communication interface K1, and an evaluation unit AE, and controls these components. The evaluation unit AE is arranged separately from the control unit C, optionally in a unit with the control unit C. All of these components are arranged in a housing for protection against the elements.

[0071] An embodiment of a forest fire cause evaluation device 10 according to the invention shows Fig. 6. The forest fire cause evaluation device 10 comprises a mesh gateway network 1 that utilizes the technology of a LoRaWAN network 1. The LoRaWAN network 1 has a star-shaped architecture in which message packets are exchanged between the evaluation units EV and a central Internet network server NS via gateways.

[0072] The forest fire cause evaluation device 10 comprises a plurality of evaluation units EV connected to first gateways G via a single-hop connection FSK. The gateways G are interconnected and partially connected 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.

[0073] The first gateways G and the border gateways BGD are interconnected via a meshed multi-hop network (MHF), so that a first gateway G does not require a direct connection to the internet network server NS. This extends the range of LoRaWAN networks by interposing a multi-hop network via the first gateways G, thus achieving full compatibility with the LoRaWAN specification.

[0074] A detailed view of a forest fire cause evaluation device 10 according to the invention shows Fig.7. The forest fire cause evaluation device 10 comprises a plurality of evaluation units EV, each of which is connected to a first gateway G via a single-hop connection FSK. The first 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. LIST OF REFERENCE SYMBOLS 1 mesh gateway network 10 Forest fire cause evaluation device EV Evaluation Unit G First Gateway BGD Border Gateway NS Internet Network Server IP Internet Protocol MHF multi-hop radio network FSK FSK modulation WN Wired connection W Forest S sensor device S1, S2 sensor element E Energy supply ES energy storage EK Energy Conversion K1 communication interface C Control unit AE evaluation unit SE, SE1, SE2, acquisition of measurement signal SE3 A, A1, A2, A3 Evaluation / Analysis of the measurement signal TR, TR1, TR2, Sending the analysis data TR3 EVA evaluation measurement signal B Determination of the cause of the fire SW1 Threshold of the first signal acquisition mode t time interval QUOTES CONTAINED IN THE DESCRIPTION

[0000] This list of documents submitted by the applicant was generated automatically and is included solely for the convenience of the reader. This list is not part of the German patent or utility model application. The DPMA assumes no liability for any errors or omissions. Cited patent literature

[0000] US 2008 / 0309502 A1

[0004]

Claims

[1] Procedure for evaluating the cause of a forest fire with the following procedural steps: • Detection (SE) of a measurement signal of a forest fire, • Analyze (A) the measurement signal, • Evaluating (EVA) the analysis data obtained from the analysis (A) of the measurement signal with an evaluation unit (EV), • Determine (B) the cause of the fire from the analysis data. [2] Method for evaluating the cause of a forest fire according to claim 1, characterized by that the detection (SE) of the measurement signal takes place during the forest fire. [3] Method for evaluating the cause of a forest fire according to claim 1 or 2, characterized by that the analysis data includes data on the substances burned by the forest fire. [4] Method for evaluating the cause of a forest fire according to claim 3, characterized by that the determination (B) of the cause of the fire is carried out from the data on the substances burned by the forest fire. [5] Method for evaluating the cause of a forest fire according to claim 4, characterized by that the determination (B) of the cause of the fire is carried out from the data on those substances burned by the forest fire which are not part of the forest. [6] Method for evaluating the cause of a forest fire according to one or more of the preceding claims, characterized by that those substances which are not part of the forest are determined from a comparison of the analysis data with library data of burnt substances from components of the forest. [7] Method for evaluating the cause of a forest fire according to claim 6, characterized by that the library data is ML / AI data. [8] Method for evaluating the cause of a forest fire according to one or more of the preceding claims, characterized by that the evaluation (EVA) of the measurement signal is started event-controlled. [9] Method for evaluating the cause of a forest fire according to claim 8, characterized by that the event that starts the evaluation (EVA) is based on the acquired measurement signals and / or their analysis (A). [10] Method for evaluating the cause of a forest fire according to claim 9, characterized by that the event that starts the evaluation (EVA) is the deviation of the acquired measurement signals and / or their analysis data from previously acquired measurement signals and / or their analysis data. [11] Method for evaluating the cause of a forest fire according to one or more of the preceding claims, characterized by that the acquisition time of the acquisition (SE) of the measurement signal is recorded. [12] Method for evaluating the cause of a forest fire according to claim 11, characterized by that the analysis data includes the acquisition time and / or time data generated from the acquisition time. [13] Method for evaluating the cause of a forest fire according to claim 12, characterized by that the evaluation (EVA) includes a temporal assessment of the analysis data. [14] Method for evaluating the cause of a forest fire according to claim 13, characterized by that the evaluation (EVA) includes analysis data that predates the event that initiated the evaluation. [15] Method for evaluating the cause of a forest fire according to one or more of the preceding claims, characterized by that the measurement signals are recorded (SE) and / or analyzed (A) by an evaluation unit (EV). [16] Method for evaluating the cause of a forest fire according to one or more of the preceding claims, characterized by that the analysis data is evaluated (EVA) by the evaluation unit (EV) and / or the cause of the fire is determined (B) by the evaluation unit (EV). [17] Method for evaluating the cause of a forest fire according to one or more of the preceding claims, characterized by that the measurement signals and / or the analysis data are sent to the network server (NS) and / or the analysis data are evaluated (EVA) by the network server (NS) and / or the cause of the fire is determined (B). [18] Method for evaluating the cause of a forest fire according to one or more of the preceding claims, characterized by that the measurement signals are analyzed (A) by several evaluation units (EV). [19] Method for evaluating the cause of a forest fire according to one or more of the preceding claims, characterized by that the analysis data from several evaluation units (EV) are used to evaluate (EVA) and / or determine (B) the cause of the fire. [20] Forest fire cause evaluation device (1) with an evaluation unit (EV) which • a sensor unit (S), • a control unit (C), • a communication unit (K1) and • has an autonomous power supply unit (E). [21] Forest fire cause evaluation device (1) with an evaluation unit (EV) according to claim 20, characterized by that the control unit (C) is intended and suitable to control the signal acquisition (SE). [22] Forest fire cause evaluation device (1) with an evaluation unit (EV) according to claim 20 or 21, characterized by that the control unit (C) is intended and suitable for recording (SE) and / or storing the detected signals in a time-resolved manner. [23] Forest fire cause evaluation device (1) with an evaluation unit (EV) according to one or more of claims 20 to 22, characterized by that the evaluation unit (EV) comprises an evaluation unit (AE) which is suitable and intended to analyse the detected signals (A). [24] Forest fire cause evaluation device (1) with an evaluation unit (EV) according to claim 23, characterized by that the evaluation unit (AE) is intended and suitable for analyzing (A) and / or storing the acquired signals in a time-resolved manner. [25] Forest fire cause evaluation device (1) with an evaluation unit (EV) according to claim 23 or 24, characterized by that the evaluation unit (AE) is part of the control unit (C). [26] Forest fire cause evaluation device (1) with an evaluation unit (EV) according to one or more of claims 20 to 25, characterized by that the sensor unit (S) has a first sensor element (S1). [27] Forest fire cause evaluation device (1) with an evaluation unit (EV) according to claim 26, characterized by that the first sensor element (S1) is a gas sensor. [28] Forest fire cause evaluation device (1) with an evaluation unit (EV) according to claim 27, characterized by that the measuring principle of the first sensor element (S1) is based on electrical signal detection or on optical signal detection. [29] Forest fire cause evaluation device (1) with an evaluation unit (EV) according to one or more of claims 20 to 28, characterized by that the evaluation unit (EV) is part of a network (10) of evaluation units (EV).

Images