A method for centralized management and control of area smoke detection devices

Abstract

This application provides a centralized management and control method for regional smoke detectors, including: comparing the current sealing margin of the smoke detector with a preset sealing threshold; when the sealing margin is lower than the sealing threshold, the current sealing status is determined to be abnormal; then, the least squares method is used to fit the correlation between the cumulative dust accumulation at the monitoring deployment point and the fluctuation of the background noise scattered in the darkroom; by comparing the change amplitude of the background noise fluctuation records in the darkroom within adjacent maintenance cycles, the fluctuation trend of noise increase is evaluated; based on the maintenance frequency and the fluctuation trend of noise increase, a classification conclusion of the inflection point of decreasing maintenance benefits for each device is obtained; the sensitivity drift compensation record in the device health database is updated according to the maintenance scheduling instruction; a feedback loop mechanism is used to calibrate the calculation benchmark of the remaining sealing margin of the rubber ring and the dynamic adjustment amplitude of the adjusted sensitivity, to obtain the final detection sensitivity maintenance scheme.

Description

Technical Field

[0001] This invention relates to the field of information technology, and in particular to a method for centralized management and control of regional smoke detectors. Background Technology

[0002] In the field of building fire safety, the centralized management and control of smoke detectors is crucial, directly impacting the timeliness and accuracy of fire early warning systems, and serving as the first line of defense for protecting life and property. As the core device for detecting fire hazards, the stability and reliability of smoke detectors' detection sensitivity play a decisive role in the effectiveness of the entire fire protection system. However, maintaining the performance of these devices during long-term operation and avoiding false alarms or missed alarms has become a major challenge that urgently needs to be addressed in the industry.

[0003] Currently, the industry commonly reduces the interference of internal dust on detection sensitivity by regularly inspecting and cleaning equipment. While this method seems intuitive and effective, it overlooks the potential damage that maintenance itself can cause to other components. Especially during frequent operation, some critical components can experience accelerated wear due to repeated disassembly and reassembly, leading to new problems. This approach often only focuses on surface issues and fails to comprehensively balance the positive and negative impacts of maintenance, making it difficult to guarantee the overall stability of the equipment's performance. In the prior art, a network-based intelligent photoelectric smoke detector and alarm control system and its control method, disclosed in CN106981166A, discloses a technology for networked digital centralized management and control of smoke detectors through a wireless gateway (base station), a service engine, and a cloud data server. The problem is that although it can remotely transmit data such as smoke concentration and improve the convenience of supervision, it still relies on regular on-site inspections and cleaning to deal with the interference of ambient air pollution on detection sensitivity during long-term operation. Frequent disassembly and assembly operations can easily lead to aging or deformation of the sealing components used to isolate external light interference, thereby damaging the sealed environment of the equipment, allowing external stray light to enter the interior and interfere with the smoke detection signal baseline, causing abnormal fluctuations in detection sensitivity, which are superimposed with the effects of dust accumulation, seriously threatening the accuracy of the alarm.

[0004] A more critical technical challenge lies in the fact that the sealing components in smoke detectors, used to isolate external light interference, are prone to aging or deformation after repeated disassembly and reassembly, thus compromising the device's airtight environment. When the sealing performance deteriorates, stray light from the outside can enter the device, interfering with the signal baseline originally used for smoke detection and causing abnormal fluctuations in detection sensitivity. This fluctuation, combined with the effect of dust accumulation inside the device, causes the device's judgment of smoke concentration to gradually deviate from the original design standard, severely threatening the accuracy of the alarm. Specifically, in actual business scenarios, maintenance personnel may clean the same smoke detector multiple times within a month to ensure its internal cleanliness, but each disassembly and reassembly causes irreversible damage to the sealing components. For example, in the fire protection system of a high-rise building, a frequently maintained smoke detector, due to aging sealing components, caused external light interference with the signal, ultimately failing to alarm in time during a small-scale smoke incident, nearly leading to a major disaster. This performance degradation caused by maintenance contradicts the initial intention of improving device reliability through maintenance.

[0005] Therefore, how to reduce the number of disassembly and reassembly while ensuring that the detection sensitivity of the smoke detector does not deviate from the standard due to the aging of the sealing components has become a key problem that this study urgently needs to overcome. Summary of the Invention

[0006] This invention provides a centralized management and control method for area smoke detectors, comprising: The cumulative dust accumulation at the deployment point is monitored in real time by a dust particle size acquisition probe. Combined with the compression and rebound of the rubber ring collected by the rubber ring deformation monitoring unit, and the background noise fluctuation record of the dark room scattering collected by the dark room photoelectric sensor, the remaining sealing margin of the rubber ring is calculated. The current sealing margin of the smoke detector is evaluated based on the remaining sealing margin of the rubber ring. When the sealing margin is lower than the preset sealing threshold, the current sealing state is determined to be abnormal. Based on the judgment result of the abnormal sealing status, the correlation between the cumulative dust accumulation at the monitoring deployment point and the fluctuation of the background noise scattered in the dark room is fitted. The sensitivity is dynamically adjusted according to the correlation to obtain the adjusted sensitivity. The inspection and cleaning history of each smoke detector is extracted, and the maintenance frequency and the corresponding aging degree of the rubber ring are determined according to the inspection and cleaning history. The fluctuation trend of noise increase is evaluated by comparing the change amplitude of the background noise fluctuation record in the anechoic chamber within adjacent maintenance cycles. Based on the maintenance frequency and the fluctuation trend of noise increase, a classification conclusion on the inflection point of decreasing maintenance benefits for each device is obtained. Based on the classification conclusion, the maintenance priority level of each device is determined. By combining the aging degree of the rubber ring with the fluctuating background noise of the darkroom, the seal fading rate is predicted, and an optimized disassembly and reassembly frequency plan is obtained based on the seal fading rate and the maintenance priority level. Based on the planned number of disassembly and assembly cycles and the preset allowable range of the calibration curve, the real-time update of the cumulative dust accumulation at the monitoring deployment point is integrated with the adjusted sensitivity. The deviation between the growth rate of the cumulative dust accumulation and the allowable range of the calibration curve is compared, and a maintenance scheduling instruction is generated. Based on the maintenance scheduling instruction, the calculation benchmark of the remaining sealing margin of the rubber ring is calibrated and the dynamic adjustment range of the adjusted sensitivity is adjusted to obtain the detection sensitivity maintenance scheme.

[0007] Furthermore, the calculation of the remaining sealing allowance of the rubber ring includes: The particle size distribution data of suspended particulate matter in the air at the deployment point is obtained by a dust particle size acquisition probe. The particles are divided into coarse particle range and fine particle range according to their diameter. The deposition mass of particles in each range is accumulated and statistically analyzed according to the sampling period to obtain the cumulative dust accumulation at the monitoring deployment point. The current thickness value of the sealing rubber ring under pressure and the recovered thickness value after pressure relief are read from the rubber ring deformation monitoring unit. The ratio of the difference between the current thickness value and the recovered thickness value to the initial thickness of the rubber ring is calculated to obtain the compression rebound amount of the rubber ring. Sampling data of scattered light intensity is extracted from the photoelectric sensor in the dark room. The average scattered light intensity under smokeless conditions is used as the background noise baseline. The difference between the scattered light intensity and the background noise baseline at each sampling time is calculated to obtain the noise deviation value. The noise deviation value is corrected by combining the compression and rebound of the rubber ring to obtain the background noise fluctuation record of the dark room. The cumulative dust accumulation, the rubber ring compression and rebound, and the background noise fluctuation record of the darkroom are input into the regression model to calculate the remaining sealing margin of the rubber ring.

[0008] Furthermore, the correlation between the cumulative dust accumulation at the fitted monitoring deployment points and the fluctuation of the anechoic chamber scattering background noise includes: Based on the judgment result of the abnormal sealing state, extract the historical cumulative dust accumulation sampling sequence and the dark room scattering background noise fluctuation sampling sequence corresponding to the current device, and pair the two sets of sampling sequences point by point according to the same sampling time to obtain the paired dataset of dust accumulation and noise fluctuation. The least squares method was used to fit the paired dataset, with the cumulative dust accumulation as the independent variable and the fluctuation of the background noise scattered in the dark room as the dependent variable, to obtain the fitted curve equation. The correlation between the cumulative dust accumulation at the monitoring deployment point and the fluctuation of the background noise scattered in the dark room was determined based on the fitted curve equation.

[0009] Furthermore, the step of dynamically adjusting the sensitivity based on the correlation to obtain the adjusted sensitivity includes: Based on the correlation between the cumulative dust accumulation at the monitoring deployment points and the fluctuation of the background noise scattered in the darkroom, the slope value in the correlation is extracted, and the slope value is compared with the standard slope value under the factory calibration conditions. The ratio between the two is calculated to obtain the slope deviation ratio. Based on the slope deviation ratio and the sensitivity benchmark value calibrated at the factory of the smoke detector, a sensitivity compensation amount is obtained. This sensitivity compensation amount is then added to the current sensitivity value of the smoke detector to obtain the adjusted sensitivity.

[0010] Furthermore, the assessment of the increasing fluctuation trend of noise includes: Extract the noise fluctuation sequence of each smoke detector in adjacent maintenance cycles from the background noise fluctuation record of the anechoic chamber, calculate the average noise fluctuation in each maintenance cycle, and subtract the average noise fluctuation of the previous maintenance cycle from the average noise fluctuation of the next maintenance cycle to obtain the noise change amplitude between adjacent maintenance cycles. When the noise change amplitude is positive and the value increases periodically, the fluctuation trend of the noise increase is determined to be an upward trend. When the noise change amplitude is negative or remains stable between adjacent weeks, the fluctuation trend is determined to be a stable trend.

[0011] Furthermore, the classification conclusion of the diminishing inflection point of maintenance revenue for each piece of equipment based on the fluctuation trend of the maintenance frequency and the increase in noise includes: The fluctuation trend is correlated with the maintenance frequency of each device. The records of the same device in multiple consecutive maintenance cycles are compared, and the decrease in noise fluctuation after each maintenance is calculated. When the maintenance frequency continues to increase but the decrease in noise fluctuation after each maintenance gradually decreases, the device is determined to have entered the interval of diminishing maintenance benefits, and the current maintenance frequency is marked as the inflection point of diminishing benefits. Equipment with a current maintenance frequency below the inflection point of diminishing returns is marked as equipment before the inflection point, and equipment with a current maintenance frequency above or equal to the inflection point of diminishing returns is marked as equipment after the inflection point, thus obtaining the classification conclusion of the inflection point of diminishing maintenance returns for each piece of equipment.

[0012] Furthermore, determining the maintenance priority level of each device based on the classification conclusion includes: Based on the classification conclusion, devices after the inflection point are set as high priority, devices before the inflection point with large noise changes are set as medium priority, and devices with small noise changes are set as low priority. Extract the compression and rebound amount of the rubber ring after the most recent disassembly and assembly of each device from the rubber ring deformation monitoring unit, and associate the compression and rebound amount with the corresponding maintenance priority level. When there are multiple devices in the same priority level, sort them from low to high according to the compression and rebound amount of the rubber ring to obtain the priority level pairing record of each device.

[0013] Furthermore, the predicted seal descent rate includes: Acquire the aging values ​​of the rubber rings of each device and the historical sequence data of the background noise fluctuation in the darkroom. Align the aging values ​​of the rubber rings and the historical sequence of noise fluctuations point by point in chronological order to form multidimensional time series data. The aligned multidimensional time series data are fitted using the vector autoregressive moving average method to obtain a fitted curve of the sealing performance changing over time. The numerical difference of the fitted curve at adjacent time points is calculated to determine the sealing descent rate.

[0014] Furthermore, the generation of maintenance scheduling instructions includes: Real-time updated data of cumulative dust accumulation is extracted from the monitoring deployment points, and the ratio of the change in cumulative dust accumulation between adjacent sampling times to the sampling time interval is calculated to obtain the growth rate of cumulative dust accumulation. The cumulative dust accumulation growth rate is compared with the allowable range of the calibration curve. When it exceeds the upper limit, the excess amplitude is calculated, and the excess amplitude is corrected in combination with the adjusted sensitivity to obtain the degree of deviation. The deviation level is matched with the disassembly and assembly time window in the disassembly and assembly number plan. When the deviation level exceeds the preset deviation threshold and the current date falls within the disassembly and assembly time window, a maintenance scheduling command is triggered.

[0015] Furthermore, the calculation benchmark for the remaining sealing margin of the calibration rubber ring and the dynamic adjustment range of the adjusted sensitivity include: Based on the equipment number in the maintenance scheduling instruction, the sensitivity drift compensation record of the corresponding equipment is read from the equipment health database, the difference in sensitivity value before and after the maintenance scheduling instruction is triggered is calculated, the sensitivity drift amount within the current maintenance cycle is obtained, and the sensitivity drift amount is written into the sensitivity drift compensation record. The sensitivity drift sequence within multiple consecutive maintenance cycles is extracted from the sensitivity drift compensation record. The mean of the drift sequence is calculated as the reference deviation value. The original calculation reference for the remaining sealing margin of the rubber ring is added to the reference deviation value to obtain the calibrated sealing margin calculation reference. Based on the calibrated sealing margin calculation benchmark, and combined with the current rubber ring compression rebound amount and anechoic chamber scattering background noise fluctuation record of each device, the detection sensitivity maintenance value of each device is determined.

[0016] The technical solutions provided by the embodiments of the present invention may include the following beneficial effects: This invention discloses a centralized management and control method for regional smoke detectors. It uses a dust particle size acquisition probe to monitor the cumulative dust accumulation at deployment points in real time, combines this with the compression and rebound of the rubber rings collected by a rubber ring deformation monitoring unit, and integrates background noise fluctuation records from a darkroom. A linear regression algorithm is then used to accurately calculate the remaining sealing margin of the rubber rings, thereby achieving real-time assessment of the current sealing status of the smoke detectors. When the sealing margin falls below a preset threshold and is deemed abnormal, the least squares method is further used to fit the correlation between the cumulative dust accumulation and the background noise fluctuation in the darkroom, dynamically adjusting the detection sensitivity. Simultaneously, the historical inspection and cleaning records of each device are extracted, and the relationship between maintenance frequency and the degree of rubber ring aging and noise fluctuation trends is analyzed. The inflection point of diminishing maintenance benefits is quantified, and high, medium, and low priority classification conclusions are formed based on this. This invention starts with the compression and rebound of the rubber ring after recent disassembly and assembly, combines aging degree and noise records, and uses time series analysis to predict the rate of seal degradation. Based on this, it integrates with maintenance priority levels to optimize the quarterly disassembly and assembly frequency plan, generating an optimal disassembly and assembly interval list. Then, it generates precise maintenance scheduling instructions based on the cumulative dust accumulation growth rate and calibration curve deviation. Finally, through a feedback loop mechanism, it calibrates the seal margin calculation benchmark and sensitivity adjustment range, achieving long-term stable maintenance of detection sensitivity. This method effectively solves the problems of seal failure and sensitivity inaccuracy in smoke detectors caused by rubber ring aging, dust accumulation, and noise drift, significantly improving the long-term reliability and maintenance efficiency of regional smoke detectors. Attached Figure Description

[0017] Figure 1 This is a flowchart of a centralized management and control method for regional smoke detectors according to the present invention.

[0018] Figure 2 This is a schematic diagram of a centralized management and control method for regional smoke detectors according to the present invention.

[0019] Figure 3 This is another schematic diagram of a centralized management and control method for regional smoke detectors according to the present invention. Detailed Implementation

[0020] The technical solutions of the embodiments of the present invention will be clearly and thoroughly described below with reference to the accompanying drawings. The described embodiments are merely some embodiments of the present invention.

[0021] like Figures 1-3 This embodiment of a centralized management and control method for regional smoke detectors may specifically include: Step S101: The cumulative dust accumulation at the deployment point is monitored in real time by the dust particle size acquisition probe. Combined with the compression and rebound of the rubber ring collected by the rubber ring deformation monitoring unit, and the background noise fluctuation record of the dark room scattering collected by the dark room photoelectric sensor, the remaining sealing margin of the rubber ring is calculated by the linear regression algorithm. The current sealing margin of the smoke detector is evaluated based on the remaining sealing margin of the rubber ring.

[0022] The particle size distribution data of suspended particulate matter in the air at the deployment point is acquired using a dust particle size acquisition probe. The particle size distribution data is divided into coarse and fine particle intervals according to particle diameter. The deposition mass of particles in each interval is accumulated and statistically analyzed according to the sampling period to obtain the cumulative dust accumulation at the monitoring deployment point. The current thickness value of the sealing rubber ring of the smoke detector under pressure and its recovered thickness value after decompression are read from the rubber ring deformation monitoring unit. The difference between the current thickness value and the recovered thickness value is divided by the initial thickness of the rubber ring to obtain the rubber ring compression rebound amount. The rubber ring elastic attenuation coefficient is determined based on the ratio of the rubber ring compression rebound amount to the factory rebound amount benchmark value. Sampling data of scattered light intensity is extracted from the photoelectric sensor in the dark room. The average scattered light intensity under smokeless conditions is used as the background noise baseline. The noise deviation value is obtained by subtracting the background noise baseline from the scattered light intensity at each sampling time. The noise deviation value is weighted and corrected using the rubber ring elastic attenuation coefficient to obtain the dark room scattered background noise fluctuation record. Using the cumulative dust accumulation, rubber ring compression and rebound, and background noise fluctuation in the darkroom as independent variables, a linear regression algorithm is used to establish a linear mapping equation between the three and the sealing performance. The remaining sealing margin of the rubber ring is calculated through the linear mapping equation, and the current sealing margin of the smoke detector is evaluated based on the remaining sealing margin of the rubber ring.

[0023] In one implementation, the assessment of the sealing margin of the smoke detector is achieved through the coordinated operation of a dust particle size acquisition probe, a rubber ring deformation monitoring unit, and a darkroom photoelectric sensor. The three types of monitoring data are fused and processed before being input into a linear regression model, which outputs a quantitative value of the remaining sealing margin of the rubber ring.

[0024] Specifically, the dust particle size acquisition probe is deployed at the monitoring point where the smoke detection equipment is located to continuously collect particle size distribution data of suspended particulate matter in the air.

[0025] For example, particles with a diameter greater than 10 micrometers are classified into the coarse particle range, while those with a diameter less than or equal to 10 micrometers are classified into the fine particle range. Particles in the fine particle range are more likely to enter the equipment and deposit due to their smaller size. The deposition mass of particles in each range is accumulated at a fixed sampling period to form a time series record of the cumulative dust accumulation at the monitoring deployment point.

[0026] In one embodiment, the rubber ring deformation monitoring unit uses a displacement sensor to measure the current thickness of the sealing rubber ring when the equipment cover is pressed, and the recovered thickness after the cover is opened and pressure is released. The rubber ring compression rebound value characterizes the deformation recovery ability of the rubber ring in a single compression and release cycle; the smaller the value, the more severe the elastic loss of the rubber ring.

[0027] It should be noted that the elastic decay coefficient of the rubber ring is obtained by calculating the ratio of the compression rebound amount of the rubber ring to the rebound amount benchmark value calibrated at the factory. This coefficient reflects the degree of decline in the elastic performance of the rubber ring material due to aging or repeated disassembly and assembly. When the value approaches zero, it indicates that the sealing ability of the rubber ring is close to failure.

[0028] Preferably, the photoelectric sensor in the darkroom collects scattered light intensity data at fixed time intervals within the darkroom of the smoke detector. The average value of the scattered light intensity in a smoke-free state serves as the background noise baseline, representing the optical background level within the darkroom when there is no external interference. The noise deviation value is obtained by subtracting the background noise baseline from the scattered light intensity at each sampling time. The fluctuation range of the noise deviation value is closely related to the sealing condition of the darkroom.

[0029] Understandably, when the sealing performance of the rubber ring deteriorates, external stray light can more easily enter the dark room, leading to increased fluctuations in the noise deviation value. The elastic attenuation coefficient of the rubber ring is used as a weighting factor to correct the noise deviation value; the lower the elastic attenuation coefficient, the larger the weighted corrected noise fluctuation value. This establishes a quantitative correlation between the physical deformation of the rubber ring and changes in optical noise.

[0030] In one possible implementation, the linear regression algorithm uses the cumulative dust amount recorded by the equipment sensors, the compression and rebound of the rubber ring, and the light scattering noise fluctuation value recorded by the optical sensor in the smoke-detecting anechoic chamber as three independent variables, and the remaining sealing margin of the rubber ring as the dependent variable. A linear mapping equation is established based on historical calibration data. The coefficients of the linear mapping equation are obtained by least squares fitting, and multiple sets of equipment monitoring data with known sealing conditions are used as training samples during the fitting process.

[0031] In this embodiment of the application, the amount of dust is the cumulative amount of dust, and the light scattering noise fluctuation value record is the darkroom scattering background noise fluctuation record.

[0032] Specifically, the linear mapping equation takes the form R = α1·D + α2·E + α3·N + β, where R is the remaining sealing allowance of the rubber ring, D is the cumulative dust accumulation, E is the compression and rebound of the rubber ring, N is the fluctuating background noise recorded in the darkroom, α1, α2, and α3 are the regression coefficients of their respective variables, and β is the intercept constant. The regression coefficients are solved using the least squares method, and the objective function is to minimize the sum of squared residuals between the predicted values ​​and the measured sealing allowances of all training samples, i.e., minΣ i (Ri -α1D i -α2E i -α3N i -(β)², take the partial derivatives of each coefficient respectively and set them to zero, and solve the system of equations simultaneously to obtain the optimal values of α1, α2, α3 and β. When the real-time monitoring data is input into the linear mapping equation, the numerical value of the remaining sealing margin of the rubber ring at the current moment is output, and whether the current sealing margin of the smoke detector is within the normal range is evaluated according to this numerical value.

[0033] Step S102, compare the current sealing margin of the smoke detector with the preset sealing threshold. When the sealing margin is lower than the sealing threshold, it is determined that the current sealing state is abnormal, and the least squares method is used to fit the correlation between the cumulative dust deposition amount at the monitoring deployment point and the fluctuation of the dark room scattering background noise.

[0034] Obtain the current sealing margin of the smoke detector and the preset sealing threshold, compare the numerical values of the current sealing margin and the sealing threshold. If the current sealing margin is lower than the sealing threshold, generate a sealing state abnormal flag. According to the sealing state abnormal flag, extract the cumulative dust deposition amount sampling sequence and the dark room scattering background noise fluctuation sampling sequence corresponding to the current device from the monitoring records, pair the two sampling sequences point by point according to the same sampling time, and obtain the paired data set of the dust deposition amount and the noise fluctuation. Use the least squares method to perform curve fitting on the paired data set, use the cumulative dust deposition amount as the independent variable and the dark room scattering background noise fluctuation as the dependent variable, solve the slope and intercept of the fitting curve, and determine the correlation between the cumulative dust deposition amount at the monitoring deployment point and the dark room scattering background noise fluctuation according to the slope and intercept.

[0035] In one implementation, the determination of the sealing state of the smoke detector is achieved by comparing the current sealing margin with the preset sealing threshold. When the comparison result shows that the sealing margin is in the abnormal range, trigger the establishment process of the correlation between the dust deposition amount and the noise fluctuation.

[0036] Specifically, the sealing threshold is pre-calibrated by the device manufacturer according to the rubber ring material characteristics and the dark room optical structure, and represents the lower limit boundary of the normal working range of the sealing performance.

[0037] In the embodiments of the present application, the sealing margin refers to the remaining sealing ability value of the rubber ring, with the unit of percentage; the threshold T is pre-calibrated by the device manufacturer according to the rubber ring material characteristics and the dark room optical structure, and the determination logic in the embodiment is to output an abnormal flag when S < T.

[0038] It should be noted that both the cumulative dust deposition amount sampling sequence and the dark room scattering background noise fluctuation sampling sequence are recorded in the local storage of the device at a fixed time interval.

[0039] In one embodiment, the sampling periods of the two sets of sampling sequences are kept consistent, and each sampling moment corresponds to a set of cumulative dust accumulation values ​​and noise fluctuation values. The two sets of sequences are matched point by point according to the same timestamp to form a paired dataset, where each pair of data contains the cumulative dust accumulation observation value and noise fluctuation observation value at the same moment.

[0040] In one possible implementation, the least squares method uses paired datasets as input samples, with cumulative dust accumulation as the independent variable and background noise fluctuation in the darkroom as the dependent variable, to construct a univariate linear fitting equation. The univariate linear fitting equation is constructed as Y = kX + b, where X is the cumulative dust accumulation, Y is the background noise fluctuation in the darkroom, k is the slope, and b is the intercept. The objective function of the fitting process is minΣ. i (Y i -kX i -b)², taking the partial derivatives with respect to k and b respectively and setting them to zero, we get the slope k = (nΣX)². i Y i -ΣX i ΣY i ) / (nΣX i ²-(ΣX i )²), intercept b=(ΣY i -kΣX i ) / n, where n is the total number of paired data points. The slope k reflects the change in noise fluctuation when the dust accumulation increases by one unit mass, and the intercept b reflects the background level of noise fluctuation when the dust accumulation is zero.

[0041] For example, when the slope value obtained from the fitting is large, it indicates that the equipment is highly sensitive to dust accumulation, and a small amount of dust accumulation can cause significant noise fluctuations; when the slope value is small, it indicates that the impact of dust accumulation on noise fluctuations is relatively mild.

[0042] It is understandable that the linear relationship between the cumulative dust accumulation and the fluctuation of the anechoic chamber's scattered background noise can be fully described by the two parameters of slope and intercept. This relationship provides a quantitative basis for sensitivity adjustment when the sealing condition is abnormal, thus providing data support for subsequent compensation operations.

[0043] Step S103: By analyzing the correlation between the cumulative dust accumulation at the monitoring deployment points and the fluctuation of the background noise scattered in the darkroom, the sensitivity is dynamically adjusted to obtain the adjusted sensitivity. The inspection and cleaning history records of each smoke detector are extracted, and the maintenance frequency and the corresponding aging degree of the rubber ring are obtained by analyzing the inspection and cleaning history records.

[0044] Based on the correlation between the cumulative dust accumulation at monitoring deployment points and the fluctuation of background noise scattered in the anechoic chamber, the slope value in the correlation is extracted as the basis for sensitivity adjustment. The slope value is normalized by dividing it by the standard slope value under factory calibration conditions to obtain the slope deviation ratio. The slope deviation ratio is multiplied by the sensitivity benchmark value of the smoke detector at the factory calibration to obtain the sensitivity compensation amount. The sensitivity compensation amount is added to the current sensitivity value to obtain the adjusted sensitivity. The inspection and cleaning history records of each device are extracted from the storage unit of the smoke detector. The inspection and cleaning history records include the execution time of each inspection and cleaning and the corresponding device number. The execution times of multiple inspections and cleanings of the same device are arranged in chronological order, and the time interval between two adjacent inspections and cleanings is calculated. The number of inspections and cleanings of each device within a fixed observation period is counted to obtain the maintenance frequency. For the maintenance frequency, the compression and rebound amount of the rubber ring after each inspection and cleaning is extracted from the rubber ring deformation monitoring unit. The attenuation range of the rubber ring compression and rebound amount after two adjacent inspections and cleanings is compared. The aging degree of the rubber ring of each device is determined according to the correspondence between the attenuation range and the maintenance frequency.

[0045] In one implementation, the dynamic adjustment of the sensitivity of the smoke detector and the assessment of the aging degree of the rubber ring are achieved through step S103, wherein the sensitivity is compensated based on the correlation between dust accumulation and noise fluctuation, and the aging degree of the rubber ring is quantified based on the inspection and cleaning history.

[0046] Specifically, the sensitivity compensation amount is determined based on the slope value in the correlation relationship. The slope value reflects the degree to which a unit change in dust accumulation affects noise fluctuation. Dividing the current slope value by the standard slope value measured under factory calibration conditions yields the dimensionless slope deviation ratio. When the slope deviation ratio is greater than 1, it indicates that the current equipment is overly sensitive to dust accumulation, and the sensitivity compensation amount is a negative correction; when the slope deviation ratio is less than 1, it indicates that the response tends to be sluggish, and the sensitivity compensation amount is a positive correction. The formula for calculating the sensitivity compensation amount is ΔS=(σ / σ0-1)·S0, where σ is the current slope value, σ0 is the standard slope value, and S0 is the factory sensitivity reference value. The adjusted sensitivity is obtained by superimposing the sensitivity compensation amount ΔS onto the current sensitivity value.

[0047] It should be noted that the inspection and cleaning history is stored in the local storage unit of the smoke detector. The execution time and device number are automatically written after each inspection and cleaning is completed.

[0048] For example, a commercial building has dozens of smoke detectors deployed. The inspection and cleaning history of each device is independent and can be indexed and retrieved according to the device number.

[0049] In one possible implementation, the maintenance frequency is statistically analyzed using a fixed observation period as the time window. Multiple inspection and cleaning execution times for the same equipment within the observation period are arranged chronologically, and the difference in days between two adjacent execution times is calculated. The total number of inspections and cleanings within the observation period is then calculated and divided by the number of days in the observation period to obtain the equipment's maintenance frequency, i.e., maintenance frequency f = C / P, where C is the total number of inspections and cleanings within the observation period, and P is the total number of days in the observation period. A higher maintenance frequency indicates that the equipment has been disassembled and cleaned more frequently within the observation period. Furthermore, the determination of the aging degree of the rubber rings is based on the correspondence between the maintenance frequency and the attenuation of the rubber ring compression and rebound amount. The rubber ring deformation monitoring unit is a sensor module built into the equipment, which collects data by real-time detection of the compression and rebound deformation of the rubber rings under pressure. The compression and rebound amount of the rubber rings after each inspection and cleaning is extracted from this unit, and the attenuation amount is obtained by subtracting the compression and rebound amounts after two adjacent inspections and cleanings.

[0050] For example, if the compression rebound of a certain piece of equipment decreases successively after three consecutive inspections and cleanings, with the rate of decline showing a progressively increasing trend, it indicates that the elasticity loss of the rubber seals in this equipment is accelerating. Correlating the rate of decline with the maintenance frequency reveals that equipment with higher maintenance frequencies and greater rate of decline exhibits more severe rubber seal aging.

[0051] Understandably, quantifying the degree of aging of the rubber rings makes it comparable between different devices, making it easier to identify individuals with higher aging risks within a cluster of devices.

[0052] Step S104: By comparing the change amplitude of the background noise fluctuation record in the anechoic chamber within adjacent maintenance cycles, the fluctuation trend of noise increase is evaluated, and the classification conclusion of the inflection point of decreasing maintenance revenue for each piece of equipment is obtained based on the maintenance frequency and the fluctuation trend of noise increase.

[0053] The noise fluctuation values ​​of each smoke detector within adjacent maintenance cycles are extracted from the background noise fluctuation records of the anechoic chamber. The average noise fluctuation value of the previous maintenance cycle is subtracted from the average noise fluctuation value of the subsequent maintenance cycle to obtain the noise change amplitude between adjacent maintenance cycles. If the noise change amplitude is positive and increases periodically, the fluctuation trend is determined to be an upward trend. If the noise change amplitude is negative or remains stable between adjacent cycles, the fluctuation trend is determined to be a stable trend. The fluctuation trend of noise increase is correlated with the maintenance frequency of each device. For the records of the same device in multiple consecutive maintenance cycles, the decrease in noise fluctuation after each maintenance is compared. If the maintenance frequency continues to increase but the decrease in noise fluctuation after each maintenance gradually decreases, the device is determined to have entered the diminishing returns interval, and the current maintenance frequency value of the device is marked as the inflection point of diminishing returns. Based on the comparison between the inflection point of diminishing returns and the maintenance frequency of each device, devices with a current maintenance frequency lower than the inflection point of diminishing returns are marked as devices before the inflection point, and devices with a current maintenance frequency higher than or equal to the inflection point of diminishing returns are marked as devices after the inflection point, thus obtaining the classification conclusion of the inflection point of diminishing maintenance returns for each device.

[0054] In one implementation, the inflection point of diminishing maintenance benefits is identified by comparing the variation patterns of noise fluctuations within adjacent maintenance cycles, and each smoke detector is classified and marked according to the location of the inflection point.

[0055] Specifically, the noise variation amplitude is determined by calculating the difference between the average noise increase during adjacent maintenance weeks.

[0056] It should be noted that diminishing maintenance benefits refer to the phenomenon that the noise reduction effect of each maintenance gradually weakens as the maintenance frequency increases.

[0057] In one possible implementation, during the early maintenance phase, a significant decrease in noise fluctuation occurs after each inspection and cleaning of a piece of equipment, indicating that the maintenance has a clear effect on restoring equipment performance. However, as maintenance frequency accumulates, the rubber seals experience irreversible elastic loss due to repeated disassembly and reassembly, leading to a decreasing decrease in noise fluctuation after each maintenance, and even a rise in noise after maintenance. This inflection point of diminishing marginal returns is called the diminishing returns inflection point. Furthermore, the diminishing returns inflection point is marked using the current maintenance frequency of the equipment as an anchor point. When it is detected that the noise reduction of a piece of equipment continuously decreases and falls below a preset improvement threshold over multiple consecutive maintenance cycles, the equipment number, the time point corresponding to the current maintenance cycle, and the current maintenance frequency value are recorded as the diminishing returns inflection point.

[0058] Understandably, the classification of equipment before and after the inflection point is based on the numerical comparison between the current maintenance frequency and the inflection point of diminishing returns. For equipment before the inflection point, the maintenance frequency has not yet reached the diminishing returns range, and increasing the number of maintenance sessions still yields positive returns. For equipment after the inflection point, the maintenance frequency has exceeded the diminishing returns inflection point, and continuing to increase the number of maintenance sessions will lead to accelerated aging of the rubber seals with limited noise improvement. Based on the above classification conclusions, maintenance strategies for different types of equipment can be formulated differently: equipment before the inflection point should maintain its current maintenance rhythm, while equipment after the inflection point should reduce the frequency of disassembly and assembly or prioritize the replacement of rubber seals.

[0059] Extract the average noise increase and rubber ring compression rebound attenuation value of each equipment within a continuous maintenance cycle, analyze the number of days between cycles when the noise increase exceeds the benchmark threshold and the noise drop after a single maintenance, evaluate the marginal benefit of noise improvement, and form a classification conclusion of high priority, medium priority, and low priority based on the equipment number and cycle node corresponding to the inflection point of decreasing maintenance benefits for each equipment when the marginal benefit is lower than the preset lower limit.

[0060] The fluctuation values ​​of the anechoic chamber scattered background noise within a continuous maintenance cycle are extracted from the monitoring records of each smoke detector. The increment of noise fluctuation during adjacent maintenance cycles is calculated and averaged to obtain the average noise increase. Simultaneously, the attenuation records of rubber ring compression and rebound within each maintenance cycle are extracted from the rubber ring deformation monitoring unit. The average noise increase is compared with a preset benchmark threshold. The number of days between maintenance cycles where the average noise increase continuously exceeds the benchmark threshold is counted. The decrease in noise fluctuation after each inspection and cleaning relative to before cleaning is extracted as the noise reduction amplitude. The noise reduction amplitude is divided by the number of days between intervals to obtain the marginal benefit value. The marginal benefit value is compared with a preset benefit lower limit. If the marginal benefit value is lower than the benefit lower limit, the equipment is determined to have entered the diminishing maintenance benefit range. The equipment number of the equipment and the time node corresponding to the current maintenance cycle are extracted and marked as the inflection point of diminishing maintenance benefits. Based on the ratio of the marginal benefit value of each device to the lower limit of benefit corresponding to the inflection point of diminishing maintenance benefits, if the ratio is lower than the preset first ratio threshold, it is marked as high priority; if the ratio is between the first ratio threshold and the second ratio threshold, it is marked as medium priority; and if the ratio is higher than the second ratio threshold, it is marked as low priority, thus forming a classification conclusion of high priority, medium priority, and low priority.

[0061] In one implementation, the refined identification of diminishing returns inflection points and the classification of equipment priority levels are achieved by extracting the mean noise increase, calculating the marginal benefit value, determining the diminishing returns interval, and dividing it into three levels based on the ratio threshold.

[0062] Specifically, the average noise increase is extracted using the fluctuating values ​​of the diffuse background noise in the darkroom for each smoke detector during a continuous maintenance cycle. This value refers to the fluctuation amplitude of the diffuse background signal obtained by the smoke detector through continuous sampling by its built-in photoelectric sensor under smoke-free conditions in a darkroom, and is typically recorded automatically at the end of each maintenance cycle. Subtracting the noise fluctuation value of the previous cycle from the noise fluctuation value of the next cycle yields multiple increment values. The arithmetic mean of these increment values ​​is the average noise increase for that device, i.e., the average noise increase ΔN = Σ i (N i+1 -N i ) / (m-1), where N i Let m be the noise fluctuation value for the i-th maintenance cycle, and m be the number of consecutive maintenance cycles.

[0063] It should be noted that the records of rubber ring compression and rebound attenuation were extracted simultaneously with the historical sequence data of background noise fluctuations in the anechoic chamber. The periodic attenuation trend of rubber ring compression and rebound corroborated the trend of the average noise increase; equipment with a larger attenuation magnitude usually also had a more significant noise increase.

[0064] In one possible implementation, the baseline threshold is pre-calibrated at the factory based on the optical structure of the darkroom and the response characteristics of standard smoke concentration, characterizing the upper limit of noise increase under normal operating conditions. The average noise increase of each maintenance cycle is compared with the baseline threshold, and the start and end cycles in which the average noise increase continuously exceeds the baseline threshold are recorded. The number of calendar days between the two cycles is calculated as the interval. Furthermore, the calculation of the marginal benefit value is the core of this scheme. The marginal benefit value is defined as the degree of noise improvement brought about by the maintenance behavior per unit time interval. Its physical meaning is the amount of noise reduction that can be obtained by one maintenance operation per day. The noise reduction amplitude is obtained by extracting the difference between the noise fluctuation value after each inspection and cleaning and the noise fluctuation value before cleaning. A positive difference indicates that the maintenance behavior has a positive improvement effect. The formula for calculating the marginal benefit value is ME=ΔN. fall / T gap , where ΔN fall T represents the noise reduction after a single maintenance. gap This refers to the number of consecutive days during which the average noise increase exceeds the baseline threshold.

[0065] Understandably, the lower bound of benefits is the critical value used to determine whether equipment has entered the diminishing returns phase of maintenance. When the marginal benefit value is lower than the lower bound, it means that the noise improvement brought about by continued maintenance is insufficient to compensate for the damage to the rubber seals caused by the maintenance itself. At this point, the equipment number and the time point corresponding to the current maintenance cycle are extracted, and the combination of the two is marked as the inflection point of diminishing returns. This inflection point records the critical moment when the equipment transitions from a positive return phase to a diminishing return phase.

[0066] In one embodiment, priority levels are determined based on the ratio of marginal benefit value to a lower limit of benefit. The ratio is calculated by dividing the marginal benefit value by the lower limit of benefit, resulting in a dimensionless relative index. This index eliminates absolute numerical deviations caused by differences in deployment environments between different devices, making smoke detectors located on different floors and in different areas comparable. A first ratio threshold and a second ratio threshold divide the ratio range into three segments: a ratio below the first ratio threshold indicates severely diminished marginal benefit; a ratio between the two thresholds indicates moderate diminished marginal benefit; and a ratio above the second ratio threshold indicates relatively minor diminished marginal benefit.

[0067] Preferably, for high-priority equipment with a ratio below the first ratio threshold, the marginal benefit of maintenance for these devices has significantly fallen below the lower limit of benefit. Continuing maintenance at the original frequency will accelerate the aging of the rubber seals with minimal benefit; therefore, rubber seal replacement should be prioritized or the maintenance frequency should be significantly reduced. For medium-priority equipment with a ratio between the two thresholds, although the marginal benefit of maintenance for these devices has entered a decreasing range, there is still some room for improvement, and the maintenance interval can be appropriately extended. For low-priority equipment with a ratio above the second ratio threshold, although the lower limit of benefit has been reached, the degree of decline is still shallow; the current maintenance rhythm can be maintained temporarily while continuous monitoring continues. Through the above three-level classification conclusions, the maintenance urgency of each smoke detector can be quantitatively distinguished, and high-priority equipment receives priority response in maintenance resource scheduling.

[0068] Step S105: Determine the maintenance priority level of each device based on the classification conclusion of the inflection point of the maintenance benefit of each device. Extract the compression rebound amount of the rubber ring after the most recent disassembly and assembly from the rubber ring deformation monitoring unit. Combine the aging degree of the rubber ring with the background noise fluctuation record of the dark room scattering. Use the vector autoregression moving average method to predict the seal drop rate. Based on the seal drop rate and the maintenance priority level, obtain the optimized disassembly and assembly number plan.

[0069] Based on the classification conclusions of the diminishing maintenance benefits inflection points for each piece of equipment, high priority, medium priority, and low priority levels are respectively assigned to different levels of maintenance urgency. The compression and rebound amount of the rubber rings after the most recent disassembly and reassembly of each piece of equipment is extracted from the rubber ring deformation monitoring unit. This compression and rebound amount is correlated with the corresponding maintenance priority level of the equipment to obtain priority level pairing records for each piece of equipment. Historical sequence data of rubber ring aging degree values ​​and anechoic chamber background noise fluctuations for each piece of equipment are obtained. The rubber ring aging degree values ​​and noise fluctuation historical sequences are aligned chronologically, and a vector autoregression moving average method is used to fit the aligned sequences. The rubber ring aging degree value sequence is used as the primary input sequence, and the noise fluctuation historical sequence is used as the auxiliary input sequence. The seal degradation rate is determined based on the numerical differences of the fitted curve at adjacent time points. The seal degradation rate is fused with the priority level pairing records. Based on the fusion result, an upper limit on the number of disassembly and reassembly attempts for each piece of equipment within a preset planning period is set. The total number of days in the planning period is divided by the upper limit on the number of disassembly and reassembly attempts to obtain the disassembly and reassembly interval days for each piece of equipment. Based on the disassembly and assembly interval days, the total number of days in the planning period is divided by the disassembly and assembly interval days to obtain the expected total number of disassembly and assembly times for each device within the planning period. The device number, disassembly and assembly interval days, and expected total number of disassembly and assembly times are combined to form the optimized disassembly and assembly frequency plan.

[0070] In one implementation, the optimized disassembly and assembly frequency planning first sets an initial disassembly and assembly interval upper limit based on the maintenance priority level, for example, a 6-month upper limit for high-priority equipment, and then dynamically adjusts the actual interval in combination with the seal descent rate.

[0071] Specifically, if the rate r (unit: % / month) exceeds the threshold of 0.5, the interval is shortened to 80% of the upper limit, thereby determining the disassembly and assembly interval and the expected total number of disassembly and assembly operations for each smoke detector within the planning cycle. High priority corresponds to the equipment with the highest maintenance urgency; the marginal benefits of such equipment have severely diminished, and continued frequent disassembly and assembly will accelerate the aging of the rubber seals. Medium priority corresponds to equipment with moderate urgency; low priority corresponds to equipment with relatively low urgency, whose maintenance marginal benefits are still within an acceptable range.

[0072] It should be noted that the extraction of the rubber ring compression rebound amount is based on the monitoring data after the most recent disassembly and assembly of each device.

[0073] For example, after a recent routine cleaning of a smoke detector, the rubber ring deformation monitoring unit recorded the thickness change of the rubber ring during the pressing and releasing process of the cover plate, thereby calculating the current compression rebound value. This value is combined with the maintenance priority level corresponding to the equipment to form a priority level pairing record. Subsequently, the seal descent rate is calculated based on this record, and an equipment maintenance plan is formulated accordingly.

[0074] In one possible implementation, the aging value of the rubber ring is aligned with the historical sequence data of the background noise fluctuation in the dark room by timestamps. The aging value of the rubber ring reflects the cumulative loss of the elastic properties of the rubber ring material, while the noise fluctuation history records the fluctuation of the scattered light intensity in the dark room at each sampling time. After the two sets of data are matched one-to-one on the time axis, a joint sequence is formed that can be processed by time series methods.

[0075] Furthermore, the vector autoregressive moving average method is a classic multivariate time series forecasting method. Its basic principle is to predict future vectors using a linear combination of historical observation vectors. In this scheme, the aligned rubber ring aging degree sequence and noise fluctuation sequence are first combined into a multidimensional joint vector sequence as input. Then, a model is constructed using the maximum likelihood method and lag order selection, and the output is the predicted sealing performance values ​​for several future time points. The autoregressive part uses the lag vector of the sequence itself to establish a regression relationship, reflecting the inertial change of sealing performance over time; the moving average part corrects the prediction error vector and eliminates the interference of random fluctuations. By fitting the prediction curve, the difference between the predicted values ​​between adjacent time points is extracted, and this difference divided by the time interval is the sealing degradation rate. The larger the sealing degradation rate, the more obvious the deterioration trend of the equipment's sealing performance.

[0076] Understandably, the fusion of seal descent rate and priority level pairing records forms the basis for determining the upper limit of disassembly / reassembly cycles. For high-priority equipment, since its marginal benefits are already severely diminishing, a lower upper limit for disassembly / reassembly cycles is set to avoid damage to the rubber seals caused by excessive disassembly / reassembly. For medium-priority equipment, a moderate upper limit for disassembly / reassembly cycles is set to strike a balance between maintenance effectiveness and rubber seal protection. For low-priority equipment, a higher upper limit for disassembly / reassembly cycles is set to allow for more routine maintenance within the planned cycle. This differentiated setting of the upper limit for disassembly / reassembly cycles reflects the differentiated treatment of equipment in different states.

[0077] In one embodiment, the number of days for disassembly and assembly intervals is obtained by dividing the total number of days in the planning period by the upper limit of the number of disassembly and assembly operations.

[0078] Preferably, the total number of disassembly and assembly operations is estimated by dividing the total number of days in the planning cycle by the number of days between disassembly and assembly operations, and taking the integer part as the final result. The equipment number of each device, the number of days between disassembly and assembly operations, and the estimated total number of disassembly and assembly operations are combined to form an optimized disassembly and assembly operation plan. This plan provides a quantitative basis for subsequent maintenance scheduling, making the timing and frequency of inspection and cleaning of each device feasible.

[0079] By using the current seal degradation rate and maintenance priority data of each smoke detector, we analyze the expected time point when the seal margin of high priority equipment will bottom out in the next quarter, assess the rubber ring replacement needs, disassembly and assembly intervals, and maximum allowable disassembly and assembly times for medium and low priority equipment, and form a quarterly disassembly and assembly frequency planning list that includes the optimal disassembly and assembly interval days and disassembly and assembly time windows.

[0080] In one implementation, the quarterly disassembly and reassembly schedule is formed by analyzing the seal degradation rate and maintenance priority of each device, and developing differentiated disassembly and reassembly arrangements for high-priority devices and medium-to-low-priority devices respectively.

[0081] Specifically, determining the maintenance priority level of each device based on the classification conclusion includes: Based on the classification conclusion, devices after the inflection point are set as high priority, devices before the inflection point with large noise changes are set as medium priority, and devices with small noise changes are set as low priority. Extract the compression and rebound amount of the rubber ring after the most recent disassembly and assembly of each device from the rubber ring deformation monitoring unit, and associate the compression and rebound amount with the corresponding maintenance priority level. When there are multiple devices in the same priority level, sort them from low to high according to the compression and rebound amount of the rubber ring to obtain the priority level pairing record of each device.

[0082] Specifically, the bottoming-out time of high-priority equipment is calculated based on the current sealing margin and the sealing descent rate.

[0083] For example, the current sealing margin of a high-priority smoke detector is a certain value. The sealing degradation rate represents the daily decline in its sealing performance. Dividing the sealing margin by the degradation rate gives the remaining number of days. Adding this remaining number of days to the start date of the quarterly planning period gives the specific date on which the sealing margin is expected to bottom out. If this date falls within the next quarter, the sealing ring replacement requirement for that device is marked as urgent.

[0084] It should be noted that the assessment of gasket replacement needs for medium-priority and low-priority equipment employs different judgment logic. The predicted seal margin at the end of the quarterly planning period is calculated based on the seal descent rate. This predicted value is then compared to a preset safety threshold, which represents the minimum permissible limit for seal performance. A predicted value higher than the safety threshold is marked as having no replacement need, while a predicted value lower than the safety threshold is marked as having a planned replacement need.

[0085] In one possible implementation, the disassembly / reassembly interval is determined based on the total number of days in the quarterly planning cycle and the maximum permissible number of disassembly / reassembly operations. The maximum permissible number of disassembly / reassembly operations is determined by the equipment's maintenance priority level; medium-priority equipment has a higher maximum permissible number of disassembly / reassembly operations than high-priority equipment, and low-priority equipment has the highest maximum permissible number of disassembly / reassembly operations. Dividing the total number of days by the maximum permissible number of disassembly / reassembly operations yields the disassembly / reassembly interval for each piece of equipment. Furthermore, the disassembly / reassembly time window refers to the expected execution date range for each disassembly / reassembly operation. Using the start date of the quarterly planning cycle as a baseline, the expected execution date for the first disassembly / reassembly, the second disassembly / reassembly, and so on, is accumulated according to the disassembly / reassembly interval, until the end of the quarterly planning cycle. The three-day range before and after each execution date constitutes the time window for that disassembly / reassembly. Maintenance personnel completing the inspection and cleaning operations within the time window are considered to have met the planning requirements.

[0086] Understandably, by combining the disassembly and assembly intervals, the time windows for each disassembly and assembly, and the equipment number, a quarterly disassembly and assembly frequency planning list is formed. This list provides a quantitative time arrangement basis for the maintenance scheduling in the next quarter.

[0087] Step S106: Based on the disassembly and assembly frequency plan and the preset calibration curve allowable range, integrate the real-time update of the cumulative dust accumulation at the monitoring deployment points with the adjusted sensitivity, generate maintenance scheduling instructions by comparing the deviation between the cumulative dust accumulation growth rate and the allowable range of the calibration curve, and determine the precise timing of the next inspection and cleaning based on the maintenance scheduling instructions.

[0088] The system obtains the planned number of disassembly and reassembly cycles for each smoke detector and the preset allowable range of the calibration curve. Real-time updated data of accumulated dust is extracted from the monitoring deployment points. The change in accumulated dust between adjacent sampling times is calculated, and this change is divided by the sampling time interval to obtain the cumulative dust accumulation growth rate. This growth rate is compared with the allowable range of the calibration curve, which represents the upper and lower limits of the dust accumulation growth rate under normal operating conditions. If the cumulative dust accumulation growth rate exceeds the upper limit, the excess value is calculated as the deviation margin. This deviation margin is then weighted and corrected using the adjusted sensitivity value to obtain the deviation degree. Based on the deviation degree and the disassembly and reassembly time window in the planned number of cycles, if the deviation degree exceeds a preset deviation threshold and the current date falls within the disassembly and reassembly time window, a maintenance scheduling instruction is triggered. This instruction includes the device number and a suggested execution date. The precise timing of the next inspection and cleaning is determined based on this instruction.

[0089] In one implementation, the precise timing of the next inspection and cleaning is achieved by integrating the number of disassembly and reassembly cycles with real-time dust monitoring data, and triggering a maintenance scheduling command after determining the degree of deviation based on the allowable range of the calibration curve.

[0090] Specifically, the calculation of the cumulative dust accumulation growth rate is based on real-time sampling data from the monitoring deployment points. The cumulative dust accumulation values ​​at two adjacent sampling times are obtained from the dust particle size acquisition probe. The change is obtained by subtracting the value of the previous time from the value of the later time. Then, the change is divided by the time interval between the two sampling times to obtain the cumulative dust accumulation growth rate.

[0091] It should be noted that the allowable range of the calibration curve is a reference range pre-calibrated by the smoke detector at the factory based on the optical structure of the darkroom and the standard testing environment. This range includes an upper limit boundary and a lower limit boundary. The upper limit boundary represents the maximum allowable value of the dust accumulation growth rate, used to determine the degree of deviation and triggering conditions. The lower limit boundary represents the minimum normal value of the dust accumulation growth rate, used to determine whether the growth rate is too low to detect potential equipment problems.

[0092] For example, when a smoke detector is deployed in an underground parking lot with poor air circulation, its dust accumulation rate is typically higher than that of a device deployed in a well-ventilated office area. The allowable range of the calibration curve takes into account the differences in different deployment environments, making the deviation judgment adaptable to various scenarios.

[0093] In one possible implementation, the degree of deviation is determined based on the extent to which the cumulative dust accumulation growth rate exceeds the upper limit boundary. The excess range is obtained by subtracting the upper limit boundary value from the growth rate, and then multiplied by an adjusted sensitivity value K as a weighting factor to obtain the degree of deviation. Here, K represents the sensitivity value derived from historical equipment response data, adjusted by multiplying the initial value by an equipment type coefficient ranging from 0.8 to 1.2. A higher sensitivity value K results in a larger weighted deviation degree, indicating that the current equipment is more sensitive to changes in dust accumulation. Furthermore, the triggering of maintenance scheduling instructions employs a dual-condition judgment. If the degree of deviation exceeds a preset deviation threshold (determined based on equipment factory testing and on-site operating condition calibration), and the current date falls within the disassembly / reassembly time window of the corresponding equipment in the disassembly / reassembly frequency planning, a maintenance scheduling instruction is triggered. The maintenance scheduling instruction records the equipment number and the suggested execution date.

[0094] Understandably, the precise timing of the next inspection and cleaning can be determined based on the suggested execution date in the maintenance scheduling instruction, so that the inspection and cleaning behavior not only responds to changes in real-time dust accumulation, but also conforms to the overall arrangement of the quarterly disassembly and assembly frequency plan.

[0095] Step S107: Update the sensitivity drift compensation record in the equipment health database according to the maintenance scheduling instruction, and use a feedback loop mechanism to calibrate the calculation benchmark of the remaining sealing margin of the rubber ring and the dynamic adjustment range of the adjusted sensitivity to obtain the final detection sensitivity maintenance scheme.

[0096] Based on the equipment number and suggested execution date in the maintenance scheduling instruction, the sensitivity drift compensation record of the corresponding equipment is read from the equipment health database. The difference between the sensitivity values ​​before and after the maintenance scheduling instruction is triggered is calculated to obtain the sensitivity drift amount within the current maintenance cycle. The sensitivity drift amount is written into the sensitivity drift compensation record and updated. A sensitivity drift sequence within multiple consecutive maintenance cycles is extracted from the sensitivity drift compensation record. The mean of the drift sequence is calculated as the baseline deviation value. The original calculation baseline for the remaining sealing margin of the rubber ring is added to the baseline deviation value to obtain the calibrated sealing margin calculation baseline. If the drift sequence shows an increasing trend, the dynamic adjustment range of sensitivity is increased; if it shows a decreasing trend, the dynamic adjustment range is decreased. Based on the calibrated sealing margin calculation baseline and the dynamic adjustment range, combined with the current rubber ring compression rebound amount and anechoic chamber background noise fluctuation record for each equipment, the calibrated sealing margin calculation baseline is multiplied by the dynamic adjustment range to obtain the detection sensitivity maintenance value for each equipment. The detection sensitivity maintenance value is combined with the corresponding equipment number and calibration timestamp to form the final detection sensitivity maintenance scheme.

[0097] In one implementation, the detection sensitivity maintenance scheme is formed by updating the sensitivity drift compensation record, calibrating the sealing margin calculation benchmark, and calculating the sensitivity maintenance value of each device in combination with the dynamic adjustment amplitude.

[0098] Specifically, the sensitivity drift is calculated based on the sensitivity values ​​before and after the maintenance scheduling command is triggered.

[0099] For example, the sensitivity value of a smoke detector before the current maintenance scheduling command is triggered is recorded in the equipment health database. The sensitivity value after the trigger is re-acquired by a photoelectric sensor in an anechoic chamber. Subtracting the value before the trigger from the value after the trigger yields the sensitivity drift during the current maintenance cycle, and this drift is added to the sensitivity drift compensation record of the corresponding equipment.

[0100] It should be noted that the baseline deviation value is obtained by averaging the drift sequence over multiple consecutive maintenance cycles. The drift values ​​from the most recent maintenance cycles are extracted from the sensitivity drift compensation records, summed, and then divided by the number of cycles to obtain the baseline deviation value.

[0101] In one possible implementation, the trend determination of the drift sequence is based on the direction of change of each value in the sequence. If the drift value of the later period is greater than that of the previous period for at least three consecutive periods, it is determined to be an increasing trend, indicating that the deviation of the equipment sensitivity is intensifying. In this case, the dynamic adjustment amplitude of the sensitivity is increased to strengthen the compensation. If the drift value of the later period is less than that of the previous period for at least three consecutive periods, it is determined to be a decreasing trend, indicating that the equipment sensitivity is gradually returning to the normal range. In this case, the dynamic adjustment amplitude is reduced to avoid overcorrection. The remaining sealing margin of the rubber ring refers to the proportion of the remaining sealing capacity of the equipment's sealing components, which is closely related to the sensitivity drift. Its original calculation benchmark plus the benchmark deviation value yields the calibrated sealing margin calculation benchmark. Further, the calculation of the detection sensitivity maintenance value is obtained by multiplying the calibrated sealing margin calculation benchmark by the dynamic adjustment amplitude, where both are dimensionless proportional values. For example, if the benchmark is 0.95 and the amplitude is 1.2, the maintenance value is 1.14. Specifically, the detection sensitivity maintenance value represents the ratio of the target sensitivity level that the device should maintain under the current sealed condition to the factory sensitivity. This maintenance value represents the target sensitivity level that each device should maintain under the current maintenance condition; a higher value indicates a lower response threshold of the device to changes in smoke concentration.

[0102] Understandably, the final detection sensitivity maintenance scheme is formed by combining the maintained detection sensitivity value with the corresponding device number and calibration timestamp. This scheme provides a quantitative reference for the sensitivity setting of each smoke detector.

[0103] If the technical solution of this application involves the processing of personal information, the relevant products have established a sound user authorization mechanism: before collecting, using, or sharing personal information, the obligation to inform is fulfilled in accordance with the law, and the individual's voluntary and explicit consent is obtained; if sensitive personal information is involved, the user's separate and explicit consent is further obtained. Specific measures include, but are not limited to: setting up prominent prompts in the information collection area, or clearly displaying the processing rules (including the processor, purpose, method, information type, etc.) through electronic interfaces such as pop-ups, checkboxes, and active submissions, to ensure that users voluntarily authorize based on their knowledge. All personal information processing activities strictly comply with national laws and regulations, especially the relevant provisions of the "Personal Information Protection Law of the People's Republic of China," to effectively safeguard the legitimate rights and interests of personal information subjects.

[0104] Obviously, those skilled in the art can make various modifications and variations to the embodiments of this application without departing from the spirit and scope of the embodiments of this application. Therefore, if these modifications and variations to the embodiments of this application fall within the scope of the claims of this application and their equivalents, this application also intends to include these modifications and variations.

Claims

1. A method for centralized management and control of regional smoke detectors, characterized in that, include: The cumulative dust accumulation at the deployment point is monitored in real time by a dust particle size acquisition probe. Combined with the compression and rebound of the rubber ring collected by the rubber ring deformation monitoring unit, and the background noise fluctuation record of the dark room scattering collected by the dark room photoelectric sensor, the remaining sealing margin of the rubber ring is calculated. The current sealing margin of the smoke detector is evaluated based on the remaining sealing margin of the rubber ring. When the sealing margin is lower than the preset sealing threshold, the current sealing state is determined to be abnormal. Based on the judgment result of the abnormal sealing status, the correlation between the cumulative dust accumulation at the monitoring deployment point and the fluctuation of the background noise scattered in the dark room is fitted. The sensitivity is dynamically adjusted according to the correlation to obtain the adjusted sensitivity. The inspection and cleaning history of each smoke detector is extracted, and the maintenance frequency and the corresponding aging degree of the rubber ring are determined according to the inspection and cleaning history. The fluctuation trend of noise increase is evaluated by comparing the change amplitude of the background noise fluctuation record in the anechoic chamber within adjacent maintenance cycles. Based on the maintenance frequency and the fluctuation trend of noise increase, a classification conclusion on the inflection point of decreasing maintenance benefits for each device is obtained. Based on the classification conclusion, the maintenance priority level of each device is determined. By combining the aging degree of the rubber ring with the fluctuating background noise of the darkroom, the seal fading rate is predicted, and an optimized disassembly and reassembly frequency plan is obtained based on the seal fading rate and the maintenance priority level. Based on the planned number of disassembly and assembly cycles and the preset allowable range of the calibration curve, the real-time update of the cumulative dust accumulation at the monitoring deployment point is integrated with the adjusted sensitivity. The deviation between the growth rate of the cumulative dust accumulation and the allowable range of the calibration curve is compared, and a maintenance scheduling instruction is generated. Based on the maintenance scheduling instruction, the calculation benchmark of the remaining sealing margin of the rubber ring is calibrated and the dynamic adjustment range of the adjusted sensitivity is adjusted to obtain the detection sensitivity maintenance scheme.

2. The centralized management and control method for regional smoke detectors according to claim 1, characterized in that, The calculation of the remaining sealing allowance of the rubber ring includes: The particle size distribution data of suspended particulate matter in the air at the deployment point is obtained by a dust particle size acquisition probe. The particles are divided into coarse particle range and fine particle range according to their diameter. The deposition mass of particles in each range is accumulated and statistically analyzed according to the sampling period to obtain the cumulative dust accumulation at the monitoring deployment point. The current thickness value of the sealing rubber ring under pressure and the recovered thickness value after pressure relief are read from the rubber ring deformation monitoring unit. The ratio of the difference between the current thickness value and the recovered thickness value to the initial thickness of the rubber ring is calculated to obtain the compression rebound amount of the rubber ring. Sampling data of scattered light intensity is extracted from the photoelectric sensor in the dark room. The average scattered light intensity under smokeless conditions is used as the background noise baseline. The difference between the scattered light intensity and the background noise baseline at each sampling time is calculated to obtain the noise deviation value. The noise deviation value is corrected by combining the compression and rebound of the rubber ring to obtain the background noise fluctuation record of the dark room. The cumulative dust accumulation, the rubber ring compression and rebound, and the background noise fluctuation record of the darkroom are input into the regression model to calculate the remaining sealing margin of the rubber ring.

3. The centralized management and control method for regional smoke detectors according to claim 1, characterized in that, The correlation between the cumulative dust accumulation at the fitted monitoring deployment points and the fluctuation of the anechoic chamber's scattered background noise includes: Based on the judgment result of the abnormal sealing state, extract the historical cumulative dust accumulation sampling sequence and the dark room scattering background noise fluctuation sampling sequence corresponding to the current device, and pair the two sets of sampling sequences point by point according to the same sampling time to obtain the paired dataset of dust accumulation and noise fluctuation. The least squares method was used to fit the paired dataset, with the cumulative dust accumulation as the independent variable and the fluctuation of the background noise scattered in the dark room as the dependent variable, to obtain the fitted curve equation. The correlation between the cumulative dust accumulation at the monitoring deployment point and the fluctuation of the background noise scattered in the dark room was determined based on the fitted curve equation.

4. The centralized management and control method for regional smoke detectors according to claim 1, characterized in that, The step of dynamically adjusting the sensitivity based on the correlation to obtain the adjusted sensitivity includes: Based on the correlation between the cumulative dust accumulation at the monitoring deployment points and the fluctuation of the background noise scattered in the darkroom, the slope value in the correlation is extracted, and the slope value is compared with the standard slope value under the factory calibration conditions. The ratio between the two is calculated to obtain the slope deviation ratio. Based on the slope deviation ratio and the sensitivity benchmark value calibrated at the factory of the smoke detector, a sensitivity compensation amount is obtained. This sensitivity compensation amount is then added to the current sensitivity value of the smoke detector to obtain the adjusted sensitivity.

5. The centralized management and control method for regional smoke detectors according to claim 1, characterized in that, The assessment of the increasing fluctuation trend of noise includes: Extract the noise fluctuation sequence of each smoke detector in adjacent maintenance cycles from the background noise fluctuation record of the anechoic chamber, calculate the average noise fluctuation in each maintenance cycle, and subtract the average noise fluctuation of the previous maintenance cycle from the average noise fluctuation of the next maintenance cycle to obtain the noise change amplitude between adjacent maintenance cycles. When the noise change amplitude is positive and the value increases periodically, the fluctuation trend of the noise increase is determined to be an upward trend. When the noise change amplitude is negative or remains stable between adjacent weeks, the fluctuation trend is determined to be a stable trend.

6. The centralized management and control method for regional smoke detectors according to claim 5, characterized in that, The classification conclusions drawn based on the fluctuation trend of the maintenance frequency and the increase in noise to obtain the inflection point of decreasing maintenance revenue for each piece of equipment include: The fluctuation trend is correlated with the maintenance frequency of each device. The records of the same device in multiple consecutive maintenance cycles are compared, and the decrease in noise fluctuation after each maintenance is calculated. When the maintenance frequency continues to increase but the decrease in noise fluctuation after each maintenance gradually decreases, the device is determined to have entered the interval of diminishing maintenance benefits, and the current maintenance frequency is marked as the inflection point of diminishing benefits. Equipment with a current maintenance frequency below the inflection point of diminishing returns is marked as equipment before the inflection point, and equipment with a current maintenance frequency above or equal to the inflection point of diminishing returns is marked as equipment after the inflection point, thus obtaining the classification conclusion of the inflection point of diminishing maintenance returns for each piece of equipment.

7. The centralized management and control method for regional smoke detectors according to claim 1, characterized in that, The process of determining the maintenance priority level of each device based on the classification conclusion includes: Based on the classification conclusion, devices after the inflection point are set as high priority, devices before the inflection point with large noise changes are set as medium priority, and devices with small noise changes are set as low priority. Extract the compression and rebound amount of the rubber ring after the most recent disassembly and assembly of each device from the rubber ring deformation monitoring unit, and associate the compression and rebound amount with the corresponding maintenance priority level. When there are multiple devices in the same priority level, sort them from low to high according to the compression and rebound amount of the rubber ring to obtain the priority level pairing record of each device.

8. The centralized management and control method for regional smoke detectors according to claim 1, characterized in that, The predicted seal descent rate includes: Acquire the aging values ​​of the rubber rings of each device and the historical sequence data of the background noise fluctuation in the darkroom. Align the aging values ​​of the rubber rings and the historical sequence of noise fluctuations point by point in chronological order to form multidimensional time series data. The aligned multidimensional time series data are fitted using the vector autoregressive moving average method to obtain a fitted curve of the sealing performance changing over time. The numerical difference of the fitted curve at adjacent time points is calculated to determine the sealing descent rate.

9. The centralized management and control method for regional smoke detectors according to claim 1, characterized in that, The generation of maintenance scheduling instructions includes: Real-time updated data of cumulative dust accumulation is extracted from the monitoring deployment points, and the ratio of the change in cumulative dust accumulation between adjacent sampling times to the sampling time interval is calculated to obtain the growth rate of cumulative dust accumulation. The cumulative dust accumulation growth rate is compared with the allowable range of the calibration curve. When it exceeds the upper limit, the excess amplitude is calculated, and the excess amplitude is corrected in combination with the adjusted sensitivity to obtain the degree of deviation. The deviation level is matched with the disassembly and assembly time window in the disassembly and assembly number plan. When the deviation level exceeds the preset deviation threshold and the current date falls within the disassembly and assembly time window, a maintenance scheduling command is triggered.

10. The centralized management and control method for regional smoke detectors according to claim 1, characterized in that, The calculation benchmark for the remaining sealing margin of the calibration rubber ring and the dynamic adjustment range of the adjusted sensitivity include: Based on the equipment number in the maintenance scheduling instruction, the sensitivity drift compensation record of the corresponding equipment is read from the equipment health database, the difference in sensitivity value before and after the maintenance scheduling instruction is triggered is calculated, the sensitivity drift amount within the current maintenance cycle is obtained, and the sensitivity drift amount is written into the sensitivity drift compensation record. The sensitivity drift sequence within multiple consecutive maintenance cycles is extracted from the sensitivity drift compensation record. The mean of the drift sequence is calculated as the reference deviation value. The original calculation reference for the remaining sealing margin of the rubber ring is added to the reference deviation value to obtain the calibrated sealing margin calculation reference. Based on the calibrated sealing margin calculation benchmark, and combined with the current rubber ring compression rebound amount and anechoic chamber scattering background noise fluctuation record of each device, the detection sensitivity maintenance value of each device is determined.

Images