A method for predicting the mean time between failures (MTBF) of a bioaerosol analyzer

Abstract

This invention relates to the field of bioaerosol analyzer technology and discloses a method for predicting the mean time between failures (MTBF) of bioaerosol analyzers. The method involves integrating operational data from the operational logs of various bioaerosol analyzers; constructing time-series sequences characterizing the operational status of key components within the same bioaerosol analyzer; aligning these time-series sequences according to the actual runtime of their respective bioaerosol analyzers to construct a time-series matrix characterizing the operational status of different bioaerosol analyzers; extracting fault data features and fault timing features from the time-series matrix; calculating the MTBF of bioaerosol analyzers under the same operating environment parameter range based on the time-series matrix; and predicting the MTBF of a target bioaerosol analyzer using a predictive learning model. This invention is beneficial for optimizing the prediction strategy for the MTBF of bioaerosol analyzers.

Description

Technical Field

[0001] This invention relates to the field of bioaerosol analyzer technology, and in particular to a method for predicting the mean time between failures (MTBF) of a bioaerosol analyzer. Background Technology

[0002] A bioaerosol analyzer is a specialized device for detecting, identifying, and analyzing organic matter in the air using bioaerosol particles. Bioaerosols are biologically derived particles suspended in the air, including bacteria, viruses, fungal spores, pollen, dust mite excrement, and plant and animal debris, typically ranging in size from 0.1 micrometers to 100 micrometers. These particles may carry pathogens or allergens, posing a potential threat to human health, the ecological environment, and biosafety. Therefore, bioaerosol analyzers have significant applications in public health, environmental monitoring, and biosafety protection.

[0003] The core function of a bioaerosol analyzer is to detect and analyze biological particles in the air (such as bacteria, viruses, fungal spores, etc.). Some bioaerosol analyzers integrate gas sensors to assist in monitoring environmental gas parameters related to bioaerosols.

[0004] Bioaerosol analyzers can monitor the concentration of bioaerosols in the air in real time, distinguish between biological and non-biological particles, and measure the particle size distribution of biological particles. Therefore, they can be widely used in hospitals, infectious disease wards, operating rooms, and other places with high cleanliness requirements, such as food processing plants and pharmaceutical workshops, for monitoring microbial contamination or for environmental and ecological research. In large monitoring areas, the density of monitoring points can be increased to form a grid-based real-time monitoring system.

[0005] Mean Time Between Failures (MTBF) refers to the average time a product or system operates correctly between two consecutive failure intervals; it is also known as the mean time between failures and is a parameter used to indicate equipment stability. MTBF prediction for bioaerosol analyzers is a core aspect of reliability engineering, primarily used to assess the long-term stability of equipment, develop maintenance strategies, and optimize design.

[0006] Bioaerosol analyzers are technology-intensive devices involving multiple disciplines such as optics, fluid mechanics, electronic control, and biosensing. Their core components (such as lasers, detectors, flow paths, and data processing modules) require high precision and environmental stability. Prolonged high-load operation, exposure to complex environmental interference, drastic changes in sampling conditions, or improper maintenance can easily lead to equipment failure. If a failure occurs, the aerosol analyzer cannot perform continuous monitoring, thus reducing the reliability of the monitoring.

[0007] In the existing technology, the prediction of the mean time between failures (MTBF) of a bioaerosol analyzer involves two key technical parameters: total runtime and total number of failures. Generally, the mean time between failures can be calculated by dividing the total runtime by the total number of failures.

[0008] When considering the total runtime of a bioaerosol analyzer, it is generally considered to be a scenario of continuous operation, that is, the total runtime is the length of the time from the start of use of the bioaerosol analyzer until the end of use.

[0009] However, bioaerosol analyzers often need to be shut down for maintenance or repair, or users may need to shut them down due to their own usage needs. As a result, the actual total running time of each bioaerosol analyzer is not consistent with the length of the time period mentioned above, which leads to a deviation in the average interval between failures calculated using the above formula.

[0010] Furthermore, in real-world usage scenarios, the failure time of a bioaerosol analyzer is often the time reported by the user to the technician, rather than the actual time the failure occurred. For example, if the user reports the failure some time after it has already occurred, the failure time recorded by the technician will not be accurate. Similarly, if the user only becomes aware of the malfunction some time after it has actually occurred, the failure time reported by the user will also be inaccurate. Therefore, the average failure interval (AUT) calculated based on this failure time will naturally not be an accurate result.

[0011] Therefore, the existing mean time between failures (MTBF) is only a rough calculation result. Using this calculation result to extrapolate the MTBF of similar bioaerosol analyzers will result in a significant deviation from the MTBF in actual use scenarios.

[0012] Therefore, there is an urgent need to propose a method for predicting the mean time between failures (MTBF) of bioaerosol analyzers in order to optimize the prediction strategy for MTBF of bioaerosol analyzers. Summary of the Invention

[0013] The main objective of this invention is to provide a method for predicting the mean time between failures (MTBF) of a bioaerosol analyzer, aiming to optimize the prediction strategy for the MTBF of a bioaerosol analyzer.

[0014] To achieve the above objectives, the present invention provides a method for predicting the mean time between failures (MTBF) of a bioaerosol analyzer, comprising the following steps:

[0015] The operating environment parameters of each bioaerosol analyzer, the operating data of each key component in each bioaerosol analyzer, and the sampling time series of each bioaerosol analyzer were obtained from the operation logs of each bioaerosol analyzer.

[0016] The system integrates the operational data of each key component of a bioaerosol analyzer under the same operating environment parameter range and sends it to the Internet of Things platform; wherein, the operating environment parameters include the test temperature, humidity, particulate matter concentration and bioaerosol concentration obtained by biosensors.

[0017] Based on the operational data of each key component of the bioaerosol analyzer, a time series sequence is constructed to characterize the operational status of each key component of the same bioaerosol analyzer.

[0018] The time series sequences were aligned according to the actual runtime of their respective bioaerosol analyzers to construct a time series matrix for characterizing the operation of different bioaerosol analyzers.

[0019] Extract fault data features and fault time point features from the time series matrix;

[0020] Based on the time series matrix, the average failure interval of the bioaerosol analyzer is calculated under the same operating environment parameter range.

[0021] The estimated operating environment parameters are input into the predictive learning model to predict the mean time between failures of the target bioaerosol analyzer.

[0022] Optionally, the step of integrating the operating data of each key component of each bioaerosol analyzer under the same operating environment parameter range includes:

[0023] Obtain the preset range of parameters for each operating environment;

[0024] Obtain operational data for each key component of each of the bioaerosol analyzers that conforms to the same operating environment parameter range;

[0025] Establish a corresponding operation sequence table for each of the key components of each of the bioaerosol analyzers;

[0026] The analyzer ID, the operating environment parameter range, and the key component ID of each bioaerosol analyzer are stored in the information unit of the operation sequence table corresponding to each key component.

[0027] Obtain the sampling time series of each of the bioaerosol analyzers, and generate a data unit for each of the operation sequence tables based on each sampling time point in the sampling time series;

[0028] The actual operating environment parameters corresponding to each sampling time point, the operating data collected by the key components at each sampling time point, and the sampling time point are stored in the data unit of the corresponding operating sequence table.

[0029] Optionally, the step of constructing a time-series sequence characterizing the operation of each key component of the same bioaerosol analyzer based on the operational data of each key component includes:

[0030] Under the same operating environment parameter range, the operating sequence list of each key component of the same bioaerosol analyzer is aligned and associated with the data units according to the sampling time point;

[0031] Based on the associated and aligned operation sequence tables, a time series is established to characterize the operation of each key component of the same bioaerosol analyzer.

[0032] Optionally, the step of aligning each of the time series according to the actual runtime of its respective bioaerosol analyzer to construct a time series matrix characterizing the operating conditions of different bioaerosol analyzers includes:

[0033] The sampling time points recorded in each data unit of the time series corresponding to different bioaerosol analyzers are converted into the actual running time of the bioaerosol analyzer, and the actual running time corresponding to each data unit is stored in the data unit;

[0034] Each of the time series sequences is aligned according to its actual runtime to construct the time series matrix.

[0035] Optionally, the step of extracting fault data features and fault time point features from the time series matrix includes:

[0036] Obtain the fault data characteristics of typical faults of each of the key components;

[0037] Fault data features of each bioaerosol analyzer are extracted from the time series matrix to determine the fault time point of each bioaerosol analyzer.

[0038] The number of failures of each bioaerosol analyzer is determined based on the failure time point of each bioaerosol analyzer.

[0039] Optionally, the step of extracting fault data features of each bioaerosol analyzer from the time series matrix to determine the fault time point of each bioaerosol analyzer includes:

[0040] Using the fault data characteristics as a scanning window, the time series of each key component of each bioaerosol analyzer is scanned to extract data units that conform to the fault data characteristics.

[0041] The sampling time point stored in the data unit from which the fault data characteristics are extracted is taken as the fault time point of the bioaerosol analyzer.

[0042] Optionally, the step of determining the number of failures of the bioaerosol analyzer based on the failure time point of each bioaerosol analyzer includes:

[0043] The fault time points extracted from the same bioaerosol analyzer are added to the same fault time point set and sorted within the fault time point set;

[0044] Based on the first fault time point in the same set of fault time points, the number of faults corresponding to the set of fault time points is set to 1;

[0045] Determine whether the time difference between each adjacent fault time point exceeds the set time difference;

[0046] If so, increment the number of faults corresponding to the set of fault time points by 1.

[0047] Optionally, the step of calculating the average time between failures of the bioaerosol analyzer under the same operating environment parameter range based on the time series matrix includes:

[0048] Based on the actual total runtime and the number of failures of each of the bioaerosol analyzers under the same operating environment parameter range, the average failure interval time of the bioaerosol analyzer under the same operating environment parameter range is calculated.

[0049] Optionally, the number of malfunctions of the bioaerosol analyzer can be determined as follows:

[0050] Generate scan window:

[0051] ;

[0052] in, The scanning window is used to scan the data unit sequences in the operation sequence table of the kth key component of the nth bioaerosol analyzer, and the scanning starts from the sampling time point t. This refers to the sequence of data units in the operation sequence table of the kth key component of the nth bioaerosol analyzer; From sampling time point t to The index;

[0053] Based on the runtime data extracted from each scanning window, the following data features are calculated:

[0054] (1) ;

[0055] in, This represents the average value of similar running data for each data unit scanned within the scanning window; A slice of running data for the data cells scanned by the scanning window; t≤i≤ i represents the sampling time point from t to... The i-th time point within;

[0056] (2) ;

[0057] in, Used to indicate the degree of fluctuation of similar running data for each data unit scanned within the scanning window;

[0058] (3) , ;

[0059] in, This represents the maximum value of similar running data for each data unit scanned within the scanning window. It represents the minimum value of similar running data for each data unit scanned within the scanning window;

[0060] (4) ;

[0061] in, This indicates the degree of multivariate overall deviation, that is, the overall deviation of the operating data of various key components of the same bioaerosol analyzer within the same scan time series. This represents the mean value within the scan window of the j-th type of running data, 1≤j≤D, where D is the number of running data types of the same bioaerosol analyzer; This is the reference mean of the j-th type of running data; This serves as a reference fluctuation for the j-th type of runtime data;

[0062] (5) A data unit is marked as conforming to the fault data characteristics if any of the following conditions are met:

[0063] or , ;

[0064] in, and These are preset thresholds;

[0065] (6) Where G is the set of fault periods extracted using the scanning window; M is the number of fault periods extracted.

[0066] After sorting G in chronological order and merging overlapping or adjacent fault periods, the result is as follows:

[0067] ;

[0068] in, This is the result of merging the fault periods. , The maximum allowed time interval is set;

[0069] Therefore, the set of failure time points obtained after merging is:

[0070] ;

[0071] ;

[0072] in, These are the time points of the failure. This is the set of failure time points obtained after merging; This represents the (s+1)th fault time point in the set of fault time points. This represents the s-th fault time point in the set of fault time points. > ; To set a time difference; This represents the number of failures.

[0073] Optionally, the key components include: a sampling system, a preprocessing system, a detection system, a power supply, and a heat dissipation device.

[0074] In the technical solution of this invention, a data storage sequence is used to integrate the data recorded in the operation logs of each bioaerosol analyzer. Obtaining data through the operation logs is simpler, more convenient, and more accurate. The operation data of each key component of each bioaerosol analyzer under the same operating environment parameter range are integrated and sent to an IoT platform, enabling the prediction of the mean time between failures (MTBF) of the bioaerosol analyzers on the IoT platform. This invention also considers the impact of different operating environments on the MTBF prediction. Furthermore, based on the operation data of each key component of the bioaerosol analyzer, a time series sequence is constructed to characterize the operation of each key component of the same bioaerosol analyzer. These time series sequences are aligned according to the actual runtime of their respective bioaerosol analyzers to construct a time series matrix characterizing the operation of different bioaerosol analyzers. Therefore, the operating parameters within each bioaerosol analyzer in this invention are aligned according to the actual runtime, thus eliminating the interference of the bioaerosol analyzer's non-operation time on the total runtime. Furthermore, this invention obtains actual operating data of each key component of the bioaerosol analyzer, extracts fault data features and fault timing features from the time series matrix, making the fault time of each bioaerosol analyzer more accurate, thus ensuring an accurate number of faults within the total operating time. Finally, the estimated operating environment parameters are input into a predictive learning model, which is then used to predict the average fault interval time (AMT) of the target bioaerosol analyzer, resulting in a more accurate AMT prediction. Therefore, this invention studies time series matrices aligned with the actual operating time of various bioaerosol analyzers under the same operating environment parameter range, achieving alignment of the operating environment and total operating time for numerous bioaerosol analyzers, and obtaining an accurate number of faults for each bioaerosol analyzer, thus making the AMT prediction more accurate. Therefore, this invention optimizes the AMT prediction strategy, which helps improve the accuracy of the prediction results. Attached Figure Description

[0075] Figure 1 This is a flowchart of the first embodiment of the method for predicting the mean time between failures (MTBF) of a bioaerosol analyzer according to the present invention;

[0076] Figure 2 This is a schematic diagram of a running sequence marker in this invention;

[0077] Figure 3 This is a schematic diagram of the time sequence used to characterize the operation of various key components of the same bioaerosol analyzer in this invention.

[0078] Figure 4 This is a schematic diagram of the timing matrix in this invention.

[0079] The objectives, features, and advantages of this invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0080] It should be understood that the specific embodiments described herein are for illustrative purposes only and are not intended to limit the scope of the invention.

[0081] In the following description, the use of suffixes such as "unit," "component," or "element" to denote elements is solely for the purpose of illustrative purposes and has no specific meaning in itself. Therefore, "unit," "component," or "element" may be used interchangeably.

[0082] Please see Figures 1 to 4 The first embodiment of the present invention provides a method for predicting the mean time between failures (MTBF) of a bioaerosol analyzer, comprising the following steps:

[0083] Step S10: Obtain the operating environment parameters of each bioaerosol analyzer, the operating data of each key component in each bioaerosol analyzer, and the sampling time series of each bioaerosol analyzer from the operation log of each bioaerosol analyzer.

[0084] Step S20: Integrate the operating data of each key component of each bioaerosol analyzer under the same operating environment parameter range, and send them to the Internet of Things platform; wherein, the operating environment parameters include the test temperature, humidity, particulate matter concentration and bioaerosol concentration obtained by the biosensor.

[0085] Step S30: Based on the operating data of each key component of the bioaerosol analyzer, construct a time series to characterize the operating status of each key component of the same bioaerosol analyzer.

[0086] Step S40: Align each of the time series according to the actual running time of the bioaerosol analyzer to construct a time series matrix for characterizing the operating conditions of different bioaerosol analyzers;

[0087] Step S50: Extract fault data features and fault time point features from the time series matrix;

[0088] Step S60: Calculate the average failure interval time of the bioaerosol analyzer under the same operating environment parameter range based on the time series matrix.

[0089] Step S70: Input the estimated operating environment parameters into the predictive learning model to predict the mean time between failures of the target bioaerosol analyzer.

[0090] In the technical solution of this invention, a data storage sequence is used to integrate the data recorded in the operation logs of each bioaerosol analyzer. Obtaining data through the operation logs is simpler, more convenient, and more accurate. The operation data of each key component of each bioaerosol analyzer under the same operating environment parameter range are integrated and sent to an IoT platform, enabling the prediction of the mean time between failures (MTBF) of the bioaerosol analyzers on the IoT platform. This invention also considers the impact of different operating environments on the MTBF prediction. Furthermore, based on the operation data of each key component of the bioaerosol analyzer, a time series sequence is constructed to characterize the operation of each key component of the same bioaerosol analyzer. These time series sequences are aligned according to the actual runtime of their respective bioaerosol analyzers to construct a time series matrix characterizing the operation of different bioaerosol analyzers. Therefore, the operating parameters within each bioaerosol analyzer in this invention are aligned according to the actual runtime, thus eliminating the interference of the bioaerosol analyzer's non-operation time on the total runtime. Furthermore, this invention obtains actual operating data of each key component of the bioaerosol analyzer, extracts fault data features and fault timing features from the time series matrix, making the fault time of each bioaerosol analyzer more accurate, thus ensuring an accurate number of faults within the total operating time. Finally, the estimated operating environment parameters are input into a predictive learning model, which is then used to predict the average fault interval time (AMT) of the target bioaerosol analyzer, resulting in a more accurate AMT prediction. Therefore, this invention studies time series matrices aligned with the actual operating time of various bioaerosol analyzers under the same operating environment parameter range, achieving alignment of the operating environment and total operating time for numerous bioaerosol analyzers, and obtaining an accurate number of faults for each bioaerosol analyzer, thus making the AMT prediction more accurate. Therefore, this invention optimizes the AMT prediction strategy, which helps improve the accuracy of the prediction results.

[0091] It is easy to understand that the bioaerosol analyzers in this invention are all of the same configuration and model. Historical data from the operation logs of these bioaerosol analyzers of the same configuration and model are used to predict the mean time between failures (MTBF) of similar bioaerosol analyzers.

[0092] Specifically, the operating environment parameters of each bioaerosol analyzer, the operating data of each key component, and the sampling time of each bioaerosol analyzer can be obtained from the operation logs of each bioaerosol analyzer. Extracting each sampling time forms a sampling time series. This allows analysis using the data already stored in the equipment, eliminating the need to acquire additional data resources and improving the data acquisition convenience of this invention.

[0093] The technical solution of this invention utilizes operational data from a bioaerosol analyzer during operation for fault extraction and analysis. Specifically, since the operating environment parameters of a bioaerosol analyzer may vary, and these differences may lead to varying degrees of instability, this invention divides the operating environment parameter range into multiple ranges. Data from bioaerosol analyzers conforming to each range are treated as a separate object for study. This allows for the investigation of the operational status of key components within the bioaerosol analyzer under similar operating environment parameters, thereby predicting the mean time between failures (MTBF) of the bioaerosol analyzer under similar operating environment parameters.

[0094] Specifically, the key components of each bioaerosol analyzer are those most susceptible to malfunction. The workflow of a bioaerosol analyzer typically includes sampling, pretreatment, detection, and data processing, each stage relying on the coordinated operation of different components. Components susceptible to malfunction in a bioaerosol analyzer may include:

[0095] Sampling system, including sampling pump;

[0096] The pretreatment system includes a particle grading device, a dehumidification module, a heating module, and a filtration device;

[0097] Detection systems, such as laser-induced fluorescence devices and biosensor devices;

[0098] The data processing and display system includes: signal amplification and filtering circuits, and a data acquisition card;

[0099] Auxiliary systems include: power supply and heat dissipation devices.

[0100] According to a first embodiment of the method for predicting the mean time between failures (MTBF) of a bioaerosol analyzer based on the present invention, and in a second embodiment of the method for predicting the mean time between failures (MTBF) of a bioaerosol analyzer based on the present invention, step S20 includes:

[0101] Step S21: Obtain the preset range of parameters for each operating environment;

[0102] Step S22: Obtain the operating data of each key component of each of the bioaerosol analyzers that conforms to the same operating environment parameter range;

[0103] Step S23: Establish a corresponding operation sequence list for each key component of each bioaerosol analyzer;

[0104] Step S24: Store the analyzer ID, the operating environment parameter range, and the key component ID corresponding to each bioaerosol analyzer into the information unit of the operation sequence table corresponding to each key component;

[0105] Step S25: Obtain the sampling time series of each of the bioaerosol analyzers, and generate a data unit for each of the operation sequence tables according to each sampling time point in the sampling time series;

[0106] Step S26: Store the actual operating environment parameters corresponding to each sampling time point, the operating data collected by the key components at each sampling time point, and the sampling time point into the data unit of the corresponding operating sequence table.

[0107] Specifically, the operating environment parameters refer to the environmental parameters that affect the performance stability and service life of the bioaerosol analyzer. In this embodiment, the selected operating environment parameters include: temperature, humidity, particulate matter concentration and bioaerosol concentration (i.e., target particle concentration). Vibration and other harmful environmental factors, such as the concentration of corrosive gases, can also be further considered.

[0108] In this embodiment, multiple operating environment parameter ranges set by the user are obtained. For example, one of the set operating environment parameter ranges can be: temperature range (20℃, 30℃), humidity range (40%, 60%), and particulate matter concentration range (0℃, 30℃, 40%). 500 ], bioaerosol concentration range ( ].

[0109] The stored operating environment parameters are obtained from the historical operation log data of the bioaerosol analyzer. It is determined whether each type of operating environment parameter meets the currently set operating environment parameter range. Bioaerosol analyzers that meet the same operating environment parameter range are regarded as the same type of object, and the operation data of each key component of each bioaerosol analyzer in the same type of object are integrated separately.

[0110] The operating data for key components is as follows:

[0111] (1) The flow rate and output pressure of the sampling pump;

[0112] (2) Particle size classification efficiency of the particle classification device;

[0113] (3) Humidity data sampled by the dehumidification module;

[0114] (4) Temperature rise rate of the heating module;

[0115] (5) Filtration efficiency and pressure difference of the filtration device;

[0116] (6) Signal response values, baseline drift, and calibration bias of biosensor devices;

[0117] (7) Gain stability, noise level and filtering effect of signal amplification and filtering circuit;

[0118] (8) Sampling rate of the data acquisition card;

[0119] (9) Voltage fluctuations and battery life of the power supply;

[0120] (10) The operating parameters of the heat dissipation device are based on the core location temperature of the bioaerosol analyzer;

[0121] Each running sequence table is divided into an information unit for the header and a data unit for recording the specific data at each sampling time point.

[0122] Taking the operation sequence table of the sampling pump of a certain bioaerosol analyzer as an example, its information unit records the operating environment parameters of the bioaerosol analyzer and the ID of the key component to which the operation sequence table belongs. This allows for the integration of various bioaerosol analyzers operating in similar operating environments through the information unit for predictive analysis, and also facilitates the integration of data from the same key component of different bioaerosol analyzers for predictive analysis.

[0123] Furthermore, each operation sequence list contains several data units, each data unit recording the sampling time point corresponding to the key component, as well as the operation data collected at that sampling time point. For example, in the operation sequence list of the sampling pump, the first data unit records the flow rate and output pressure at the first sampling time; the second data unit records the flow rate and output pressure at the second sampling time.

[0124] By using the operation sequence tables corresponding to the key components of the same bioaerosol analyzer, the fluctuation of the operation status of each key component over time under the same operating environment parameters is recorded, so as to determine the failure time of each key component of the same bioaerosol analyzer. Furthermore, by using the fluctuation of the operation data of the same key component of different bioaerosol analyzers within the same operating environment parameter range over time, it is helpful to analyze the similarities and differences in the operation status of the same key component.

[0125] According to a second embodiment of the method for predicting the mean time between failures (MTBF) of a bioaerosol analyzer based on the present invention, and in a third embodiment of the method for predicting the mean time between failures (MTBF) of a bioaerosol analyzer based on the present invention, step S30 includes:

[0126] Step S31: Under the same operating environment parameter range, the operation sequence list of each key component of the same bioaerosol analyzer is aligned and associated with the data units according to the sampling time point.

[0127] Step S32: Based on the associated and aligned operation sequence tables, establish a time series to characterize the operation of each key component of the same bioaerosol analyzer.

[0128] Specifically, the operational sequence lists of each key component of each bioaerosol analyzer are aligned according to the sampling time points. For example, the information units of the sampling pump's operational sequence list, the particle classifier's information units, and the dehumidification module's information units are aligned (e.g., vertically or horizontally). Then, the first data unit of the sampling pump's operational sequence list, the first data unit of the particle classifier, and the first data unit of the dehumidification module are aligned; the second data unit of the sampling pump's operational sequence list, the second data unit of the particle classifier, and the second data unit of the dehumidification module are aligned, and so on, until each data unit of each key component is aligned. The alignment result is the time sequence of the same bioaerosol analyzer.

[0129] By using time-series data from the same bioaerosol analyzer, the operational status of each key component under the same operating environmental parameters can be observed over time. Furthermore, by analyzing the normal operating range set for each key component, it can be determined whether the operational data stored in each data unit has reached the fault range. This allows for the identification of the fluctuations in operating parameters and their cycles before failure in each key component of the same bioaerosol analyzer, as well as the sequence of failures of each key component across different bioaerosol analyzers.

[0130] In a third embodiment of the method for predicting the mean time between failures (MTBF) of a bioaerosol analyzer according to the present invention, and in a fourth embodiment of the method for predicting the mean time between failures (MTBF) of a bioaerosol analyzer according to the present invention, step S40 includes:

[0131] Step S41: Convert the sampling time points recorded in each data unit in the time series corresponding to different bioaerosol analyzers into the actual running time of the bioaerosol analyzer, and store the actual running time corresponding to each data unit into the data unit;

[0132] Step S42: Align each of the time series sequences according to the actual runtime to construct the time series matrix.

[0133] Because this invention aims to integrate and study data from bioaerosol analyzers with similar operating environment parameters, and because different bioaerosol analyzers have different start-up times, runtimes, and shutdown times—for example, analyzer A might run for 10 days, then shut down, and restart after 15 days; or analyzer A might run for 3 days, then shut down, and restart after 4 days—if the total runtime is calculated based on the initial start-up time, the runtime will not actually align due to different shutdowns and other factors affecting different bioaerosol analyzers after a period of use. The difference in the actual runtime of each bioaerosol analyzer will lead to a large error in the mean time between failures (MTBF).

[0134] Therefore, in this embodiment, the time series of different bioaerosol analyzers are aligned according to their actual operating time. For example, the operating data of analyzer A for 100 hours of actual operating time is aligned with the operating data of analyzer B for 100 hours of actual operating time to delete the actual unrunning segments that are meaningless for predicting the mean time between failures and cause significant prediction bias. The actual unrunning segments are the operation interruption events.

[0135] Operational interruption events include: analyzer shutdown events, analyzer maintenance events, and calibration events. As needed, operational interruption events can be further included to include low-power standby (no sampling) events.

[0136] However, it is quite difficult to align the time series of different bioaerosol analyzers according to their actual running time, and there is currently no effective method for statistical analysis of actual running time.

[0137] In this invention, the sampling timestamps in the operation log are obtained, and each sampling timestamp is recorded in the corresponding data unit. However, the recording format of the sampling timestamp is: year-month-day, hour-minute-second. Therefore, the total runtime cannot be obtained directly from the sampling timestamps. Instead, the total runtime of the biocolloid analyzer can be calculated by accumulating the sampling time intervals of each data unit in the same biocolloid analyzer and subtracting the time intervals that significantly exceed the sampling time period (for example, when two adjacent sampling times are 2020-06-25 08:00:00 and 2020-06-28 08:00:00, the interval between them is too long, so this sampling time interval can be considered as downtime and excluded from the total runtime).

[0138] Meanwhile, this invention employs time series sequences and time series matrices, which can clearly define the sampling order of each data unit in each time series sequence, and also obtain the specific sampling time point through the sampling time point stored in each data unit.

[0139] In a fourth embodiment of the method for predicting the mean time between failures (MTBF) of a bioaerosol analyzer according to the present invention, and in a fifth embodiment of the method for predicting the mean time between failures (MTBF) of a bioaerosol analyzer according to the present invention, step S50 includes:

[0140] Step S51: Obtain the fault data characteristics of typical faults of each of the key components;

[0141] Step S52: Extract the fault data features of each bioaerosol analyzer from the time series matrix to determine the fault time point of each bioaerosol analyzer;

[0142] Step S53: Determine the number of failures of each bioaerosol analyzer based on the failure time point of each bioaerosol analyzer.

[0143] In a fifth embodiment of the method for predicting the mean time between failures (MTBF) of a bioaerosol analyzer according to the present invention, and in a sixth embodiment of the method for predicting the mean time between failures (MTBF) of a bioaerosol analyzer according to the present invention, step S52 includes:

[0144] Step S521: Using the fault data features as a scanning window, the time sequence of each key component of each bioaerosol analyzer is scanned to extract data units that conform to the fault data features.

[0145] Step S522: The sampling time point stored in the data unit from which the fault data features are extracted is taken as the fault time point of the bioaerosol analyzer.

[0146] In a sixth embodiment of the method for predicting the mean time between failures (MTBF) of a bioaerosol analyzer according to the present invention, and in a seventh embodiment of the method for predicting the mean time between failures (MTBF) of a bioaerosol analyzer according to the present invention, step S53 includes:

[0147] Step S531: Add the fault time points extracted from the same bioaerosol analyzer to the same fault time point set and sort them within the fault time point set;

[0148] Step S532: Based on the first fault time point in the same set of fault time points, set the number of faults corresponding to the set of fault time points to 1;

[0149] Step S533: Determine whether the time difference between each adjacent fault time point exceeds the set time difference;

[0150] Step S534: If yes, increment the number of faults corresponding to the set of fault time points by 1.

[0151] Specifically, the number of malfunctions of the bioaerosol analyzer is determined as follows:

[0152] Generate scan window:

[0153] ;

[0154] in, The scanning window is used to scan the data unit sequences in the operation sequence table of the kth key component of the nth bioaerosol analyzer, and the scanning starts from the sampling time point t. This refers to the sequence of data units in the operation sequence table of the kth key component of the nth bioaerosol analyzer; From sampling time point t to The index;

[0155] Based on the runtime data extracted from each scanning window, the following data features are calculated:

[0156] (1) ;

[0157] in, This represents the average value of similar running data for each data unit scanned within the scanning window; A slice of running data for the data cells scanned by the scanning window; t≤i≤ i represents the sampling time point from t to... The i-th time point within;

[0158] (2) ;

[0159] in, Used to indicate the degree of fluctuation of similar running data for each data unit scanned within the scanning window;

[0160] (3) , ;

[0161] in, This represents the maximum value of similar running data for each data unit scanned within the scanning window. It represents the minimum value of similar running data for each data unit scanned within the scanning window;

[0162] (4) ;

[0163] in, This indicates the degree of multivariate overall deviation, that is, the overall deviation of the operating data of various key components of the same bioaerosol analyzer within the same scan time series. This represents the mean value within the scan window of the j-th type of running data, 1≤j≤D, where D is the number of running data types of the same bioaerosol analyzer; This is the reference mean of the j-th type of running data; This serves as a reference fluctuation for the j-th type of runtime data;

[0164] (5) A data unit is marked as conforming to the fault data characteristics if any of the following conditions are met:

[0165] or , ;

[0166] in, and These are preset thresholds;

[0167] (6) Where G is the set of fault periods extracted using the scanning window; M is the number of fault periods extracted.

[0168] After sorting G in chronological order and merging overlapping or adjacent fault periods, the result is as follows:

[0169] ;

[0170] in, This is the result of merging the fault periods. , The maximum allowed time interval is set;

[0171] Therefore, the set of failure time points obtained after merging is:

[0172] ;

[0173] ;

[0174] in, These are the time points of the failure. This is the set of failure time points obtained after merging; This represents the (s+1)th fault time point in the set of fault time points. This represents the s-th fault time point in the set of fault time points. > ; To set a time difference; This represents the number of failures.

[0175] In a seventh embodiment of the method for predicting the mean time between failures (MTBF) of a bioaerosol analyzer according to the present invention, and in an eighth embodiment of the method for predicting the mean time between failures (MTBF) of a bioaerosol analyzer according to the present invention, step S60 includes:

[0176] Step S61: Based on the actual total running time and the number of failures of each of the bioaerosol analyzers under the same operating environment parameter range, calculate the average failure interval time of the bioaerosol analyzer under the same operating environment parameter range.

[0177] By using the actual total runtime and the number of failures for each of the aforementioned bioaerosol analyzers, the average interval between failures (AQI) of the bioaerosol analyzers under the same operating environment parameter range can be calculated. For example:

[0178] ;

[0179] in, N represents the average time between failures for similar bioaerosol analyzers, and N represents the number of similar bioaerosol analyzers. Let n be the actual total runtime of the nth bioaerosol analyzer. This represents the number of malfunctions of the nth bioaerosol analyzer.

[0180] Further, step S70 includes:

[0181] Based on the above method, the mean time between failures (MTBF) is calculated for each type of bioaerosol analyzer under each range of operating environmental parameters.

[0182] This allows us to establish a mapping relationship between the type of bioaerosol analyzer, the range of operating environment parameters, and the mean time between failures, in order to build a predictive learning model.

[0183] By inputting the target bioaerosol analyzer and the estimated operating environment parameters into the predictive learning model, the mean time between failures (MTBF) of the target bioaerosol analyzer can be output, thereby enabling the prediction of the MTBF of the target bioaerosol analyzer.

[0184] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a computer-readable storage medium (such as ROM / RAM, magnetic disk, optical disk) as described above, and includes several instructions to cause a terminal device to enter the methods described in the various embodiments of the present invention.

[0185] In the description of this specification, references to terms such as "one embodiment," "another embodiment," "other embodiments," or "first embodiment to Xth embodiment," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the present invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, method steps, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0186] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or system that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or system. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or system that includes that element.

[0187] The sequence numbers of the above embodiments of the present invention are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments.

[0188] The above are merely preferred embodiments of the present invention and do not limit the patent scope of the present invention. Any equivalent structural or procedural transformations made based on the content of the present invention's specification and drawings, or direct or indirect applications in other related technical fields, are similarly included within the patent protection scope of the present invention.

Claims

1. A bioaerosol analyzer mean time between failure prediction method, characterized by, The method comprises the following steps: obtain the operation environment parameters of each bioaerosol analyzer, the operation data of each key component in each bioaerosol analyzer, and the sampling time sequence of each bioaerosol analyzer from the operation log of each bioaerosol analyzer; integrate the operation data of each key component of each bioaerosol analyzer under the same operation environment parameter interval and send to the Internet of Things platform; wherein the operation environment parameters include the temperature, humidity, particulate matter concentration and bioaerosol concentration obtained by the biological sensor of the test; construct a time sequence sequence for representing the operation condition of each key component of the same bioaerosol analyzer according to the operation data of each key component of the bioaerosol analyzer; align each time sequence sequence according to the actual operation duration of the bioaerosol analyzer to which it belongs, and construct a time sequence matrix for representing the operation condition of different bioaerosol analyzers; extract fault data features and fault time point features from the time sequence matrix; calculate the average fault interval time of the bioaerosol analyzer under the same operation environment parameter interval according to the time sequence matrix; input the estimated operation environment parameters into the prediction learning model to predict the average fault interval time of the target bioaerosol analyzer.

2. The bioaerosol analyzer mean time between failure prediction method of claim 1, wherein, The step of integrating the operation data of each key component of each bioaerosol analyzer under the same operation environment parameter interval comprises: obtain each operation environment parameter interval; obtain the operation data of each key component of each bioaerosol analyzer under the same operation environment parameter interval; establish a corresponding operation sequence list for each key component of each bioaerosol analyzer; store the analyzer ID corresponding to each bioaerosol analyzer, the operation environment parameter interval corresponding to each bioaerosol analyzer and the key component ID belonging to it into the information unit of the operation sequence list corresponding to each key component; obtain the sampling time sequence of each bioaerosol analyzer, and generate a data unit for each operation sequence list according to each sampling time point in the sampling time sequence; store the actual operation environment parameters corresponding to each sampling time point, the operation data collected by the key component at each sampling time point, and the sampling time point into the data unit of the corresponding operation sequence list.

3. The bioaerosol analyzer mean time between failure prediction method of claim 2, wherein, The step of constructing a time sequence sequence for representing the operation condition of each key component of the same bioaerosol analyzer according to the operation data of each key component of the bioaerosol analyzer comprises: align the operation sequence lists of each key component of the same bioaerosol analyzer under the same operation environment parameter interval according to the sampling time points of the data units; establish a time sequence sequence for representing the operation condition of each key component of the same bioaerosol analyzer according to the aligned operation sequence lists.

4. The bioaerosol analyzer mean time between failure prediction method of claim 3, wherein, The step of aligning each time sequence sequence according to the actual operation duration of the bioaerosol analyzer to which it belongs to construct a time sequence matrix for representing the operation condition of different bioaerosol analyzers comprises: Converting the sampling time points recorded by each data unit in the time sequence corresponding to each bioaerosol analyzer into actual running time of the bioaerosol analyzer, and storing the actual running time corresponding to each data unit into the data unit; Aligning each time sequence according to the actual running time to construct the time matrix.

5. The bioaerosol analyzer mean time between failure prediction method of claim 4, wherein, The step of extracting the fault data features and the fault time points from the time matrix comprises: obtaining the fault data features of the typical fault of each key component; extracting the fault data features of each bioaerosol analyzer from the time matrix to determine the fault time points of each bioaerosol analyzer; determining the fault times of the bioaerosol analyzer according to the fault time points of each bioaerosol analyzer.

6. The bioaerosol analyzer mean time between failure prediction method of claim 5, wherein, The step of extracting the fault data features of each bioaerosol analyzer from the time matrix to determine the fault time points of each bioaerosol analyzer comprises: using the fault data features as a scanning window to scan the time sequence of each key component of each bioaerosol analyzer respectively to extract data units meeting the fault data features; storing the sampling time points stored in the data units extracted to the fault data features as the fault time points of the bioaerosol analyzer.

7. The bioaerosol analyzer mean time between failure prediction method of claim 6, wherein, The step of determining the fault times of the bioaerosol analyzer according to the fault time points of each bioaerosol analyzer comprises: adding the fault time points extracted from the same bioaerosol analyzer to the same fault time point set and sorting them in the fault time point set; setting the fault times corresponding to the fault time point set as 1 according to the first fault time point in the same fault time point set; judging whether the time difference between each adjacent fault time point exceeds the set time difference; if so, performing a plus 1 operation on the fault times corresponding to the fault time point set.

8. The bioaerosol analyzer mean time between failure prediction method of claim 7, wherein, The step of calculating the average fault interval time of the bioaerosol analyzer under the same running environment parameter interval according to the time matrix comprises: based on the actual total running time and the fault times of each bioaerosol analyzer corresponding to the same running environment parameter interval, calculating the average fault interval time of the bioaerosol analyzer under the same running environment parameter interval.

9. The bioaerosol analyzer mean time between failure prediction method of claim 8, wherein, The fault times of the bioaerosol analyzer are determined in the following manner: Generate a scanning window: ； wherein, is a scan window for scanning each data unit sequence in the run sequence list of the kth critical component of the nth bioaerosol analyzer, and the scanning starts from the sampling time point t; is each data unit sequence in the run sequence list of the kth critical component of the nth bioaerosol analyzer; is an index of the sampling time point t to . According to the running data extracted by each scanning window, the following data feature calculation is performed: （1） ； wherein, is an average of the same kind of running data of each data unit scanned within the scan window; is a running data slice of the data unit scanned by the scan window; t≤i≤ , i represents the i-th time point within the sampling time point t to . （2） ； wherein, to indicate the degree of variability of the homogenous run data for each data unit scanned within the scan window; （3） ， ； wherein, is a maximum value of the homogeneous running data of each data unit scanned within the scan window, is a minimum value of the homogeneous running data of each data unit scanned within the scan window; （4） ； wherein, represents the multivariate comprehensive deviation degree, i.e. the comprehensive deviation of the operation data of each key component of the same bioaerosol analyzer occurring within the same scanning timing, represents the mean value within the scanning window of the jth type of operation data, 1≤j≤D, D being the number of types of operation data of the same bioaerosol analyzer; is the reference mean value of the jth type of operation data; is the reference fluctuation of the jth type of operation data; (5) If any of the following conditions is met, it is marked as a data unit meeting the fault data features: or , ; wherein, and are preset threshold values, respectively. (6) where G is the set of failure time periods extracted using the scan window; and M is the number of failure time periods extracted. After sorting G in chronological order, the result obtained by merging overlapping or adjacent fault periods is: ； wherein, is the fault period merge result, , is the set maximum allowed time interval; Thus, the fault time point set obtained after merging is: ； ； wherein, respectively are fault time points, is a set of fault time points obtained after merging; denotes the s+1th fault time point in the set of fault time points, denotes the sth fault time point in the set of fault time points, > ; is a set time difference; is a fault number.

10. The bioaerosol analyzer mean time between failure prediction method of any of claims 1 to 9, wherein, The key components include: sampling system, pretreatment system, detection system, power supply and heat dissipation device.

Images