Monitoring and early warning method and system based on laboratory operation

Abstract

The invention discloses a monitoring and early warning method and system based on laboratory operation, and the method comprises the steps: obtaining and processing laboratory equipment state data and personnel operation records, and constructing a dynamic asymmetric Bayesian network which reflects the personnel operation intention, the equipment state and the interference relation; outputting probability distribution from the equipment state to the personnel operation intention, monitoring whether the equipment state is abnormal or not, and updating the probability distribution; performing judgment and early warning on the updated probability distribution based on an ROC curve, and defining a loss function to dynamically adjust the dynamic asymmetric Bayesian network; according to the method, the dynamic asymmetric Bayesian network is constructed, and parameters are dynamically adjusted according to the real-time change of the equipment state, so that the model shows higher robustness in a dynamically changing laboratory environment; by establishing a backward inference mechanism of the dynamic asymmetric Bayesian network, the operation intention of the personnel is inferred from the equipment state, the perception level of potential risks is improved, and the defects of a traditional system in deep inference are overcome.

Description

Technical Field

[0001] The present invention relates to the technical field of laboratory monitoring and early warning, and particularly to a monitoring and early warning method and system based on laboratory operations. Background Art

[0002] Laboratory safety monitoring and early warning technology, as an indispensable guarantee means in the modern scientific research environment, has made great progress in terms of technological progress and application expansion in recent years. Early laboratory monitoring mainly relied on video monitoring equipment to achieve preliminary safety status assessment and early warning functions by monitoring the images of personnel activities and the operating status of equipment. With the popularization of the Internet of Things (IoT) technology, remote status monitoring and management in the laboratory environment have gradually emerged, making it possible to identify equipment operation anomalies at an early stage and perform preventive maintenance. In addition, the introduction of artificial intelligence and machine learning technologies has further promoted technological innovation. For example, deep learning algorithms are used to classify personnel behavior patterns, or time series data analysis is used to predict potential risks. However, when facing the complex and dynamic laboratory environment, these technologies often have difficulty in performing correlation analysis of the laboratory environment, especially when monitoring the interaction relationship between the operation intention of personnel, the equipment status, and interference, showing certain limitations.

[0003] Among them, this is particularly obvious when dealing with high-complexity and high-risk scenarios. Traditional monitoring systems are difficult to adapt to the variability of personnel behavior and the dynamic evolution of equipment status in laboratory operations, and cannot accurately obtain the equipment status, personnel intention, and interference relationship during the experiment process. In addition, existing technologies lack the ability to integrate and analyze multi-source heterogeneous data, and fail to fully explore the deep associations among personnel operations, equipment operating status, and hidden interference factors (such as someone attempting to deliberately block the monitoring equipment), limiting the system's comprehensive perception and accurate early warning of potential risks. Summary of the Invention

[0004] The purpose of this part is to outline some aspects of the embodiments of the present invention and briefly introduce some preferred embodiments. Simplifications or omissions may be made in this part, as well as in the abstract and title of the present application, to avoid obscuring the purpose of this part, the abstract, and the title. However, such simplifications or omissions shall not be used to limit the scope of the present invention.

[0005] In view of the above existing problems, the present invention is proposed. Therefore, the present invention provides a monitoring and early warning method based on laboratory operations to solve the problems mentioned in the background art.

[0006] To solve the above technical problems, the present invention provides the following technical solutions:

[0007] In a first aspect, the present invention provides a monitoring and early warning method based on laboratory operations, including:

[0008] Through the sensor network and the laboratory management system, obtain and process the laboratory equipment status data and personnel operation records, and construct a dynamic asymmetric Bayesian network that reflects the relationship among personnel operation intentions, equipment status, and interference relationships;

[0009] Utilize the dynamic asymmetric Bayesian network to output the probability distribution that infers from the equipment status back to the personnel operation intention, monitor whether the equipment status is abnormal, and update the probability distribution according to the equipment status monitoring result;

[0010] Based on the ROC curve, judge the updated probability distribution, give an early warning according to the judgment result, and define a loss function to dynamically adjust the dynamic asymmetric Bayesian network, so as to realize the intelligent monitoring and early warning of user violation behaviors under laboratory operations.

[0011] As a preferred scheme of the monitoring and early warning method based on laboratory operations according to the present invention, wherein: constructing a dynamic asymmetric Bayesian network that reflects the relationship among personnel operation intentions, equipment status, and interference relationships includes:

[0012] Define the nodes in the dynamic Bayesian network as the personnel operation intention O, the equipment status S, and the interference relationship H respectively;

[0013] The value of the personnel operation intention O is compliance operation and violation operation, the value of the equipment status S is normal operation and abnormal operation, and the value of the interference relationship H is no interference state and interference state;

[0014] Define the dynamic Bayesian network as an asymmetric form, where the edge relationship of the nodes is O→S, indicating the direct impact of the personnel operation intention on the equipment status; H→S, indicating the direct impact of the interference relationship on the equipment status; there is no O→H edge, indicating that the personnel operation intention does not directly affect the interference relationship.

[0015] As a preferred scheme of the monitoring and early warning method based on laboratory operations according to the present invention, wherein: it further includes:

[0016] According to the switching frequency of the equipment status, dynamically adjust the weight of H→S, and the formula is expressed as:

[0017] w H→S (t)=α·Var(S t-k:t )+(1-α)·w H→S (t-1)

[0018] where, w H→S (t) represents the edge weight from the interference relationship H to the equipment status S at time t; α represents the smoothing factor, which is used to balance the importance of the variance at time t-k and the edge weight at time t-1; Var(S t-k:t) is represented as the variance of the device state S from time t - k to time t; w H→S (t - 1) is represented as the edge weight from H to S at time t - 1.

[0019] As a preferred solution of the monitoring and warning method based on laboratory operations described in the present invention, wherein: using the dynamic asymmetric Bayesian network, the probability distribution of reverse inferring the personnel operation intention from the device state is output, including:

[0020] According to the personnel operation intention O and the interference relationship H, calculate the probability distribution of the device state S, and construct a pseudo-likelihood function through the prior probability of the set personnel operation intention;

[0021] Calculate the posterior probability of the interference relationship H, and the posterior probability of the interference relationship H is obtained by marginalization calculation through the posterior probability of the personnel operation intention O;

[0022] According to the pseudo-likelihood function, the posterior probability of the interference relationship H, and the probability distribution property rules, output the probability distribution of reverse inferring the personnel operation intention from the device state by an iterative method, that is, the posterior probability.

[0023] As a preferred solution of the monitoring and warning method based on laboratory operations described in the present invention, wherein: the iterative method includes:

[0024] Start from the initial probability distribution being the prior probability of the set personnel operation intention, that is, use the prior probability of the set personnel operation intention as the iterative starting point;

[0025] In each iteration, use the pseudo-likelihood function and the posterior probability of the interference relationship H to calculate the new posterior probability, that is, the posterior probability of reverse inferring the personnel operation intention from the device state, and update the posterior probability of the interference relationship H through the personnel operation intention O;

[0026] As the number of iterations increases, make the posterior probability of reverse inferring the personnel operation intention from the device state approach the prior probability of the personnel operation intention until the absolute value of the probability change between two consecutive iterations is less than 0.01, and then stop the iteration.

[0027] As a preferred solution of the monitoring and warning method based on laboratory operations described in the present invention, wherein: monitor whether the device state is abnormal, and update the probability distribution according to the device state monitoring result, including:

[0028] Obtain the device state monitoring result by counting the number of times the device state switches from normal to abnormal and the maximum value of the duration in seconds of each abnormal state;

[0029] If it is determined that the monitoring result of the device status is abnormal, then according to the abnormal value of the monitoring result of the device status, update the posterior probability of inferring the personnel operation intention from the slave device status.

[0030] As a preferred solution of the monitoring and warning method based on laboratory operations described in the present invention, wherein: judging the updated probability distribution based on the ROC curve, giving a warning according to the judgment result, and defining a loss function to dynamically adjust the dynamic asymmetric Bayesian network, including:

[0031] Draw an ROC curve through the updated probability distribution. If the posterior probability of inferring the personnel operation intention from the slave device status after update is greater than the optimal threshold in the generated ROC curve, trigger a local acoustic and optical warning, and at the same time send a text message for notification, and upload the warning record and the text message notification record to the laboratory management system;

[0032] At the same time, define a loss function according to the non-warning situation, and perform gradient descent on the edge weights in the dynamic asymmetric Bayesian network.

[0033] In a second aspect, the present invention provides a monitoring and warning system based on laboratory operations, which includes:

[0034] A dynamic asymmetric Bayesian network construction module, configured to obtain and process laboratory equipment status data and personnel operation records through a sensor network and a laboratory management system, and construct a dynamic asymmetric Bayesian network reflecting the relationship between personnel operation intention, equipment status and interference;

[0035] A device status monitoring module, configured to use the dynamic asymmetric Bayesian network to output the probability distribution of inferring the personnel operation intention from the device status, monitor whether the device status is abnormal, and update the probability distribution according to the device status monitoring result;

[0036] A warning and dynamic asymmetric Bayesian network dynamic adjustment module, configured to judge the updated probability distribution based on the ROC curve, give a warning according to the judgment result, and define a loss function to dynamically adjust the dynamic asymmetric Bayesian network, so as to realize intelligent monitoring and warning under laboratory operations.

[0037] In a third aspect, the present invention provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and: when the processor executes the computer program, any step of the above method is implemented.

[0038] In a fourth aspect, the present invention provides a computer-readable storage medium, on which a computer program is stored, and: when the computer program is executed by a processor, any step of the above method is implemented.

[0039] Compared with the prior art, the beneficial effects of the invention are as follows:

[0040] 1. The present invention constructs a dynamic asymmetric Bayesian network through personnel operation intentions, equipment states, and covert interference. This network model can dynamically adjust parameters according to the real-time changes in equipment states, accurately capture the interactions between operations and states, improve the detection accuracy of abnormal situations in the laboratory, and enable the model to exhibit stronger robustness in a dynamically changing laboratory environment. In addition, by establishing a reverse inference mechanism in the dynamic asymmetric Bayesian network, inferring personnel operation intentions from equipment states, and iteratively optimizing to identify the subtle effects of covert interference, the perception and early warning levels of potential risks are enhanced, making up for the deficiencies of traditional systems in deep inference.

[0041] 2. Using the sensor network and the laboratory management system, collecting equipment state data and personnel operation records, and integrating them into the dynamic asymmetric Bayesian network for analysis, provides a comprehensive view in the actual laboratory operation environment, enabling the laboratory management system to issue early warnings in a timely and accurate manner, effectively reducing the possibility of early warning omissions due to data fragmentation. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the following drawings are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings. Among them:

[0043] Figure 1 is the overall flowchart of the monitoring and early warning method based on laboratory operations according to an embodiment of the present invention;

[0044] Figure 2 is the iterative flowchart of the probability distribution for inferring personnel operation intentions from equipment states in the monitoring and early warning method based on laboratory operations according to an embodiment of the present invention;

[0045] Figure 3 is the comparison chart of the influence of dynamic weight adjustment on the warning accuracy rate in the monitoring and early warning method based on laboratory operations according to an embodiment of the present invention;

[0046] Figure 4 is the ROC curve and iterative convergence process (personnel operation detection scenario) chart of the monitoring and early warning method based on laboratory operations according to an embodiment of the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0047] To make the above objects, features, and advantages of the present invention more apparent and understandable, the following provides a detailed description of the specific embodiments of the present invention in conjunction with the accompanying drawings of the specification. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the scope of protection of the present invention.

[0048] In the following description, numerous specific details are set forth to facilitate a thorough understanding of the present invention. However, the present invention may be practiced in other ways different from those described herein. Those skilled in the art can make similar extensions without departing from the connotation of the present invention. Therefore, the present invention is not limited by the specific embodiments disclosed below.

[0049] Secondly, the so-called "one embodiment" or "embodiment" herein refers to specific features, structures, or characteristics that may be included in at least one implementation manner of the present invention. The phrase "in one embodiment" appearing in different places in this specification does not necessarily refer to the same embodiment, nor is it an embodiment that is separate or selectively exclusive of other embodiments.

[0050] The present invention is described in detail in conjunction with schematic diagrams. When elaborating on the embodiments of the present invention, for ease of explanation, the cross-sectional views showing the device structure will be enlarged locally not in accordance with the general scale, and the schematic diagrams are only examples and should not limit the scope of protection of the present invention herein. In addition, in actual production, three-dimensional spatial dimensions including length, width, and depth should be included.

[0051] Meanwhile, in the description of the present invention, it should be noted that the orientation or positional relationships indicated by terms such as "upper, lower, inner, and outer" are based on the orientation or positional relationships shown in the drawings. This is only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation. Therefore, it should not be construed as a limitation to the present invention. In addition, the terms "first, second, or third" are only used for descriptive purposes and cannot be construed as indicating or implying relative importance.

[0052] Unless otherwise clearly defined and limited in the present invention, the terms "installation, connection, and coupling" should be understood in a broad sense. For example, it can be a fixed connection, a detachable connection, or an integral connection; it can also be a mechanical connection, an electrical connection, or a direct connection, or can be indirectly connected through an intermediate medium, or can be the communication inside two elements. For those of ordinary skill in the art, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0053] Embodiment 1

[0054] Refer to Figure 1 and Figure 2, which is the first embodiment of the present invention. This embodiment provides a monitoring and early warning method based on laboratory operations, including:

[0055] S1. Through the sensor network and the laboratory management system, obtain and process the laboratory equipment status data and personnel operation records, and construct a dynamic asymmetric Bayesian network reflecting the personnel operation intention, equipment status, and interference relationship.

[0056] Specifically, the sensor network is used to obtain the laboratory equipment status data in real time. The obtained data includes temperature, pressure, current, etc. And the collected data is received through the streaming processing framework Apache Kafka, generating a data packet per second, whose format is JSON, containing a timestamp and a status value. For example, {"timestamp":"2025-03-03 10:00:01","temperature":25.3,"pressure":101.2}.

[0057] It should be noted that when obtaining the equipment status data, it is necessary to perform denoising processing on the equipment status data. Among them, the denoising adopts the moving average filtering formula, and the purpose is to align the timestamp of the equipment data after removing the disturbance with the data in the laboratory management system.

[0058] Specifically, the laboratory management system records the operation behaviors (equipment switch time, equipment parameter adjustment records, current laboratory users, and violation records of corresponding personnel, etc.), and aligns the timestamps with the equipment status data (the error is less than 0.1 second).

[0059] Furthermore, the nodes in the dynamic Bayesian network are respectively defined as the personnel operation intention O, the equipment status S, and the interference relationship H.

[0060] It should be explained that the traditional Bayesian network is a graphical model used to represent random variables and their probabilistic dependence relationships. In this model, nodes represent random variables, edges represent the causal or dependence relationships between variables, and the direction of the edges usually reflects the direction of causality. In addition, the edge relationships in the traditional Bayesian network are bidirectional or symmetric, while in the solution of the present invention, constructing an asymmetric Bayesian network means that the direction of the edges is unidirectional and the edge relationships are not completely equivalent.

[0061] Specifically, the personnel operation intention O takes values of compliant operation and violation operation, the equipment status S takes values of normal operation and abnormal operation, and the interference relationship H takes values of no interference state and interference state.

[0062] Furthermore, define the dynamic Bayesian network in an asymmetric form, where the edge relationship between nodes is O→S, indicating the direct impact of the personnel's operation intention on the device state (meaning that whether the personnel's operation intention is compliant or non-compliant will directly cause changes in the device state; for example, compliant operations may keep the device normal, while non-compliant operations may cause the device to malfunction); H→S, indicating the direct impact of the interference relationship on the device state (for example, blocking the camera or tampering with data, which will affect the observed state of the device. Suppose the device is in an abnormal state, but it will be covered up as normal); there is no O→H edge, indicating that the personnel's operation intention does not directly affect the interference relationship (indicating that the operation intention and the interference relationship are independent variables. In other words, the presence or absence of interference is not directly determined by the personnel's operation intention, but occurs independently).

[0063] Specifically, explain the values of the personnel's operation intention O, the device state S, and the interference relationship H, as well as the edge relationships between them; for the edge relationship O→S, simply put, if the personnel's operation intention is compliant (the experimenter operates the instrument according to the regulations), then usually the device state will remain normal because compliant operations do not damage the device; if the personnel's operation intention is non-compliant, it may cause the device state to become abnormal (the experimenter secretly adjusts the instrument parameters); however, O→S is only the case usually because the device state S is also affected by the interference relationship H; for H→S, interference also affects the device state. Among them, the interference may be an intentional act (such as blocking the surveillance camera), or it may be an accidental situation (such as a device failure caused by its own reasons); then in the absence of interference, the device state can directly reflect the result of the personnel's operation intention (the experimenter operates non-compliance, resulting in device abnormality); in the presence of interference, the interference may cover up or distort the device state; (for example, the experimenter operates non-compliance, resulting in device abnormality, but if he also blocks the surveillance camera at the same time, then the surveillance camera may observe that the device state is normal because the device abnormality is hidden); for no O→H, it means that the presence or absence of interference does not depend on the personnel's operation intention, that is, when the personnel's operation intention is compliant, there may or may not be interference; when the personnel's operation intention is non-compliant, there may or may not be interference; in addition, if there is O→H, it means that the personnel's operation intention directly determines the interference, that is, a non-compliant intention always leads to interference, and a compliant intention always has no interference; but the actual situation is not like this. For example, a person with a compliant operation intention may accidentally block the camera, and a person with a non-compliant operation intention may not take any interference measures (block the surveillance camera).

[0064] It should be noted that since the laboratory environment is not static, the state of the equipment may be affected by various factors, such as equipment aging, external interference, or changes in operating habits. If fixed edge weights are used, the dynamic asymmetric Bayesian network cannot reflect these changes in a timely manner, resulting in a decrease in the accuracy of prediction or detection. If the weight of H→S is adjusted dynamically, the model can update according to the real-time equipment state data, making the edge weights more in line with the current laboratory environment.

[0065] Furthermore, according to the switching frequency of the equipment state, the weight of H→S is adjusted dynamically, and the formula is expressed as:

[0066] w H→S (t) = α·Var(S t-k:t )+(1-α)·w H→S (t-1)

[0067] Among them, w H→S (t) represents the edge weight from the interference relationship H to the equipment state S at time t; α represents the smoothing factor, which is used to balance the importance of the variance at time t-k and the edge weight at time t-1; Var(S t-k:t ) represents the variance of the equipment state S from time t-k to time t; w H→S (t-1) represents the edge weight from H to S at time t-1.

[0068] It should be noted that if the switching frequency of the equipment state is high, it indicates that the influence of interference on the equipment state is enhanced. At this time, by increasing the weight of H→S, the sensitivity of the model to anomalies can be improved; when the switching frequency of the equipment state is low, it indicates that the influence of interference on the equipment state is weak, and the weight adjustment range can be appropriately reduced to make the model respond quickly to the equipment state.

[0069] S2. Using the dynamic asymmetric Bayesian network, output the probability distribution of inferring the personnel operation intention from the equipment state, monitor whether the equipment state is abnormal, and update the probability distribution according to the monitoring result of the equipment state.

[0070] It should be noted that due to the asymmetry of the Bayesian network structure and the unidirectional introduction of the interference relationship H, traditional Bayesian inference methods are difficult to apply directly. The method proposed in this invention provides a probability reconstruction mechanism based on the pseudo-likelihood function and iterative update to achieve reverse inference from the equipment state to the personnel operation intention.

[0071] Furthermore, according to the personnel operation intention O and the interference relationship H, calculate the probability distribution of the equipment state S, and construct a pseudo-likelihood function through the prior probability of the set personnel operation intention.

[0072] Specifically, the constructed pseudo-likelihood function \(L(O|S,H)\) can be expressed by the mathematical formula as follows:

[0073] \(L(O|S,H)=P(S|O,H)\cdot P(O)\)

[0074] Among them, \(P(O)\) represents the prior probability of the set personnel operation intention, and the set initial value follows a uniform distribution (for example, the probabilities of the compliance personnel operation intention and the non-compliance personnel operation intention are equal); \(P(S|O,H)\) represents the probability distribution of the device state \(S\) obtained under the combined action of the personnel operation intention \(O\) and the interference relationship \(H\); assuming that \(O\) takes the value of compliance and \(H\) takes the value of no interference, then the obtained result is the probability distribution of the device state under the condition that the personnel operation intention is compliant and there is no interference.

[0075] Furthermore, the posterior probability of the interference relationship \(H\) is calculated. The posterior probability of the interference relationship \(H\) is obtained through marginalization calculation using the posterior probability of the personnel operation intention \(O\).

[0076] Specifically, the marginalization calculation is performed through the posterior probability of the operation intention \(O\):

[0077]

[0078] Among them, \(P\) (k) (H|S) represents the posterior probability of the interference relationship \(H\) in the \(k\)-th iteration; \(P\) (k) (O|S) represents the posterior probability of the personnel operation intention \(O\) in the \(k\)-th iteration.

[0079] Furthermore, according to the pseudo-likelihood function, the posterior probability of the interference relationship \(H\), and the probability distribution property rules, the probability distribution from the device state to the personnel operation intention is output through an iterative method, that is, the posterior probability.

[0080] Specifically, the posterior probability output from the device state to the personnel operation intention is:

[0081]

[0082] Among them, \(\sum\) O \(L(O|S,H)\cdot P\) (k) (H|S) represents a normalization factor to ensure that \(P\) (k+1) (O|S) satisfies the probability distribution property rules, that is, the sum is 1.

[0083] Specifically, referring to Figure 2 , the steps of the iterative method are as follows:

[0084] S201. Starting from the prior probability of the set personnel operation intention as the initial probability distribution, that is, using the prior probability of the set personnel operation intention as the iteration starting point: \(P\) (0)(O|S) = P(O);

[0085] S202. In each iteration, use the pseudo-likelihood function and the posterior probability of the interference relationship H to calculate the new posterior probability P (k+1) (O|S), that is, the posterior probability of inferring from the device state to the human operation intention, and update the posterior probability P (k) (H|S) of the interference relationship H through the human operation intention O;

[0086] S203. As the number of iterations increases, make the posterior probability of inferring from the device state to the human operation intention approach the prior probability of the human operation intention until the absolute value of the probability change between two consecutive iterations is less than 0.01, that is, |P (k+1) (O|S) - P (k) (O|S)| < 0.01, and stop the iteration;

[0087] It should be noted that through iterative operations, P(O|S) is gradually optimized, avoiding directly solving the normalization constant P(S) in the Bayesian network and overcoming the limitations of traditional Bayesian inference methods in non-symmetric networks and interference relationship scenarios;

[0088] Furthermore, by counting the number of times the device state switches from normal to abnormal and the maximum value of the duration in seconds of each abnormal state, the device state monitoring result is obtained;

[0089] Specifically, within a time window of 60s, count the number of times the device state switches from normal to abnormal, define the switching condition as the normal range exceeded by specific device state parameters (such as temperature > 30°C), and record the duration in seconds of each abnormal state, and take the maximum value;

[0090] It should be noted that if a data transmission interruption occurs in the sensor, mark the device state as missing, suspend the calculation process, and issue a warning, indicating that the device state monitoring result is abnormal;

[0091] Even further, if it is determined that the device state monitoring result is abnormal, update the posterior probability of inferring from the device state to the human operation intention according to the abnormal value of the device state monitoring result;

[0092] Specifically, the update uses the Sigmoid function to smoothly adjust the posterior probability P (k +1) (O|S), that is where ΔF is the abnormal value of the device state monitoring result;

[0093] S3. Judge the updated probability distribution based on the ROC curve, issue an early warning based on the judgment result, and define the loss function to dynamically adjust the dynamic asymmetric Bayesian network, so as to realize intelligent monitoring and early warning of user violations under laboratory operation;

[0094] Furthermore, the ROC curve is drawn through the updated probability distribution. If the posterior probability of the personnel operation intention inferred from the equipment status after the update is greater than the optimal threshold in the generated ROC curve, a local sound and light warning is triggered, and a text message is sent for notification. The warning record and the text message notification record are uploaded to the laboratory management system.

[0095] Specifically, through the updated P (k+1) (O|S), draw the ROC curve, where each point on the ROC curve corresponds to a specific θ value, and each point on the ROC curve is obtained by (FPR, TPR) as coordinates, that is, the false positive rate and the true positive rate. Through the Youden index, the θ that maximizes the FPR and TPR is selected as the optimal threshold;

[0096] Specifically, the local sound and light warning shows that the laboratory lights are flashing red, and sends text messages and broadcasts. At the same time, the warning records and text message notification records are uploaded to the laboratory management system; the format is: warning: warning wording + warning timestamp; for example, "Warning: Laboratory equipment abnormality, time: 2025-03-0310:00:01";

[0097] It should be noted that by selecting the optimal threshold on the ROC curve, the model can ensure a high true positive rate when issuing an early warning, while keeping the false positive rate within an acceptable range; on the one hand, it can improve the reliability of the early warning, and on the other hand, it can avoid the problem of wasting resources due to misjudgment of early warnings;

[0098] Furthermore, the loss function is defined according to the non-warning situation, and the edge weights in the dynamic asymmetric Bayesian network are gradient-decreased;

[0099] It should be noted that, if there is no warning, the edge weights in the dynamic asymmetric Bayesian network need to be gradient-decreased to ensure that the model maintains a high level of inference accuracy in the idle state;

[0100] Specifically, the loss function L is defined as:

[0101] L=-∑logP(S|O,H,w H→S )

[0102] Specifically, the gradient descent formula is expressed as:

[0103]

[0104] Among them, is represented as the updated w H→S , is represented as the w before update H→S , and η is the learning rate.

[0105] Furthermore, this embodiment also provides a monitoring and early warning system based on laboratory operations, including:

[0106] A dynamic asymmetric Bayesian network construction module, configured to obtain and process laboratory equipment status data and personnel operation records through a sensor network and a laboratory management system, and construct a dynamic asymmetric Bayesian network reflecting the relationship between personnel operation intentions, equipment status, and interference;

[0107] An equipment status monitoring module, configured to use the dynamic asymmetric Bayesian network to output the probability distribution of inferring personnel operation intentions from the equipment status, monitor whether the equipment status is abnormal, and update the probability distribution according to the equipment status monitoring results;

[0108] An early warning and dynamic adjustment module for the dynamic asymmetric Bayesian network, configured to judge the updated probability distribution based on the ROC curve, give an early warning according to the judgment result, and define a loss function to dynamically adjust the dynamic asymmetric Bayesian network, so as to achieve intelligent monitoring and early warning under laboratory operations.

[0109] This embodiment also provides a computer device applicable to the monitoring and early warning method based on laboratory operations, including:

[0110] A memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions to implement the monitoring and early warning method based on laboratory operations as proposed in the above embodiment.

[0111] The computer device may be a terminal, which includes a processor, a memory, a communication interface, a display screen, and an input device connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage medium. The communication interface of the computer device is used to communicate with external terminals in a wired or wireless manner, and the wireless manner can be achieved through WIFI, operator networks, NFC (Near Field Communication), or other technologies. The display screen of the computer device can be a liquid crystal display screen or an electronic ink display screen. The input device of the computer device can be a touch layer covering the display screen, or a button, trackball, or touchpad set on the outer shell of the computer device, or an external keyboard, touchpad, or mouse, etc.

[0112] This embodiment also provides a storage medium, on which a computer program is stored. When the program is executed by a processor, it implements the monitoring and warning method based on laboratory operations proposed in the above embodiment.

[0113] The storage medium proposed in this embodiment and the data storage method proposed in the above embodiment belong to the same inventive concept. Technical details not described in detail in this embodiment can be referred to the above embodiment, and this embodiment has the same beneficial effects as the above embodiment.

[0114] Embodiment 2

[0115] Refer to Figure 3 and Figure 4 , which is the second embodiment of the present invention. This embodiment provides a monitoring and warning method based on laboratory operations, including: verifying the beneficial effects in the solution of the present invention through experiments; selecting a chemistry laboratory of a certain university as the experimental scenario, and the experimental objects include a high-pressure reactor (model: Parr 4560), a gas chromatograph (model: Agilent 7890B), and personnel operation behaviors; the sensor network is composed of the following devices: temperature sensor: Omega HH314 (accuracy ±0.1°C, sampling frequency 1Hz); pressure sensor: Honeywell TruStability HSC series (accuracy ±0.5% FS, range 0-10MPa); current monitoring module: NI cDAQ-9188 (accuracy ±0.01A); camera: FLIR A315 thermal imaging camera (resolution 640×480, frame rate 30fps);

[0116] The prior art control group adopted: fixed threshold alarm and traditional Bayesian network; among them, the fixed threshold alarm is to set a static threshold for device parameters (such as temperature > 80°C to trigger an alarm); among them, the traditional Bayesian network is a Bayesian network based on fixed edge weights (O→S = 0.7, H→S = 0.3).

[0117] The experimental results are shown as Figure 3 and Figure 4 shown; it can be seen from Figure 3 that the dynamic adjustment method of the solution of the present invention can still maintain a higher accuracy rate when the number of samples increases, verifying its advantages in the dynamic environment of the laboratory; secondly, it can be seen from Figure 4 (left figure) that the AUC of the solution of the present invention is approximately 0.97, significantly higher than the AUC of the traditional method which is approximately 0.82; at the same time, it can be found from Figure 4 (right figure) that the highest value of the posterior probability reaches 1 after 5 iterations, indicating that the stability of the posterior probability can be achieved through a short number of iterations;

[0118] In summary, the network model established by the solution of the present invention can dynamically adjust parameters according to the real-time changes of the device state when the number of samples increases, and the stability of the model can be achieved through a short number of iterations, providing technical support for monitoring and early warning in the actual laboratory operation environment.

[0119] Those skilled in the art should understand that the embodiments of the present invention can be provided as a method, a system, or a computer program product. Therefore, the present application can adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present application can adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code. The solutions in the embodiments of the present application can be implemented in various computer languages, for example, object-oriented programming languages such as Java and interpreted scripting languages such as JavaScript.

[0120] The present application is described with reference to the flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or block in the flowcharts and / or block diagrams, and the combination of processes and / or blocks in the flowcharts and / or block diagrams can be realized by computer program instructions. These computer program instructions can be provided to the processors of general-purpose computers, special-purpose computers, embedded processors, or other programmable data processing devices to generate a machine, so that the instructions executed by the processors of the computer or other programmable data processing devices generate for realizing in the process Figure 1 one process or multiple processes and / or blocks Figure 1means for the functions specified in one or more boxes.

[0121] These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to operate in a particular manner, such that the instructions stored in the computer-readable memory produce a manufacture including an instruction means that implements the functions specified in one Figure 1 or more processes and / or boxes Figure 1 or more boxes.

[0122] These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer-implemented process, so that the instructions executed on the computer or other programmable apparatus provide steps for implementing the functions specified in one Figure 1 or more processes and / or boxes Figure 1 or more boxes.

[0123] Although the preferred embodiments of the present application have been described, those skilled in the art can make additional changes and modifications once they learn the basic creative concepts. Therefore, the appended claims are intended to be construed to include the preferred embodiments and all changes and modifications that fall within the scope of the present application.

[0124] Obviously, those skilled in the art can make various changes and variations to the present application without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the claims of the present application and their equivalent technologies, the present application is also intended to include these modifications and variations.

Claims

1. A monitoring and warning method based on laboratory operations, characterized in that, Including: Obtain and process the status data of laboratory equipment and personnel operation records through a sensor network and a laboratory management system, and construct a dynamic asymmetric Bayesian network reflecting the relationship between personnel operation intentions, equipment status, and interference relationships; Using the dynamic asymmetric Bayesian network, output the probability distribution of inferring personnel operation intentions from the equipment status, monitor whether the equipment status is abnormal, and update the probability distribution according to the monitoring result of the equipment status; Based on the ROC curve, judge the updated probability distribution, give an early warning according to the judgment result, and define a loss function to dynamically adjust the dynamic asymmetric Bayesian network, so as to realize the intelligent monitoring and early warning of user violations under laboratory operations.

2. The monitoring and warning method based on laboratory operations according to claim 1, characterized in that, Constructing a dynamic asymmetric Bayesian network reflecting the relationship between personnel operation intentions, equipment status, and interference relationships, including: Define the nodes in the dynamic Bayesian network as personnel operation intention O, equipment status S, and interference relationship H respectively; The value of the personnel operation intention O is compliant operation and violation operation, the value of the equipment status S is normal operation and abnormal operation, and the value of the interference relationship H is no interference state and interference state; Define the dynamic Bayesian network as an asymmetric form, where the edge relationship of the nodes is O→S, indicating the direct impact of personnel operation intentions on the equipment status; H→S, indicating the direct impact of interference relationships on the equipment status; there is no O→H edge, indicating that personnel operation intentions do not directly affect interference relationships.

3. The monitoring and warning method based on laboratory operations according to claim 2, wherein, Also including: Dynamically adjust the weight of H→S according to the switching frequency of the equipment status, and the formula is expressed as: w H→S w(t) = α·Var(S t-k:r ) + (1 - α)·w H→S (t - 1) where, w H→S (t) represents the edge weight from interference relationship H to device state S at time t; α represents the smoothing factor, which is used to balance the variance at time t - k and the importance of the edge weight at time t - 1; Var(S t-k:t ) represents the variance of device state S from time t - k to time t; w H→S (t - 1) represents the edge weight from H to S at time t - 1.

4. The monitoring and early warning method based on laboratory operations according to claim 2, characterized in that, Using the dynamic asymmetric Bayesian network, output the probability distribution of inferring personnel operation intentions from the equipment status, including: According to the personnel operation intention O and the interference relationship H, calculate the probability distribution of the equipment status S, and construct a pseudo-likelihood function through the prior probability of the set personnel operation intention; Calculate the posterior probability of the interference relationship H, and the posterior probability of the interference relationship H is obtained by marginalization calculation through the posterior probability of the personnel operation intention O; According to the pseudo-likelihood function, the posterior probability of the interference relationship H, and the probability distribution property rules, output the probability distribution of inferring personnel operation intentions from the equipment status, that is, the posterior probability, through an iterative method.

5. The monitoring and early warning method based on laboratory operations according to claim 4, wherein, The iterative method includes: Start from the initial probability distribution as the prior probability of the set personnel operation intention, that is, use the prior probability of the set personnel operation intention as the iterative starting point; In each iteration, use the pseudo-likelihood function and the posterior probability of the interference relationship H to calculate the new posterior probability, that is, the posterior probability of inferring personnel operation intentions from the equipment status, and update the posterior probability of the interference relationship H through the personnel operation intention O; As the number of iterations increases, make the posterior probability of inferring personnel operation intentions from the equipment status approach the prior probability of the personnel operation intention until the absolute value of the probability change between two consecutive iterations is less than 0.01, and then stop the iteration.

6. The monitoring and early warning method based on laboratory operations according to claim 4, wherein, Monitor whether the equipment status is abnormal, and update the probability distribution according to the monitoring result of the equipment status, including: The device status monitoring result is obtained by counting the number of times the device status switches from normal to abnormal and the maximum value of the duration in seconds of each abnormal state. If it is determined that the device status monitoring result is abnormal, then based on the abnormal value of the device status monitoring result, the posterior probability of inferring the personnel operation intention from the device status is updated.

7. The monitoring and early warning method based on laboratory operations according to claim 3 or 6, characterized in that Based on the ROC curve, the updated probability distribution is judged, and early warnings are issued according to the judgment results, and a loss function is defined to dynamically adjust the dynamic asymmetric Bayesian network, including: The ROC curve is drawn through the updated probability distribution. If the posterior probability of inferring the personnel operation intention from the device status after the update is greater than the optimal threshold in the generated ROC curve, a local audible and visual warning is triggered, and a text message is sent for notification at the same time. The warning record and the text message notification record are uploaded to the laboratory management system. At the same time, a loss function is defined according to the non-warning situation, and gradient descent is performed on the edge weights in the dynamic asymmetric Bayesian network.

8. A monitoring and warning system based on laboratory operations, based on the monitoring and warning method based on laboratory operations according to any one of claims 1 to 7, characterized in that, Including: A dynamic asymmetric Bayesian network construction module, configured to obtain and process laboratory device status data and personnel operation records through a sensor network and a laboratory management system, and construct a dynamic asymmetric Bayesian network reflecting the relationship between personnel operation intention, device status, and interference. A device status monitoring module, configured to use the dynamic asymmetric Bayesian network to output the probability distribution of inferring the personnel operation intention from the device status, monitor whether the device status is abnormal, and update the probability distribution according to the device status monitoring result. An early warning and dynamic adjustment module of the dynamic asymmetric Bayesian network, configured to judge the updated probability distribution based on the ROC curve, issue early warnings according to the judgment results, and define a loss function to dynamically adjust the dynamic asymmetric Bayesian network, so as to realize intelligent monitoring and early warning under laboratory operations.

9. A computer device, comprising a memory and a processor, the memory storing a computer program, characterized in that: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by the processor, the steps of the method according to any one of claims 1 to 7 are implemented.

Images