Multi-dimensional linkage dwelling house whole-house safety and health examination system and multi-dimensional linkage dwelling house whole-house safety and health examination device

Abstract

The invention relates to the technical field of smart home, and discloses a multi-dimensional linkage residence whole house safety and health examination system and device, and the system comprises a detection device layer, a transmission network layer, a cloud platform layer, and an interaction application layer. The detection device layer is integrated with a glass explosion-proof detection unit, a kitchen fire utilization detection unit, a home-entry power supply safety detection unit, an indoor air and deodorization detection unit, an indoor sound environment detection unit and a key area water utilization detection unit. And the transmission network layer executes a heterogeneous network convergence strategy through the communication link arbitration unit, and integrates wireless and wired communication links. The cloud platform layer comprises a linkage analysis grading early warning module, and the linkage analysis grading early warning module operates a light and force fusion judgment model of a glass state and a micro-leakage diagnosis control process. According to the invention, the false alarm is eliminated through multi-source data fusion, and the crossing from passive alarm to active safety intervention and beforehand health early warning is realized.

Description

Technical Field

[0001] The present invention relates to the technical field of smart home, and particularly to a multi-dimensional linkage residential whole-house safety and health examination system and device. Background Art

[0002] Existing residential safety monitoring systems usually rely on a single wireless communication protocol to build a data transmission network. In complex residential building structures, reinforced concrete walls have a shielding and attenuation effect on high-frequency wireless signals, which easily leads to communication blind spots or unstable connections. Especially in serious disasters such as fires, the power supply lines are often damaged first, resulting in the power-off failure of the wireless router. Once a single communication link is interrupted, the alarm information cannot be uploaded to the cloud server, making the survival ability and data transmission reliability of the system under extreme working conditions unable to be guaranteed.

[0003] In terms of the accuracy of security detection and linkage control, traditional monitoring devices mainly rely on single-dimensional sensing signals for judgment. Existing glass break detectors usually only rely on detecting the vibration amplitude to trigger an alarm, and non-destructive vibrations generated by external thunder, heavy object falling or door closing are extremely likely to cause false alarms. Frequent false alarms will reduce users' trust in the system. At the same time, common gas leakage alarm devices usually only have local audible and visual warning functions, and cannot actively cut off the indoor power supply circuit when detecting that the gas leakage concentration exceeds the standard. At this time, if indoor electrical equipment automatically runs or the switch operates to generate electric sparks, it is extremely easy to ignite the accumulated gas and cause a secondary explosion accident. This passive monitoring mode lacks a key active safety intervention mechanism.

[0004] In terms of hidden danger investigation and health management, existing waterway and circuit monitoring technologies mainly focus on post-event alarms for sudden failures. Conventional turbine water flow sensors have a starting flow blind spot and cannot identify tiny leaks or drips with extremely small flow rates in the main water pipeline. Long-term micro-leaks not only cause water resource waste but may also lead to wall dampness and mildew. In addition, existing systems lack the ability to analyze the long-term trend of equipment operating status, and cannot effectively identify the flow attenuation trend caused by micro-blockage of the pipeline or the tiny abnormal current fluctuations caused by aging of the line insulation. They can only issue an alarm after the water supply pipe bursts or the circuit is short-circuited and burned out, and it is difficult to discover potential hidden dangers in advance through in-depth mining of historical data, and cannot meet the user's need for pre-event preventive maintenance of the health status of residential facilities. Summary of the Invention

[0005] Aiming at the deficiencies of the prior art, the present invention provides a multi-dimensional linkage residential whole-house safety and health examination system and device, which solves the problem that the existing residential safety monitoring system has unreliable data transmission under complex building structures or fire power-off conditions due to a single communication link.

[0006] To achieve the above objectives, the present invention provides the following technical solution:

[0007] The first aspect of this invention provides a multi-dimensional linkage whole-house safety and health check system, including a kitchen fire safety detection unit, which is configured in the kitchen area of ​​the residence for collecting methane concentration data;

[0008] The power supply safety detection unit is installed in the power distribution box and is used to collect circuit status data and perform power-off operations; the glass explosion-proof safety detection unit is installed in the window area and is used to collect mechanical vibration status and impact acceleration data.

[0009] A key area water use detection unit is configured in the main water inlet pipeline to collect instantaneous water flow and static water pressure data; a network communication transmission layer is used to transmit data and instructions; a cloud platform layer includes a linkage analysis and hierarchical early warning module; the linkage analysis and hierarchical early warning module is used to receive and parse the methane concentration data, the circuit status data, the mechanical vibration status, the impact acceleration data, the instantaneous water flow and the static water pressure data, and generate control instructions based on the parsing results.

[0010] In the specific configuration of the transport network layer, the transport network layer includes a data transmission center module. This module contains a communication link arbitration unit, which executes a heterogeneous network fusion strategy. Logically, this strategy integrates three physical communication links: WiFi 6 wireless LAN, LoRaWAN low-power wide area network, and power line carrier wired network. The communication link arbitration unit evaluates the channel quality of each physical communication link in real time and dynamically allocates transmission paths based on channel quality and the type of data being transmitted.

[0011] The kitchen fire safety detection unit includes a catalytic combustion gas sensor and a K-type thermocouple temperature sensor; the catalytic combustion gas sensor and the K-type thermocouple temperature sensor are physically integrated into the fire safety composite detection component; the fire safety composite detection component is physically installed on the lower edge of the range hood cover. The key area water usage detection unit includes a turbine-type water flow sensor, a diffused silicon water pressure sensor, and a contact humidity sensor; the turbine-type water flow sensor and the diffused silicon water pressure sensor are physically installed on the main water pipe inside the cabinet. The indoor air and odor detection unit includes an odor sensor; the odor sensor is physically installed on the surface of the kitchen ceiling panel or the top edge of the wall cabinet. The incoming power supply safety detection unit includes a Hall current sensor, a voltage sensor, and a residual current sensor. The Hall current sensor and the residual current sensor adopt a split-type openable magnetic ring physical structure. The Hall current sensor is physically sleeved on the outer insulation layer of the phase wire of the incoming main power supply conductor. The residual current sensor is also physically sleeved on the outside of the assembly of the phase wire and the neutral wire of the incoming main power supply conductor. The voltage sensor adopts an insulation piercing clamp physical structure. The insulation piercing clamp is physically clamped on the incoming main power supply conductor.

[0012] The living room acoustic environment detection unit is physically deployed at the center of the living room ceiling; this unit includes an omnidirectional condenser microphone, which is physically fixed to the ceiling surface using a ceiling mount. The glass explosion-proof safety detection unit is physically deployed on the tempered glass window sash; it includes a polarized optical stress sensor and a MEMS impact sensor; the polarized optical stress sensor is physically bonded to the center of the inner surface of the tempered glass window sash; the MEMS impact sensor is physically bonded to the side of the polarized optical stress sensor's housing or to the aluminum alloy frame of the window sash. The odor sensor in the indoor air and odor detection unit is physically installed on the upper part of the bathroom wall or the surface of the ceiling panel; it collects data on the concentration of odor gases accumulated in the upper part of the bathroom. In the key area water use detection unit, the contact humidity sensor is physically attached to the outer area of ​​the flange ring of the toilet base, the inspection port area at the bottom of the bathtub, or the low-lying tile surface around the floor drain. The contact humidity sensor adopts a flexible flat cable structure or a thin film encapsulation structure, and the sensing surface of the contact humidity sensor directly establishes physical contact with the surface of the ground material. The contact humidity sensor detects the change in dielectric constant of the contact surface.

[0013] In terms of logic control and linkage, the system achieves safety judgment in multiple scenarios through the cloud platform layer. The linkage analysis and hierarchical early warning module receives and parses methane concentration data from the kitchen fire safety detection unit; the linkage analysis and hierarchical early warning module compares the methane concentration data with the explosion-proof safety threshold; when the methane concentration data continuously exceeds the explosion-proof safety threshold for a duration that reaches the confirmation window length, the linkage analysis and hierarchical early warning module determines that a confirmed gas leak event has occurred and generates a power outage trigger signal; the linkage analysis and hierarchical early warning module generates an emergency fuse command based on the power outage trigger signal; the emergency fuse command is sent to the power supply safety detection unit via the transmission network layer; the power supply safety detection unit receives the emergency fuse command and drives the shunt trip coil of the intelligent circuit breaker to cut off the power supply circuit.

[0014] A second aspect of this invention provides a multi-dimensional interconnected whole-house safety and health checkup device, which utilizes the multi-dimensional interconnected whole-house safety and health checkup system provided in the first aspect. The multi-dimensional interconnected whole-house safety and health checkup device includes a catalytic combustion gas sensor assembly installed in the kitchen area to collect methane concentration data in the environment; and a smart circuit breaker assembly installed in the main electrical distribution box to receive commands and drive the shunt trip coil to cut off the power supply circuit.

[0015] A microelectromechanical system (MEMS) impact sensor assembly is installed on the surface of a window glass to collect instantaneous acceleration amplitude; a turbine-type water flow sensor assembly and a diffused silicon water pressure sensor assembly are installed on the main water inlet pipe to collect instantaneous water flow and static water pressure, respectively; and a composite transmission gateway assembly is installed inside the residence to transmit the data collected by the components to the cloud platform layer via a power line carrier transmission module.

[0016] This invention provides a multi-dimensional, interconnected whole-house safety and health check system and device. It has the following beneficial effects:

[0017] 1. This invention configures a communication link arbitration unit in the transmission network layer, integrating WiFi 6, LoRaWAN, and power line carrier networks through a heterogeneous network fusion strategy. The communication link arbitration unit dynamically allocates transmission paths based on channel quality, and the multi-link redundancy mechanism effectively overcomes the weakness of single wireless signals in penetrating walls. In extreme situations where a fire causes a power outage in the wireless router, alarm data can still be uploaded to the cloud platform layer through the backup link, improving the communication reliability of the whole-house security monitoring system.

[0018] 2. This invention utilizes a cloud-based platform to run a light and force fusion judgment model for glass states, combined with high-frequency audio energy ratio analysis to achieve multi-dimensional cross-validation. The system only confirms an alarm when both physical impact and audio characteristics simultaneously meet the conditions, eliminating false alarms caused by thunder or falling heavy objects. Furthermore, based on methane concentration data, the system triggers an emergency fuse command in the in-home power supply safety detection unit, proactively cutting off the power supply circuit to prevent secondary explosions caused by electrical sparks after a gas leak, thus achieving a leap from passive monitoring to proactive safety protection.

[0019] 3. This invention utilizes a linkage analysis and hierarchical early warning module to execute a micro-leakage diagnosis and control process based on water pressure fluctuations. During periods of water inactivity, it combines the pressure decay rate with the humidity change rate to sensitively detect minute leaks in the main water pipeline, solving the problem that traditional flow meters cannot detect dripping leaks. Simultaneously, the system runs a trend prediction algorithm based on a long short-term memory network, analyzing time series data to identify micro-blockages or aging trends in the pipeline, issuing health warnings before faults occur, thus achieving proactive maintenance. Attached Figure Description

[0020] Figure 1 This is a system block diagram of the present invention;

[0021] Figure 2 This is a system architecture diagram of the present invention;

[0022] Figure 3 This is a layout diagram of the kitchen scene transmission device of the present invention;

[0023] Figure 4 This is a layout diagram of the bedroom and main electrical box scene transmission device of the present invention;

[0024] Figure 5 This is a layout diagram of the living room and balcony scene transmission device of the present invention;

[0025] Figure 6 This is a layout diagram of the bathroom scene transmission device of the present invention.

[0026] Legend

[0027] 1. Data transmission center module; 2. In-home distribution box; 4. Odor sensor; 5. Communication module; 6. Power line carrier transmission module; 7. Catalytic combustion gas sensor; 8. K-type thermocouple temperature sensor; 9. Turbine-type water flow sensor; 10. Main water pipe; 11. Diffused silicon water pressure sensor; 12. Omnidirectional capacitive microphone; 13. Hall current sensor; 14. Voltage sensor; 15. Residual current sensor; 16. Four-in-one air quality sensor; 17. MEMS impact sensor; 18. Polarized optical stress sensor; 19. Contact humidity sensor. Detailed Implementation

[0028] The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0029] See attached document Figure 1 and attached Figure 2 This invention provides a multi-dimensional, interconnected whole-house safety and health check system. It consists of four layers: a detection device layer, a transmission network layer, a cloud platform layer, and an interactive application layer. The detection device layer, located at the bottom of the architecture, is responsible for collecting multi-dimensional analog signals from the physical environment and converting them into digital signals. The transmission network layer, located in the middle layer, constructs the data uplink and command downlink channels. The cloud platform layer, located at the core of the architecture, performs data storage, fusion calculations, and logical judgments. The interactive application layer, located at the top layer, performs visualization and human-computer interaction.

[0030] The detection device layer includes a glass explosion-proof safety detection unit, a kitchen fire safety detection unit, a key area water usage detection unit, a power supply safety detection unit, an indoor air and odor detection unit, and a living room acoustic environment detection unit. The glass explosion-proof safety detection unit is physically deployed on the inner surface of the residential window glass and includes a polarized optical stress sensor 18 and a MEMS (Micro-Electro-Mechanical Systems) impact sensor 17. The kitchen fire safety detection unit is physically deployed above the kitchen gas stove and includes a catalytic combustion gas sensor 7 and a K-type thermocouple temperature sensor 8. The key area water usage detection unit is physically deployed at key nodes of the water supply pipeline and on the ground, and includes a turbine-type water flow sensor 9, a diffused silicon water pressure sensor 11, and a contact humidity sensor 19. The power supply safety detection unit is physically deployed inside the power distribution box 2 and includes a Hall current sensor 13, a voltage sensor 14, and a residual current sensor 15. The indoor air quality and odor detection unit is physically deployed at the breathing zone height and odor source areas in each room. The indoor air quality and odor detection unit includes a four-in-one air quality sensor 16 and an odor sensor 4. The living room acoustic environment detection unit is physically deployed in the living room ceiling area. The living room acoustic environment detection unit includes an omnidirectional condenser microphone 12.

[0031] The transmission network layer consists of a power line carrier transmission module, a WiFi 6 wireless communication module 5, and a LoRaWAN wireless communication module 5. The power line carrier transmission module is physically connected to the 220V AC power line in the residence, using the power line as the signal transmission medium. The WiFi 6 wireless communication module 5 establishes high-bandwidth wireless LAN coverage. The LoRaWAN wireless communication module 5 establishes low-power wide-area network coverage. Data collected by the detection device layer is uploaded via one or more paths selected from the power line carrier transmission module, WiFi 6 wireless communication module 5, or LoRaWAN wireless communication module 5 according to a preset routing strategy. The cloud platform layer includes a multi-source data fusion storage module and a linkage analysis and hierarchical early warning module. The multi-source data fusion storage module receives data packets from the transmission network layer, parses, cleans, and persistently stores the data packets. The linkage analysis and hierarchical early warning module reads historical and real-time data from the multi-source data fusion storage module, runs a preset logical algorithm model, and generates early warning signals and control commands.

[0032] The interactive application layer includes a visualized health profile module, a multi-channel early warning receiving module, and a personalized control module. The visualized health profile module retrieves data processed by the cloud platform layer to generate graphical reports. The multi-channel early warning receiving module receives alarm information output by the linkage analysis and hierarchical early warning module. The personalized control module sends user-defined threshold parameters and control commands to the cloud platform layer; the control commands are then transmitted via the transmission network layer to the in-home power supply safety detection unit or associated actuators. To quantitatively describe the convergence state of multi-source data in the system topology, a full-dimensional state vector S of the system at time t is defined. topology (t). Full-dimensional state vector S topology (t) represents the real-time data set of all detection nodes in the system topology:

[0033] S topology (t)=

[0034] {U glass (t),U fire (t),U water (t),U power (t),U air (t),U sound (t)};

[0035] In the formula: U glass (t) is defined as the subset of stress and impact data uploaded by the glass explosion-proof safety testing unit at time t; U fire (t) is defined as a subset of gas concentration and temperature data uploaded by the kitchen fire safety detection unit at time t; U water(t) is defined as a subset of flow, pressure, and humidity data uploaded by the water use monitoring unit in the key area at time t; U power (t) is defined as a subset of current, voltage, and leakage current data uploaded by the power supply safety detection unit at time t; U air (t) is defined as a subset of gas concentration data uploaded by the indoor air and odor detection unit at time t; U sound (t) is defined as a subset of sound pressure level and spectrum data uploaded by the living room sound environment detection unit at time t.

[0036] See attached document Figure 2 and appendix Figure 3 To be continued Figure 6 The glass explosion-proof safety testing unit, kitchen fire safety testing unit, key area water use testing unit, household power supply safety testing unit, indoor air and odor prevention testing unit, and living room sound environment testing unit are physically distributed in different spatial nodes according to the functional division of the residence, forming a whole-house covering sensor network.

[0037] The data transmission center module 1 is physically deployed in the geometric center area of ​​the residential floor plan. It is connected to a wall power outlet or placed inside a low-voltage distribution box. The data transmission center module 1 establishes a wireless communication signal field covering the interior space of the residence. The kitchen fire safety detection unit is physically deployed in the kitchen area. The catalytic combustion gas sensor 7 and the K-type thermocouple temperature sensor 8 are fixedly installed on the surface of the range hood above the gas stove using a magnetic base or adhesive backing. The installation positions of the catalytic combustion gas sensor 7 and the K-type thermocouple temperature sensor 8 are within a spatial range with a radius of 30 to 50 centimeters centered on the center point of the gas stove.

[0038] The water usage detection unit for key areas is physically deployed around the water supply facilities in the kitchen, bathroom, and balcony areas. In the kitchen area, a turbine-type water flow sensor 9 and a diffused silicon water pressure sensor 11 are installed in series or in parallel on the main water inlet pipe inside the cabinet below the sink. In the bathroom area, a contact-type humidity sensor 19 is physically attached to the joint between the toilet base and the floor tiles or in a low-lying area around the drain. In the balcony area, the diffused silicon water pressure sensor 11 is physically connected to the water inlet of the washing machine faucet. The power supply safety detection unit is physically deployed inside the enclosed space of the main power distribution box 2. The Hall current sensor 13 and the residual current sensor 15 adopt an open-loop ring structure. The Hall current sensor 13 and the residual current sensor 15 are directly sleeved on the outside of the insulation layer of the main power supply conductor, without cutting off the main power supply conductor. The voltage sensor 14 physically contacts the metal conductor of the main power supply conductor through an insulation piercing clamp or a magnetic probe.

[0039] The glass explosion-proof safety detection unit is physically deployed on the inner surface of the window glass in the bedroom and balcony areas. A polarized optical stress sensor 18 is physically attached to the geometric center of the window glass using an optically transparent adhesive. A MEMS impact sensor 17 is physically attached to the side of the polarized optical stress sensor 18 or to the corner structure of the window frame. The indoor air and odor detection unit is physically deployed in the bedroom, living room, and bathroom areas. In the bedroom area, a four-in-one air quality sensor 16 is placed on a bedside table or at breathing zone height. In the bathroom area, an odor sensor 4 is fixedly installed on the upper part of the wall, at least 2 meters above the ground, or near the exhaust fan vent.

[0040] The living room acoustic environment detection unit is physically deployed at the center of the ceiling in the living room area. An omnidirectional condenser microphone 12 is physically fixed to the ceiling surface using a ceiling mount. The omnidirectional condenser microphone 12 collects omnidirectional sound wave signals within the living room area. Power line carrier transmission modules are distributed across wall-mounted power outlets in various functional areas. These modules are physically plugged into the wall-mounted power outlets via standard power plugs, establishing an electrical connection with the building's internal power line network. The data output ports of each detection unit are connected to the nearest power line carrier transmission module's data input port via wired cables or short-range wireless radio frequency signals.

[0041] See attached document Figure 3 The kitchen multi-dimensional safety monitoring subsystem adopts a spatial three-dimensional distributed architecture, physically divided into an upper fire monitoring area, a lower water monitoring area, and a top air monitoring area. In the upper fire monitoring area, a catalytic combustion gas sensor 7 and a K-type thermocouple temperature sensor 8 are physically integrated on the same mounting bracket, forming a fire safety composite detection component. This component is fixed to the lower edge of the range hood's cover using high-temperature resistant double-sided adhesive or a magnetic base. The physical installation position of the fire safety composite detection component is vertically above the center of the gas stove burner, with a vertical distance between the component and the burner surface maintained between 300 mm and 500 mm. The catalytic combustion gas sensor 7 collects methane and carbon monoxide concentration signals from the environment. The K-type thermocouple temperature sensor 8 collects real-time thermal field temperature signals above the gas stove.

[0042] In the lower water monitoring area, a turbine-type water flow sensor 9 and a diffused silicon water pressure sensor 11 are physically installed on the main water pipe 10 inside the cabinet. The main water pipe 10 is the cold water inlet pipe below the kitchen sink. The turbine-type water flow sensor 9 is connected to the main water pipe 10 in series, and the water flow in the main water pipe 10 directly drives the impeller of the turbine-type water flow sensor 9 to rotate. The diffused silicon water pressure sensor 11 is connected to the main water pipe 10 via a bypass using a T-connector, and the pressure-sensing diaphragm of the diffused silicon water pressure sensor 11 directly contacts the water in the main water pipe 10. In the upper air monitoring area, an odor sensor 4 is physically installed on the surface of the kitchen ceiling panel or the top edge of the wall cabinet. The odor sensor 4 is located in an area 100 mm to 200 mm below the kitchen ceiling. The odor sensor 4 collects residual odors from cooking fumes or putrid odors from kitchen waste.

[0043] The kitchen multi-dimensional safety monitoring subsystem includes a communication module 5 and a power line carrier transmission module 6. These modules are physically integrated into a single power adapter housing, forming a composite transmission gateway. The composite transmission gateway is physically plugged into a 220V power outlet on the kitchen wall. A catalytic combustion gas sensor 7, a K-type thermocouple temperature sensor 8, and an odor sensor 4 are connected to the composite transmission gateway via signal wires. A turbine-type water flow sensor 9 and a diffused silicon water pressure sensor 11 are connected to the composite transmission gateway via waterproof signal lines routed along the back panel of the cabinet. To accurately quantify the real-time flow status of the kitchen water system, the turbine-type water flow sensor 9 outputs a pulse signal frequency f. flow With instantaneous water flow rate Q water Follows a linear transformation relationship:

[0044]

[0045] In the formula: f flow Defined as the frequency of the pulse signal output by the turbine-type water flow sensor 9, measured in Hertz (Hz); K meter The meter coefficient of the turbine-type water flow sensor 9 is defined as the number of pulses per liter, and the meter coefficient is determined by the sensor's factory calibration parameters; 60 is defined as the time conversion constant for converting flow rate per second to flow rate per minute. The microprocessor in the composite transmission gateway converts the pulse signal frequency into an instantaneous water flow rate value according to the above formula, and sends the instantaneous water flow rate value to the system cloud via the power line carrier transmission module.

[0046] See attached document Figure 4 The non-intrusive monitoring subsystem for household power supply mainly consists of a Hall current sensor 13, a residual current sensor 15, and a voltage sensor 14.

[0047] The Hall current sensor 13 and the residual current sensor 15 adopt a split-type opening and closing magnetic ring physical structure. This structure includes a fixed half-ring and a movable half-ring, which are physically connected by a mechanical hinge. In the installed state, the movable half-ring is closed and locked by a snap-fit, forming a closed magnetic circuit for the magnetic core of either the Hall current sensor 13 or the residual current sensor 15. The Hall current sensor 13 is physically fitted onto the outer insulation layer of the phase wire of the main power supply conductor. The Hall current sensor 13 utilizes the Hall effect to sense the intensity of the alternating magnetic field flowing through the phase wire, thereby outputting an induced electromotive force signal. The residual current sensor 15 is simultaneously physically fitted onto the outer side of the assembly of the phase and neutral wires of the main power supply conductor. The residual current sensor 15 detects the vector sum of the phase current and the neutral current. When the circuit is normal, the vector sum is close to zero; when leakage occurs, the residual current sensor 15 outputs a non-zero induced current signal. The installation process of Hall current sensor 13 and residual current sensor 15 does not require cutting off the main power supply line or stripping the insulation layer of the main power supply line.

[0048] Voltage sensor 14 employs an insulation piercing clamp physical structure. The insulation piercing clamp is physically held onto the main power supply conductor entering the household. A metal piercing probe is installed inside the clamp. During clamping, the metal piercing probe pierces the outer insulation layer of the main power supply conductor and establishes electrical contact with the internal metal conductor. Voltage sensor 14 acquires the real-time voltage signal of the incoming power supply. The signal output terminals of Hall current sensor 13, residual current sensor 15, and voltage sensor 14 are connected to a local data acquisition module via shielded cables. The data acquisition module has a built-in analog-to-digital converter that synchronously and discretely samples the analog voltage and analog current signals. To accurately quantify the real-time active power of the household electrical load, the data acquisition module performs discrete integration calculations. Active power P active The calculation logic is as follows:

[0049]

[0050] In the formula: N is defined as the total number of discrete sampling points within a single power frequency cycle, and the total number of sampling points is determined by the system sampling frequency; u[n] is defined as the instantaneous voltage value of the nth sampling point, in volts; i[n] is defined as the instantaneous current value of the nth sampling point, in amperes. The data acquisition module uses the above formula to calculate the active power data, and the active power data and the leakage current data output by the residual current sensor 15 are sent to the cloud platform layer via the power line carrier transmission module 6.

[0051] See attached document Figure 5The glass explosion-proof and acoustic environment monitoring subsystem is physically divided into a ceiling sound field monitoring node and a window glass status monitoring node. The ceiling sound field monitoring node is physically deployed at or near the geometric center of the living room ceiling. An omnidirectional condenser microphone 12 is physically fixed to the surface of the ceiling plasterboard or concrete slab using a ceiling-mounted bracket. The microphone's pickup port faces downwards, and it collects 360-degree omnidirectional audio signals within the living room space. The omnidirectional condenser microphone 12 is connected to the pre-installed ceiling light fixture power line via a power adapter or to a power outlet on the upper part of the wall via an extension cable. The power line carrier transmission module is physically integrated inside the power adapter of the omnidirectional condenser microphone 12.

[0052] Window glass condition monitoring nodes are physically deployed on tempered glass window sashes in balconies or bedrooms. A polarized optical stress sensor 18 is physically bonded to the central area of ​​the inner surface of the tempered glass window sash using optical-grade epoxy resin adhesive. The polarized optical stress sensor 18 includes a polarized light source emitter and a light intensity receiver, and detects the birefringence optical path difference after the light beam passes through the glass medium. A MEMS impact sensor 17 is physically bonded to the side of the housing of the polarized optical stress sensor 18 or to the aluminum alloy frame of the window sash. The MEMS impact sensor 17 contains a triaxial microelectromechanical accelerometer, and collects transient vibration acceleration signals when the glass is impacted by an external force.

[0053] The signal output terminals of the polarized optical stress sensor 18 and the MEMS impact sensor 17 are connected to the local signal processing unit via a flexible ribbon cable. The local signal processing unit is physically fixed to the edge of the window frame. The local signal processing unit contains a built-in microprocessor that digitizes the acquired raw analog signals. To quantify the residual stress state inside the glass, the microprocessor performs calculations based on the photoelastic principle. The stress value σ inside the glass is... glass The calculation logic is as follows:

[0054]

[0055] In the formula: R opt Defined as the optical path difference detected by the polarized optical stress sensor 18, in nanometers; C mat Defined as the photoelastic coefficient of glass materials, measured in nanometers per centimeter and megapascals (MPa), the photoelastic coefficient is determined by the chemical composition of the glass; L path Defined as the effective optical path length of the detection beam propagating inside the glass, measured in centimeters, the effective optical path length is equal to the physical thickness of the glass. To quantify the instantaneous impact intensity on the glass, the microprocessor performs vector synthesis of the triaxial acceleration data. The synthesized acceleration a resultant The calculation logic is as follows:

[0056]

[0057] In the formula: a x Defined as the X-axis acceleration component value acquired by MEMS impact sensor 17; a y Defined as the Y-axis acceleration component value acquired by MEMS impact sensor 17; a z Defined as the Z-axis acceleration component value acquired by the MEMS impact sensor 17. The microprocessor transmits the calculated stress value inside the glass and the composite acceleration data to the power line carrier transmission module in the living room via a wireless radio frequency signal, and then uploads them to the cloud platform.

[0058] See attached document Figure 6 The bathroom vertical layer monitoring subsystem divides the space vertically into an upper gas sensing zone and a lower liquid sensing zone based on the physical density characteristics of the substances being measured. In the upper gas sensing zone, the odor sensor 4 is physically installed on the upper part of the bathroom wall or the surface of the ceiling panel. The physical installation height of the odor sensor 4 is within a range of at least 2000 mm from the ground and within 500 mm horizontally from the exhaust fan inlet. The odor sensor 4 integrates a metal oxide semiconductor gas-sensitive element, which is sensitive to the adsorption of ammonia, hydrogen sulfide, and volatile organic compounds. The odor sensor 4 collects data on the concentration of odor gases accumulated in the upper part of the bathroom.

[0059] In the lower liquid sensing area, the contact humidity sensor 19 is physically attached to the outer area of ​​the flange ring on the toilet seat, the access panel area at the bottom of the bathtub, or the low-lying tile surface around the floor drain. The contact humidity sensor 19 employs a flexible flat cable structure or a thin-film encapsulation structure, with its sensing surface directly in physical contact with the floor surface. Unlike traditional probe-type water immersion sensors, the contact humidity sensor 19 senses minute water molecule penetration by measuring changes in the dielectric constant of the contact surface. It can detect micro-leakage occurring within the structural layer even when no water has formed on the floor.

[0060] The communication link of the bathroom vertical layer monitoring subsystem adopts power line carrier technology. Communication module 5 is physically integrated with the power line carrier transmission module and plugged into a splash-proof power socket next to the bathroom sink. Since bathroom walls are typically covered with ceramic tiles, which shield and attenuate wireless radio frequency signals, the power line carrier transmission module utilizes a 220V power line embedded in the wall to transmit digital signals, thus avoiding the wireless signal shielding problem. Odor sensor 4 and contact humidity sensor 19 are connected to communication module 5 via concealed wiring. To accurately distinguish between water vapor condensation from normal bathing and structural leaks caused by aging pipe flanges, the local microprocessor performs a judgment calculation based on the humidity change rate. The humidity change rate R output by contact humidity sensor 19... leak The calculation logic is as follows:

[0061]

[0062] Where: M surf (t) is defined as the relative moisture content of the contact surface measured by the contact humidity sensor 19 at the current sampling time t, in percentage; M surf (t-Δt) is defined as the relative moisture content of the contact surface measured by the contact humidity sensor 19 at the previous sampling time t-Δt, in percentage form; Δt is defined as the sampling time interval in minutes. The microprocessor packages the calculated humidity change rate data with the gas concentration data collected by the odor sensor 4, and the data packet is sent to the system cloud platform via the power line carrier transmission module. The system cloud platform determines whether there is a risk of microleakage based on the persistence of the humidity change rate.

[0063] See attached document Figure 1 Appendix Figure 2 and appendix Figure 6 The power line carrier transmission module is physically connected to the AC power distribution network inside the residence. The power line carrier transmission module uses the power frequency of 50 Hz or 60 Hz power lines as the transmission medium for high-frequency data carriers.

[0064] The power line carrier transmission module internally includes a signal coupling circuit, a bandpass filter, and an orthogonal frequency division multiplexing (OFDM) modem. The signal coupling circuit uses a high-voltage ceramic capacitor and a high-frequency signal transformer to construct an isolated coupling channel. The signal coupling circuit physically isolates the power frequency high voltage, allowing only the high-frequency carrier signal to pass. The bandpass filter is physically connected between the signal coupling circuit and the OFDM modem, filtering out power frequency harmonic interference and impulse noise interference on the power line. The OFDM modem performs data modulation and demodulation operations. At the transmitting end, the OFDM modem maps the serial digital signal to be transmitted onto multiple orthogonal subcarriers through serial-to-parallel conversion. The power line carrier transmission module superimposes the modulated high-frequency signal onto the power frequency voltage waveform. The superimposed total power line voltage signal V... total The mathematical expression logic of (t) is as follows:

[0065]

[0066] In the formula: V grid Defined as the voltage amplitude of industrial frequency alternating current, measured in volts; F grid Defined as the fundamental frequency of the power frequency AC, with a value of 50 or 60 Hz; t is defined as the time variable, with a unit of seconds; M is defined as the total number of effective subcarriers used in the orthogonal frequency division multiplexing system; k is defined as the index number of the subcarrier, ranging from 1 to M; A k f is defined as the signal amplitude of the k-th subcarrier, in volts; k Defined as the center frequency of the kth subcarrier, in Hertz, with the center frequency located in the frequency band of 2 MHz to 30 MHz; Defined as the initial phase of the k-th subcarrier, in radians, the initial phase carries the modulated data information.

[0067] At the network topology control level, the system adopts a master-slave communication architecture. The power line carrier unit located in the data transmission center module 1 is configured as the central coordinator, and the power line carrier transmission modules located in each scenario-specific detection device are configured as site agents. The central coordinator periodically broadcasts beacon frames on the power lines. After power-on reset, the site agents automatically search for beacon frames and initiate network access requests to the central coordinator. The central coordinator assigns a unique short logical address to each legitimate site agent and establishes a logical mapping table. To ensure data packet integrity in complex power line noise environments, the power line carrier transmission module employs a forward error correction coding mechanism. The orthogonal frequency division multiplexing modem adds Reed-Solomon codes or convolutional codes as redundancy check bits before transmitting data. The power line carrier transmission module at the receiving end uses the redundancy check bits to detect and correct bit errors caused by pulse interference during transmission. When the bit error rate at the receiving end exceeds a preset threshold, the power line carrier transmission module at the receiving end sends a retransmission request command to the sending end, triggering the automatic retransmission mechanism.

[0068] See attached document Figure 1 and appendix Figure 2 The heterogeneous network fusion strategy logically integrates three physical communication links: WiFi 6 wireless LAN, LoRaWAN low-power wide area network, and power line carrier wired network. This strategy is executed by the communication link arbitration unit within the data transmission center module 1. The communication link arbitration unit is responsible for real-time evaluation of the channel quality of each physical communication link and dynamically allocates transmission paths based on channel quality and the type of transmitted data.

[0069] The WiFi 6 wireless communication module 5 constructs a high-speed data transmission channel, used to transmit broadband audio data and system firmware update packages collected by the living room sound environment detection unit. The LoRaWAN wireless communication module 5 constructs a high-penetration control command channel, used to transmit sensor status data and control commands distributed in the bedroom and kitchen corners. The power line carrier transmission module constructs an anti-shielding backup channel, used to maintain data connectivity when the wireless radio frequency signal is blocked by metal or severely attenuated by walls. The communication link arbitration unit periodically reads the physical layer parameters of each communication interface. Based on the physical layer parameters, the communication link arbitration unit calculates the real-time link quality score S for each communication link. link (k). Real-time link quality score S link The calculation logic for (k) is as follows:

[0070]

[0071] In the formula: P rssi (k) is defined as the current received signal strength indication value of the k-th physical communication link, in decibels and milliwatts; P refDefined as the absolute value of the reference signal strength, which represents the maximum received power under ideal conditions, and is measured in decibels (dB) and milliwatts (mW); R loss (k) is defined as the packet loss rate of the k-th physical communication link in the most recent statistical window, and its value ranges from 0 to 1; ω rssi Defined as a weighting coefficient for signal strength, this coefficient is used to adjust the weighting of signal strength in the score; ω plr Defined as a weighting coefficient for packet loss rate, and satisfying ω rssi +ω plr =1.

[0072] The communication link arbitration unit has a built-in path selection decision table. When the data to be transmitted is audio stream data, the communication link arbitration unit first checks the real-time link quality score of the Wi-Fi link. If the real-time link quality score of the Wi-Fi link is lower than a preset bandwidth threshold, the communication link arbitration unit switches the data stream to the power line carrier transmission module for transmission. When the data to be transmitted is sensor status packet data, the communication link arbitration unit simultaneously sends data to the LoRaWAN wireless communication module 5 and the power line carrier transmission module. The data transmission center module 1 processes the received data packets according to the first-come, first-served principle, thereby eliminating multipath effects and improving the reliability of data arrival. This multimodal parallel transmission mechanism ensures that the whole-house security monitoring network remains logically online even if a single network fails or is interfered with.

[0073] See attached document Figure 1 and appendix Figure 5 The optical-force fusion judgment model for glass states receives raw sensor data from the glass explosion-proof safety detection unit. By calculating the combined acceleration value of the residual stress inside the glass and the external impact, the model classifies the physical state of the glass in real time. The optical-force fusion judgment model for glass states first performs optical calculations of the stress data. The model reads the phase delay data uploaded by the polarized optical stress sensor 18. According to the laws of photoelasticity, the average plane stress S inside the glass... glass There is a linear mapping relationship between (t) and the phase delay data. Mean plane stress S glass The calculation logic for (t) is as follows:

[0074]

[0075] In the formula: Defined as the phase delay detected by the polarized optical stress sensor 18, in radians; λ is defined as the operating wavelength of the polarized light source, in nanometers; C opt Defined as the photoelastic coefficient of glass materials, with units of nanometers per centimeter and megapascals; d glass2π is defined as the physical path length of the light beam penetrating the glass, i.e., the glass thickness, in centimeters; 2π is defined as the geometric constant for converting phase and optical path difference.

[0076] The light and force fusion judgment model in the glass state simultaneously performs vector synthesis of impact data. The model reads the three-axis orthogonal acceleration components uploaded by MEMS impact sensor 17. To eliminate the influence of sensor installation angle deviation on impact amplitude judgment, the model calculates the Euclidean norm of the three-axis acceleration. The synthesized impact acceleration A... shock The calculation logic for (t) is as follows:

[0077]

[0078] In the formula: a x (t),a y (t),a z (t) is defined as the raw real-time acceleration value output by the MEMS impact sensor 17 on the X, Y, and Z axes; g x ,g y ,g z The gravitational acceleration component is defined as the component calibrated by the MEMS impact sensor 17 in a static state. Introducing the gravitational acceleration component is to remove the interference of static gravity on dynamic impact calculations. The core of the light and force fusion judgment model for the glass state lies in the decision logic of executing multi-parameter fusion. The model defines two state indicator variables: a self-explosion risk warning signal W. risk With broken intrusion alarm signal A break The model compares the mean plane stress S... glass (t) and the preset safety stress threshold S th To generate a self-destruction risk warning signal W risk If the internal stress of the glass remains high for an extended period, it indicates a risk of spontaneous breakage. Spontaneous breakage risk warning signal W risk The logical expression for the decision is as follows:

[0079]

[0080] In the formula: S th Defined as the critical threshold of safe stress for glass materials, the critical threshold of safe stress is set according to the national standard for tempered glass. The model synthesizes the impact acceleration A through joint analysis. shock (t) and the rate of stress change are used to generate a broken intrusion alarm signal A. break A simple impact signal is insufficient to determine glass breakage; such an impact could be from a non-destructive event like a soccer ball collision. The physical characteristic of broken glass is the instantaneous release of internal stress accompanied by a strong impact. Breakage Intrusion Alarm Signal A break The logical expression for the decision is as follows:

[0081]

[0082] In the formula: τ is defined as the length of the time window for stress change detection, in seconds; ΔS drop Defined as the drop threshold for instantaneous stress release, in megapascals (MPa); ∧ is defined as the logical AND operator; |·| is defined as the absolute value operator. The light and force fusion judgment model for glass only outputs a breakage intrusion alarm signal with a logic value of 1 when the combined impact acceleration exceeds the threshold and a sudden stress change is detected simultaneously. This dual-modal fusion mechanism filters out false alarms that only involve impact without stress release, as well as events involving only slow thermal expansion and contraction causing stress changes.

[0083] See attached document Figure 1 A trend prediction algorithm based on long short-term memory networks is used to process water flow time-series data uploaded by water use monitoring units in key areas and current time-series data uploaded by household power supply safety monitoring units. The linkage analysis and hierarchical early warning module identifies flow attenuation trends caused by minor pipe blockages or abnormal current fluctuation trends caused by line aging by analyzing long-term dependencies in the time series.

[0084] The linkage analysis hierarchical early warning module first constructs a time sliding window, serializing continuously acquired sensor values ​​into a model input vector. For the current time t, the model input vector contains historical observations from time tw to time t, where w is the length of the sliding window. The model input vector is then fed into the storage units of the Long Short-Term Memory (LSTM) network. The LTM network regulates information flow through three gating mechanisms: the forget gate, the input gate, and the output gate. The linkage analysis hierarchical early warning module performs update calculations within the LTM units. The forget gate determines which information from the previous time step is discarded, the input gate determines which new information is updated to the cell state, and the output gate determines what value is output based on the current cell state. Cell state C t With the hidden layer output h t The update logic is as follows:

[0085] f t =σ(W f ·[h t-1 ,x t ]+b f );

[0086] i t =σ(W i ·[h t-1 ,x t ]+b i );

[0087]

[0088] ot =σ(W o ·[h t-1 ,x t ]+b o );

[0089] h t =o t ⊙tanh(C t );

[0090] In the formula: t is defined as the current time step number; f t Defined as the activation vector of the forget gate, with values ​​ranging from 0 to 1; i t Defined as the activation vector of the input gate, with values ​​ranging from 0 to 1; t Defined as the activation vector of the output gate, with values ​​ranging from 0 to 1; Defined as the candidate cell state vector at the current moment, with values ​​ranging from -1 to 1; C t Defined as the cell state vector updated at the current time; C t -1 is defined as the cell state vector of the previous time step; h t Defined as the hidden layer output vector at the current time step, this vector is the predicted feature output at the current time step; h t-1 Defined as the hidden layer output vector of the previous time step; x t Defined as the sensor input data at the current moment, including instantaneous water flow rate or RMS current value; W f W i W c W o Defined as the weight matrix corresponding to the forget gate, input gate, candidate states, and output gate, the weight matrix is ​​obtained through training on historical data; b f ,b i ,b c ,b o Defined as the corresponding bias vector; σ is defined as the Sigmoid activation function; tanh is defined as the hyperbolic tangent activation function; ⊙ is defined as the Hadamard product. The linkage analysis and hierarchical early warning module uses a trained long short-term memory network model to output predicted values ​​for the next k time steps. To quantify the degree of functional degradation of physical facilities, the linkage analysis and hierarchical early warning module calculates the trend decay rate index. Taking water pipe flow monitoring as an example, the trend decay rate D... decay The calculation logic is as follows:

[0091]

[0092] In the formula: Defined as the average flow rate statistics within the historical baseline period, which is selected as the stable operation period during the initial stage of equipment installation; Defined as the average of the traffic forecasts for the next k time steps output by the model. The linked analysis and hierarchical early warning module monitors the trend decay rate in real time. When the trend decay rate D... decay When the value shows a monotonically increasing trend and exceeds the preset maintenance threshold (e.g., 15%), the linkage analysis and hierarchical early warning module generates a maintenance suggestion signal for slow pipe blockage or line overload aging. The maintenance suggestion signal does not trigger an emergency audible and visual alarm, and is only pushed to the user's visual health profile module.

[0093] See attached document Figure 1 Appendix Figure 3 and appendix Figure 4 The gas concentration data collected by the kitchen fire safety detection unit is logically linked to the actuator of the power supply safety detection unit to eliminate the potential hazard of electrical sparks igniting leaked gas. The linkage control process begins with the kitchen fire safety detection unit. The catalytic combustion gas sensor 7 collects methane concentration data in the kitchen environment at a sampling frequency of once per second. The catalytic combustion gas sensor 7 converts the analog voltage signal into a digital concentration value. The digital concentration value is packaged via a composite transmission gateway and sent to the cloud platform layer via a power line carrier transmission module.

[0094] The cloud platform layer's linkage analysis and hierarchical early warning module receives and parses methane concentration data. This module compares real-time methane concentration data with a preset explosion-proof safety threshold. To prevent false alarms, the linkage analysis and hierarchical early warning module introduces a time-integration confirmation mechanism. When the methane concentration data continuously exceeds the explosion-proof safety threshold for a duration equal to the preset confirmation window length, the linkage analysis and hierarchical early warning module determines that a confirmed gas leak event has occurred. To quantify the judgment logic, a power-off trigger signal S is defined. trip (t). Power-off trigger signal S trip The generation logic of (t) follows the following Boolean expression:

[0095]

[0096] In the formula: C gas (t) is defined as the real-time methane concentration value uploaded by the catalytic combustion gas sensor 7 at time t, in percentage LEL (lower explosive limit); C limit Defined as a preset explosion-proof safety threshold, such as 10% LEL; W is defined as the total number of time steps in the confirmation window; k is defined as the index variable for time backtracking. Defined as an indicator function, the function value is 1 when the condition within the parentheses is met, and 0 otherwise; N confirm Defined as the minimum number of samples that must meet the exceeding condition within the confirmation window; this minimum number of samples is used to filter out transient interference signals; ∧ is defined as the logical AND operator. Once power is off, the trigger signal S... tripWhen the value of (t) changes to 1, the linkage analysis and hierarchical early warning module immediately generates the highest priority emergency circuit breaker command. The emergency circuit breaker command includes the target device address code and the action operation code. The target device address code points to the intelligent circuit breaker control unit physically deployed in the inlet distribution box 2.

[0097] The emergency circuit breaker command is transmitted to the in-home power supply safety detection unit via the power line carrier transmission module. The microprocessor within the in-home power supply safety detection unit receives and decodes the emergency circuit breaker command. The microprocessor outputs a high-level drive signal to the shunt trip coil of the smart circuit breaker. The shunt trip coil is energized, generating mechanical thrust, which drives the circuit breaker operating handle to trip instantly, cutting off the power supply circuit to the kitchen area. After the in-home power supply safety detection unit performs the tripping action, Hall current sensor 13 and voltage sensor 14 immediately detect the circuit status. If the current and voltage in the kitchen circuit are detected to be zero, the in-home power supply safety detection unit sends a circuit breaker confirmation message to the cloud platform layer. Upon receiving the circuit breaker confirmation message, the cloud platform layer pushes a strong alarm message for automatic power-off protection against gas leaks to the user's mobile phone through the interactive application layer. The entire linkage process is completed within milliseconds, ensuring that electrical ignition sources are eliminated before the gas concentration reaches the explosive limit.

Claims

1. A multi-dimensional, interconnected whole-house residential safety and health check system, characterized in that, include: A kitchen fire safety detection unit is installed in the kitchen area of ​​a residence to collect methane concentration data; The power supply safety detection unit is installed in the power distribution box and is used to collect circuit status data and perform power-off operations. The glass explosion-proof safety detection unit is installed in the window area to collect mechanical vibration and impact acceleration data. The key area water use detection unit is installed in the main water inlet pipe to collect instantaneous water flow and static water pressure data; An indoor air and odor detection unit is installed in the kitchen and bathroom areas to collect data on specific indoor air components. An indoor acoustic environment detection unit is installed in the living room and bedroom areas to collect indoor noise data; The network communication transport layer is used to transmit data and instructions; The cloud platform layer includes a linkage analysis and hierarchical early warning module; the linkage analysis and hierarchical early warning module is used to receive and parse the methane concentration data, the circuit status data, the mechanical vibration status, the impact acceleration data, the instantaneous water flow rate and the static water pressure data, and generate control commands based on the parsing results.

2. The multi-dimensional interconnected whole-house safety and health check system for residences according to claim 1, characterized in that, The kitchen fire safety detection unit includes a catalytic combustion gas sensor. The linkage analysis and hierarchical early warning module is used to compare the methane concentration data with the preset explosion-proof safety threshold. The linkage analysis and hierarchical early warning module is used to introduce a time integration confirmation mechanism. When the duration for which the methane concentration data continuously exceeds the explosion-proof safety threshold reaches the preset confirmation window length, a definite gas leak event is determined to have occurred. The linkage analysis and hierarchical early warning module is used to generate an emergency circuit breaker command after a confirmed gas leak event is determined.

3. The multi-dimensional interconnected whole-house safety and health check system for residences according to claim 2, characterized in that, The household power supply safety detection unit includes a smart circuit breaker control unit and a smart circuit breaker. The intelligent circuit breaker has a shunt trip coil; the intelligent circuit breaker control unit is used to receive the emergency fuse command and output a drive signal to the shunt trip coil; the shunt trip coil is used to generate mechanical thrust after being energized to drive the intelligent circuit breaker to cut off the power supply circuit.

4. The multi-dimensional interconnected whole-house residential safety and health check system according to claim 3, characterized in that, The power supply safety detection unit also includes a Hall current sensor and a voltage sensor. The Hall current sensor and the voltage sensor are used to detect the circuit status after the smart circuit breaker performs an action. The power supply safety detection unit is used to send a fuse confirmation message to the cloud platform layer when it detects that the current and voltage have both returned to zero.

5. The multi-dimensional interconnected whole-house safety and health check system for residences according to claim 1, characterized in that, It also includes a multi-source data fusion storage module and a living room acoustic environment detection unit; The glass explosion-proof safety detection unit includes a microelectromechanical system (MEMS) impact sensor. The microelectromechanical system impact sensor is used to record the timestamp of the impact when it detects that the window glass has been hit by an external force, and to send a suspected impact event message to the cloud platform layer. The linkage analysis and hierarchical early warning module is used to send a data retrieval instruction to the multi-source data fusion and storage module based on the suspected collision event message; The multi-source data fusion storage module is used to extract audio data streams that cover the timestamp from the audio cache queue uploaded by the living room sound environment detection unit.

6. The multi-dimensional interconnected whole-house safety and health check system for residences according to claim 5, characterized in that, The linkage analysis and hierarchical early warning module is used to perform a fast Fourier transform on the audio data stream to obtain the spectrum signal; The linkage analysis and hierarchical early warning module is used to calculate the high-frequency energy ratio of the spectrum signal; the linkage analysis and hierarchical early warning module is used to simultaneously check whether the physical impact intensity exceeds the preset destructive impact acceleration threshold, and whether the high-frequency energy ratio exceeds the preset glass breakage high-frequency energy ratio threshold. The linkage analysis and hierarchical early warning module is used to confirm an illegal intrusion or sabotage event when both the physical impact intensity exceeds the destructive impact acceleration threshold and the high-frequency energy ratio exceeds the high-frequency energy ratio threshold for glass breakage are met simultaneously.

7. The multi-dimensional interconnected whole-house safety and health check system for residences according to claim 1, characterized in that, The water usage detection unit for key areas includes a turbine-type water flow sensor and a diffused silicon water pressure sensor. The linkage analysis and hierarchical early warning module is used to enter the micro-leakage diagnosis mode when the instantaneous water flow rate collected by the turbine water flow sensor remains at zero for a period of time exceeding a preset silent window. The diffused silicon water pressure sensor is used to collect the static water pressure in the micro-leakage diagnostic mode; the linkage analysis and hierarchical early warning module is used to calculate the pressure decay rate of the static water pressure.

8. A multi-dimensional, interconnected whole-house residential safety and health check-up system according to claim 7, characterized in that, It also includes a bathroom vertical layer monitoring subsystem, which contains a contact humidity sensor; The contact humidity sensor is used to monitor the trend of relative permittivity change to reflect the rate of humidity change. The linkage analysis and hierarchical early warning module is used to invoke the humidity change rate when the pressure decay rate exceeds the preset natural fluctuation threshold. The linkage analysis and hierarchical early warning module is also used to generate a micro-leakage diagnosis signal based on the pressure decay rate and the humidity change rate.

9. A multi-dimensional, interconnected residential whole-house safety and health check system according to claim 8, characterized in that, The linkage analysis and hierarchical early warning module is also used to execute the following judgment logic: When the pressure decay rate exceeds the natural fluctuation threshold and the humidity change rate shows an increase in humidity, a physical leak is determined to have occurred and a high-risk diagnostic signal is generated. When the pressure decay rate exceeds the natural fluctuation threshold but the humidity change rate shows no significant change in ambient humidity, a latent pressure loss is determined and a potential risk warning message is generated.

10. A multi-dimensional, interconnected whole-house residential safety and health check-up device, characterized in that, A multi-dimensional, interconnected whole-house safety and health check system for residences, applicable to any one of claims 1-9, comprising: A catalytic combustion gas sensor assembly, installed in the kitchen area, is used to collect methane concentration data in the environment; The intelligent circuit breaker assembly, installed in the household distribution box, is used to receive commands and drive the shunt trip coil to cut off the power supply circuit; A microelectromechanical system (MEMS) impact sensor assembly is mounted on the surface of a window glass to collect instantaneous acceleration amplitude. A turbine-type water flow sensor assembly and a diffused silicon water pressure sensor assembly are installed on the main water inlet pipe and are used to collect instantaneous water flow and static water pressure, respectively. A composite transmission gateway component, installed in a residence, is used to send data collected by the component to the cloud platform layer via a power line carrier transmission module.

Images