A laboratory clean attenuation image-based environmental risk early warning method and device

Abstract

The application discloses an environment risk early warning method and device based on a laboratory clean attenuation image, belongs to the technical field of environment risk early warning, obtains and pre-processes multi-dimensional operation data, constructs a clean ability attenuation image based on the pre-processed multi-dimensional operation data, obtains a cleanliness level and a clean ability attenuation trend according to the clean ability attenuation image, maps the cleanliness level to a current risk level, and performs dynamic correction, performs maintenance node prediction based on the clean ability attenuation trend and the corrected risk level, so as to perform environment risk early warning. The clean ability change trend is directly displayed, remote monitoring and real-time early warning of laboratory risks are realized, the clean ability attenuation image is constructed through multi-dimensional operation data, dynamic analysis and pre-judgment of the environment parameter change trend are realized, the risks are identified in the early stage of risk incubation, the attenuation process of the clean environment gradually deteriorating with the lapse of time is fully depicted, and environment risk early warning is performed in time.

Description

Technical Field

[0001] This invention belongs to the field of environmental risk early warning technology, specifically relating to an environmental risk early warning method and device based on laboratory cleanliness degradation profiles. Background Technology

[0002] Laboratories are places for conducting experiments, encompassing teaching and research activities. They can be categorized by affiliation into university laboratories, national institutions, and industrial enterprises; by industry into disease control and medical testing laboratories; and by hazard type into chemical, biological, and electromechanical laboratories. Laboratory cleanliness testing is a crucial step in ensuring that the laboratory environment meets specific standards, especially in demanding research, production, and testing environments, where cleanliness directly impacts the reliability and accuracy of experimental results. Cleanliness typically refers to the content of contaminants such as particulate matter, microorganisms, and chemical substances in the air. The level of cleanliness directly affects the precision of experiments. Cleanliness levels include Class I, Class II, and Class III cleanrooms. Current cleanliness testing mainly includes static and dynamic testing. Static testing is usually conducted when the laboratory environment is stable, using specialized cleanliness testing instruments to monitor the content of airborne particles. Dynamic testing is conducted during normal laboratory operation, typically using instruments such as anemometers and particle counters to monitor airflow and particle changes. Dynamic testing reflects the cleanliness of the laboratory in actual use and helps identify potential sources of contamination.

[0003] However, existing laboratory cleanliness testing has corresponding defects, including: (1) Most monitoring methods are based on fixed threshold trigger alarms, which lack dynamic analysis and prediction of environmental parameter change trends; (2) The early warning timeliness is poor, and alarms are usually triggered only when the risk has occurred or is close to the critical value, which cannot achieve early identification in the incubation stage of the risk; (3) Most existing detection technologies only focus on the static environmental state at a single moment, which cannot fully depict the decay process of the clean environment gradually deteriorating over time, and it is difficult to predict the long-term evolution trend of laboratory cleanliness.

[0004] Therefore, how to provide an effective technical solution to address the problems existing in current technologies, such as the lack of dynamic analysis and prediction of environmental parameter change trends, poor early warning timeliness, inability to achieve early identification in the incubation stage of risks, inability to fully depict the gradual deterioration process of clean environment over time, and difficulty in predicting the long-term evolution trend of laboratory cleanliness, has become an urgent technical challenge to be solved. Summary of the Invention

[0005] The purpose of this invention is to provide an environmental risk early warning method and device based on laboratory cleanliness degradation profiles, in order to solve the above-mentioned problems existing in the prior art.

[0006] To achieve the above objectives, the present invention adopts the following technical solution: In a first aspect, the present invention provides an environmental risk early warning method based on laboratory cleanliness degradation profiles, comprising: The laboratory acquires multi-dimensional operational data collected by a multi-modal detection instrument. The multi-dimensional operational data includes filter element pressure difference, filtration efficiency, exhaust fan power, pollutant concentration, running time, and alarm frequency. The multi-dimensional operational data is preprocessed to obtain preprocessed multi-dimensional operational data. A cleanliness capacity decay profile is constructed based on preprocessed multi-dimensional operational data. The cleanliness capacity decay profile is used to characterize the result of the laboratory cleanliness decaying over time. The cleanliness level is obtained based on the cleanliness capacity decay profile, and the cleanliness capacity decay trend is also obtained based on the cleanliness capacity decay profile. The cleanliness level is mapped to the current risk level, and the current risk level is dynamically corrected based on the pre-processed multi-dimensional operational data to obtain the corrected risk level. Maintenance nodes are predicted based on the cleanliness capacity decay trend and the corrected risk level to obtain the corresponding maintenance priority; Environmental risk warnings are issued based on cleanliness levels, revised risk levels, and maintenance priorities.

[0007] In one possible design, multi-dimensional operational data from the laboratory, acquired by a multimodal testing instrument, is obtained, including: The LoRa wireless communication network is used to collect raw data from multiple sensors in the laboratory. The LoRa wireless communication network includes a master node and sub-nodes. The sub-nodes are equipped with at least one of the following: temperature sensor, humidity sensor, flame sensor, carbon monoxide sensor, alcohol sensor, smoke sensor, natural gas sensor, and liquefied gas sensor. The master control node is based on an NB-IoT module and uses the MQTT protocol to upload raw data from multiple sensors to the cloud and display it synchronously on the user terminal.

[0008] In one possible design, the multi-dimensional operational data is preprocessed to obtain preprocessed multi-dimensional operational data, including: Data cleaning is performed on the multi-dimensional operational data to handle outliers and missing values. An outlier detection algorithm is used to identify outliers in the multi-dimensional operational data and remove them to obtain cleaned multi-dimensional operational data. The cleaned multi-dimensional operational data is normalized to unify the dimensions among the cleaned multi-dimensional operational data, resulting in normalized multi-dimensional operational data, which is then used as preprocessed multi-dimensional operational data.

[0009] In one possible design, a profile of cleanroom capability degradation is constructed based on preprocessed multi-dimensional operational data, including: An input feature vector is constructed based on the preprocessed multi-dimensional operational data, and the input feature vector is input into a pre-trained cleanroom assessment model to obtain the comprehensive cleanroom capability assessment value at the current moment. Based on the current comprehensive cleanroom capability assessment value, the comprehensive cleanroom capability assessment values ​​of multiple historical moments are obtained. Based on the current comprehensive cleanroom capability assessment value and the comprehensive cleanroom capability assessment values ​​of multiple historical moments, combined with the filter element pressure difference, exhaust fan power and pollutant concentration at the corresponding moment, a change trend curve is constructed using time series analysis method to obtain the cleanroom capability decay curve over time. The cleanliness degradation curves are compared to obtain cleanliness degradation labels, which include degradation rate, rebound threshold and fluctuation variance. Based on the cleanliness capacity decay curve and cleanliness decay label, a cleanliness capacity decay profile is obtained.

[0010] In one possible design, the trend of cleaning capacity decay is obtained based on the cleaning capacity decay profile, including: Obtain the historical comprehensive assessment value of cleanliness in the cleanliness capacity decay profile, and arrange the historical comprehensive assessment value of cleanliness in chronological order to form time series data; Based on preset prediction compensation, rolling prediction is performed on the time series data to obtain predicted values ​​of cleanliness capacity at multiple future times. The type of decline trend is predicted based on the cleanliness capacity prediction values ​​at multiple future time points. The decline trend type is determined by the slope of the change in the cleanliness capacity prediction value over time, and is used as the cleanliness capacity decline trend.

[0011] In one possible design, the current risk level is dynamically adjusted based on preprocessed multi-dimensional operational data to obtain a revised risk level, including: The pre-processed multi-dimensional running data is input into a pre-trained risk correction model to obtain the risk probability vector at the current moment, which includes probability values ​​corresponding to different risk levels. The risk level corresponding to the highest probability value in the risk probability vector is used as the initial revised risk level; The dynamic weighting coefficient is obtained based on the real-time filter element pressure difference change rate. Based on the dynamic weighting coefficient, the preliminary corrected risk level is weighted and fused with the current risk level to obtain the corrected risk level.

[0012] In one possible design, maintenance nodes are predicted based on the cleanliness capacity decay trend and the revised risk level to obtain the corresponding maintenance priorities, including: Based on the cleaning capacity decay trend, the decay rate is obtained. Then, based on the decay rate and the current comprehensive evaluation value of the cleaning capacity, a pre-trained cleaning assessment model is used to predict the remaining time required for the cleaning capacity to decline to the preset failure threshold, and the remaining time is used as the maintenance window period. Based on the revised risk level and the preset risk level weight table, the corresponding risk weight coefficient is obtained; Based on the maintenance window period and risk weight coefficient, a maintenance urgency score is obtained, and the corresponding maintenance priority is determined according to the maintenance urgency score.

[0013] In one possible design, environmental risk warnings are generated based on cleanliness levels, revised risk levels, and maintenance priorities, including: Based on the revised risk level, a warning signal code is obtained, which includes a red warning code, an orange warning code, a yellow warning code, and a blue warning code, corresponding to the four revised risk levels from high to low, respectively. The cleanliness level and maintenance priority are integrated to obtain a maintenance recommendation text; The warning signal encoding and maintenance suggestion text are pushed to each sub-node via the LoRa wireless communication network, and then pushed to the user terminal for display via the cloud.

[0014] Secondly, the present invention provides an environmental risk early warning device based on laboratory cleanliness degradation profiles, comprising: The data acquisition unit is used to acquire multi-dimensional operational data of the laboratory collected by the multi-modal detection instrument. The multi-dimensional operational data includes filter element pressure difference, filtration efficiency, exhaust fan power, pollutant concentration, running time and alarm frequency. The multi-dimensional operational data is preprocessed to obtain preprocessed multi-dimensional operational data. The profile construction unit is used to construct a cleanliness capacity decay profile based on preprocessed multi-dimensional operational data. The cleanliness capacity decay profile is used to characterize the result of the laboratory cleanliness decaying over time. The cleanliness level is obtained based on the cleanliness capacity decay profile, and the cleanliness capacity decay trend is also obtained based on the cleanliness capacity decay profile. The level correction unit is used to map the cleanliness level to the current risk level, and dynamically correct the current risk level based on the pre-processed multi-dimensional operational data to obtain the corrected risk level. The node prediction module is used to predict maintenance nodes based on the cleanliness capacity decay trend and the corrected risk level, so as to obtain the corresponding maintenance priority. The risk warning module is used to provide environmental risk warnings based on cleanliness level, revised risk level, and maintenance priority.

[0015] Thirdly, the present invention provides a computer device comprising a memory, a processor, and a transceiver connected in sequence and communication, wherein the memory is used to store a computer program, the transceiver is used to send and receive messages, and the processor is used to read the computer program and execute the environmental risk early warning method based on laboratory cleanliness degradation profile as described in the first aspect above.

[0016] Fourthly, the present invention provides a computer-readable storage medium storing instructions that, when executed on a computer, perform the environmental risk early warning method based on laboratory cleanliness degradation profiles as described in the first aspect above.

[0017] Fifthly, the present invention provides a computer program product containing instructions that, when the instructions are executed on a computer, cause the computer to perform the environmental risk early warning method based on laboratory cleanliness degradation profiles as described in the first aspect above.

[0018] The beneficial effects of this invention are as follows: This invention discloses an environmental risk early warning method and device based on a laboratory cleanliness degradation profile. First, multi-dimensional operational data of the laboratory is acquired by a multi-modal detection instrument, and the multi-dimensional operational data is preprocessed to obtain preprocessed multi-dimensional operational data. A cleanliness degradation profile is constructed based on the preprocessed multi-dimensional operational data. The cleanliness degradation profile is used to characterize the result of the laboratory cleanliness degradation over time. A cleanliness level is obtained based on the cleanliness degradation profile, and a cleanliness degradation trend is obtained based on the cleanliness degradation profile. The cleanliness level is mapped to the current risk level, and the current risk level is dynamically corrected based on the preprocessed multi-dimensional operational data to obtain a corrected risk level. Maintenance nodes are predicted based on the cleanliness degradation trend and the corrected risk level to obtain the corresponding maintenance priority. Environmental risk early warning is performed based on the cleanliness level, the corrected risk level, and the maintenance priority. This invention constructs a cleanliness capability decay profile based on multi-dimensional operational data, thereby intuitively displaying the changing trend of laboratory cleanliness capability, outputting the current risk level, and realizing remote monitoring and real-time early warning of laboratory risks. By constructing a cleanliness capability decay profile based on multi-dimensional operational data, it enables dynamic analysis and prediction of environmental parameter change trends, and can identify risks in the early stages of risk development. It fully depicts the gradual deterioration process of the clean environment over time, predicts the long-term evolution trend of laboratory cleanliness, and provides timely environmental risk warnings, making it easy to apply and promote. Attached Figure Description

[0019] Figure 1 A flowchart illustrating an environmental risk early warning method based on laboratory cleanliness degradation profiles provided in this embodiment of the invention; Figure 2This is a structural diagram of an environmental risk early warning device based on laboratory cleanliness degradation profiles provided in an embodiment of the present invention. Detailed Implementation

[0020] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the present invention will be briefly introduced below in conjunction with the accompanying drawings and descriptions of the embodiments or the prior art. Obviously, the following description of the structure of the accompanying drawings is only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort. It should be noted that the description of these embodiments is for the purpose of helping to understand the present invention, but does not constitute a limitation of the present invention.

[0021] It should be understood that although the terms first, second, etc., may be used herein to describe various units, these units should not be limited by these terms. These terms are only used to distinguish one unit from another. For example, a first unit may be referred to as a second unit, and similarly, a second unit may be referred to as a first unit, without departing from the scope of the exemplary embodiments of the invention.

[0022] It should be understood that the term "and / or" that may appear in this document is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can mean: A exists alone, B exists alone, and A and B exist simultaneously. The term " / and" that may appear in this document describes another relationship between related objects, indicating that two relationships can exist. For example, A / and B can mean: A exists alone, and A and B exist alone. In addition, the character " / " that may appear in this document generally indicates that the related objects before and after it are in an "or" relationship.

[0023] Example: like Figure 1 As shown, the first aspect of this embodiment provides an environmental risk early warning method based on laboratory cleanliness degradation profiles. This method can be executed, but is not limited to, by a computer device or virtual machine with certain computing resources, such as a personal computer or smartphone, or by a virtual machine. The environmental risk early warning method includes, but is not limited to, the following steps: S1. Acquire multi-dimensional operational data of the laboratory collected by the multi-modal detection instrument. The multi-dimensional operational data includes filter element pressure difference, filtration efficiency, exhaust fan power, pollutant concentration, running time and alarm frequency. Preprocess the multi-dimensional operational data to obtain preprocessed multi-dimensional operational data. Specifically, in step S1, the multi-dimensional operational data of the laboratory collected by the multi-modal testing instrument is acquired, including: S11. Use a LoRa wireless communication network to collect raw data from multiple sensors in the laboratory. The LoRa wireless communication network includes a master node and sub-nodes. The sub-nodes are equipped with at least one of the following: temperature sensor, humidity sensor, flame sensor, carbon monoxide sensor, alcohol sensor, smoke sensor, natural gas sensor, and liquefied gas sensor. S12. The master control node is based on the NB-IoT module and uses the MQTT protocol to upload raw data from multiple sensors to the cloud and display it synchronously on the user terminal.

[0024] It should be noted that the LoRa (Long Range Radio) wireless communication network is characterized by its longer propagation distance compared to other wireless methods under the same power consumption conditions, and it allows devices to operate under low signal strength and has high anti-interference performance. In step S12, the raw data from multiple sensors is also uploaded to the cloud based on the WiFi module. The user terminal includes the Alibaba Cloud platform and a web page. First, the NB-IoT module is initialized, then a server IP and port connection is established, and the connection is established by sending instructions through the main control microcontroller. The raw data from multiple sensors is encapsulated, and the main control node sends the encapsulated data through the NB-IoT module. The child node communicates with the main control node through the LoRa wireless communication network, sets the LoRa wireless communication network at the main control node to broadcast listening mode, and sends the character 'A' to the child node when the timer interrupt occurs at the main control node. When the LoRa wireless communication network of the child node receives the data and judges the data, it sends the raw data from multiple sensors to the LoRa wireless communication network of the main control node.

[0025] Furthermore, the multi-dimensional operational data is preprocessed to obtain preprocessed multi-dimensional operational data, including: S13. Perform data cleaning on the multi-dimensional operational data to handle outliers and missing values ​​in the multi-dimensional operational data. Use an outlier detection algorithm to identify outliers in the multi-dimensional operational data and remove outliers to obtain cleaned multi-dimensional operational data. S14. Normalize the cleaned multi-dimensional operational data to unify the dimensions among the cleaned multi-dimensional operational data, and obtain normalized multi-dimensional operational data. Use the normalized multi-dimensional operational data as the preprocessed multi-dimensional operational data.

[0026] It should be noted that the Empirical Cumulative Distribution-based Outlier Detection (ECOD) algorithm estimates the variable distribution in a non-parametric manner and identifies outliers using tail probabilities. Specifically, it constructs an empirical cumulative distribution function for each dimension to estimate the position of the observed value in the distribution. It can automatically determine the skewness of the data distribution and derive a comprehensive outlier score for the observed value, thereby determining whether the point deviates from the main distribution. In this embodiment, the specific handling of missing values ​​involves replacing the missing value with the mean of the data surrounding the missing value. This cleans and standardizes the multi-dimensional running data, eliminating the differences caused by variations in data from different sensors, in order to facilitate subsequent environmental risk prediction.

[0027] S2. Construct a cleanliness capacity decay profile based on the preprocessed multi-dimensional operational data. The cleanliness capacity decay profile is used to characterize the result of the laboratory cleanliness decaying over time. The cleanliness level is obtained based on the cleanliness capacity decay profile, and the cleanliness capacity decay trend is also obtained based on the cleanliness capacity decay profile. In step S2, a cleanroom capability degradation profile is constructed based on the preprocessed multi-dimensional operational data, including: S21. Construct an input feature vector based on the preprocessed multi-dimensional operational data, and input the input feature vector into the pre-trained cleanroom assessment model to obtain the comprehensive cleanroom capability assessment value at the current moment; S22. Based on the current comprehensive cleanroom capability assessment value, obtain the corresponding comprehensive cleanroom capability assessment values ​​for multiple historical moments. Based on the current comprehensive cleanroom capability assessment value and the comprehensive cleanroom capability assessment values ​​for multiple historical moments, and combined with the filter element pressure difference, exhaust fan power and pollutant concentration at the corresponding moment, construct a trend curve using time series analysis to obtain the cleanroom capability decay curve over time. S23. Compare the cleanliness capacity decay curves to obtain a cleanliness decay label, which includes the decay rate, rebound threshold and fluctuation variance. S24. Based on the cleanliness capacity decay curve and cleanliness decay label, a cleanliness capacity decay profile is obtained.

[0028] It should be noted that the construction process of the pre-trained cleanroom assessment model includes: first, preparing training and testing sets; then, selecting a kernel function, such as a radial basis function kernel; subsequently, setting the kernel function, regularization parameters, and parameter optimization algorithm; calculating the kernel values ​​between all pairs of samples in the training set according to the selected kernel function to obtain a symmetric kernel matrix; obtaining the output weights based on the symmetric kernel matrix; calculating the kernel vectors of the test samples and all training samples in the training set; obtaining the predicted values ​​based on the kernel vectors and output weights; normalizing the predicted values ​​to convert them back to the original data scale; and then using the test set to evaluate the converted predicted values. If the predicted values ​​are within the preset error range, the trained cleanroom assessment model is obtained.

[0029] Furthermore, based on the cleanroom capacity decay profile, the cleanroom capacity decay trend is obtained, including: S25. Obtain the historical comprehensive assessment value of cleanliness in the cleanliness capacity decay profile, and arrange the historical comprehensive assessment value of cleanliness in chronological order to form time series data; S26. Based on preset prediction compensation, perform rolling prediction on the time series data to obtain predicted values ​​of cleanliness capacity at multiple future times; S27. Based on the predicted cleanliness capacity values ​​at multiple future time points, predict the type of decline trend, and determine the type of decline trend according to the slope of the change of the predicted cleanliness capacity values ​​over time, so as to serve as the cleanliness capacity decline trend.

[0030] In practice, when the absolute value of the slope of the cleanliness capacity decay curve increases, it is determined to be an accelerated decay trend; when the absolute value of the slope is constant, it is determined to be a uniform decay trend; and when the absolute value of the slope decreases, it is determined to be a decelerated decay trend.

[0031] S3. Map the cleanliness level to the current risk level, and dynamically correct the current risk level based on the pre-processed multi-dimensional operational data to obtain the corrected risk level; Specifically, in step S3, the current risk level is dynamically adjusted based on the preprocessed multi-dimensional operational data to obtain the adjusted risk level, including: S31. Input the preprocessed multi-dimensional running data into the pre-trained risk correction model to obtain the risk probability vector at the current moment, wherein the risk probability vector includes probability values ​​corresponding to different risk levels; S32. The risk level corresponding to the highest probability value in the risk probability vector is taken as the initial revised risk level; S33. Based on the real-time filter element pressure difference change rate, obtain the dynamic weighting coefficient. Based on the dynamic weighting coefficient, weight and fuse the preliminary corrected risk level with the current risk level to obtain the corrected risk level.

[0032] It should be noted that the risk correction model is a kernel extreme learning machine classifier with parameters optimized by the improved Heron Eagle Optimization Algorithm. The improved Heron Eagle Optimization Algorithm is used to optimize the regularization parameters and kernel parameters of the kernel extreme learning machine classifier. The initialization of the improved Heron Eagle Optimization Algorithm adopts the Tent chaotic mapping strategy to generate an initial population that uniformly covers the search space. The position update of the improved Heron Eagle Optimization Algorithm introduces the Cauchy mutation operator to expand the search range. The escape phase of the improved Heron Eagle Optimization Algorithm adopts the cosine similarity strategy to measure the consistency of individual orientations in order to escape local optima.

[0033] In practice, the construction process of the pre-trained risk correction model includes: acquiring pre-processed historical multi-dimensional operational data; encoding the pre-processed historical multi-dimensional operational data; dividing the encoded data into training and testing sets; constructing an initial kernel extreme learning machine classifier based on the training set; subsequently, initializing the population and setting initial parameters, including population size, maximum number of iterations, optimization dimension, search boundary, and kernel function parameters and regularization parameters in the initial kernel extreme learning machine classifier, where the kernel function parameters and regularization parameters of each individual correspond to one kernel extreme learning machine classifier training session; constructing the risk correction model, mapping the accuracy of the risk correction model to the fitness value of the heron, and calculating the fitness value of each heron. Based on the improved heron-eagle optimization algorithm, the system enters the exploration node and updates the individual heron-eagle's position. Then, it enters the development phase. When the random number is greater than a preset number, an environment-based cooperative camouflage strategy is executed to update the optimal position; when the random number is less than the preset number, a cosine similarity-based strategy is executed to update the optimal position. It then checks if the maximum number of iterations has been reached. If so, the optimal value is output to obtain the kernel parameters and regularization parameters of the optimized kernel extreme learning machine classifier. Otherwise, a feedback adjustment mechanism is executed, and a risk correction model is reconstructed, mapping the accuracy of the risk correction model to the heron-eagle's fitness value. Finally, a risk correction model is constructed based on the kernel parameters and regularization parameters of the optimized kernel extreme learning machine classifier.

[0034] S4. Based on the cleanliness capacity decay trend and the corrected risk level, predict maintenance nodes to obtain the corresponding maintenance priorities; Specifically, in step S4, maintenance nodes are predicted based on the cleanliness capacity decay trend and the corrected risk level to obtain the corresponding maintenance priorities, including: S41. Based on the cleaning capacity decay trend, obtain the decay rate. Then, based on the decay rate and the current comprehensive cleaning capacity assessment value, use a pre-trained cleaning assessment model to predict the remaining time required for the cleaning capacity to decline to the preset failure threshold, and use the remaining time as the maintenance window period. S42. Based on the revised risk level and the preset risk level weight table, obtain the corresponding risk weight coefficient; S43. Based on the maintenance window period and risk weight coefficient, obtain the maintenance urgency score, and then determine the corresponding maintenance priority based on the maintenance urgency score.

[0035] It should be noted that the decay rate is based on the slope of the change in the cleanliness capability prediction value over time. In specific implementation, the current comprehensive cleanliness capability assessment value is compared with a preset set of cleanliness level judgment thresholds. The preset set of cleanliness level judgment thresholds includes a first threshold, a second threshold, and a third threshold. The first threshold is greater than the second threshold, and the second threshold is greater than the third threshold. When the current comprehensive cleanliness capability assessment value is greater than the first threshold, it is judged as a normal level. When the current comprehensive cleanliness capability assessment value is less than the first threshold but greater than the second threshold, it is judged as a concern level. When the current comprehensive cleanliness capability assessment value is less than the second threshold but greater than the third threshold, it is judged as a warning level. When the current comprehensive cleanliness capability assessment value is less than the third threshold, it is judged as a dangerous level.

[0036] In practice, the maintenance priority is determined based on the maintenance urgency score by comparing it with a preset set of priority intervals. These intervals include a first interval, a second interval, a third interval, and a fourth interval, with the order decreasing from first to fourth. If the maintenance urgency score is in the first interval, the maintenance priority is determined as urgent maintenance; if it is in the second interval, it is determined as high-priority maintenance; if it is in the third interval, it is determined as medium-priority maintenance; and if it is in the fourth interval, it is determined as low-priority maintenance.

[0037] S5. Conduct environmental risk warnings based on cleanliness level, revised risk level, and maintenance priority.

[0038] In step S5, environmental risk warnings are issued based on the cleanliness level, the revised risk level, and the maintenance priority, including: S51. Based on the corrected risk level, a warning signal code is obtained, which includes a red warning code, an orange warning code, a yellow warning code, and a blue warning code, corresponding to the four levels of the corrected risk level from high to low, respectively; S52. Integrate cleanliness levels and maintenance priorities to obtain maintenance recommendation text; S53. The warning signal encoding and maintenance suggestion text are pushed to each sub-node through the LoRa wireless communication network, and then pushed to the user terminal for display via the cloud.

[0039] It should be noted that when the warning signal is coded as a red warning code or an orange warning code, an emergency alarm signal is generated, and the main control node is used to control the relay to shut down the electrical equipment.

[0040] Furthermore, the cleanroom capacity decay profile is constructed and updated cyclically according to a preset time period. The preset time period is adaptively adjusted according to the laboratory's operating status. In this embodiment, the time period is shortened when the real-time pollutant concentration exceeds a preset concentration threshold; the time period is shortened when the exhaust fan power fluctuation exceeds a preset fluctuation threshold; and the time period is shortened when the filter element pressure difference change rate exceeds a preset change rate threshold. The preset concentration threshold, preset fluctuation threshold, and preset change rate threshold can all be set according to actual conditions, and this embodiment does not limit them.

[0041] like Figure 2 As shown, the second aspect of this embodiment provides an environmental risk early warning device based on laboratory cleanliness degradation profiles, comprising: The data acquisition unit is used to acquire multi-dimensional operational data of the laboratory collected by the multi-modal detection instrument. The multi-dimensional operational data includes filter element pressure difference, filtration efficiency, exhaust fan power, pollutant concentration, running time and alarm frequency. The multi-dimensional operational data is preprocessed to obtain preprocessed multi-dimensional operational data. The profile construction unit is used to construct a cleanliness capacity decay profile based on preprocessed multi-dimensional operational data. The cleanliness capacity decay profile is used to characterize the result of the laboratory cleanliness decaying over time. The cleanliness level is obtained based on the cleanliness capacity decay profile, and the cleanliness capacity decay trend is also obtained based on the cleanliness capacity decay profile. The level correction unit is used to map the cleanliness level to the current risk level, and dynamically correct the current risk level based on the pre-processed multi-dimensional operational data to obtain the corrected risk level. The node prediction module is used to predict maintenance nodes based on the cleanliness capacity decay trend and the corrected risk level, so as to obtain the corresponding maintenance priority. The risk warning module is used to provide environmental risk warnings based on cleanliness level, revised risk level, and maintenance priority.

[0042] The working process, working details and technical effects of the environmental risk early warning device based on laboratory cleanliness degradation profile provided in the second aspect of this embodiment can be found in the environmental risk early warning method based on laboratory cleanliness degradation profile described in the first aspect, and will not be repeated here.

[0043] This embodiment provides a computer device including a memory, a processor, and a transceiver connected in sequence. The memory stores a computer program, the transceiver sends and receives messages, and the processor reads the computer program to execute the environmental risk warning method based on laboratory cleanroom degradation profiles as described in the first aspect. Specifically, the memory may include, but is not limited to, random access memory (RAM), read-only memory (ROM), flash memory, first-in-first-out (FIFO) memory, and / or first-in-last-out (FILO) memory, etc.; the processor may include, but is not limited to, an STM32F105 series microprocessor. Furthermore, the computer device may also include, but is not limited to, a power module, a display screen, and other necessary components.

[0044] The working process, working details and technical effects of the aforementioned computer equipment provided in the third aspect of this embodiment can be found in the environmental risk early warning method based on laboratory cleanliness degradation profiles described in the first aspect, and will not be repeated here.

[0045] The fourth aspect of this embodiment provides a computer-readable storage medium, wherein the computer-readable storage medium stores instructions, and when the instructions are executed on a computer, the environmental risk early warning method based on laboratory cleanliness degradation profiles as described in the first aspect is performed. The computer-readable storage medium refers to a data storage medium, which may include, but is not limited to, floppy disks, optical disks, hard disks, flash memory, USB flash drives, and / or Memory Sticks. The computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices.

[0046] The working process, working details and technical effects of the aforementioned computer-readable storage medium provided in the fourth aspect of this embodiment can be found in the environmental risk early warning method based on laboratory cleanliness degradation profile as described in the first aspect, and will not be repeated here.

[0047] The fifth aspect of this embodiment provides a computer program product, including a computer program or instructions, which, when executed by a computer, are used to implement the environmental risk early warning method based on laboratory cleanliness degradation profiles as described in the first aspect.

[0048] The working process, working details and technical effects of the aforementioned computer program product provided in this embodiment can be found in the environmental risk early warning method based on laboratory cleanliness degradation profile as described in the first aspect, and will not be repeated here.

[0049] Finally, it should be noted that the above description is merely a preferred embodiment of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. An environmental risk early warning method based on laboratory cleanliness degradation profiles, characterized in that, include: The laboratory acquires multi-dimensional operational data collected by a multi-modal detection instrument. The multi-dimensional operational data includes filter element pressure difference, filtration efficiency, exhaust fan power, pollutant concentration, running time, and alarm frequency. The multi-dimensional operational data is preprocessed to obtain preprocessed multi-dimensional operational data. A cleanliness capacity decay profile is constructed based on preprocessed multi-dimensional operational data. The cleanliness capacity decay profile is used to characterize the result of the laboratory cleanliness decaying over time. The cleanliness level is obtained based on the cleanliness capacity decay profile, and the cleanliness capacity decay trend is also obtained based on the cleanliness capacity decay profile. The cleanliness level is mapped to the current risk level, and the current risk level is dynamically corrected based on the pre-processed multi-dimensional operational data to obtain the corrected risk level. Maintenance nodes are predicted based on the cleanliness capacity decay trend and the corrected risk level to obtain the corresponding maintenance priority; Environmental risk warnings are issued based on cleanliness levels, revised risk levels, and maintenance priorities.

2. The environmental risk early warning method based on laboratory cleanliness degradation profiles according to claim 1, characterized in that, Acquire multi-dimensional operational data from the laboratory collected by multimodal testing instruments, including: The LoRa wireless communication network is used to collect raw data from multiple sensors in the laboratory. The LoRa wireless communication network includes a master node and sub-nodes. The sub-nodes are equipped with at least one of the following: temperature sensor, humidity sensor, flame sensor, carbon monoxide sensor, alcohol sensor, smoke sensor, natural gas sensor, and liquefied gas sensor. The master control node is based on an NB-IoT module and uses the MQTT protocol to upload raw data from multiple sensors to the cloud and display it synchronously on the user terminal.

3. The environmental risk early warning method based on laboratory cleanliness degradation profiles according to claim 1, characterized in that, The multi-dimensional operational data is preprocessed to obtain preprocessed multi-dimensional operational data, including: Data cleaning is performed on the multi-dimensional operational data to handle outliers and missing values. An outlier detection algorithm is used to identify outliers in the multi-dimensional operational data and remove them to obtain cleaned multi-dimensional operational data. The cleaned multi-dimensional operational data is normalized to unify the dimensions among the cleaned multi-dimensional operational data, resulting in normalized multi-dimensional operational data, which is then used as preprocessed multi-dimensional operational data.

4. The environmental risk early warning method based on laboratory cleanliness degradation profiles according to claim 1, characterized in that, A profile of cleanroom capability degradation is constructed based on preprocessed multi-dimensional operational data, including: An input feature vector is constructed based on the preprocessed multi-dimensional operational data, and the input feature vector is input into a pre-trained cleanroom assessment model to obtain the comprehensive cleanroom capability assessment value at the current moment. Based on the current comprehensive cleanroom capability assessment value, the comprehensive cleanroom capability assessment values ​​of multiple historical moments are obtained. Based on the current comprehensive cleanroom capability assessment value and the comprehensive cleanroom capability assessment values ​​of multiple historical moments, combined with the filter element pressure difference, exhaust fan power and pollutant concentration at the corresponding moment, a change trend curve is constructed using time series analysis method to obtain the cleanroom capability decay curve over time. The cleanliness degradation curves are compared to obtain cleanliness degradation labels, which include degradation rate, rebound threshold and fluctuation variance. Based on the cleanliness capacity decay curve and cleanliness decay label, a cleanliness capacity decay profile is obtained.

5. The environmental risk early warning method based on laboratory cleanliness degradation profiles according to claim 1, characterized in that, Based on the cleanroom capacity decay profile, the cleanroom capacity decay trend is obtained, including: Obtain the historical comprehensive assessment value of cleanliness in the cleanliness capacity decay profile, and arrange the historical comprehensive assessment value of cleanliness in chronological order to form time series data; Based on preset prediction compensation, rolling prediction is performed on the time series data to obtain predicted values ​​of cleanliness capacity at multiple future times. The type of decline trend is predicted based on the cleanliness capacity prediction values ​​at multiple future time points. The decline trend type is determined by the slope of the change in the cleanliness capacity prediction value over time, and is used as the cleanliness capacity decline trend.

6. The environmental risk early warning method based on laboratory cleanliness degradation profiles according to claim 1, characterized in that, The current risk level is dynamically adjusted based on the preprocessed multi-dimensional operational data to obtain the adjusted risk level, including: The pre-processed multi-dimensional running data is input into a pre-trained risk correction model to obtain the risk probability vector at the current moment, which includes probability values ​​corresponding to different risk levels. The risk level corresponding to the highest probability value in the risk probability vector is used as the initial revised risk level; The dynamic weighting coefficient is obtained based on the real-time filter element pressure difference change rate. Based on the dynamic weighting coefficient, the preliminary corrected risk level is weighted and fused with the current risk level to obtain the corrected risk level.

7. The environmental risk early warning method based on laboratory cleanliness degradation profiles according to claim 1, characterized in that, Maintenance nodes are predicted based on the cleanliness capacity decay trend and the corrected risk level to obtain the corresponding maintenance priorities, including: Based on the cleaning capacity decay trend, the decay rate is obtained. Then, based on the decay rate and the current comprehensive evaluation value of the cleaning capacity, a pre-trained cleaning assessment model is used to predict the remaining time required for the cleaning capacity to decline to the preset failure threshold, and the remaining time is used as the maintenance window period. Based on the revised risk level and the preset risk level weight table, the corresponding risk weight coefficient is obtained; Based on the maintenance window period and risk weight coefficient, a maintenance urgency score is obtained, and the corresponding maintenance priority is determined according to the maintenance urgency score.

8. The environmental risk early warning method based on laboratory cleanliness degradation profiles according to claim 2, characterized in that, Environmental risk warnings are issued based on cleanliness levels, revised risk levels, and maintenance priorities, including: Based on the revised risk level, a warning signal code is obtained, which includes a red warning code, an orange warning code, a yellow warning code, and a blue warning code, corresponding to the four revised risk levels from high to low, respectively. The cleanliness level and maintenance priority are integrated to obtain a maintenance recommendation text; The warning signal encoding and maintenance suggestion text are pushed to each sub-node via the LoRa wireless communication network, and then pushed to the user terminal for display via the cloud.

9. An environmental risk early warning device based on laboratory cleanliness degradation profiles, used to implement the method according to any one of claims 1-8, characterized in that, include: The data acquisition unit is used to acquire multi-dimensional operational data of the laboratory collected by the multi-modal detection instrument. The multi-dimensional operational data includes filter element pressure difference, filtration efficiency, exhaust fan power, pollutant concentration, running time and alarm frequency. The multi-dimensional operational data is preprocessed to obtain preprocessed multi-dimensional operational data. The profile construction unit is used to construct a cleanliness capacity decay profile based on preprocessed multi-dimensional operational data. The cleanliness capacity decay profile is used to characterize the result of the laboratory cleanliness decaying over time. The cleanliness level is obtained based on the cleanliness capacity decay profile, and the cleanliness capacity decay trend is also obtained based on the cleanliness capacity decay profile. The level correction unit is used to map the cleanliness level to the current risk level, and dynamically correct the current risk level based on the pre-processed multi-dimensional operational data to obtain the corrected risk level. The node prediction module is used to predict maintenance nodes based on the cleanliness capacity decay trend and the corrected risk level, so as to obtain the corresponding maintenance priority. The risk warning module is used to provide environmental risk warnings based on cleanliness level, revised risk level, and maintenance priority.

10. A computer program product, comprising a computer program or instructions, characterized in that, When the computer program or the instructions are executed by the computer, they implement the environmental risk early warning method based on laboratory cleanliness degradation profiles as described in any one of claims 1 to 7.

Images