Formaldehyde and total volatile organic compound detection device and detection method

Abstract

The invention relates to the technical field of intelligent sensing systems, and discloses a formaldehyde and total volatile organic compound detection device and detection method.The formaldehyde and total volatile organic compound detection device comprises a pretreatment filter screen with a hydrolyzable hydrophilic coating on the surface; the pressure difference monitoring module is used for outputting a pressure difference monitoring value reflecting the permeability of the pretreatment filter screen; the reference gas providing module is used for continuously generating and conveying clean air; the differential sensing module is used for providing a background reference signal for the processor; and the processor is used for dynamically compensating the detection value of the gas sensor according to the physical blockage state of the pretreatment filter screen. According to the invention, hydrated dust is actively captured and removed, an error is judged by analyzing a physical signal provided by the differential pressure monitoring module and an electrical signal provided by the differential sensing module, and finally, a compensation coefficient is dynamically generated to correct the reading of the detection device in real time, so that the credibility of a detection result of the detection device is improved.

Description

Technical Field

[0001] This invention relates to the field of intelligent sensing system technology, and more specifically, to a formaldehyde and total volatile organic compound detection device and method. Background Technology

[0002] During home renovation, various engineered wood products, adhesives, paints, and other materials continuously release formaldehyde, benzene, and other total volatile organic compounds (TVOCs). Therefore, to ensure residential safety, users often use formaldehyde and TVOC testing devices after renovation to assess the level of indoor pollution and decide when to move in based on the readings.

[0003] However, an often overlooked fact is that the evaluation process itself may become ineffective due to the side effects of renovation. For example, the renovation process often includes intricate finishing work such as on-site polishing and cutting of marble or artificial stone countertops for kitchen cabinets, windowsills, or floors. Although this process does not directly release formaldehyde, it generates a large amount of micron-sized calcium carbonate dust. After the renovation is completed, users usually open windows for ventilation and place detection devices to continuously monitor the gas concentration. Since the calcium carbonate dust has not completely dissipated, it will inevitably adhere to the air inlet of the working detection device.

[0004] The key issue is that this dust is not ordinary dust. When the indoor environment changes drastically, for example, within a few days after polishing, if the local weather changes from dry sunny days to rainy days, the indoor humidity may rise sharply. The high humidity environment will cause a large number of water molecules to be strongly adsorbed on the surface of the calcium carbonate dust attached to the air inlet of the detection device. Under the action of the adsorbed water film, the dust particles agglomerate and stick together through capillary force, and their physical properties change from loose particles to dense, hardened physical blocking layer.

[0005] This barrier, formed by the blocking layer, severely hinders the normal passage of target gas molecules such as formaldehyde in the air to the internal sensor unit. This causes the device readings to initially peak, then gradually decrease as the blockage worsens, eventually stabilizing within a lower, theoretically safe range. Users may mistakenly believe the pollution has been eliminated and move in. In reality, the sensor's detection capability is severely weakened, unable to respond to true concentration exceedances, resulting in a serious risk of false negatives and posing a threat to the user's health and safety. Summary of the Invention

[0006] To address the above problems, the present invention provides the following technical solution:

[0007] In a first aspect, the present invention discloses a formaldehyde and total volatile organic compound (TVOC) detection device, comprising a housing, a gas sensor, and a processor, the device further comprising:

[0008] A pretreatment filter screen is disposed on the air intake path of the device, and its surface has a hydrolytic hydrophilic coating. The hydrophilic coating is used to undergo a hydrolysis reaction and partially dissolve after absorbing moisture to remove hydrated dust attached to its surface.

[0009] A differential pressure monitoring module is located downstream of the pretreatment filter and is used to output a differential pressure monitoring value that reflects the permeability of the pretreatment filter.

[0010] A reference gas supply module is used to continuously generate and deliver clean air;

[0011] A differential sensing module includes a main sensing channel and a reference sensing channel. The main sensing channel detects the sampled gas from the air intake path and provides a detection signal to the processor. The reference sensing channel continuously receives and detects clean air from the reference gas supply module to provide a real-time updated background reference signal to the processor.

[0012] The processor is configured to:

[0013] By continuously comparing the detection signal with the background reference signal and combining it with the differential pressure monitoring value, the physical blockage state caused by the hydration and adhesion of dust on the surface of the pre-treated filter is determined.

[0014] Based on the physical blockage state, the detection value of the gas sensor is dynamically compensated to correct the measurement deviation caused by the physical blockage.

[0015] Furthermore, the processor is also configured to:

[0016] The physical blockage state is determined by analyzing whether the continuous rise of the differential pressure monitoring value and the continuous attenuation of the detection signal relative to the background reference signal satisfy a preset coordinated change relationship.

[0017] Furthermore, the processor is configured to execute a blockage diagnostic model, the input of which includes at least differential pressure monitoring values ​​and differential signal values ​​that vary over time, and the output is a quantitative assessment of the physical blockage status.

[0018] Furthermore, the construction of the blockage diagnostic model includes:

[0019] During the device's lifespan or in a pre-set accelerated aging experiment, the sequence data of the pressure difference monitoring value and differential signal value changing over time are continuously collected, and these are correlated with the known true values ​​of the physical blockage state simulating the dust hydration process.

[0020] Furthermore, the method for dynamically compensating the detected values ​​of the gas sensor includes:

[0021] Based on the quantitative assessment results of the physical blockage state, a real-time compensation coefficient is generated through a predefined mapping rule, and the detection value of the gas sensor is corrected.

[0022] Furthermore, the main sensing channel and the reference sensing channel in the differential sensing module have the same air path structure, sensing element and operating parameters, except for the air intake source.

[0023] Furthermore, the reference gas supply module includes a purification unit configured to reduce the concentrations of formaldehyde and total volatile organic compounds in the input air to below the detection limit of the gas sensor through adsorption, catalysis, or sieving.

[0024] Furthermore, the hydrolyzable hydrophilic coating is at least one of polyvinyl alcohol, hydroxypropyl methylcellulose, or alginate.

[0025] Furthermore, the device also includes:

[0026] The status indication module is used to output status information and maintenance instructions corresponding to different blockage levels based on the determined physical blockage status.

[0027] Secondly, the present invention also discloses a method for detecting formaldehyde and total volatile organic compounds, applied in the aforementioned formaldehyde and total volatile organic compound detection device, the method comprising:

[0028] Continuously acquire differential pressure monitoring values ​​that reflect the permeability of the pre-treatment filter, as well as the detection signal of the main sensing channel and the background reference signal of the reference sensing channel generated by the differential sensing module;

[0029] Based on the trend of the differential pressure monitoring value and the trend of the difference between the detection signal and the background reference signal, the physical blockage state of the pretreatment filter caused by dust hydration and adhesion is diagnosed.

[0030] Based on the diagnosed physical blockage, the raw readings of the gas sensor are dynamically compensated to output a corrected pollutant concentration value.

[0031] Compared with related technologies, the present invention has the following beneficial effects:

[0032] This invention constructs an intelligent sensing system integrating active protection, multi-dimensional sensing, intelligent diagnosis, and measurement correction functions. This system uses a pre-treated filter with a hydrophilic surface to actively manage the key chemical process of dust hydration, which leads to blockage. It also utilizes the properties of the coating material to promote the early conversion and removal of dust, inhibiting the formation of a dense blockage layer at its source. Secondly, by simultaneously monitoring two different chains of evidence—physical pressure difference and electrochemical differential signals—the pressure difference monitoring module provides a direct physical signal indicating blockage, while the differential sensing module, supported by a reference gas module, provides a pure electrical signal with relatively reduced sensor sensitivity. These two signals together verify the real-time degree of blockage in the intake path. The core function of the processor is to continuously compare and fuse these two signals. By analyzing their coordinated change trends, it can accurately isolate the signal attenuation caused by the adhesion of hydrated dust from complex environmental interference and background noise from the natural aging of the sensor, achieving precise fault location. Based on this diagnostic result, it dynamically generates compensation coefficients to correct the readings in real time. Thus, even when the hardware performance has deteriorated, it can still maintain the accuracy of the final output at the software level. This solves the problem of inaccurate detection results of formaldehyde and total volatile organic compounds caused by physical blockage due to the coupling of specific home decoration dust and sudden changes in humidity. It improves the reliability of the detection results and ensures the health and safety of users.

[0033] This invention uses a status indication module to detect the invisible sensor performance degradation (passivation) process caused by physical blockage inside the detection device, and uses multi-dimensional data (CI, ΔP) to indicate this process. t The status is mapped to a clear level that is visible, audible, and readable and displayed to the user, reducing the security risks of silent device failure. Attached Figure Description

[0034] Figure 1 A schematic diagram of the data processing flow of a formaldehyde and total volatile organic compound detection device provided by the present invention;

[0035] Figure 2 This is a schematic flowchart of the steps for a method for detecting formaldehyde and total volatile organic compounds provided by the present invention. Detailed Implementation

[0036] The technical solutions of 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.

[0037] Before describing the embodiments in detail, the air quality detection device will be briefly described first, which will serve as the basis for those skilled in the art to understand the subsequent technical solutions.

[0038] Air quality detection devices typically include a housing, gas sensors, and a processor. The housing surface has a main air inlet (inlet path) and a main air outlet for gas inflow and outflow. The housing surface is also usually equipped with a display screen and status indicator lights for human-machine interaction. The gas sensor is mainly used to detect the concentration of harmful gases in the air. The processor is generally responsible for data acquisition, calculation, logical judgment, and output control, such as issuing an alarm when the concentration of harmful gases is high.

[0039] Please see Figure 1 As shown, this embodiment provides a formaldehyde and total volatile organic compound (TVOC) detection device. The device includes a housing, a gas sensor, a processor, a pretreatment filter, a differential pressure monitoring module, a reference gas supply module, a differential sensing module, and a status indication module. The pretreatment filter actively adsorbs moisture from the environment; the differential pressure monitoring module reflects the permeability of the pretreatment filter; the reference gas supply module continuously generates and delivers clean air; the differential sensing module provides an electrical signal indicating the relative decrease in the sensitivity of the gas sensor; the processor combines the filter's permeability and the electrical signal to determine the physical blockage state and dynamically compensates for the gas sensor's detection value to correct measurement deviations caused by physical blockage; and the status indication module provides early warnings based on the physical blockage state.

[0040] In this embodiment, the housing may be a rectangular or cylindrical shell made of engineering plastic, and its interior is divided into interconnected air intake chambers, sensing and processing chambers, and air exhaust chambers by partitions; multiple air intake holes communicating with the air intake chambers may be opened on the left side or front panel of the housing to form the main air intake port; multiple air exhaust holes communicating with the air exhaust chambers may be opened on the right side or rear panel of the housing to form the main air exhaust port; an air intake path is formed between the main air intake port and the main air exhaust port, penetrating the housing; a display screen and status indicator lights may be provided on the top or front panel of the housing.

[0041] Since the detection device mentioned in this embodiment is mainly for formaldehyde and total volatile organic compounds (TVOC), the gas sensor can be an electrochemical formaldehyde sensor and a metal oxide semiconductor TVOC sensor (in other embodiments, sensors for different gases can be selected). They can be installed side by side on a common circuit board in the device, with the detection surface of their sensitive element facing the airflow direction.

[0042] The processor can be a microcontroller unit based on the ARM-Cortex-M series, such as the STM32 series chip. The microcontroller unit can be mounted on a common circuit board and electrically connected to all peripheral modules such as gas sensors and differential pressure sensors through the printed circuits on the circuit board. It is mainly responsible for data acquisition, calculation, logic judgment and output control.

[0043] A pretreatment filter is disposed in the air intake path of the detection device, and its surface has a hydrolytically hydrophilic coating. The hydrophilic coating is designed to undergo a hydrolysis reaction and partially dissolve after absorbing moisture to remove hydrated dust adhering to its surface. For example, the hydrolytically hydrophilic coating is at least one of polyvinyl alcohol, hydroxypropyl methylcellulose, or alginate.

[0044] Specifically, for the structural design of the filter module, a rectangular plastic frame can be selected, with the functional filter material clamped and fixed inside the frame. The module can be horizontally inserted into the housing through a guide rail or snap-fit ​​structure into the pre-set opening of the partition between the air intake chamber and the sensing and processing chamber. Its installation position ensures that all air entering from the main air intake must pass through the filter 100% before entering the subsequent air path.

[0045] For the filter substrate, non-woven fabric or sintered mesh material with a porosity of more than 85% and an average pore size of less than 10 micrometers can be selected, such as stainless steel mesh or polyethylene mesh, to serve as a supporting skeleton (frame) to ensure that the initial air resistance is less than 50Pa. A dirt collection tank can be placed at the bottom of the filter frame.

[0046] For the application of the hydrophilic coating, in this embodiment, the coating can be polyvinyl alcohol (in other embodiments, hydroxypropyl methylcellulose or alginate can also be selected). The hydrolyzable polymer polyvinyl alcohol is prepared into an aqueous solution with a concentration of 5%-10%. The nonwoven fabric substrate is completely immersed in the solution through an impregnation process, and then cured in an oven at 75°C for 30 minutes, thereby forming a continuous PVA film with a thickness of about 2-3 micrometers on the surface of each fiber. This film is the hydrolyzable coating and gives the filter a lasting hydrophilicity.

[0047] Through the above design, when air containing fine calcium carbonate dust generated during marble polishing flows through a filter with a hydrophilic coating, in a high-humidity environment, water molecules are strongly adsorbed on the surface of the dust particles to form a water film. Within this adsorbed water film, surface hydration and ion dissolution occur, meaning that trace amounts of calcium ions (Ca) will remain on the surface of the calcium carbonate particles. 2+ ) and bicarbonate ions (HC03) - Dissolution causes changes in its surface charge and physicochemical properties.

[0048] The water film covering the surface of dust particles forms capillary bridges at multiple particle contact points, generating strong capillary forces that cause dust particles to adhere to and agglomerate. Furthermore, this continuous, water-rich microenvironment provides the necessary conditions for subsequent physicochemical changes in dust (transformation from a loose calcite form to other forms) and promotes its adhesion to filter fibers or other contaminants.

[0049] Ultimately, this strong adsorption, agglomeration, and potential morphological changes induced by high humidity together lead to the formation of a dense physical blockage layer of dust on the filter surface.

[0050] Under continuous high humidity, polyvinyl alcohol (PVA) coatings undergo significant swelling and dissolution. The mechanism is as follows: numerous hydroxyl groups on the PVA molecular chains form strong hydrogen bonds with water molecules, causing the polymer network to absorb water and swell. As water molecules continue to penetrate, the interaction forces between polymer chains weaken, eventually causing the coating to transform from a gel state into a flowable viscous solution (this process is mainly regulated by the degree of alcoholysis, molecular weight, and crystallinity of PVA; by selecting appropriate PVA specifications, controllable dissolution of the coating can be achieved under a preset continuous high humidity threshold). The viscous solution formed by the coating dissolution encapsulates the hydrated dust, which is then carried away from the filter surface by the shear force of the subsequent airflow and falls into the dust collection tank at the bottom of the filter frame.

[0051] Since the ambient humidity data inside the device needs to be continuously collected throughout the process to determine whether the continuous high humidity conditions that trigger coating hydrolysis and self-cleaning have been reached, a digital humidity sensor can be installed inside the device to collect this data. At the same time, it is also necessary to collect the pressure difference data before and after the filter to be a direct indicator for evaluating the filter clogging status and self-cleaning effect. This data can be provided by the subsequent pressure difference monitoring module.

[0052] In this embodiment, the self-cleaning mechanism ensures that the pressure difference of the filter screen remains at a low level for a long time, guaranteeing smooth airflow for sensor sampling. This is the physical basis for measurement accuracy. Furthermore, when the self-cleaning mechanism is insufficient to cope with extreme dust loads, the pressure difference of the filter screen will show a detectable increase, which provides a key input signal for subsequent intelligent diagnosis.

[0053] The differential pressure monitoring module is located downstream of the pretreatment filter and outputs a differential pressure monitoring value that reflects the permeability of the pretreatment filter. Specifically, this module is a sensing unit used to measure the gas pressure difference between the inlet and outlet sides of the pretreatment filter. Its output value directly and linearly reflects the resistance of the filter to airflow, i.e., permeability.

[0054] In this embodiment, the core of the differential pressure monitoring module is a miniature differential pressure sensor, which can be an integrated MEMS (Micro-Electro-Mechanical Systems) differential pressure sensor chip, such as Infineon's DPS422 or a similar model. This chip can be packaged in a small electronic package, which has two pressure ports (high-pressure port P+ and low-pressure port P-) and digital interface pins (such as I). 2 C).

[0055] The differential pressure monitoring module can be mounted on a circuit board located in the sensing and processing chamber. The digital interface pins of the differential pressure sensor are directly connected to the corresponding I / O pins of the processor (MCU) via traces on the circuit board. 2 The C communication pin is connected to enable power supply and data transmission.

[0056] A flexible micro silicone tube with an inner diameter of approximately 1 mm can be used for the air path connection. One tube guides the air pressure at the front end of the pretreatment filter (inlet chamber) to the P+ port of the differential pressure sensor; the other tube guides the air pressure at the rear end of the pretreatment filter (inlet of the sensor processing chamber) to the P- port. This connection ensures that the differential pressure sensor can directly and accurately measure the pressure loss across the pretreatment filter.

[0057] After the device is calibrated at the factory or a brand new pre-treatment filter is installed, the processor records the pressure difference value at this time in a clean air environment as the initial reference pressure difference value (ΔP0). At this time, ΔP0 is mainly determined by the inherent airflow resistance of the filter substrate, and the value is very small (e.g., <30Pa).

[0058] During operation, the built-in micro-fan maintains a constant flow of air throughout the air path. However, as dust begins to adhere to and hydrate the surface of the pretreatment filter, the effective air passage area decreases, and airflow resistance increases. This causes the pressure at the rear of the filter (measured by the P- port) to decrease relative to the pressure at the front (measured by the P+ port), resulting in a decrease in the real-time differential pressure monitoring value (ΔP) measured by the differential pressure sensor. t The differential pressure increases accordingly, and finally the differential pressure sensor continuously increases ΔP. t It is sent to the processor in the form of a digital signal.

[0059] ΔP is calculated by the processor. t The increase relative to ΔP0 (ΔP t -ΔP0), and analyze its rate of change over time (d(ΔP)). t -ΔP0) / dt), and the continuous and accelerating upward trend is direct evidence of physical blockage.

[0060] In this embodiment, the differential pressure monitoring module is not used for ordinary flow control, but is specially configured to monitor the health status of the pretreatment filter and convert the chemical and physical process of dust hydration and adhesion into a key engineering parameter, namely differential pressure, that can be continuously tracked and quantified.

[0061] Compared to relying solely on the attenuation of gas sensor signals for inference, differential pressure data provides direct physical evidence of filter blockage, improving the reliability of the diagnosis.

[0062] Furthermore, dust adhesion and pressure differential increase are gradual processes, which can be monitored by ΔP. t By detecting subtle growth trends, the system can provide early warnings of filter performance degradation before a significant decrease in the sensitivity of the gas sensor, which is beneficial for subsequent preventative maintenance.

[0063] The reference gas supply module is used to continuously generate and deliver clean air. In this embodiment, the reference gas supply module refers to an independent gas path subsystem capable of continuously providing clean air to the reference sensing channel (located in the subsequent differential sensing module). Its core requirement is that the concentration of the target pollutants (formaldehyde, TVOC) in the provided gas is extremely low and stable.

[0064] Therefore, in this embodiment, the reference gas supply module includes a purification unit, which is configured to reduce the concentration of formaldehyde and total volatile organic compounds in the input air to below the detection limit of the gas sensor through adsorption, catalysis or sieving.

[0065] Specifically, the reference gas supply module includes a purification chamber filled with composite purification packing material. This packing material is composed of a high-efficiency particulate air filter layer, a high specific surface area activated carbon and modified molecular sieve adsorption layer, and a selective catalytic oxidation coating layer, arranged sequentially. The activated carbon and molecular sieves are responsible for the physical adsorption and sieving of most TVOCs, while the catalytic oxidation coating can catalytically decompose small-molecule polar substances such as formaldehyde into carbon dioxide and water at room temperature. This deeply purifies the air entering the reference sensing channel, almost completely removing formaldehyde, total volatile organic compounds, and dust, thus generating clean air.

[0066] In other embodiments, specialized membrane materials with extremely low permeability to the target gas (e.g., certain composite liquid crystal polymer membranes or metal-organic framework material-supported membranes with high barrier properties against organic vapors) can be used to achieve molecular-level selective separation. However, the key is that the unit must be based on a reliable separation mechanism (adsorption, catalysis, or sieving) to ensure that the concentration of the target pollutant in the output gas is extremely low and stable.

[0067] The inlet of the reference gas supply module can be connected to a separate, miniature background air inlet on the housing of the device via a thin conduit. The position of this inlet is designed to avoid drawing in the same airflow as the main inlet, so as to obtain a relatively stable background air. The outlet of the module is directly and uniquely connected to the reference sensing channel inlet of the differential sensing module via another conduit.

[0068] The differential sensing module includes a main sensing channel and a reference sensing channel. The main sensing channel detects the sampled gas from the intake path and provides a detection signal to the processor. The reference sensing channel continuously receives and detects clean air from the reference gas supply module to provide the processor with a real-time updated background reference signal.

[0069] For example, the main sensing channel and the reference sensing channel in the differential sensing module have the same air path structure, sensing element and operating parameters, except for the air intake source.

[0070] Specifically, the differential sensing module is not a standalone component, but rather consists of two spatially adjacent and structurally highly symmetrical sensing chambers and their associated circuitry, integrated onto the same circuit board at the core of the sensing processing cavity. The air inlet of the main sensing channel connects to the main airflow path after the pretreatment filter and differential pressure monitoring module; the air inlet of the reference sensing channel is uniquely connected to the air outlet of the reference gas supply module.

[0071] In this embodiment, the air path length, pipe diameter, and number of bends of the two channels from their respective air intake ports to the surface of the sensor sensing element are completely identical to ensure that the airflow dynamic characteristics are the same.

[0072] Two electrochemical formaldehyde sensors of identical model and batch can be used, one for the main sensing channel and the other for the reference sensing channel, with no difference in their working electrode, counter electrode, or electrolyte. Furthermore, the power supply circuits and signal amplification and conditioning circuits (such as operational amplifier model and gain resistor value) of the two sensors are also completely identical to eliminate system errors introduced by the circuitry.

[0073] Since the two identical sensors operate synchronously under symmetrical conditions, the detection signal V output by the main sensing channel sensor... a This signal includes the actual pollutant concentration, environmental interference such as temperature and humidity, sensor drift, and gas transmission attenuation due to dust blockage. The reference sensor channel output signal V... b Theoretically, it only includes environmental interference and sensor drift.

[0074] Therefore, the processor calculates the differential signal ΔV=V in real time. a -V b This operation can effectively cancel out V a and V b The common-mode component (i.e., environmental interference and synchronization drift) in the signal is crucial. Therefore, the ΔV signal mainly consists of two parts: one is the actual pollutant concentration information; the other is the information on the decrease in sensitivity due to dust blockage (i.e., the blockage effect). Analyzing the long-term decay trend of ΔV is key to diagnosing blockage.

[0075] In this embodiment, a hardware symmetry design ensures that the physical, chemical, and electrical environments of the two sensors have only one controlled difference (control variable): whether the air intake is blocked by dust. This allows the differential signal ΔV to reflect the impact of the specific fault of dust blockage with extreme sensitivity and specificity, providing a high-quality signal input for subsequent accurate diagnostic algorithms.

[0076] Regarding the processor, the selection of processor type has been introduced above and will not be repeated here. In this embodiment, the processor is configured to: continuously compare the signal of the main sensing channel with the real-time updated background reference signal, and combine the differential pressure monitoring value to determine the physical blockage state caused by dust hydration adhesion on the surface of the pre-treatment filter; based on the physical blockage state, dynamically compensate the detection value of the gas sensor to correct the measurement deviation caused by physical blockage.

[0077] In this embodiment, the core data stream received by the processor must first be defined:

[0078] ΔP t Sequence: A sequence of differential pressure monitoring values ​​from the differential pressure monitoring module as a function of time t, reflecting the degree of physical blockage.

[0079] V b Sequence: Background reference signal sequence from the reference sensing channel in the differential sensing module.

[0080] V a Sequence: The sequence of detection signals from the main sensing channel in the differential sensing module.

[0081] ΔV sequence: The differential signal sequence ΔV=V calculated in real time by the processor. a -V b Its changing trend is directly related to the relative decay of the sensitivity of the main sensing channel.

[0082] The processor then samples the above data once per second, and to improve robustness, it also samples ΔP. t The sequence and the ΔV sequence are both filtered using a 10-minute moving average to obtain the smoothed trend sequence ΔP. s (t) and ΔV s (t), then, the slopes of these two smoothed trend sequences over the most recent hour are calculated and denoted as S. ΔP and S ΔV .

[0083] For example, the processor is configured to determine the physical blockage state by analyzing whether a preset co-change relationship is satisfied between the continuous rise of the differential pressure monitoring value and the continuous decay of the detection signal relative to the background reference signal.

[0084] Specifically, to achieve primary diagnosis based on cooperative change rules, a primary diagnostic rule can be built into the processor to quickly identify typical congestion patterns. The rule is as follows:

[0085] If S is satisfied at the same time ΔP > Threshold Th1 (pressure differential continues to rise) and S ΔV If the value is less than the threshold Th2 (differential signal continues to decay) and this state lasts for more than 30 minutes, it is initially determined to be a state of "physical blockage in progress".

[0086] The thresholds Th1 and Th2 can be determined through basic experiments before shipment. This embodiment provides a concise and reproducible experimental procedure to ensure that those skilled in the art can understand and implement it.

[0087] The experiment was conducted in a cleanroom environment with an ambient temperature of 25±2℃ and a relative humidity of 50±10%RH, where a prototype of the device of this invention, which had already undergone aging and calibration, could be deployed. The experimental environment was ensured to be free of any dust sources, and the background concentrations of formaldehyde and TVOC were below the sensor's detection limit. A high-efficiency air filter was installed at the device's air inlet to ensure that the air entering the air path was clean. Simultaneously, a data recording device was used to synchronously acquire the differential pressure monitoring value ΔP and the differential signal ΔV output by the device.

[0088] First, establish a noise baseline for the equipment. Under clean air conditions, ensure the device operates continuously and stably for at least 24 hours. During this period, no dust load is applied to the device, and no human interference is allowed. Record the raw time-series data (ΔP) of ΔP and ΔV over time at 1-second sampling intervals. t (Sequence and ΔV sequence).

[0089] Then, noise feature values ​​are extracted, and the collected 24-hour ΔP t The sequence and the ΔV sequence are both zero-mean processed, i.e., their respective averages are subtracted. Then, the standard deviation σ of each sequence is calculated. ΔP and σ ΔV These two standard deviations directly reflect the normal signal fluctuation amplitude caused by factors such as sensor electronic noise, minor airflow fluctuations, and environmental disturbances when the equipment is in a fault-free state, i.e., the equipment noise level.

[0090] Next, the threshold value is determined. In this embodiment, the threshold value of the pressure difference rise slope Th1 is set to 3 times σ. ΔP The differential signal attenuation slope threshold Th2 is set to -3 times σ. ΔVThe reason for choosing 3 times the standard deviation is that, according to the statistical principle of normal distribution, the probability of signal fluctuations exceeding 3 times the standard deviation under normal circumstances is extremely low (usually less than 0.3%). Therefore, when the upward slope of ΔP consistently exceeds 3 times σ... ΔP Furthermore, the decay slope of ΔV remains consistently lower than -3 times σ. ΔV In this case, it can be determined with high confidence that the coordinated change is not caused by random noise, but by a deterministic fault of physical blockage. If higher diagnostic sensitivity is desired, the factor can be appropriately reduced (e.g., set to 2.5 times); if higher reliability and fewer false alarms are desired, the factor can be appropriately increased (e.g., set to 4 times). The specific factor can be determined through cross-validation during the R&D phase based on product positioning and reliability requirements.

[0091] Finally, the threshold parameters must be solidified, and the finalized Th1 and Th2 values ​​are embedded in the device's factory firmware as the default judgment boundaries for the primary diagnostic rules. Once set, these thresholds remain unchanged throughout the device's lifecycle, unless optimized and adjusted based on large-scale field data during subsequent software upgrades.

[0092] Furthermore, the processor is also configured to execute a blockage diagnostic model, the input of which includes at least differential pressure monitoring values ​​and differential signal values ​​that vary over time, and the output is a quantitative assessment of the physical blockage status.

[0093] Specifically, the blockage diagnosis model used in this embodiment is a lightweight hybrid neural network structure designed specifically for time-series data. Its core is the extraction of the pressure difference sequence (ΔP). t The dynamic correlation characteristics between the sequence and the differential signal sequence (ΔV sequence).

[0094] For the model's input, it can accept two parallel one-dimensional time series of length T (e.g., T=1440, representing data points from the past 24 hours at 1-minute intervals) as input:

[0095] Input sequence A: Normalized pressure difference trend sequence ΔP n (t).

[0096] Input sequence B: Normalized differential signal trend sequence ΔV n (t).

[0097] Regarding the model architecture and hierarchy, the model can be constructed by connecting the following modules in sequence:

[0098] Feature extraction layer: Consists of two independent one-dimensional convolutional (Conv1D) modules, which process the two input sequences respectively. Each Conv1D module contains: one convolutional layer (kernel size = 5, number of filters = 16, activation function = ReLU), followed by a max pooling layer (pooling size = 2).

[0099] Temporal correlation layer: The outputs of the two feature extraction layers are concatenated along the feature dimension and then fed into a bidirectional long short-term memory (Bi-LSTM) layer. This layer has 32 units and is used to learn the dynamic temporal dependencies between the two sequences.

[0100] Fully Connected Decision Layer: The final output of the Bi-LSTM layer is fed into a fully connected network with the following structure: Dense(16, ReLU) → Dropout(0.2) → Dense(1, Sigmoid). The model ultimately outputs a scalar between 0 and 1, namely the congestion index CI.

[0101] Furthermore, the construction of the blockage diagnosis model includes: continuously collecting the sequence data of differential pressure monitoring values ​​and differential signal values ​​changing over time during the device's life cycle or in a pre-set accelerated aging experiment, and training them by associating them with the known true values ​​of the physical blockage state that simulate the dust hydration process.

[0102] Specifically, the first step is the preparation of training data: In a standard environmental chamber, the prototype device to be trained is connected in parallel with a metrologically calibrated, high-precision reference-grade gas analyzer (such as a gas chromatograph or photochemical sensor). Both share the same stable and known-concentration standard formaldehyde / TVOC gas source via a three-way valve, and this gas source has been thoroughly mixed before entering the two branches.

[0103] Calcium carbonate powder was applied to the pretreatment filter of the prototype device according to a predetermined procedure to simulate marble polishing dust. Humidity was adjusted, and one or more humidity shock events were set to simulate a real progressive clogging process. Throughout the process, the following three sets of time-series data were collected and recorded synchronously and continuously:

[0104] a. The prototype's own ΔP t And the ΔV sequence (model input features).

[0105] b. Initial readings C of the prototype gas sensor a (t).

[0106] c. Readings from the reference-level analyzer C b (t) (as the actual concentration of the gas source at this moment).

[0107] Calculate the gas transport efficiency attenuation rate (true value): For each data acquisition time t, calculate the apparent gas transport efficiency factor η(t) of the computing device under the current blockage state, using the following formula:

[0108] η(t)=C a (t) / C b (t);

[0109] η(t) directly characterizes the ratio of the gas concentration reaching the sensor of the device to the actual concentration due to filter clogging. η=1 indicates no attenuation, while a decrease in the value of η directly indicates a loss of transmission efficiency.

[0110] The calculated η(t) sequence is used as the corresponding time ΔP. t The ground truth labels for the ΔV data pairs are used for model training.

[0111] Next, the data is preprocessed: the collected long-term time-series data is divided into multiple sample pairs using a sliding window (length T, step size S) [(ΔP q ,ΔV q ),Label], where ΔP q This represents a complete differential pressure sequence ΔP t A subsequence of length T extracted from the data; ΔV q ΔP represents a subsequence of length T extracted from the corresponding differential signal sequence ΔV(t) within exactly the same time interval; q and ΔV q Strictly aligned in time; Label represents the truth value of the physical blockage state within that time period. In this embodiment, this label is the average or representative value of the gas transport efficiency attenuation rate η obtained by synchronous measurement in a comparative calibration experiment within that time period (e.g., the instantaneous value at the midpoint of the time period).

[0112] Then, the sequences in each sample are normalized so that their mean is 0 and their standard deviation is 1.

[0113] Next are the specific training parameters and steps:

[0114] Loss function: The mean squared error loss function can be used.

[0115] Optimizer: The Adam optimizer can be used, with an initial learning rate set to 0.001.

[0116] Regularization: L2 weight regularization (coefficient = 1e-4) and Dropout (ratio = 0.2) can be used in fully connected layers to prevent overfitting.

[0117] Training cycle: The training is conducted for a total of 100 rounds. If the validation set loss does not decrease for 10 consecutive rounds, the learning rate is reduced to 0.5 of its original value.

[0118] Training objective: To optimize network parameters through backpropagation so that the CI prediction value output by the model is as close as possible to the actual filter pore blockage rate label.

[0119] Based on the above description of the specific and feasible model architecture and training steps, those skilled in the art can understand and reproduce this blockage diagnosis model. This model utilizes pairwise time-series data (ΔP) with clear physical meaning collected from real physical failure processes. t The model is trained using sequences (ΔV sequences) and their labels, enabling it to learn the unique dynamic patterns of a specific fault, dust hydration blockage, thereby achieving high-precision state assessment.

[0120] Furthermore, the method for dynamically compensating the gas sensor's detection value includes: generating a real-time compensation coefficient based on a quantitative assessment of the physical blockage state using predefined mapping rules, and then correcting the gas sensor's detection value.

[0121] Specifically, since the ultimate goal of diagnosis is to correct measured values, the processor can perform dynamic compensation based on the quantitative evaluation result CI and generate real-time compensation coefficients K. A compensation mapping table is pre-stored within the processor, defining the mapping relationship from CI values ​​to compensation coefficients K. For example:

[0122] CI < 0.3 → K = 1.0 (No compensation required)

[0123] 0.3≤CI<0.7→K=1.0 / (1-0.5×CI) (mild to moderate compensation)

[0124] CI≥0.7→K=2.5 (Severe compensation, and trigger advanced alert)

[0125] This mapping, based on experimental calibration, aims to counteract the decrease in gas transport efficiency caused by blockage. Finally, a correction is performed, and the processor acquires the raw concentration reading C from the gas sensor in the main sensing channel. d (This value has been underestimated due to congestion), and the final output to the user is the corrected concentration value C. f Calculate using the following formula:

[0126] C f =K×C d ;

[0127] This operation allows the system to restore the gas concentration to near-real levels even if the sensor detects less gas due to front-end blockage.

[0128] In this embodiment, the direct physical quantity (pressure difference) and the indirect electrochemical effect (differential signal attenuation) are fused and analyzed through an algorithm model, which realizes accurate and quantitative diagnosis of the specific fault of physical blockage caused by dust hydration. Based on the diagnosis results, the output is automatically adjusted through predefined mapping rules, thereby improving the reliability of the measurement results output by the device.

[0129] To better understand how to use this measuring device, please refer to [link / reference]. Figure 2 As shown in the figure, this embodiment also provides a method for detecting formaldehyde and total volatile organic compounds, the method steps of which are as follows:

[0130] Step 1: Continuously acquire differential pressure monitoring values ​​that reflect the permeability of the pre-treatment filter, as well as the detection signal of the main sensing channel and the background reference signal of the reference sensing channel generated by the differential sensing module;

[0131] Specifically, the processor within the device continuously reads and records the differential pressure monitoring value ΔP output by the differential pressure monitoring module. t and the detection signal V from the main sensing channel output by the differential sensing module. a Background reference signal V of the reference sensing channel b And calculate the difference ΔV(t) between the two in real time.

[0132] Step 2: Based on the trend of differential pressure monitoring values ​​and the trend of the difference between the detection signal and the background reference signal, diagnose the physical blockage state of the pretreatment filter caused by dust hydration and adhesion.

[0133] Specifically, the processor analyzes the data trends obtained in the first step, and sets ΔP... t The continuous upward trend of ΔV(t) is compared with the continuous downward trend of ΔV(t), and then input into a pre-trained clogging diagnosis model. This model analyzes the dynamic correlation between these two time-series signals and outputs a quantified clogging index (CI), thereby accurately diagnosing the physical clogging state and severity of the filter caused by dust hydration and adhesion.

[0134] Step 3: Based on the diagnosed physical blockage, the raw readings of the gas sensor are dynamically compensated to output the corrected pollutant concentration value.

[0135] Specifically, based on the blockage status diagnosed in the second step, the processor determines a real-time compensation coefficient K using predefined mapping rules. These predefined mapping rules are established before the device leaves the factory or is calibrated, through systematic calibration experiments, to predetermine the quantitative relationship between the blockage index CI and the compensation coefficient K. This relationship is then embedded in the processor's algorithm or lookup table. Generally, a larger CI value indicates a more severe blockage, and a larger K is required. Subsequently, the compensation coefficient K is compared with the initial reading C of the gas sensor. dMultiply by the product to obtain and output the final corrected concentration value C. f .

[0136] To allow users to more intuitively observe the severity of the blockage, in this embodiment, the status indication module outputs status information and maintenance instructions corresponding to different blockage levels based on the determined physical blockage status. This module serves as the user interface, transforming complex internal diagnostic conclusions into clear and actionable status information.

[0137] Specifically, the status prompt module is a hardware assembly used to convey information to the user, and in this embodiment it may include:

[0138] A color LCD screen embedded in the front panel of the housing and a set of multi-color light-emitting diode (LED) status indicators (green / yellow / red) serve as visual units; a miniature buzzer serves as an auditory unit; the data interface can also use a Wi-Fi / Bluetooth module to push information to the accompanying mobile application (App).

[0139] The processor's early warning logic can be based on the congestion index (CI) and real-time differential pressure (ΔP). t The absolute value of ) determines the specific state levels and trigger conditions, which can be set as follows:

[0140] Level 1: Normal state (green light always on); Triggering condition: CI < 0.3 and ΔP t <1.5×ΔP0.

[0141] Output: The display screen shows the current pollutant concentration value; the App displays "The equipment is working normally".

[0142] Level 2: Observation suggestion (green light flashing / App push notification); Triggering condition: 0.3≤CI<0.5 or ΔP t ≥1.5×ΔP0.

[0143] Output: The display shows the concentration while scrolling a message saying "The filter is purifying, please keep an eye on it"; the app pushes the same message. This level is designed to subtly remind the user that the device is undergoing self-cleaning (coating hydrolysis), which is part of the normal operating process.

[0144] Level 3: Maintenance prompt (yellow light on / prominent push notification in the app); Triggering condition: 0.5 ≤ CI < 0.7 or ΔP t ≥2.0×ΔP0.

[0145] Output: The display shows an icon and text "Intake filter cleaning recommended," with an "(!)" mark next to the concentration value; the app pushes detailed cleaning instructions with pictures and text. This level indicates that the self-cleaning mechanism is insufficient to maintain performance and manual intervention is required.

[0146] Level 4: Failure Warning (Red light flashing and beeping / App emergency alarm); Triggering condition: CI ≥ 0.7 or ΔP t ≥3.0×ΔP0 or compensation coefficient K>2.5.

[0147] Output: The device's red light flashes and emits intermittent beeping; the display prominently displays "Warning: Detection function has severely degraded, readings may be low, please maintain immediately!"; the app sends a high-priority alarm. This level means that blockage has severely affected measurement accuracy, and the device can no longer provide reliable data.

[0148] The processor can determine the current level based on the above conditions and call the pre-stored corresponding text, icon, and audio-visual mode instructions to drive the various units of the status prompt module to output in coordination. All status events and key data (CI, ΔP) triggered at that time are also included. t The K value is recorded in the internal memory and can be queried.

[0149] In this embodiment, the status indication module detects the invisible sensor performance degradation (passivation) process caused by physical blockage within the detection device through multi-dimensional data (CI, ΔP). t The mapping is defined into clear status levels that are visible, audible, and readable, thus solving the security risks of silent failure of traditional equipment.

[0150] In summary, by using a pre-treated filter with a hydrophilic surface, the key chemical process of dust hydration, which leads to blockage, is transformed from passively being endured to being actively managed. Furthermore, the coating material properties are utilized to promote the early conversion and removal of dust, thus inhibiting the formation of a dense blockage layer at its source. Secondly, by simultaneously monitoring two different chains of evidence—physical pressure difference and electrochemical differential signals—the pressure difference monitoring module provides a direct physical signal indicating blockage, while the differential sensing module, supported by a reference gas supply module, provides a pure electrical signal with relatively reduced sensor sensitivity. These two signals together verify the real-time degree of blockage in the intake path. The core function of the processor is to continuously compare and fuse these two signals. By analyzing their coordinated change trends, it can accurately isolate the signal attenuation caused by the adhesion of hydrated dust from complex environmental interference and background noise from the natural aging of the sensor, achieving precise fault location. Based on this diagnostic result, it dynamically generates compensation coefficients to correct the readings in real time. Thus, even when the hardware performance has deteriorated, it can still maintain the accuracy of the final output at the software level. This solves the problem of inaccurate detection results of formaldehyde and total volatile organic compounds caused by physical blockage due to the coupling of specific home decoration dust and sudden changes in humidity. It improves the reliability of the detection results and ensures the health and safety of users.

[0151] Furthermore, the status indication module reveals the invisible sensor performance degradation (passivation) process caused by physical blockage within the detection device, which is not visible inside the device, through multi-dimensional data (CI, ΔP). t The status is mapped to a clear level that is visible, audible, and readable and displayed to the user, reducing the security risks of silent device failure.

[0152] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the aforementioned scope.

[0153] In conclusion, the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A formaldehyde and total volatile organic compound (TVOC) detection device, comprising a housing, a gas sensor, and a processor, characterized in that, The device further includes: A pretreatment filter screen is disposed on the air intake path of the device, and its surface has a hydrolytic hydrophilic coating. The hydrophilic coating is used to undergo a hydrolysis reaction and partially dissolve after absorbing moisture to remove hydrated dust attached to its surface. A differential pressure monitoring module is located downstream of the pretreatment filter and is used to output a differential pressure monitoring value that reflects the permeability of the pretreatment filter. A reference gas supply module is used to continuously generate and deliver clean air; A differential sensing module includes a main sensing channel and a reference sensing channel. The main sensing channel detects the sampled gas from the air intake path and provides a detection signal to the processor. The reference sensing channel continuously receives and detects clean air from the reference gas supply module to provide a real-time updated background reference signal to the processor. The processor is configured to: By continuously comparing the detection signal with the background reference signal and combining it with the differential pressure monitoring value, the physical blockage state caused by the hydration and adhesion of dust on the surface of the pre-treated filter is determined. Based on the physical blockage state, the detection value of the gas sensor is dynamically compensated to correct the measurement deviation caused by the physical blockage.

2. The formaldehyde and total volatile organic compound detection device according to claim 1, characterized in that, The processor is also configured to: The physical blockage state is determined by analyzing whether the continuous rise of the differential pressure monitoring value and the continuous attenuation of the detection signal relative to the background reference signal satisfy a preset coordinated change relationship.

3. The formaldehyde and total volatile organic compound detection device according to claim 1, characterized in that, The processor is configured to execute a blockage diagnostic model, the input of which includes at least differential pressure monitoring values ​​and differential signal values ​​that vary over time, and the output is a quantitative assessment of the physical blockage status.

4. The formaldehyde and total volatile organic compound detection device according to claim 3, characterized in that, The construction of the blockage diagnosis model includes: During the device's lifespan or in a pre-set accelerated aging experiment, the sequence data of the pressure difference monitoring value and differential signal value changing over time are continuously collected, and these are correlated with the known true values ​​of the physical blockage state simulating the dust hydration process.

5. The formaldehyde and total volatile organic compound detection device according to claim 3, characterized in that, The method for dynamically compensating the detected value of the gas sensor includes: Based on the quantitative assessment results of the physical blockage state, a real-time compensation coefficient is generated through a predefined mapping rule, and the detection value of the gas sensor is corrected.

6. The formaldehyde and total volatile organic compound detection device according to claim 1, characterized in that, The main sensing channel and the reference sensing channel in the differential sensing module have the same air path structure, sensing element and operating parameters, except for the air intake source.

7. The formaldehyde and total volatile organic compound detection device according to claim 1, characterized in that, The reference gas supply module includes a purification unit configured to reduce the concentrations of formaldehyde and total volatile organic compounds in the input air to below the detection limit of the gas sensor through adsorption, catalysis, or sieving.

8. The formaldehyde and total volatile organic compound detection device according to claim 1, characterized in that, The hydrolyzable hydrophilic coating is at least one of polyvinyl alcohol, hydroxypropyl methylcellulose, or alginate.

9. The formaldehyde and total volatile organic compound detection device according to claim 1, characterized in that, The device further includes: The status indication module is used to output status information and maintenance instructions corresponding to different blockage levels based on the determined physical blockage status.

10. A method for detecting formaldehyde and total volatile organic compounds, characterized in that, The method, applied in any one of claims 1-8, to a formaldehyde and total volatile organic compound detection device, comprises: Continuously acquire differential pressure monitoring values ​​that reflect the permeability of the pre-treatment filter, as well as the detection signal of the main sensing channel and the background reference signal of the reference sensing channel generated by the differential sensing module; Based on the trend of the differential pressure monitoring value and the trend of the difference between the detection signal and the background reference signal, the physical blockage state of the pretreatment filter caused by dust hydration and adhesion is diagnosed. Based on the diagnosed physical blockage, the raw readings of the gas sensor are dynamically compensated to output a corrected pollutant concentration value.

Images