Environment early warning method and system based on multi-source sensor data

Abstract

The present application relates to the technical field of environmental monitoring and alarm, and particularly relates to an environmental early warning method and system based on multi-source sensor data, comprising collecting real-time environmental parameter data sets in the laboratory before the start of the target experimental project reservation period, and collecting equipment operation data sets of the laboratory in the past set time period; identifying environmental constraint features of the target experimental project to obtain multiple experimental environmental constraint features; based on the equipment operation data sets, calculating the equipment adjustment change rate of each experimental environmental constraint feature per unit time; based on the real-time environmental parameter data sets, calculating the initial abnormality degree of each experimental environmental constraint feature; according to the type of the target experimental project, determining at least one experimental self-influence feature; for each experimental environmental constraint feature, inputting the corresponding equipment adjustment change rate, initial abnormality degree and determined experimental self-influence feature into a pre-trained laboratory environmental early warning analysis model.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of environmental monitoring and alarm, and in particular to an environmental early warning method and system based on multi-source sensor data. BACKGROUND

[0002] In the experimental research in the fields of biology, chemistry, precision instruments, etc., the stability and compliance of the laboratory environment parameters are crucial to the accuracy and repeatability of the experimental results. Such laboratories usually need to arrange the use period through a reservation mechanism, and most experiments have strict constraints on the environment, such as specific temperature, humidity, cleanliness, gas concentration, etc. The parameters need to be kept within the limited range throughout the experiment.

[0003] The existing laboratory environment management mostly adopts a real-time monitoring mode, that is, the environmental data is collected in real time through sensors, and an alarm is triggered when the monitoring value exceeds the threshold required by the experiment. However, the existing method often alarms during the experiment, at which time the experiment has been started. If the experiment is terminated due to the non-compliance of the environment, it will cause waste of samples, reagents, and time. Even if emergency environmental intervention measures are taken, the fluctuations in the parameter adjustment process may still interfere with the experimental process, leading to distorted results and seriously affecting the experimental results.

[0004] Therefore, there is an urgent need for an early warning method that can evaluate the environmental adaptability before the start of the experimental reservation period and predict potential risks in combination with the characteristics of the experiment itself, so as to avoid experimental interruption or result deviation caused by environmental problems. SUMMARY

[0005] The present application provides an environmental early warning method based on multi-source sensor data, which can effectively solve the problems in the background art and improve the accuracy and repeatability of experimental results.

[0006] To achieve the above purpose, in a first aspect, the present application provides an environmental early warning method based on multi-source sensor data, comprising:

[0007] Before the start of the reservation period of the target experimental project, collecting a real-time environmental parameter data set in the laboratory and collecting a device operation data set of the laboratory in a set time period in the past;

[0008] Performing environmental constraint feature recognition on the target experimental project to obtain multiple experimental environmental constraint features;

[0009] Based on the device operation data set, calculating the device adjustment change rate of each experimental environmental constraint feature per unit time;

[0010] Based on the real-time environmental parameter data set, calculating the initial abnormality degree of each experimental environmental constraint feature;

[0011] determine at least one experiment self-influence characteristic according to a target experiment project type;

[0012] For each experiment environment constraint characteristic, input the corresponding device adjustment rate, initial abnormality degree, and determined experiment self-influence characteristic into a pre-trained laboratory environment early warning analysis model to obtain an experiment environment adaptation risk score of the target experiment project, and trigger a hierarchical early warning based on the score.

[0013] With reference to the first aspect, in a possible design, the device operation data set includes a historical environment parameter data set and an operation state of the environment adjustment device.

[0014] With reference to the first aspect, in a possible design, the environment parameter data set includes an environment temperature, an environment humidity, an environment air pressure, an illumination intensity, a suspended particulate matter concentration, a microorganism concentration, an oxygen concentration, and a harmful gas concentration, which are collected by a multi-source sensor array deployed in the laboratory.

[0015] With reference to the first aspect, in a possible design, the experiment environment constraint characteristic includes at least one of a temperature constraint, a humidity constraint, an air pressure constraint, an illumination intensity constraint, a suspended particulate matter concentration constraint, a microorganism concentration constraint, an oxygen concentration constraint, and a harmful gas concentration constraint.

[0016] With reference to the first aspect, in a possible design, the basis for identifying the environment constraint characteristic of the target experiment project includes an experiment type and nature, an experiment operation process, and an experiment sample characteristic.

[0017] With reference to the first aspect, in a possible design, a formula for calculating the device adjustment rate of each experiment environment constraint characteristic within a unit time is as follows:

[0018] wherein R P represents the device adjustment rate, n represents a total number of effective device adjustment events for the environment constraint characteristic P within a set time period, P ie represents a target value of the environment parameter P when the device adjusts the environment parameter P to be within the experiment constraint range for the first time in the i th adjustment event, P is represents an initial value of the environment parameter P when the device starts active adjustment in the i th adjustment event, t ie represents a time when the environment parameter P reaches the experiment constraint range for the first time in the i th adjustment event, t is represents a time when the device starts active adjustment in the i th adjustment event.

[0019] With reference to the first aspect, in a possible design, the experiment self-influence characteristic represents an experiment environment constraint characteristic generated by the experiment itself during the experiment.

[0020] In a second aspect, the present application also provides an environment early warning system based on multi-source sensor data, comprising:

[0021] A data collection module collects real-time environment parameter data sets in the laboratory before the start of the target experimental project reservation period, and collects equipment operation data sets of the laboratory in the past set time period;

[0022] An experimental environment constraint feature identification module identifies the environmental constraint features of the target experimental project to obtain multiple experimental environment constraint features;

[0023] An equipment adjustment change rate calculation module calculates the equipment adjustment change rate of each experimental environment constraint feature per unit time based on the equipment operation data set;

[0024] An initial abnormality degree calculation module calculates the initial abnormality degree of each experimental environment constraint feature based on the real-time environment parameter data set;

[0025] An experimental self-influence feature determination module determines at least one experimental self-influence feature according to the type of the target experimental project;

[0026] An experimental environment adaptation risk score and early warning module inputs the corresponding equipment adjustment change rate, initial abnormality degree, and determined experimental self-influence feature of each experimental environment constraint feature into a pre-trained laboratory environment early warning analysis model to obtain the experimental environment adaptation risk score of the target experimental project, and triggers a graded early warning based on the score.

[0027] In combination with the second aspect, in a possible design, the data collection module further comprises a sensor anomaly detection unit for detecting anomalies in the sensors collecting real-time environment parameter data sets.

[0028] In combination with the second aspect, in a possible design, the experimental self-influence feature determination module further comprises a self-influence feature dynamic updating unit for dynamically updating the determined experimental self-influence feature according to real-time progress data of the experiment during the target experimental project.

[0029] The technical solutions of the present application can achieve the following technical effects:

[0030] Performing environment adaptation evaluation before the start of the target experimental project reservation period can discover potential environmental problems before the experiment starts, take adjustment measures in advance, avoid experiment interruption due to substandard environment, and ensure the continuity of the experiment; Experiment interruption will cause waste of samples and reagents and invalid consumption of time; Early evaluation of environment adaptation can prevent such situations from occurring, reduce experimental costs, and improve resource utilization efficiency;

[0031] Through the multi-source sensor array deployed in the laboratory, a rich set of real-time environmental parameter data sets including environmental temperature, humidity, air pressure, light intensity, suspended particulate matter concentration, microorganism concentration, oxygen concentration and harmful gas concentration are collected, and device operation data sets containing historical environmental parameter data sets and environmental regulation device operation states in a set time period are collected at the same time; the multi-source data complement and verify each other, which can more comprehensively and accurately reflect the laboratory environment, and provide a solid foundation for subsequent analysis;

[0032] The environmental regulation device operation state information in the device operation data set is helpful to understand the working capacity and efficiency of the device; combined with the historical environmental parameter data, the adjustment effect of the device on the environmental parameters under different working conditions can be analyzed, so that the guarantee capacity of the current environmental regulation device can be more accurately evaluated, and the environmental adaptability evaluation is more in line with the actual situation;

[0033] The environmental constraint characteristics of the target experiment project are identified, and the specific requirements of the experiment on temperature, humidity and other environmental parameters are determined; the constraint degree and range of environmental parameters are different for different experiments, and accurate identification of these characteristics can make the environmental evaluation more targeted and improve the accuracy of the warning; according to the type of the target experiment project, the experiment self-influence characteristics, i.e. the influence of the experiment itself on the environmental constraint characteristics during the experiment process, are determined, which can more comprehensively evaluate the change trend of the environmental parameters during the experiment process, predict potential risks in advance, and make the warning more scientific and reasonable;

[0034] The device adjustment rate, initial abnormality degree and experiment self-influence characteristics are input into the pre-trained laboratory environment warning analysis model to obtain the experiment environment adaptation risk score of the target experiment project; the multiple influence factors are comprehensively considered to reflect the environmental adaptation risk degree with specific numerical value, so that the risk evaluation is more scientific, objective and intuitive; based on the risk score, a graded warning is triggered, different levels of warning correspond to different risk degrees and processing measures, which improves the efficiency and flexibility of dealing with environmental problems;

[0035] Through early evaluation and warning, it is ensured that the environmental parameters can be stabilized in the range meeting the requirements of the experiment at the beginning of the reservation period and during the entire experiment process, the influence of environmental fluctuations on the experiment results is reduced, the accuracy and repeatability of the experiment results are improved, and the reliability and scientificity of the experiment results are improved. BRIEF DESCRIPTION OF DRAWINGS

[0036] Figure 1 The flowchart of the present application;

[0037] Figure 2 The structural diagram of the environment warning system based on multi-source sensor data. DETAILED DESCRIPTION

[0038] The present application will be described below in conjunction with the drawings in the present application.

[0039] As Figure 1 shown, the environment early warning method based on multi-source sensor data of the application specifically comprises the following steps:

[0040] S1, collecting real-time environmental parameter data set in the laboratory before the target experimental project reservation period starts, and collecting equipment operation data set of the laboratory in the past set time period;

[0041] S2, identifying the environmental constraint characteristics of the target experimental project to obtain multiple experimental environmental constraint characteristics;

[0042] S3, calculating the equipment adjustment change rate of each experimental environmental constraint characteristic per unit time based on the equipment operation data set;

[0043] S4, calculating the initial abnormality degree of each experimental environmental constraint characteristic based on the real-time environmental parameter data set;

[0044] S5, determining at least one experimental self-influence characteristic according to the type of the target experimental project;

[0045] S6, for each experimental environmental constraint characteristic, inputting the corresponding equipment adjustment change rate, initial abnormality degree and determined experimental self-influence characteristic into the pre-trained laboratory environment early warning analysis model to obtain the experimental environment adaptation risk score of the target experimental project, and triggering a graded warning based on the score.

[0046] In this embodiment, the environmental adaptability is evaluated before the target experimental project reservation period starts, which can find potential environmental problems before the experiment starts, take adjustment measures in advance, avoid experiment interruption due to substandard environment, and ensure the continuity of the experiment; experiment interruption will cause waste of samples and reagents and invalid consumption of time; early evaluation of environmental adaptability can prevent such situations from happening, reduce experimental cost, and improve resource utilization efficiency;

[0047] Through the multi-source sensor array deployed in the laboratory, a rich real-time environmental parameter data set including environmental temperature, humidity, air pressure, light intensity, suspended particulate matter concentration, microorganism concentration, oxygen concentration and harmful gas concentration, etc. is collected, and the equipment operation data set containing historical environmental parameter data set and environmental adjustment equipment operation state in the past set time period is collected; the multi-source data complement and verify each other, which can more comprehensively and accurately reflect the laboratory environment, and provide a solid foundation for subsequent analysis;

[0048] The environment adjusting device running state information in the device running data set helps to understand the working capacity and efficiency of the device; in combination with the historical environment parameter data, the adjusting effect of the device on the environment parameter under different working conditions can be analyzed, so that the guarantee capacity of the current environment adjusting device can be more accurately evaluated, and the environment adaptability evaluation is more in line with the actual situation;

[0049] The environment constraint characteristics of the target experiment project are identified, and the specific requirements of the experiment on temperature, humidity and other environment parameters are determined; the constraint degree and range of environment parameters of different experiments are very different, and accurate identification of these characteristics can make the environment evaluation more targeted and improve the accuracy of the warning; the experiment self-influence characteristics, i.e., the influence of the experiment itself on the environment constraint characteristics during the experiment process, are determined according to the type of the target experiment project, so that the change trend of the environment parameter during the experiment process can be more comprehensively evaluated, the potential risk can be predicted in advance, and the warning is more scientific and reasonable;

[0050] The device adjusting rate, initial abnormality degree and experiment self-influence characteristics are input into the pre-trained laboratory environment warning analysis model to obtain the experiment environment adaptation risk score of the target experiment project; the multiple influence factors are comprehensively considered to reflect the environment adaptation risk degree with specific numerical values, so that the risk evaluation is more scientific, objective and intuitive; the hierarchical warning is triggered based on the risk score, different levels of warning correspond to different risk degrees and processing measures, and the efficiency and flexibility of dealing with environmental problems are improved;

[0051] Through early evaluation and warning, it is ensured that the environment parameter can be stabilized in the range meeting the requirements of the experiment at the beginning of the reservation period and during the entire experiment process, the influence of environmental fluctuations on the experiment result is reduced, the accuracy and repeatability of the experiment result are improved, and the experiment result is more reliable and scientific.

[0052] In some embodiments of the present application, for step S1, real-time environment parameter data set in the laboratory is collected before the reservation period of the target experiment project starts, and device running data set of the laboratory in the past set time period is collected;

[0053] The environment parameter data set includes:

[0054] The environment temperature, the temperature is a key factor affecting many biological and chemical reactions; in biological experiments, the activity of enzymes will change significantly with the change of temperature, and too high or too low temperature can cause enzyme deactivation, thereby affecting the experiment result; in chemical experiments, temperature also affects the reaction rate and equilibrium; through a high-precision temperature sensor, the temperature value in the laboratory can be obtained in real time;

[0055] Environmental humidity, in precision instrument experiments, too high humidity will cause the internal electronic components of the instrument to be damp, affect its performance and life; while in some biological culture experiments, the appropriate humidity is the necessary condition to maintain the normal growth of cells or tissues; humidity sensor can measure the air humidity in the laboratory in real time, and express it in percentage form;

[0056] Environmental air pressure, changes in air pressure will affect some experiments involving gas reactions or requiring precise control of the gas environment; air pressure sensor can monitor the air pressure in the laboratory in real time;

[0057] Light intensity, for light sensitive experiments, light intensity is an important environmental parameter; appropriate light intensity can ensure the smooth progress of the experiment, while too strong or too weak light may cause the experimental results to deviate; light intensity sensor can measure the light intensity in the laboratory;

[0058] Concentration of suspended particulate matter, suspended particulate matter will contaminate experimental samples or interfere with the normal operation of experimental instruments; in precision instrument experiments, particulate matter will adhere to the surface of optical elements, affecting the accuracy of the instrument; in biological experiments, particulate matter may carry bacteria or viruses, causing contamination of experimental samples; particulate matter sensor can monitor the concentration of suspended particulate matter in the laboratory in real time;

[0059] Microbial concentration, in biological experiments, the presence of microorganisms will interfere with the experimental results; in cell culture experiments, microbial contamination may cause cell death or abnormal growth; microbial sensor or by culture counting method can obtain the information of microbial concentration in the laboratory;

[0060] Oxygen concentration, many biological and chemical reactions need to be carried out under specific oxygen concentration; for example, in cell respiration experiments, changes in oxygen concentration will affect the respiration rate of cells; in chemical oxidation reactions, oxygen is an important reactant; oxygen sensor can measure the oxygen concentration in the laboratory in real time;

[0061] Harmful gas concentration, some chemical experiments may produce harmful gases, harmful gases not only harm the health of experimental personnel, but also affect the experimental results; harmful gas sensor can monitor the harmful gas concentration in the laboratory in real time.

[0062] The device operation data set includes historical environmental parameter data set and operation state of environmental regulation device;

[0063] Historical environmental parameter data set, collect historical environmental parameter data in the past set period, can understand the trend and rule of laboratory environmental parameter change; help to predict the possible change of environmental parameter in the future period, provide reference for environmental adaptability evaluation;

[0064] The running state of the environmental regulation device, including air conditioners, humidifiers, dehumidifiers, air purifiers, etc., directly affects the stability of the laboratory environmental parameters; the running time, power, fault information, etc. of the device are recorded, which can evaluate the performance and reliability of the device.

[0065] In this embodiment, by collecting real-time environmental parameter data set before the start of the target experiment project reservation period, the current environmental state in the laboratory can be grasped in advance; the accurate collection of environmental temperature, humidity, air pressure and other parameters can adapt to the characteristics of different experiments, provide direct basis for judging whether the current environment meets the basic conditions for starting the experiment, and reduce the probability of experimental risk caused by unsuitable initial environment;

[0066] On the data support level, the device running data set of the laboratory in the past set time period is collected, which has important reference value; the historical environmental parameter data set can reflect the change trend and rule of the laboratory environmental parameters, which is helpful to predict the future environmental change trend combined with real-time data, and improve the cognition of environmental stability; and the running state data of the environmental regulation device can directly reflect the performance, reliability and environmental regulation ability of the device.

[0067] In some embodiments of the present application, for step S2, the environmental constraint feature recognition is performed on the target experiment project to obtain a plurality of experimental environmental constraint features.

[0068] Different types of experiments have different and strict requirements for environmental parameters; for example, cell culture in biological experiments requires specific temperature, humidity and gas concentration environment; some reactions in chemical experiments are extremely sensitive to light intensity, air pressure and harmful gas concentration; precision instrument experiments have strict restrictions on suspended particulate matter concentration and microorganism concentration; by identifying the environmental constraint features of the target experiment project, the key environmental conditions that must be met during the experiment can be determined, so that the laboratory environment can be monitored and controlled in a targeted manner, ensuring the accuracy, repeatability and reliability of the experimental results, and avoiding experimental failure or result deviation caused by unsuitable environment;

[0069] The basis for identifying the environmental constraint features of the target experiment project includes:

[0070] The type and nature of the experiment, different types of experiments have different reaction mechanisms and operation requirements, which determine their unique needs for environmental parameters; for example, enzyme reaction experiments require very precise temperature control, as temperature can affect enzyme activity and reaction rate; plant tissue culture experiments require specific light intensity and duration to promote photosynthesis and growth; in chemical experiments, redox reactions may be sensitive to oxygen concentration, and photochemical reactions require specific wavelengths of light; precision instrument experiments, such as optical instrument experiments, require very low concentrations of suspended particulate matter, as particulate matter can scatter light and affect instrument accuracy;

[0071] The experimental operation process also affects the environmental constraint characteristics; for example, during a chemical titration experiment, a relatively stable environment is required to avoid external interference that can affect the accuracy of the titration endpoint; if the temperature changes significantly during the titration process, it can affect the volume of the solution and the reaction rate, thereby affecting the experimental results; during sample collection and processing in biological experiments, a clean environment is required to prevent contamination of the sample;

[0072] The characteristics of the experimental sample are an important basis for identifying environmental constraint characteristics; biological samples such as cells and tissues are very sensitive to environmental conditions and require specific temperature, humidity, and gas concentration for storage and culture; chemical samples such as volatile and oxidizable substances have strict requirements for temperature, humidity, and oxygen concentration to prevent sample deterioration and affect experimental results;

[0073] The experimental environment constraint characteristics include at least one of temperature constraints, humidity constraints, air pressure constraints, light intensity constraints, suspended particulate matter concentration constraints, microorganism concentration constraints, oxygen concentration constraints, and harmful gas concentration constraints.

[0074] In this embodiment, by carefully identifying the environmental constraint characteristics of the target experimental project, the key environmental conditions that must be met during the experiment can be accurately located; a clear and accurate target is provided for subsequent laboratory environment monitoring and control; according to the identified environmental constraint characteristics, environmental monitoring equipment can be deployed and monitoring parameters can be determined; only the environmental parameters closely related to the experiment are monitored in real time, avoiding blind monitoring of all possible environmental parameters, which improves monitoring efficiency and ensures accurate control of key environmental factors; according to the specific requirements of each constraint characteristic, the operating parameters of the environmental adjustment equipment are reasonably adjusted, so that the laboratory environment quickly and accurately reaches and maintains the appropriate range, improving the accuracy and effectiveness of environmental control; accurate identification of environmental constraint characteristics and environmental management accordingly can effectively avoid experimental failure or result deviation caused by environmental inadaptation; in biological experiments, appropriate temperature, humidity and gas concentration environments can ensure normal cell growth and metabolism, thereby obtaining reliable experimental data; in chemical experiments, appropriate light intensity, air pressure and harmful gas concentration conditions can ensure that chemical reactions proceed as expected, improving product purity and yield; in precision instrument experiments, low suspended particulate matter concentration and microorganism concentration environments can reduce interference with instrument precision and performance, ensuring the accuracy of measurement results; by controlling environmental factors from the source, a solid guarantee is provided for the accuracy and reliability of experimental results; by identifying environmental constraint characteristics and strictly controlling laboratory environment, each experiment is carried out under the same or similar environmental conditions, reducing the random influence of environmental factors on experimental results; whether the experiment is carried out at different times or in different batches, consistent results can be obtained, improving the repeatability of the experiment.

[0075] In some embodiments of the present application, for step S3, based on the device operation data set, the device adjustment rate of each experimental environmental constraint characteristic per unit time is calculated;

[0076] The device adjustment rate of each experimental environmental constraint characteristic per unit time can quantify the dynamic performance of the device when adjusting the environmental parameters; it is helpful to predict in advance whether the device has enough capacity to adjust the environmental parameters to meet the experimental requirements before the experimental reservation period starts, and the fluctuation that may occur during the adjustment process, providing key data support for subsequent evaluation of experimental environment adaptation risk, thereby effectively avoiding the problem of experimental environment not meeting the standard due to insufficient device adjustment capacity or large adjustment process fluctuations;

[0077] From the collected device operation data set of the laboratory in the past set time period, the historical environmental parameter data related to each experimental environmental constraint characteristic and the operating state data of the environmental adjustment equipment are extracted;

[0078] A suitable time period is selected to calculate the device adjustment rate; the time period should be able to reflect the dynamic characteristics of the device in the normal adjustment process, while considering the preparation time before the start of the experimental reservation period; according to the actual operation and historical data of the laboratory, a relatively stable and representative time period can be selected;

[0079] The formula for calculating the device adjustment rate of each experimental environment constraint feature in a unit of time is:

[0080] Wherein, R P represents the device adjustment rate; n represents the total number of valid device adjustment events for the environmental constraint feature P in the past set time period; P ie represents the target value of the environmental parameter P first adjusted by the device to the experimental constraint range in the i th adjustment event; P is represents the initial value of the environmental parameter P when the device starts active adjustment in the i th adjustment event; t ie represents the time when the environmental parameter P first reaches the experimental constraint range in the i th adjustment event; t is represents the time when the device starts active adjustment in the i th adjustment event.

[0081] In this embodiment, the dynamic performance of the device when adjusting the environmental parameter is quantified by calculating the device adjustment rate; the evaluation of the device adjustment capacity is no longer limited to qualitative description, but has specific and measurable indicators; before the start of the experimental reservation period, according to the calculated device adjustment rate, it can be predicted in advance whether the device has enough capacity to adjust the environmental parameter to the range required by the experiment; it helps the experimental personnel to make plans and preparations in advance, avoids the situation that the environmental parameter cannot meet the standard at the beginning of the experiment due to insufficient device capacity, and thus causes waste of time and resources; the calculation result of the device adjustment rate can also reflect the fluctuation that may occur in the adjustment process of the device, which helps to take measures in advance to reduce the influence of fluctuation on the experimental results and ensure the accuracy and repeatability of the experiment; the calculated device adjustment rate provides key data support for subsequent evaluation of experimental environment adaptation risk; combined with other related data, the adaptation degree of the experimental environment to the experimental requirements can be more comprehensively and accurately evaluated; through the calculation and analysis of the device adjustment rate, the problem of non-standard experimental environment caused by insufficient device adjustment capacity or large adjustment process fluctuation can be effectively avoided; it helps to improve the success rate of the experiment, ensures the smooth progress of the experimental research, and reduces the loss and risk caused by environmental problems.

[0082] In some embodiments of the present application, for step S4, based on the real-time environmental parameter data set, the initial abnormality degree of each experimental environment constraint feature is calculated;

[0083] The initial abnormality degree is a quantitative index for measuring the matching degree of real-time environmental parameters and experimental environmental constraint characteristics, and reflects the severity of deviation of each environmental parameter from the experimental requirements before the start of the experimental reservation period.

[0084] For environmental parameters with clear upper and lower limits, including temperature and humidity, the calculation method is as follows: when the real-time parameter is lower than the required lower limit, the difference between the lower limit value and the real-time value is subtracted, and the difference between the upper limit and the lower limit of the required range is divided by the result, which is the initial abnormality degree; when the real-time parameter is higher than the required upper limit, the difference between the real-time value and the upper limit value is subtracted, and the difference between the upper limit and the lower limit of the required range is divided by the result; if the real-time parameter is between the upper and lower limits, then the initial abnormality degree is zero, indicating that the parameter is not abnormal at this time;

[0085] For environmental parameters that only require not to exceed a certain upper limit, including suspended particulate matter concentration and harmful gas concentration, the calculation method is to compare the real-time parameter value with the required upper limit value, and determine the initial abnormality degree through the proportional relationship between the two, the larger the proportion, the more serious the abnormality; when the real-time parameter value reaches or exceeds the upper limit, it is considered to be a serious abnormality; for environmental parameters that only require not to be lower than a certain lower limit, including oxygen concentration, the calculation method is to compare the required lower limit value with the real-time parameter value, and determine the initial abnormality degree through the proportional relationship between the two, the larger the proportion, the more serious the abnormality; when the real-time parameter value is lower than or equal to the lower limit value, it is considered to be a serious abnormality;

[0086] For cases where not only the parameter is required to be within a certain range, but also the fluctuation range of the parameter is required, including air pressure in precision instrument experiments, when calculating the initial abnormality degree, it is necessary to consider whether the average level of the parameter is within the required range and whether the fluctuation amplitude of the parameter is within the allowed range, and by reasonably allocating the influence weight of the two aspects, a comprehensive abnormality degree value is obtained.

[0087] In the embodiment, the matching degree of real-time environmental parameters and experimental environmental constraint characteristics is quantified, and the severity of the deviation of each environmental parameter from the experimental requirements before the start of the experimental reservation period is clearly reflected. The quantified evaluation makes it specific and perceptible whether the environment meets the requirements, and the experimenters can intuitively understand the initial state of each parameter, know which parameters have abnormalities and the severity of the abnormalities. This helps to discover potential environmental problems in advance and provides a basis for timely intervention. By clearly defining the abnormality degree of each parameter, experimenters can adjust the abnormal parameters in a targeted manner, prioritize parameters with higher abnormality degrees, ensure that the environment is adjusted to meet the requirements before the experiment starts, avoid experiment interruption or result deviation due to initial environmental abnormalities, and reduce waste of samples, reagents, and time. Different types of environmental parameters use corresponding calculation methods to ensure the rationality and accuracy of the initial abnormality degree calculation, and can comprehensively cover various experimental environmental constraint characteristics, making the evaluation results more in line with the actual situation.

[0088] In some embodiments of the application, for step S5, at least one experimental self-influence characteristic is determined according to the target experimental project type;

[0089] The experimental self-influence characteristic represents the experimental environmental constraint characteristics generated by the experiment itself during the experiment;

[0090] The basis for determining the experimental self-influence characteristic is the type of the target experimental project. Due to differences in reaction principles and operation methods, different types of experiments produce completely different self-influence characteristics;

[0091] Biological experiments often involve cell culture, microbial culture, etc. Experimental self-influence characteristics include oxygen consumption, carbon dioxide emission, temperature fluctuation, humidity change, etc.

[0092] In chemical experiments, changes in temperature, air pressure, humidity, etc. may occur during chemical reactions. For example, strong acid and strong base reactions may release heat or gas, thereby changing the temperature and gas concentration in the laboratory.

[0093] The self-influence characteristics of precision instrument experiments include vibration, temperature change, and air flow change of precision instruments during the experiment, which will affect the stability of the equipment and the accuracy of the experiment.

[0094] The target experimental project type is input into the pre-constructed experimental type and self-influence characteristic database, and the corresponding experimental self-influence characteristics are automatically matched by analyzing the category and characteristics.

[0095] In the present embodiment, the limitations of traditional environmental early warning, which only focuses on the influence of external environment on the experiment, are made up, and the reaction of the experiment itself on the environment is taken into consideration, thus improving the dimension of environmental risk assessment. The self-influence characteristics of the experiment reflect the changes of environmental parameters caused by the operation and reaction of the experiment itself during the experiment, which enables the risk prediction to expand from one-way environmental influence on the experiment to two-way interaction between the environment and the experiment, making the assessment more comprehensive. The combination of the self-influence characteristics of the experiment, the device adjustment change rate and the initial abnormality degree enables the early warning analysis model to more accurately predict the trend of changes in environmental parameters during the experiment, avoids the risk misjudgment caused by ignoring the influence of the experiment itself on the environment, and improves the accuracy and reliability of the early warning. By clearly defining the self-influence characteristics of different types of experiments, experimenters can develop environmental intervention plans and adjust the air conditioning adjustment strategy in advance. For oxygen consumption in biological experiments, oxygen supplement equipment is prepared in advance, so that preparations are made in advance to ensure the stability of the experimental environment. With the pre-constructed experiment type and self-influence characteristic database, automatic matching of the self-influence characteristics is realized, which not only ensures the efficiency of feature determination, but also reduces the subjectivity of manual judgment, making the determination of the self-influence characteristics of the experiment more standardized and scientific.

[0096] In some embodiments of the present application, for step S6, for each experimental environmental constraint feature, the corresponding device adjustment change rate, initial abnormality degree and determined experimental self-influence characteristic are input into the pre-trained laboratory environment early warning analysis model to obtain the experimental environment adaptation risk score of the target experimental project, and the score is used to trigger a graded early warning;

[0097] A large amount of historical data of different types of experiments under various environmental conditions is collected, including device adjustment change rate, initial abnormality degree, experimental self-influence characteristic and corresponding experimental environment adaptation result. These data are cleaned to remove errors or outliers, and normalized to make the data of different characteristics have the same scale, avoiding the excessive influence of some characteristics with large value range on model training;

[0098] The prepared data is divided into training set, validation set and test set; the training set data is input as input, and the corresponding experimental environment adaptation result is input as label to input into the selected basic model architecture for training; during the training process, the parameters of the model are adjusted to make the prediction result of the model on the training set as close as possible to the actual label; the validation set is used to evaluate the performance of the model during the training process to prevent overfitting of the model; when the performance of the model on the validation set no longer improves, the training is stopped; finally, the test set is used to evaluate the trained model, and the accuracy, recall rate, F1 value and other indicators of the model are calculated to determine the generalization ability and reliability of the model;

[0099] The device adjusts the change rate, the initial abnormal degree and the determined experimental self-influence characteristics need data processing before input;

[0100] Processing of device adjustment change rate: the device adjustment change rate of each experimental environment constraint characteristic in unit time, which reflects the adjustment ability and speed of the environmental adjustment device to the environmental parameters; before inputting the model, the accuracy and stability of the data need to be ensured; if the data has large fluctuations or outliers, filtering algorithm is used for smoothing processing to remove noise interference, so that the device adjustment change rate data can truly reflect the running state and adjustment trend of the environmental adjustment device;

[0101] Processing of initial abnormal degree: the initial abnormal degree of each experimental environment constraint characteristic calculated based on the real-time environmental parameter data set, which represents the deviation between the current real-time environmental parameters and the ideal environmental parameters required by the experiment; when processing the initial abnormal degree data, the initial abnormal degree of different characteristics can be weighted according to the sensitivity of the experiment to each environmental constraint characteristic, highlighting the environmental factors that have greater impact on the experimental results;

[0102] Processing of experimental self-influence characteristics: experimental self-influence characteristics represent the experimental environmental constraint characteristics generated by the experiment itself during the experiment; different types of experiments have different self-influence characteristics; for experimental self-influence characteristics, accurate identification and quantification need to be carried out according to the type of experiment and related knowledge;

[0103] When inputting the model, the experimental self-influence characteristics are fused with the device adjustment change rate and the initial abnormal degree data to form a complete input vector;

[0104] The device adjustment change rate, the initial abnormal degree and the experimental self-influence characteristics data processed and fused are input into the laboratory environment early warning analysis model trained in advance; the model analyzes and predicts the input data according to the learned patterns and rules, and outputs the experimental environment adaptation risk score; the risk score output by the model is a comprehensive evaluation of the current state and future trend of the experimental environment, considering many factors such as the current situation of the environmental parameters, the adjustment ability of the environmental adjustment device and the influence of the experiment on the environment, etc.

[0105] According to the pre-set risk score rule, the risk level is divided into different levels, including low risk, medium risk and high risk; setting risk score 0-3 as low risk, at this time the experimental environment is basically adapted, no need to warn; 4-6 is medium risk, the system can remind the experimental personnel through the short message, email or laboratory management system pop-up window and so on, inform that there is a certain risk in the current experimental environment, suggest to pay close attention or take some preventive measures; 7-10 is high risk, the system immediately issues an emergency warning, informs the experimental personnel to stop the experiment reservation or takes the emergency environmental intervention measure, in order to avoid the experiment interruption or result deviation caused by environmental problems.

[0106] In the embodiment, rich historical data of multiple types of experiments under different environmental conditions is collected, and cleaning and normalization processing is carried out, error outliers are removed, data scales are unified, feature value range differences are avoided to affect model training, high-quality, standardized data basis is provided for the model, which is helpful to improve model training effect and prediction accuracy; training set, validation set and test set are reasonably divided, the model is trained through the training set, the validation set prevents overfitting, the test set evaluates the generalization ability and reliability of the model, the accuracy, recall rate, F1 value and other indicators are used to comprehensively measure the performance of the model, and the laboratory environment early warning analysis model with excellent performance and reliability is ensured to be trained; the device adjustment change rate data in unit time is filtered and smoothed to remove noise interference, ensure data accuracy and stability, and truly reflect the operation state and adjustment trend of the environmental adjustment device, and provide reliable input for the model; according to the sensitivity of the experiment to each environmental constraint feature, the initial abnormal degree of different features is weighted, and the key environmental factors are highlighted, so that the model pays more attention to the factors that have greater influence on the experimental results, and the prediction pertinence and accuracy are improved; the self-influence characteristics of different types of experiments are accurately identified and quantified, and are fused with the device adjustment change rate and the initial abnormal degree data to form a complete input vector, so that the influence of the experiment on the environment is fully considered, and the model analysis is more comprehensive and accurate; the model outputs the experimental environment adaptation risk score, comprehensively considers environmental parameter present situation, environmental adjustment device adjustment ability and the influence of the experiment on the environment and other factors, and comprehensively evaluates the current state and future trend of the experimental environment; according to the pre-set risk score rule, the risk level is divided and different warning measures are taken, low risk does not need to be warned, medium risk reminds the experimental personnel to pay attention, high risk is urgently warned and measures are taken, effectively preventing the experiment interruption or result deviation caused by environmental problems, and ensuring the smooth progress of the experiment.

[0107] As shown in Figure 2 The application also provides an environment early warning system based on multi-source sensor data, which specifically comprises the following modules.

[0108] a data collection module, which collects a real-time environmental parameter data set in the laboratory before the start of a target experiment project reservation period, and collects a device operation data set of the laboratory in a past set time period;

[0109] an experimental environment constraint feature identification module, which identifies environmental constraint features of the target experiment project to obtain multiple experimental environment constraint features;

[0110] a device adjustment change rate calculation module, which calculates a device adjustment change rate of each experimental environment constraint feature per unit time based on the device operation data set;

[0111] an initial abnormality degree calculation module, which calculates an initial abnormality degree of each experimental environment constraint feature based on the real-time environmental parameter data set;

[0112] an experimental self-influence feature determination module, which determines at least one experimental self-influence feature according to a target experiment project type;

[0113] an experimental environment adaptation risk score and early warning module, which inputs, for each experimental environment constraint feature, the corresponding device adjustment change rate, initial abnormality degree, and determined experimental self-influence feature into a pre-trained laboratory environment early warning analysis model to obtain an experimental environment adaptation risk score of the target experiment project, and triggers a graded early warning based on the score.

[0114] In this embodiment, the environmental adaptation evaluation is performed before the start of the target experiment project reservation period, rather than monitoring and alarming during the experiment, which can discover possible substandard problems in the environment in advance, so that the experiment personnel have sufficient time to adjust and process before the experiment starts, avoiding the situation that the experiment is forced to terminate due to substandard environment during the experiment, and reducing the possibility of experiment interruption from the source; since the environmental adaptation is evaluated and predicted in advance, the waste of materials such as samples and reagents caused by termination of the experiment in the middle can be effectively avoided, and the time cost caused by the interruption of the experiment is also reduced, improving the utilization efficiency of laboratory resources;

[0115] The system predicts potential risks in combination with the characteristics of the experiment itself, and ensures that the environmental parameters are in a stable state that meets the experimental requirements before the experiment starts; the system avoids the interference on the experimental process caused by the fluctuation in the environmental parameter adjustment process in the existing real-time monitoring mode, thereby reducing the probability of distortion of the experimental results and significantly improving the reliability and accuracy of the experimental results; the system collects multiple environmental parameters through a multi-source sensor array and analyzes in combination with the past device operation data set of the laboratory; at the same time, the experimental self-influence features are considered, and the environmental adaptation risk is evaluated in multiple dimensions and in all directions, so that the evaluation result is more comprehensive and accurate, and the potential environmental risks can be more effectively predicted;

[0116] The system triggers a hierarchical early warning through an experimental environment adaptation risk score, and the experimental personnel can take corresponding processing measures according to the early warning level; for low-risk conditions, simple adjustments can be made; for high-risk conditions, more sufficient preparation or changes in the experimental plan can be made in advance, so that the response measures are more targeted and effective; the experimental self-influence characteristics are determined according to the target experimental project type, that is, the influence of the experiment on the environment during the experiment process is considered; in combination with the risk prediction method of the experimental self characteristics, the early warning is more in line with the actual situation of the experiment, the accuracy of the potential risk prediction is improved, and the smooth progress of the experiment and the reliability of the results are further ensured.

[0117] In a specific implementation, as an embodiment, the data acquisition module further includes a sensor anomaly detection unit for anomaly detection on the sensors collecting real-time environmental parameter data sets;

[0118] The core role of the sensor anomaly detection unit is to continuously and dynamically monitor the abnormal state of each sensor participating in data collection during the collection of real-time environmental parameter data sets; its detection trigger time is throughout the whole process of data collection, including both centralized detection before the start of the target experimental project reservation period and real-time inspection during the real-time environmental parameter collection process, to ensure that the abnormal conditions of the sensors in different stages can be found in time.

[0119] In a specific implementation, as an embodiment, the experimental self-influence characteristic determination module further includes a self-influence characteristic dynamic updating unit for dynamically updating the determined experimental self-influence characteristics according to the real-time progress data of the experiment during the process of the target experimental project, and the updating frequency matches the interval of the key steps of the experiment;

[0120] By introducing the self-influence characteristic dynamic updating unit, the stability of the environmental parameters during the experiment is significantly improved, the cell growth state is better, and the repeatability and accuracy of the experimental results are effectively guaranteed; at the same time, since the potential environmental risks can be predicted in advance and intervention measures can be taken in time, the experiment interruption and sample waste caused by environmental problems are avoided, and the experimental efficiency and economic benefits are improved.

[0121] The basic principles, main features and advantages of the present application are shown and described above. Those skilled in the art should understand that the present application is not limited by the above embodiments, and the above embodiments and descriptions in the specification are only to illustrate the principles of the present application. Without departing from the spirit and scope of the present application, various changes and improvements can be made to the present application, and these changes and improvements all fall within the scope of the claimed present application. The scope of protection of the present application is defined by the appended claims and their equivalents.

Claims

1. A method for environment warning based on multi-source sensor data, characterized in that, The method comprises the following steps: Before the start of the target experiment project reservation period, collect real-time environmental parameter data sets in the laboratory, and collect equipment operation data sets of the laboratory in a set time period in the past; Identify the environmental constraint features of the target experiment project, and obtain multiple experiment environmental constraint features; Based on the equipment operation data set, calculate the equipment adjustment change rate of each experiment environmental constraint feature per unit time; Based on the real-time environmental parameter data set, calculate the initial abnormality degree of each experiment environmental constraint feature; According to the type of the target experiment project, determine at least one experiment self-influence feature; the experiment self-influence feature represents the experiment environmental constraint feature generated by the experiment itself during the experiment; For each experiment environmental constraint feature, input the corresponding device adjustment change rate, initial abnormality degree and determined experiment self-influence feature into the pre-trained laboratory environment early warning analysis model to obtain the experiment environment adaptation risk score of the target experiment project, and trigger a hierarchical warning based on the score. The formula for calculating the equipment adjustment change rate of each experiment environmental constraint feature per unit time is: ， wherein R P represents the device adjustment rate; n represents the total number of valid device adjustment events for the environmental constraint feature P in the past set time period; P ie represents the target value of the environmental constraint feature P when the device first adjusts the environmental constraint feature P to within the experimental constraint range in the i adjustment event; P is represents the initial value of the environmental constraint feature P when the device starts to actively adjust in the i adjustment event; t ie represents the time when the environmental constraint feature P first reaches within the experimental constraint range in the i adjustment event; t is represents the time when the device starts to actively adjust in the i adjustment event. 2.The environment warning method based on multi-source sensor data according to claim 1, characterized in that, The equipment operation data set includes historical environmental parameter data set and running state of environmental adjustment equipment. 3.The environment warning method based on multi-source sensor data according to claim 1, characterized in that, The environmental parameter data set includes environmental temperature, environmental humidity, environmental air pressure, light intensity, suspended particulate matter concentration, microorganism concentration, oxygen concentration and harmful gas concentration, which are collected by deploying a multi-source sensor array in the laboratory. 4.The environment warning method based on multi-source sensor data according to claim 1, characterized in that, The experiment environmental constraint features include at least one of temperature constraint, humidity constraint, air pressure constraint, light intensity constraint, suspended particulate matter concentration constraint, microorganism concentration constraint, oxygen concentration constraint and harmful gas concentration constraint. 5.The multi-source sensor data based environmental warning method of claim 1, wherein, The basis for identifying the environmental constraint features of the target experiment project includes experiment type and nature, experiment operation process and experiment sample characteristics.

6. A multi-source sensor data based environment warning system, the system applied to the multi-source sensor data based environment warning method of claim 1, characterized in that, The method comprises the following steps: A data collection module collects real-time environmental parameter data sets in the laboratory before the start of the target experiment project reservation period, and collects equipment operation data sets of the laboratory in a set time period in the past; An experiment environment constraint feature identification module identifies the environmental constraint features of the target experiment project, and obtains multiple experiment environmental constraint features; An equipment adjustment change rate calculation module calculates the equipment adjustment change rate of each experiment environmental constraint feature per unit time based on the equipment operation data set; An initial abnormality degree calculation module calculates the initial abnormality degree of each experiment environmental constraint feature based on the real-time environmental parameter data set; An experiment self-influence feature determination module determines at least one experiment self-influence feature according to the type of the target experiment project; An experiment environment adaptation risk score and early warning module inputs the corresponding device adjustment change rate, initial abnormality degree and determined experiment self-influence feature of each experiment environmental constraint feature into the pre-trained laboratory environment early warning analysis model to obtain the experiment environment adaptation risk score of the target experiment project, and triggers a hierarchical warning based on the score.

7. The multi-source sensor data based environmental warning system of claim 6, wherein, The data collection module further comprises a sensor anomaly detection unit for detecting anomalies in the real-time environmental parameter data collection sensors.

8. The multi-source sensor data based environmental warning system of claim 6, wherein, The experiment self-influence characteristic determining module further comprises a self-influence characteristic dynamic updating unit, configured to dynamically update the determined experiment self-influence characteristic according to real-time progress data of the experiment during the process of the target experiment project.

Images