Self-energized environmental parameter self-adaptive electronic alarm and alarm method thereof

Abstract

The invention relates to the technical field of electronic alarms, and discloses a self-energized environmental parameter adaptive electronic alarm and an alarm method thereof, and the method comprises the steps: triggering the online learning of environmental parameters to construct a dynamic baseline model after a system starts multi-source energy collection and builds a power supply capability; entering a low-power-consumption monitoring state based on the baseline model, and when abnormal disturbance is detected, awakening a master controller and generating a multi-dimensional abnormal evidence set; through collaborative collection and intelligent management of multi-source energy, long-term stable operation of equipment under the condition of no external power supply is realized, and the deployment and maintenance cost is remarkably reduced; a dynamic baseline is constructed through online learning of an incremental Gaussian mixture model, and omen identification and accurate judgment of environmental abnormality are realized by combining multi-dimensional evidence fusion of environmental gradient tensor, acoustic Doppler and the like, so that the false alarm rate and the missing report rate are greatly reduced; through a hierarchical response mechanism integrating abnormal index calculation and context awareness, intelligent adjustment of alarm intensity and a communication strategy is realized.

Description

Technical Field

[0001] This invention relates to the technical field of electronic alarms, and more particularly to a self-powered environmental parameter adaptive electronic alarm and its alarm method. Background Technology

[0002] With the rapid development of IoT technology, environmental monitoring systems are increasingly widely used in fields such as industrial safety, smart homes, smart agriculture, warehousing and logistics, underground utility tunnels, and field facilities. Electronic alarms, as core devices for anomaly early warning, require intelligent features, reliability, and deployment flexibility as key technical indicators. Traditional environmental parameter alarms typically rely on mains power or periodic battery replacements, resulting in complex wiring, high maintenance costs, and difficulty in deployment in remote or concealed areas. Especially in scenarios with unstable power supply or difficult maintenance, the devices are prone to failure due to power outages, severely impacting monitoring continuity and safety.

[0003] Traditional environmental parameter alarms generally suffer from several problems: reliance on external power sources or batteries limits deployment and increases maintenance costs; fixed alarm thresholds fail to adapt to complex and changing environments, leading to frequent false alarms or missed alarms; lack of intelligent learning capabilities makes it difficult to distinguish between normal fluctuations and genuine anomalies; a significant conflict between system power consumption and performance makes it difficult for self-powered devices to operate stably for extended periods; and the absence of self-diagnostic functions prevents timely calibration or early warning when sensors drift or energy harvesting efficiency declines. Furthermore, existing technologies are mostly limited to single-parameter monitoring and passive response mechanisms, lacking multi-source information fusion, precursor recognition, contextual awareness, and closed-loop optimization capabilities, resulting in a low overall level of intelligence and making it difficult to meet the application requirements of unattended operation, remote monitoring, and long-term maintenance-free operation. Summary of the Invention

[0004] In view of the problems existing in the above-mentioned self-powered adaptive environmental parameter electronic alarm and its alarm method, the present invention is proposed.

[0005] Therefore, the purpose of this invention is to provide a self-powered environmental parameter adaptive electronic alarm and its alarm method, the purpose of which is: To solve the above-mentioned technical problems, the present invention provides the following technical solution: including, after the system starts multi-source energy acquisition and establishes power supply capability, triggering online learning of environmental parameters to construct a dynamic baseline model; Based on the baseline model, it enters a low-power monitoring state. When an abnormal disturbance is detected, it wakes up the main controller and generates a multi-dimensional abnormal evidence set. A comprehensive anomaly index is calculated based on the aforementioned abnormal evidence set, and corresponding graded alarm instructions are generated based on the index results. The alarm command is executed and combined with context information to perform multimodal response and remote communication, while generating a complete event log; The system uses the event log to perform self-diagnosis and model updates, optimizes the baseline and strategy, and then returns to the listening state to achieve closed-loop adaptive operation.

[0006] As a preferred embodiment of the self-powered environmental parameter adaptive electronic alarm and its alarm method described in this invention, the multi-source energy acquisition includes one or more of solar photovoltaic, thermoelectric power generation, piezoelectric vibration energy and radio frequency energy, and the acquired energy is rectified, boosted and dynamically routed by an intelligent power management unit, with real-time energy being prioritized for system operation, and redundant energy stored in a hybrid energy storage system of supercapacitor and lithium battery.

[0007] As a preferred embodiment of the self-powered environmental parameter adaptive electronic alarm and its alarm method described in this invention, the intelligent power management unit further predicts the future energy supply trend based on ambient light, temperature gradient and vibration intensity, and dynamically adjusts the system sampling frequency and communication strategy to achieve energy supply and demand balance.

[0008] As a preferred embodiment of the self-powered environmental parameter adaptive electronic alarm and its alarm method described in this invention, the online learning of the environmental parameters adopts an incremental Gaussian mixture model (GMM), continuously updating the mean and variance of each parameter through a sliding time window, and combining time tags to establish day-night, weekly cycle and seasonal change patterns to form a dynamic baseline model.

[0009] As a preferred embodiment of the self-powered environmental parameter adaptive electronic alarm and its alarm method described in this invention, the dynamic baseline model is used to generate an adaptive alarm threshold band, the upper and lower limits of which are dynamically adjusted according to the current time and historical statistical characteristics, and supports adaptive optimization of the safety factor based on user feedback or scene type.

[0010] As a preferred embodiment of the self-powered environmental parameter adaptive electronic alarm and its alarm method described in this invention, the multidimensional anomaly evidence set includes parameter deviation, environmental gradient tensor change rate, acoustic Doppler frequency shift, and energy acquisition disturbance signal. The environmental gradient tensor is constructed by the spatial gradient and time change rate of distributed nodes and is used to identify precursors of abnormal evolution.

[0011] As a preferred embodiment of the self-powered environmental parameter adaptive electronic alarm and its alarm method described in this invention, the comprehensive anomaly index is calculated by weighted fusion of various evidence items. The weights are dynamically configured based on sensor reliability, historical false alarm rate and current situation, and a nonlinear mapping function is introduced to enhance the response of high deviation items.

[0012] As a preferred embodiment of the self-powered environmental parameter adaptive electronic alarm and its alarm method described in this invention, the context information includes the current time, personnel presence status, user alarm response history, and ambient light intensity. The system dynamically adjusts the audible and visual alarm intensity, communication priority, and local prompting mode according to the context.

[0013] As a preferred embodiment of the self-powered environmental parameter adaptive electronic alarm and its alarm method described in this invention, the system self-diagnosis includes sensor zero-point drift detection, communication link success rate analysis, energy storage unit health assessment and photovoltaic surface contamination judgment, and triggers self-calibration, alarm or maintenance prompt based on the diagnostic results.

[0014] As a preferred embodiment of the self-powered environmental parameter adaptive electronic alarm and its alarm method described in this invention, it includes a multi-source energy acquisition module, a hybrid energy storage unit, a multi-parameter sensor array, an edge intelligent processing unit, and a multi-modal alarm and communication module. The alarm is configured to execute the alarm method as described in any one of claims 1 to 9, achieving external power supply-free, adaptive decision-making, and closed-loop optimized operation.

[0015] The beneficial effects of this invention are as follows: Through multi-source energy collaborative acquisition and intelligent management, the device achieves long-term stable operation without external power supply, significantly reducing deployment and maintenance costs; by constructing a dynamic baseline through online learning using an incremental Gaussian mixture model, and combining environmental gradient tensor and acoustic Doppler multi-dimensional evidence fusion, it achieves early warning identification and accurate judgment of environmental anomalies, greatly reducing false alarm and missed alarm rates; through a comprehensive anomaly index calculation and context-aware hierarchical response mechanism, it achieves intelligent adjustment of alarm intensity and communication strategy, improving user experience and system reliability; through an event-driven low-power architecture and a closed-loop self-diagnosis and self-updating mechanism, it achieves adaptive optimization and maintenance-free operation of the system in complex environments. The overall technical solution is highly intelligent, integrated, and practical, suitable for various scenarios such as industrial monitoring, smart homes, and field security. Attached Figure Description

[0016] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort. Wherein: Figure 1 This is a schematic diagram of the process structure of the present invention. Detailed Implementation

[0017] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0018] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of the invention. Therefore, the invention is not limited to the specific embodiments disclosed below.

[0019] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in different places in this specification does not necessarily refer to the same embodiment, nor is it a single or selective embodiment that is mutually exclusive with other embodiments.

[0020] Secondly, the present invention is described in detail with reference to the schematic diagrams. When detailing the embodiments of the present invention, for ease of explanation, the cross-sectional views illustrating the device structure may be partially enlarged, not according to the usual scale. Furthermore, the schematic diagrams are merely examples and should not limit the scope of protection of the present invention. In addition, actual fabrication should include three-dimensional spatial dimensions of length, width, and depth. Example

[0021] Reference Figure 1 As an embodiment of the present invention, a self-powered environmental parameter adaptive electronic alarm and its alarm method are provided. This device includes...

[0022] After the system initiates multi-source energy harvesting and establishes power supply capability, it triggers online learning of environmental parameters to build a dynamic baseline model. Based on the baseline model, it enters a low-power monitoring state. When an abnormal disturbance is detected, it wakes up the main controller and generates a multi-dimensional abnormal evidence set. A comprehensive anomaly index is calculated based on the aforementioned abnormal evidence set, and corresponding graded alarm instructions are generated based on the index results. The alarm command is executed and combined with context information to perform multimodal response and remote communication, while generating a complete event log; The system uses the event log to perform self-diagnosis and model updates, optimizes the baseline and strategy, and then returns to the listening state to achieve closed-loop adaptive operation.

[0023] After system startup, the multi-source energy harvesting module is first activated, including solar photovoltaic, thermoelectric power generation, piezoelectric vibration, and radio frequency energy harvesting units. The intelligent power management unit rectifies, boosts, and dynamically routes the harvested, weak energy, prioritizing the use of real-time generated electricity to drive system operation. Excess energy is stored in a hybrid energy storage system composed of supercapacitors and lithium batteries. When the energy storage voltage reaches the main control unit's startup threshold, the system determines it has stable power supply capability and then triggers an online learning process for environmental parameters. The multi-parameter sensor array begins collecting environmental data such as temperature, humidity, gas concentration, light intensity, and noise at preset intervals. The edge intelligent processing unit uses an incremental Gaussian mixture model (GMM) to continuously learn from the collected data, updating each parameter through a sliding time window. By combining the mean and variance with time labels to establish diurnal, weekly, and seasonal variation patterns, a baseline model reflecting the dynamic characteristics of the real environment is constructed. After modeling, the system automatically enters a low-power monitoring state, with the main control core entering deep sleep and only the co-core remaining to monitor sensor front-end signals and energy fluctuations at microwatt-level power consumption. When abnormal disturbances such as parameter rate of change exceeding limits, acoustic Doppler frequency shift anomalies, or energy harvesting disturbances are detected, the co-core immediately wakes up the main control unit and initiates rapid synchronous sampling to acquire the current environmental state vector. Simultaneously, it combines the spatial gradient information of distributed nodes to calculate the Frobenius norm of the environmental gradient tensor to identify abnormal evolution trends. Finally, a multi-dimensional anomaly evidence set containing parameter deviation, gradient rate of change, motion characteristics, and energy disturbances is generated as... The decision input control unit runs a weighted fusion algorithm based on this evidence set to calculate a comprehensive anomaly index. The weight of each evidence item is dynamically adjusted according to sensor reliability, historical false alarm rate, and current situation, and a nonlinear mapping function is introduced to enhance the response sensitivity to high deviation items. The system generates corresponding graded alarm commands based on the numerical range of the comprehensive anomaly index. When the index is at a medium level, a level one warning is triggered with slow LED flashing. When the index reaches a high threshold, a level two alarm is activated with audible and visual warnings and a warning message is sent via LoRa. When the index exceeds the emergency threshold, a level three response is executed, including high-intensity flashing, high-frequency buzzer, and concurrent alarm via LoRa and NB-IoT dual channels. While executing the alarm command, the system combines the current time, personnel presence status, light intensity, and user history. The system dynamically adjusts the intensity of sound and light and communication strategies based on contextual information such as historical response behavior to avoid invalid alarms in nighttime or unattended scenarios. It generates a complete event log containing time, location, parameter values, anomaly index, response actions, and communication results. After the alarm process ends, the system reads this event log for self-diagnosis and analysis, assessing whether the sensor has zero-point drift, whether the communication link is stable, the health status of the energy storage unit, and whether there is dust accumulation on the photovoltaic surface. Based on the diagnostic results, it executes a self-calibration procedure or issues a maintenance prompt. Simultaneously, the event data is incorporated into the training set to update the statistical parameters and alarm strategy weights of the dynamic baseline model, optimizing the system's adaptability to environmental changes. After completing the diagnosis and update, the system re-enters a low-power monitoring state, forming a comprehensive system encompassing energy supply, environmental perception, intelligent decision-making, and...The complete closed-loop operation process, from graded response to self-optimization, achieves true self-powering, self-adaptation, self-diagnosis, and sustainable operation.

[0024] Specifically, the multi-source energy harvesting includes one or more of solar photovoltaic, thermoelectric power generation, piezoelectric vibration energy and radio frequency energy. The harvested energy is rectified, boosted and dynamically routed through an intelligent power management unit. Real-time energy is prioritized for system operation, and redundant energy is stored in a hybrid energy storage system of supercapacitors and lithium batteries.

[0025] The intelligent power management unit further predicts future energy supply trends based on ambient light, temperature gradient, and vibration intensity, and dynamically adjusts the system sampling frequency and communication strategy to achieve energy supply and demand balance.

[0026] The online learning of environmental parameters adopts an incremental Gaussian mixture model (GMM), which continuously updates the mean and variance of each parameter through a sliding time window, and combines time tags to establish diurnal, weekly, and seasonal variation patterns to form a dynamic baseline model.

[0027] Furthermore, the dynamic baseline model is used to generate adaptive alarm threshold bands, whose upper and lower limits are dynamically adjusted according to the current time and historical statistical characteristics, and supports adaptive optimization of the safety factor based on user feedback or scenario type.

[0028] The multidimensional anomaly evidence set includes parameter deviation, environmental gradient tensor change rate, acoustic Doppler frequency shift, and energy harvesting perturbation signal. The environmental gradient tensor is constructed by the spatial gradient and temporal change rate of distributed nodes and is used to identify precursors of anomalous evolution.

[0029] Ideally, the comprehensive anomaly index is calculated by weighted fusion of various evidence items, with the weights dynamically configured based on sensor reliability, historical false alarm rate, and current context, and a nonlinear mapping function is introduced to enhance the response of high deviation items.

[0030] Contextual information includes current time, personnel presence status, user alarm response history, and ambient light intensity. The system dynamically adjusts the intensity of the audible and visual alarms, communication priority, and local prompting mode based on the context.

[0031] It should be noted that the system self-diagnosis includes sensor zero-point drift detection, communication link success rate analysis, energy storage unit health assessment, and photovoltaic surface contamination judgment, and triggers self-calibration, alarms, or maintenance prompts based on the diagnostic results.

[0032] It includes a multi-source energy harvesting module, a hybrid energy storage unit, a multi-parameter sensor array, an edge intelligent processing unit, and a multi-modal alarm and communication module. The alarm is configured to execute the alarm method as described in any one of claims 1 to 9, achieving external power supply-free, adaptive decision-making, and closed-loop optimized operation.

[0033] After system startup, the multi-source energy harvesting module is activated first. This module integrates various energy acquisition methods, including solar photovoltaic, thermoelectric power generation, piezoelectric vibration energy harvesting, and radio frequency energy reception, and can adapt to various installation environments, such as indoors and outdoors, fixed or vibrating. The harvested electrical energy is usually an unstable, low-voltage AC or pulsating DC signal, which needs to be rectified, boosted, and regulated by the intelligent power management unit. This management unit adopts a dynamic routing strategy, prioritizing the direct supply of real-time generated energy to the system to reduce energy storage conversion losses. When energy production exceeds immediate demand, excess electrical energy is stored in a hybrid energy storage system consisting of supercapacitors and rechargeable lithium batteries. The supercapacitors are used to handle instantaneous high-power operations (such as communication transmission), while the lithium batteries provide long-term stable energy support, thereby ensuring continuous operation of the system without external power supply.

[0034] After the energy storage system voltage reaches the main control unit's startup threshold, the system enters the initialization phase, activating a multi-parameter sensor array to periodically sample environmental parameters such as temperature, humidity, gas concentration, light intensity, and noise. The edge intelligent processing unit uses an incremental Gaussian mixture model (GMM) to learn the collected data online. By setting a sliding time window, it continuously updates the probability distribution characteristics of each parameter, including mean and variance, and combines time labels (such as hour, day of the week, and season) to establish the diurnal pattern, weekly fluctuation, and seasonal variation patterns of environmental parameters. Ultimately, it constructs a baseline model that can dynamically reflect the normal state of the environment. This model not only captures static thresholds but also identifies "normal changing trends," providing a scientific basis for subsequent adaptive alarms.

[0035] Based on the constructed dynamic baseline model, the system enters a low-power monitoring mode, with the main control MCU core entering deep sleep and only the low-power co-core monitoring critical events. When the sensor front-end detects an excessive rate of change in parameters, the acoustic Doppler module identifies unexpected movement (such as intrusion or loose equipment), or abnormal disturbances occur in the energy acquisition signal, the co-core immediately wakes up the main control unit. After startup, the main control unit quickly acquires current environmental data and, combined with the spatial distribution information of multiple distributed nodes, calculates the joint evolution characteristics of environmental parameters in terms of spatial gradient and temporal rate of change, forming an environmental gradient tensor, and using its Frobenius norm as an indicator of anomaly evolution intensity. Finally, the system generates a multi-dimensional anomaly evidence set, including parameter deviation, gradient rate of change, Doppler frequency shift, and energy disturbance signal, to comprehensively determine the potentiality and severity of the anomaly.

[0036] After obtaining a multidimensional anomaly evidence set, the system calculates the Comprehensive Anomaly Index (WAI) and uses a weighted fusion algorithm to quantify and integrate the various pieces of evidence. The weights of each evidence item are not fixed but dynamically adjusted based on the sensor's historical reliability, false alarm rate, and the current context (e.g., reducing the weight of PM2.5 in a high-pollution environment). Nonlinear mapping functions such as the Sigmoid function are introduced to exponentially enhance the response to significant deviations, ensuring that even early, subtle anomalies can be effectively identified. Based on the WAI value, the system generates tiered alarm commands: a low index triggers a Level 1 warning (e.g., flashing yellow LED), a medium index triggers a Level 2 alarm (rapid red light flashing + buzzer), and a high index triggers a Level 3 emergency response (high-intensity flashing + concurrent multi-channel communication), achieving a progressive response mechanism from mild to severe.

[0037] When executing alarm commands, the system further incorporates contextual information for intelligent adjustment. Contextual information includes the current time (e.g., automatically reducing sound at night), the presence of personnel (via Wi-Fi probes or infrared sensors), user response behavior to historical alarms (e.g., automatically increasing the threshold after frequent false alarms), and ambient light intensity (enhancing flashing lights in dark environments). Based on this information, the system dynamically adjusts the intensity of the audible and visual alarms, selects the optimal communication channel (LoRa or NB-IoT), sets retransmission strategies, and generates a complete event log containing time, location, parameter values, anomaly index, response actions, and communication status. This ensures that the alarm is effective without causing excessive interference, improving user experience and the system's intelligence level.

[0038] After the alarm is executed, the system enters the self-diagnosis and model update phase. By analyzing the event logs, the system assesses the health status of each module: it uses reference source comparison to detect sensor zero-point drift, statistically analyzes communication ACK success rate to determine link stability, monitors charge / discharge curves to assess battery / capacitor health, and analyzes the continuous decline trend of power generation efficiency over multiple days to determine whether dust has accumulated on the photovoltaic surface. Based on the diagnostic results, the system can automatically trigger zero-point calibration, issue maintenance reminders, or temporarily disable faulty sensors. Simultaneously, the event data is incorporated into the training set to update GMM model parameters, and alarm weights and safety coefficients are adjusted based on user feedback to achieve continuous model optimization and policy adaptation, ensuring the accuracy and robustness of the system's long-term operation.

[0039] During operation, the device achieves self-powered operation through multi-source energy harvesting and hybrid energy storage technology. After power supply is established, online learning of environmental parameters is initiated. An incremental Gaussian mixture model is used to construct a dynamic baseline model that integrates time patterns, serving as a reference benchmark for anomaly detection. Based on this model, the system enters a low-power monitoring state. When abnormal signals such as parameter mutations, spatial gradient evolution, and motion disturbances are detected, the main controller is awakened, generating an anomaly evidence set containing multi-dimensional evidence. A weighted fusion algorithm is used to calculate a comprehensive anomaly index, enabling a quantitative assessment of the severity of the anomaly. Based on the index results, a graded alarm command is generated. Combined with contextual information such as current time, personnel status, and user habits, the system intelligently adjusts the sound and light response and communication strategy to improve alarm effectiveness and user experience. After the alarm is completed, the system uses event logs for self-diagnosis, detecting issues such as sensor drift, communication failure, and energy storage attenuation. It then performs self-calibration and model updates to optimize baseline parameters and alarm strategies before finally returning to the monitoring state.

[0040] In summary, multi-source energy harvesting and hybrid energy storage technologies enable long-term stable operation of the equipment without external power supply, significantly reducing deployment and maintenance costs. An online learning mechanism based on an incremental Gaussian mixture model constructs a dynamic baseline model that integrates time-varying patterns, accurately identifying normal environmental fluctuations and abnormal changes, effectively reducing false alarm rates caused by periodic environmental variations. By introducing multi-dimensional evidence such as environmental gradient tensors and acoustic Doppler effects, early perception and comprehensive judgment of precursors to abnormal evolution are achieved, improving alarm sensitivity and foresight. Combined with a context-aware hierarchical response mechanism, alarm intensity and communication strategies can be intelligently adjusted based on factors such as time and personnel presence, balancing warning effectiveness and user experience. Closed-loop self-diagnosis and model update functions enable the system to perform sensor calibration, fault warning, and strategy self-optimization, ensuring long-term reliability and adaptability.

[0041] It is important to note that the constructions and arrangements of this application shown in several different exemplary embodiments are merely illustrative. Although only a few embodiments are described in detail in this disclosure, those who consult this disclosure will readily understand that many modifications are possible (e.g., changes in the size, dimensions, structure, shape, and proportions of various elements, as well as parameter values ​​(e.g., temperature, pressure, etc.), mounting arrangements, use of materials, color, orientation, etc.) without substantially departing from the novel teachings and advantages of the subject matter described in this application). For example, an element shown as integrally formed may be composed of multiple parts or elements, the position of elements may be inverted or otherwise altered, and the nature or number or position of discrete elements may be changed or altered. Therefore, all such modifications are intended to be included within the scope of the invention. The order or sequence of any process or method steps may be changed or rearranged according to alternative embodiments. In the claims, any "device plus function" clause is intended to cover the structure described herein that performs the function, and not only structurally equivalent but also equivalent in structure. Other substitutions, modifications, alterations, and omissions may be made in the design, operation, and arrangement of the exemplary embodiments without departing from the scope of the invention. Therefore, the present invention is not limited to the specific embodiments, but extends to various modifications that still fall within the scope of the appended claims.

[0042] Furthermore, in order to provide a concise description of exemplary embodiments, not all features of actual embodiments (i.e., those features that are not relevant to the currently considered best mode for carrying out the invention, or those features that are not relevant to implementing the invention) may be omitted.

[0043] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A self-powered adaptive environmental parameter electronic alarm and its alarm method, characterized in that: Comprising, After the system starts multi-source energy harvesting and establishes power supply capability, online learning of environmental parameters is triggered to build a dynamic baseline model; Based on the baseline model, a low-power listening state is entered, and when an abnormal disturbance is detected, the main control is awakened and a multi-dimensional abnormal evidence set is generated; According to the abnormal evidence set, a comprehensive abnormality index is calculated, and corresponding hierarchical alarm instructions are generated according to the index results; The alarm instructions are executed, and multi-modal responses and remote communications are carried out in combination with situational information, while complete event logs are generated; Using the event logs, system self-diagnosis and model updating are completed, and after optimizing the baseline and strategy, the system returns to the listening state, realizing closed-loop adaptive operation.

2. The self-powered, self-adapting environmental parameter alarm and its alarm method according to claim 1, characterized in that: The multi-source energy harvesting includes one or more of solar photovoltaic, temperature difference power generation, piezoelectric vibration energy and radio frequency energy, and the collected energy is rectified, boosted and dynamically routed by an intelligent power management unit. The real-time energy is preferentially used for system operation, and the redundant energy is stored in a super capacitor and lithium battery hybrid energy storage system.

3. The self-powered, self-adapting environmental parameter alarm and its alarm method according to claim 2, characterized in that: The intelligent power management unit further predicts future energy supply trends based on environmental light, temperature gradient and vibration intensity, and dynamically adjusts the system sampling frequency and communication strategy to achieve energy supply and demand balance.

4. The self-powered, self-adapting environmental parameter alarm and its alarm method according to claim 3, characterized in that: The online learning of environmental parameters uses an incremental Gaussian mixture model (GMM), which continuously updates the mean and variance of each parameter through a sliding time window, and establishes day-night, week cycle and seasonal change patterns in combination with time labels to form a dynamic baseline model.

5. The self-powered, self-adapting environmental parameter alarm and its alarm method according to claim 4, characterized in that: The dynamic baseline model is used to generate an adaptive alarm threshold band, the upper and lower limits of which are dynamically adjusted according to the current time and historical statistical characteristics, and supports adaptive optimization of safety factors based on user feedback or scene type.

6. A self-powered environmental parameter adaptive electronic alarm and its alarming method according to claim 5, characterized in that: The multi-dimensional abnormal evidence set includes parameter deviation, environmental gradient tensor change rate, acoustic Doppler frequency shift amount and energy harvesting disturbance signal, wherein the environmental gradient tensor is constructed by the spatial gradient and time change rate of the distributed nodes, and is used to identify abnormal evolution precursors.

7. A self-powered environmental parameter adaptive electronic alarm and its alarming method according to claim 6, characterized in that: The comprehensive abnormality index is calculated by weighted fusion of each evidence item, the weight is dynamically configured according to sensor reliability, historical false alarm rate and current situation, and a nonlinear mapping function is introduced to enhance the response of high deviation items.

8. The self-powered, self-adapting environmental parameter alarm and its alarm method according to claim 7, characterized in that: The situational information includes current time, personnel presence status, user alarm response history and environmental light intensity, and the system dynamically adjusts the sound and light alarm intensity, communication priority and local prompt mode according to the situation.

9. The self-powered, environmentally parameter-adaptable electronic alarm and its alarming method according to claim 8, characterized in that: The system self-diagnosis includes sensor zero drift detection, communication link success rate analysis, energy storage unit health evaluation and photovoltaic surface pollution judgment, and triggers self-calibration, alarm or maintenance prompt according to the diagnosis results.

10. The self-powered, self-adapting environmental parameter alarm and its alarm method according to claim 9, characterized in that: Comprising a multi-source energy harvesting module, a hybrid energy storage unit, a multi-parameter sensor array, an edge intelligent processing unit, a multi-modal alarm and communication module, the alarm is configured to execute the alarm method as claimed in any one of claims 1 to 9, realizing external power-free, adaptive decision and closed-loop optimized operation.

Images