Safety early warning method and system

Abstract

The invention discloses a safety early warning method and system, and the method comprises the steps: obtaining environment data and equipment data corresponding to each piece of laboratory equipment, the environment data and the equipment data being data collected by a detection sensor corresponding to the laboratory equipment, and deploying at least one piece of laboratory equipment in a laboratory, a single laboratory device is correspondingly provided with at least one detection sensor, and the detection sensor is detachably installed in the communication network of the access server; comparing the environment data and the equipment data with a preset parameter threshold to obtain a comparison result; inputting the equipment data and the environment data into a pre-trained anomaly evaluation model for identification to obtain an anomaly score; determining an abnormal event and an early warning strategy corresponding to the abnormal event according to the comparison result and the abnormal score; and sending the early warning information to the target early warning equipment to indicate the target early warning equipment to execute the early warning strategy. According to the invention, the timeliness and accuracy of the detection and response of the laboratory abnormal event are improved.

Description

Technical Field

[0001] This application relates to the field of laboratory safety testing technology, specifically to a safety early warning method and system. Background Technology

[0002] In biomedical laboratory research settings, various abnormal events can easily occur, such as abnormal temperature, insufficient equipment pressure, and the release of toxic or harmful gases, affecting experimental results and threatening the lives of laboratory personnel. Currently, single-point alarm methods are commonly used to detect these abnormal events. However, existing single-point alarm methods mainly rely on a fixed threshold for a single alarm, and then depend on a manual confirmation mechanism for the confirmation and response to abnormal events. This results in poor accuracy in detecting abnormal events, and because the manual response time is uncontrollable, abnormal events still pose a significant threat to experimental results and personnel safety. Summary of the Invention

[0003] This application provides a safety early warning method and system to improve the timeliness and accuracy of detection and response to abnormal events in the laboratory, thereby ensuring experimental safety.

[0004] In a first aspect, embodiments of this application provide a security early warning method applied to a server of a security management platform, the security early warning method comprising:

[0005] The system acquires environmental data and equipment data corresponding to each laboratory device. The environmental data and equipment data are data collected by the detection sensors corresponding to the laboratory devices. At least one laboratory device is deployed in the laboratory, and each laboratory device is configured with at least one detection sensor. The detection sensor can be detachably installed and connected to the communication network of the server.

[0006] The environmental data and the device data are compared with preset parameter thresholds to obtain the comparison results;

[0007] The device data and the environmental data are input into a pre-trained anomaly assessment model for identification, and an anomaly score is obtained.

[0008] Based on the comparison results and the anomaly score, determine the abnormal events and the corresponding early warning strategies for the abnormal events;

[0009] Send warning information to the target warning device to instruct the target warning device to execute the warning strategy.

[0010] In one possible example of the first aspect, determining the abnormal event and the corresponding early warning strategy based on the comparison result and the anomaly score includes:

[0011] Based on the comparison results, at least one target parameter type exceeding the preset parameter threshold is determined, wherein the parameter type is an environmental parameter or a device parameter;

[0012] Based on the type of at least one target parameter, determine the abnormal event;

[0013] By comparing the abnormal scores with multiple preset score thresholds, a score level is obtained;

[0014] The early warning strategy corresponding to the abnormal event is determined based on the abnormal event and the rating level.

[0015] In one possible example of the first aspect, before comparing the environmental data and the device data with a preset parameter threshold to obtain the comparison result, the method further includes:

[0016] Acquire historical operating data, fault records, and preset operating specification data of the laboratory equipment;

[0017] An initial prediction model is trained based on the historical operating data of the laboratory equipment, the fault records of the laboratory equipment, and the preset operating specification data to obtain a threshold prediction model;

[0018] Acquire target environmental data and target equipment data of the laboratory equipment within a target time period;

[0019] The target environment data and the target device data are input into the threshold prediction model for calculation to generate the preset parameter threshold.

[0020] In one possible example of the first aspect, the step of inputting the target environment data and the target device data into the threshold prediction model for calculation to generate the preset parameter threshold includes:

[0021] Get the current time data;

[0022] Determine whether the model update conditions are met based on the environmental data and the time data;

[0023] When the model update conditions are met, the target weight is obtained based on the time data;

[0024] The weight parameters of the threshold prediction model are updated to the target weights, and the updated training is performed to obtain the updated threshold prediction model.

[0025] The target environment data and the target device data are input into the updated threshold prediction model for calculation to generate the preset parameter threshold.

[0026] In one possible example of the first aspect, the step of inputting the device data and the environmental data into a pre-trained anomaly assessment model for identification to obtain an anomaly score includes:

[0027] The device data and the environmental data are input into the anomaly assessment model to calculate the reconstruction error.

[0028] The reconstruction error is normalized to obtain the anomaly score.

[0029] In one possible example of the first aspect, acquiring the environmental data and equipment data corresponding to each laboratory device includes:

[0030] Acquire signal data collected by the target sensor, as well as environmental or equipment data;

[0031] By comparing the signal data with the fingerprint database, the target laboratory equipment corresponding to the target sensor is determined;

[0032] The environmental data or the equipment data is labeled as the data corresponding to the target laboratory equipment.

[0033] In one possible example of the first aspect, the detection sensor is detachably connected to the server's communication network via magnetic mounting; the method further includes:

[0034] When a target device is detected to be connected, the device information of the target device is obtained, wherein the target device is the detection sensor or the early warning device;

[0035] Obtain the driver program for the target device based on the device information;

[0036] Load the driver and drive the target device according to the driver.

[0037] In one possible example of the first aspect, after loading the driver and driving the target device according to the driver, the method further includes:

[0038] When an upgrade instruction for the target device is detected, the first firmware package carried by the upgrade instruction is obtained;

[0039] Obtain the second firmware package currently used by the target device;

[0040] By comparing the first firmware package and the second firmware package, the firmware to be upgraded is determined.

[0041] Update the driver for the target device according to the firmware to be upgraded.

[0042] In one possible example of the first aspect, updating the driver for the target device according to the firmware to be upgraded includes:

[0043] The driver for the target device stored in the first partition is updated according to the firmware to be upgraded, and the driver image of the target device is stored in the first partition and the second partition;

[0044] After the driver stored in the first partition is updated, the driver stored in the second partition of the target device is updated according to the firmware to be upgraded.

[0045] Secondly, embodiments of this application provide a security management system applied to a server of a security management platform, the security management system comprising:

[0046] The acquisition unit is used to acquire environmental data and equipment data corresponding to each laboratory device. The environmental data and equipment data are data collected by the detection sensors corresponding to the laboratory devices. At least one laboratory device is deployed in the laboratory, and each laboratory device is configured with at least one detection sensor. The detection sensor can be detachably installed and connected to the communication network of the server.

[0047] The comparison unit is used to compare the environmental data and the device data with a preset parameter threshold to obtain the comparison result;

[0048] The identification unit is used to input the device data and the environmental data into a pre-trained anomaly assessment model for identification and to obtain an anomaly score.

[0049] The determining unit is used to determine the abnormal event and the corresponding early warning strategy based on the comparison result and the abnormality score.

[0050] The sending unit is used to send warning information to the target warning device to instruct the target warning device to execute the warning strategy.

[0051] Thirdly, embodiments of this application provide an electronic device including a processor, a memory, a communication interface, and one or more programs, the one or more programs being stored in the memory and configured to be executed by the processor, the programs including instructions for performing the steps in the first aspect of embodiments of this application.

[0052] Fourthly, embodiments of this application provide a computer-readable storage medium storing a computer program for electronic data interchange, wherein the computer program causes a computer to perform some or all of the steps described in the first aspect of this embodiment.

[0053] Fifthly, this application provides a computer program product, wherein the computer program product includes a non-transitory computer-readable storage medium storing a computer program operable to cause a computer to perform some or all of the steps described in the first aspect of this application. The computer program product may be a software installation package.

[0054] As can be seen, in this application, the server acquires environmental and equipment data corresponding to each laboratory device. This environmental and equipment data consists of data collected by the detection sensors corresponding to the laboratory devices. At least one laboratory device is deployed within the laboratory, and each device is configured with at least one detection sensor. These sensors are detachably connected to the server's communication network. The server compares the environmental and equipment data with preset parameter thresholds to obtain comparison results. It then inputs the equipment and environmental data into a pre-trained anomaly assessment model for identification, obtaining an anomaly score. Based on the comparison results and anomaly scores, it determines the abnormal event and its corresponding early warning strategy. Finally, it sends an early warning message to the target early warning device to instruct it to execute the early warning strategy. Therefore, this application's multi-dimensional detection and analysis of the laboratory by combining environmental and equipment data is beneficial for improving the accuracy of anomaly event detection. Furthermore, this application further determines abnormal events and corresponding early warning strategies through preset parameter thresholds and anomaly scores, and sends early warning information to the target early warning device to instruct it to execute the early warning strategy. This not only improves detection accuracy but also enables timely and accurate responses to abnormal events, eliminating their impact on the experiment and thus improving experimental safety and the reliability of experimental results. Attached Figure Description

[0055] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0056] Figure 1 This is a schematic diagram of the architecture of a security management platform provided in an embodiment of this application;

[0057] Figure 2 This is a schematic diagram illustrating an application scenario of a security management platform provided in an embodiment of this application;

[0058] Figure 3 This is a flowchart illustrating a security early warning method provided in an embodiment of this application;

[0059] Figure 4This is a block diagram of the functional units of a security management system provided in an embodiment of this application;

[0060] Figure 5 This is a block diagram of the functional units of another security management system provided in this application embodiment;

[0061] Figure 6 This is a structural block diagram of an electronic device provided in an embodiment of this application. Detailed Implementation

[0062] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, and not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present application.

[0063] The terms "first," "second," etc., in the specification, claims, and accompanying drawings of this application are used to distinguish different objects, not to describe a specific order. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or apparatus that includes a series of steps or units is not limited to the listed steps or units, but may optionally include steps or units not listed, or may optionally include other steps or units inherent to these processes, methods, products, or apparatuses.

[0064] In this document, the term "embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of this application. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0065] In the field of laboratory equipment operation monitoring and safety protection, existing monitoring systems suffer from significant technical bottlenecks. Specifically, existing monitoring systems mostly employ a single-point monitoring mode, resulting in generally low monitoring coverage and an inability to uncover potential risks from multi-dimensional data. Furthermore, the accuracy and timeliness of fault warnings are poor. Simultaneously, regarding alarm response mechanisms, existing monitoring systems rely on manual confirmation processes, with an average response time of 8 to 12 minutes from the occurrence of an anomaly to the initiation of a response. For critical equipment such as autoclaves and freeze dryers, which require extremely high response speeds, this delay can easily lead to irreversible losses such as equipment damage and sample failure, seriously threatening laboratory research safety and data integrity. Moreover, with the continuously shortening cycle of laboratory equipment upgrades, existing monitoring systems also have significant deficiencies in scalability. When expanding the system to adapt to new equipment, downtime accounts for a large proportion of the total maintenance time, significantly impacting the normal operating efficiency of the laboratory and failing to meet the rapidly evolving technological demands of modern laboratories.

[0066] To address the aforementioned issues, this application provides a safety early warning method and system. The server acquires environmental and equipment data corresponding to each laboratory device. This data is collected by the corresponding detection sensors of the laboratory devices. At least one laboratory device is deployed within the laboratory, and each device is equipped with at least one detection sensor. These sensors are detachably connected to the server's communication network. The environmental and equipment data are compared with preset parameter thresholds to obtain comparison results. The equipment and environmental data are then input into a pre-trained anomaly assessment model for identification, resulting in an anomaly score. Based on the comparison results and anomaly scores, anomaly events and corresponding early warning strategies are determined. Early warning information is sent to the target early warning device to instruct it to execute the early warning strategy. Therefore, this application, by combining environmental and equipment data for multi-dimensional detection and analysis of the laboratory, improves the accuracy of anomaly event detection. Furthermore, this application determines anomaly events and corresponding early warning strategies through preset parameter thresholds and anomaly scores, and sends early warning information to the target early warning device to instruct it to execute the early warning strategy. This improves detection accuracy while also enabling timely and accurate responses to anomaly events, eliminating their impact on the experiment and thus enhancing experimental safety and the reliability of experimental results.

[0067] The technical solution of this application and how the technical solution of this application solves the above-mentioned technical problems are described in detail below with specific embodiments. These specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described again in some embodiments. The embodiments of this application will now be described with reference to the accompanying drawings.

[0068] The technical solution of this application can be applied to, for example... Figure 1 The security management platform 10 shown includes a server 100, a detection sensor 200, an early warning device 300, and laboratory equipment 400. The server 100 is communicatively connected to the detection sensor 200 and the early warning device 300, respectively.

[0069] See also Figure 1 and Figure 2 The safety management platform can deploy one or more laboratory devices. These devices can be biosafety cabinets, autoclaves, freeze dryers, centrifuges, or other laboratory equipment.

[0070] Each laboratory device can be configured with a corresponding sensor array module. The individual sensors within this module can be modularly designed and support hot-swappable replacement, allowing for sensor insertion (installation) or removal (removal) without interrupting platform system operation. This eliminates the need for power shutdown or shutdown, enabling module replacement or upgrades and shortening platform system deployment time. A single laboratory device's sensor array module includes one or more sensors. For example, see [link to example]. Figure 2 The laboratory equipment includes a first laboratory device, a second laboratory device, ..., an Nth laboratory device. The first laboratory device is equipped with a first, second, and third detection sensor; the second laboratory device is equipped with a fourth and fifth detection sensor; and the Nth laboratory device is equipped with an N1, N2, and N3 detection sensor. These detection sensors include, but are not limited to, temperature sensors, humidity sensors, pressure sensors, motion detection modules (such as gyroscopes), gas concentration sensors, vibration accelerometers, or current sensors. The type and number of detection sensors deployed on different laboratory devices can be the same or different, depending on actual needs. The detection sensors can be deployed on critical parts of the laboratory equipment, such as centrifuge bearings or high-pressure reactor seals. For example, when a centrifuge rotates at high speed, the bearings are subjected to mechanical stress and are prone to wear, which may cause abnormal vibration, temperature rise, or even shaft breakage or equipment malfunction. Therefore, deploying vibration accelerometers and temperature sensors on the centrifuge bearings can monitor motor current fluctuations and the internal state of the equipment. For example, sealing rings are weak points in high-pressure environments; aging or improper installation can lead to sudden pressure drops / rises and toxic gas leaks. Deploying pressure sensors and gas concentration sensors on the sealing rings of high-pressure reactors can monitor the internal temperature gradient and sealing failure characteristics of the pressure vessel. Alternatively, detection sensors can also be deployed in critical areas of the laboratory environment, such as ventilation openings and reagent storage areas.

[0071] Specifically, the laboratory can be equipped with multiple early warning devices (such as...). Figure 2 The warning devices shown (first warning device, second warning device, ..., Mth warning device) can be of different types, such as ventilation systems, fire extinguishing systems, cooling systems, audible and visual alarm systems, rich media push systems, etc. The server can configure different warning devices according to the type of abnormal event to execute the corresponding warning strategy.

[0072] Specifically, server 100 refers to a remote computer used to process large amounts of computing tasks and store data. Server 100 can be a physical server, a server cluster consisting of multiple physical servers, or a distributed system. In this embodiment, the number of servers is not specifically limited. Alternatively, server 100 can also be a cloud server providing basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, content delivery networks (CDNs), and big data and artificial intelligence platforms. No further restrictions are imposed here. In a specific implementation, server 100 in the aforementioned security management platform 10 can also be a server cluster. In one possible example, server 100 can be configured on different servers within a server cluster, thereby reducing the configuration requirements when a single server is processing data.

[0073] This application integrates multiple detection sensors to construct a composite sensor network, and enables communication between the detection sensors, early warning devices, and the server via a Controller Area Network (CAN) bus. This facilitates data transmission and heterogeneous data fusion, overcoming the limitations of traditional single-point detection. It allows for three-dimensional monitoring of environmental parameters, improving the accuracy and reliability of monitoring results. Furthermore, this application establishes standardized device access specifications, enabling plug-and-play integration of new devices (such as detection sensors or early warning devices), shortening platform system deployment time.

[0074] Please see Figure 3 , Figure 3 This is a flowchart illustrating a security early warning method provided in an embodiment of this application. This method can be applied to applications such as... Figure 1 or Figure 2 The server in the security management platform shown, such as Figure 3 As shown, this safety early warning method includes:

[0075] S310 acquires environmental and equipment data for each laboratory device.

[0076] Among them, environmental data and equipment data are data collected by the detection sensors corresponding to the laboratory equipment. At least one laboratory device is deployed in the laboratory, and each laboratory device is configured with at least one detection sensor. The detection sensor can be detachably installed and connected to the server's communication network.

[0077] Specifically, environmental data refers to the data collected by the detection sensors configured for laboratory equipment, which are used to detect the value of at least one environmental parameter associated with the laboratory equipment. For example, the environmental parameter associated with the laboratory equipment may include ambient temperature. Equipment data refers to the data collected by the detection sensors configured for laboratory equipment, which are used to detect the value of at least one equipment parameter associated with the laboratory equipment. For example, the equipment parameter associated with the laboratory equipment may include equipment temperature and equipment rotation speed.

[0078] Taking the detection sensors configured for laboratory equipment, including pressure sensors, temperature sensors, and motion detection modules, as an example, each detection sensor in this laboratory equipment can have its own dedicated signal conditioning circuit developed and designed. For example, for the pressure sensor, a pressure data acquisition channel can be established using a programmable gain amplifier and an analog-to-digital converter (such as a 16-bit ADC); for the temperature sensor, a temperature data acquisition channel can be established using a low-pass filter (such as a cutoff frequency of 10Hz) and an instrumentation amplifier; and for the motion detection module, a motion data acquisition channel can be established by combining a six-axis data fusion algorithm (such as the Mahony complementary filtering algorithm). The environmental and equipment data associated with the laboratory equipment can be collected through the above channels and then uploaded to the server in real time via low-power IoT protocols (such as LoRa / NB-IoT), enabling the server to obtain the environmental and equipment data corresponding to each laboratory device.

[0079] In practice, after receiving the data uploaded by the detection sensor, the server can first determine whether it can find the corresponding laboratory equipment. If the corresponding laboratory equipment is not found, that is, when the detection sensor is uploading data for the first time, the server can first determine the corresponding laboratory equipment.

[0080] Specifically, when the detection sensor uploads the environmental data or equipment data it is detecting, it can also upload the equipment identification information of the corresponding laboratory equipment. The server can use this equipment identification information to determine the laboratory equipment corresponding to the detection sensor.

[0081] Alternatively, in one possible example, acquiring the environmental data and equipment data corresponding to each of the laboratory devices includes: acquiring signal data collected by the target sensor and environmental data or equipment data; comparing the signal data with a fingerprint database to determine the target laboratory device corresponding to the target sensor; and marking the environmental data or equipment data as data corresponding to the target laboratory device.

[0082] The target sensor can be any detection sensor deployed within the laboratory. In this example, it could be a newly deployed detection sensor.

[0083] The signal data refers to the Received Signal Strength Indicator (RSSI) values ​​collected by the detection sensors from various signal sources (such as Wi-Fi access points, Bluetooth beacons, etc.). The fingerprint database includes the coordinates and corresponding signal strength values ​​of all reference points within the laboratory. Reference points are defined based on the location area and are known locations; for example, a reference point can be set at regular intervals (e.g., 2 meters). Laboratory equipment can be deployed at different reference point locations within the laboratory.

[0084] In practice, if the server does not find the corresponding laboratory equipment, it can use a matching algorithm (such as nearest neighbor (NN), k-nearest neighbor (KNN), or weighted k-nearest neighbor (WKNN)) to match the signal data uploaded by the detection sensor in the fingerprint database to find the reference point most similar to the signal data collected by the detection sensor. The laboratory equipment deployed at the coordinates of the reference point is then identified as the target laboratory equipment corresponding to the detection sensor. This establishes and stores the correspondence between the detection sensor and the target laboratory equipment, and identifies the environmental data or equipment data collected by the detection sensor as the data corresponding to the target laboratory equipment. This helps to ensure the accuracy and reliability of the data source.

[0085] In practice, steps S320 to S350 can be performed separately for each laboratory device. The steps performed on multiple laboratory devices can be performed simultaneously to improve identification efficiency.

[0086] S320, compare the environmental data and the device data with preset parameter thresholds to obtain the comparison result.

[0087] The preset parameter threshold is not necessarily a single threshold; it can include the threshold range corresponding to various parameter types associated with the laboratory equipment when completing the current batch of experiments. The parameter types associated with the laboratory equipment when completing the current batch of experiments include at least one parameter type, which can be an environmental parameter or a device parameter. For example, the parameter types associated with the laboratory equipment when completing the current batch of experiments may include ambient temperature, equipment rotation speed, and equipment temperature.

[0088] In practice, after the server obtains the environmental and equipment data corresponding to the laboratory equipment, it compares the actual values ​​of each parameter type in the environmental and equipment data with the threshold range corresponding to that parameter type in the preset parameter thresholds to obtain the comparison results. This determines whether the actual value of the parameter type exceeds the threshold range corresponding to that parameter type in the preset parameter thresholds. For example, the threshold range for the parameter type "equipment rotation speed" in the preset parameter thresholds can be 2190 rpm to 2210 rpm, and the threshold range for the parameter type "equipment temperature" can be 47.5℃ to 56.5℃. Based on the equipment data, the server can obtain that the actual value corresponding to the laboratory equipment's rotation speed is 2200 rpm, and the actual value corresponding to the equipment temperature is 51.2℃. The comparison result shows that the actual values ​​corresponding to both the equipment rotation speed and equipment temperature do not exceed their respective threshold ranges in the preset parameter thresholds.

[0089] In one possible example, before comparing the environmental data and the equipment data with a preset parameter threshold to obtain the comparison result, the method further includes: acquiring historical operating data of the laboratory equipment, fault records of the laboratory equipment, and preset operating specification data; training an initial prediction model based on the historical operating data, fault records, and preset operating specification data of the laboratory equipment to obtain a threshold prediction model; acquiring target environmental data and target equipment data of the laboratory equipment within a target time period; and inputting the target environmental data and target equipment data into the threshold prediction model for calculation to generate the preset parameter threshold.

[0090] Historical operational data includes environmental and equipment data for all laboratory equipment within the laboratory during historical periods. Alternatively, historical operational data may also include environmental and equipment data for all laboratory equipment of the same type as the target laboratory equipment currently being monitored during historical periods. This allows for the training of corresponding threshold prediction models based on the equipment type of each laboratory device, thereby further improving the accuracy and reliability of the threshold prediction model's prediction results and enhancing the accuracy of abnormal event detection.

[0091] The historical period is a pre-set time period, which can be a large span of time such as a quarter before the current time, a year before the current time, or five years before the current time, to ensure that there is enough training data, thereby ensuring the accuracy and reliability of the preset parameter thresholds predicted by the threshold prediction model obtained after subsequent training.

[0092] The laboratory equipment fault log records abnormal events that occurred in multiple laboratory devices within a historical time period, along with the environmental and equipment data corresponding to each abnormal event at the same timestamp. For example, an abnormal event could be "centrifuge speed abnormal due to bearing wear" or "sterilizer pressure abnormality due to seal failure," etc.

[0093] The preset operating specification data refers to the data content constrained by the experimental operating standards or specifications. For example, preset operating specification data may include: the maximum speed constrained by the centrifuge experimental operating specifications is 13300 rpm. If the detection sensor detects that the centrifuge's operating speed exceeds this maximum speed, it can be marked as a fault record.

[0094] The target time period is a preset time period, which refers to the time period adjacent to the current time but preceding the current time. The length of the target time period is shorter than the historical time period. For example, if the length of the target time period is 24 hours and the current time is 21:00 on June 2nd, then the target time period is the period from 21:00 on June 1st to 21:00 on June 2nd of the same year.

[0095] Specifically, historical operational data and laboratory equipment fault records can be tagged with timestamps to generate time-series tables. In the time-series tables, the content of abnormal events in the fault records can be recorded using codes (such as marking seal failure as 1), which is beneficial for data recording and identification.

[0096] In practice, the initial prediction model can be a time series model, such as LSTM. The trained threshold prediction model is used to predict the preset parameter thresholds of laboratory equipment in the first future time period. Specifically, a training sample set can be generated based on historical operating data, laboratory equipment fault records, and preset operating specifications using methods such as feature cross-validation, label verification, or stratified sampling. Then, the training samples in the training sample set are input into the initial prediction model to be trained in batches. Specifically, after each training sample is fed into the initial prediction model, it outputs a probability distribution prediction result for the prediction time. Based on the values ​​of each parameter type in the probability distribution prediction result, the corresponding preset parameter threshold can be generated. For example, the threshold range of the equipment temperature within the preset parameter threshold can be generated based on the equipment temperature value in the probability distribution prediction result, i.e., ±3σ, where σ can be the square root of the variance of the temperature predicted by the model. Then, the actual value within the prediction time corresponding to the preset parameter threshold in the training sample can be compared with the preset parameter threshold. If the actual value is not within the preset parameter threshold, training and optimization can continue until the training converges to obtain the threshold prediction model.

[0097] Specifically, multiple sensors within the laboratory collect and report environmental and equipment data in real time. The server inputs data from past target time periods (e.g., the past 24 hours) into a trained threshold prediction model to output a preset parameter threshold for the predicted time period (e.g., the next hour). When the server receives environmental and equipment data reported by the sensors within the predicted time period, it compares this data with the preset parameter threshold to obtain the comparison result.

[0098] In practice, the server can also use environmental and equipment data collected and reported by the detection sensors at preset intervals (e.g., 24 hours) as training data to periodically update and optimize the threshold prediction model. Alternatively, the server can use environmental and equipment data collected and reported by the detection sensors in real time as training data to update and optimize the threshold prediction model in real time, so that the prediction results of the threshold prediction model are more adapted to the current experimental environment, which is beneficial to improving the accuracy of abnormal event detection and reducing the false alarm rate.

[0099] As can be seen, in this example, the threshold prediction model trained by historical operating data, fault records, and preset operating specification data predicts the preset parameter threshold, which can realize the dynamic updating of the preset parameter threshold and help improve the accuracy and reliability of abnormal event identification.

[0100] In one possible example, the step of inputting the target environment data and the target device data into the threshold prediction model for calculation to generate the preset parameter threshold includes: obtaining the current time data; determining whether the model update condition is met based on the environment data and the time data; if the model update condition is met, obtaining the target weights based on the time data; updating the weight parameters of the threshold prediction model to the target weights and performing update training to obtain the updated threshold prediction model; and inputting the target environment data and the target device data into the updated threshold prediction model for calculation to generate the preset parameter threshold.

[0101] Time data is used to represent the current time, and it can be expressed as a date, month, or quarter.

[0102] Specifically, model update conditions are used to constrain the values ​​of environmental parameters corresponding to different times. For example, model update conditions can constrain the ambient temperature in summer (June to August) to be above 22°C. In specific implementation, after obtaining the current time data, the server can determine the current time range based on the time data and determine the values ​​of various environmental parameters based on the environmental data, thereby determining whether the model update conditions are met based on the current time range and the values ​​of the environmental parameters. For example, if the model update conditions are: in spring (March to May) and the ambient temperature is between 10°C and 22°C; or in summer and the ambient temperature is above 22°C; or in autumn (September to November) and the ambient temperature is between 10°C and 25°C; or in winter (December to February) and the ambient temperature is below 10°C, then if the current time data determines that it is summer and the current ambient temperature is 28°C, then the model update conditions are met.

[0103] Alternatively, specifically, model update conditions can be further used to constrain environmental data corresponding to different regions at different times. For example, the model update condition could be: in city A, it is summer and the ambient temperature is above 25°C. In this case, the server also needs to obtain the laboratory's location information to determine its geographical location, and then combine the time data, location information, and environmental data to determine whether the current model update condition is met, thereby further improving the accuracy of model update node determination.

[0104] In practice, the server can be configured with multiple different threshold prediction models based on time differences, and the weight parameters of these models will vary. When the model update conditions are met, the target weights of the threshold prediction model that should be configured for the current time can be obtained first. Then, the weight parameters of the threshold prediction model are updated to these target weights, and the model is updated and trained based on historical operating data to obtain an updated threshold prediction model that conforms to the current time. The target environment data and target device data for the target time period can then be input into this updated threshold prediction model for calculation, thereby obtaining the preset parameter threshold that should correspond to the current time.

[0105] Specifically, when updating and training the threshold prediction model, it can be trained based on data from the same historical time period. For example, it can be updated and trained using data from the past five summers. Alternatively, when updating and training the threshold prediction model, data from the same historical time period can be identified as core data, and data from the period between the current time and the previous same time period (e.g., between 21:00 on June 2nd last year and 21:00 on June 2nd now) can be identified as auxiliary data. These mixed periodic data are then integrated to obtain mixed periodic data, and the threshold prediction model is updated and trained based on this mixed periodic data. This ensures that the updated threshold prediction model can adapt to the current environment while ensuring that the model's prediction results reflect recent changes in the laboratory, thereby improving the accuracy of the model's predictions.

[0106] As can be seen in this example, the model update conditions are determined by the current time data and environmental data. When the model update conditions are met, the threshold prediction model is updated and trained according to the target weights. The updated threshold prediction model is then used to predict the current preset parameter threshold. This helps to improve the fit between the predicted preset parameter threshold and the current time, and can reduce the impact of experimental differences caused by environmental factors, thereby improving the accuracy of abnormal event detection.

[0107] S330, the device data and the environmental data are input into a pre-trained anomaly assessment model for identification, and an anomaly score is obtained.

[0108] Among them, the anomaly score is used to evaluate the predictive quality of the anomaly assessment model.

[0109] In specific implementation, the anomaly assessment model can be a probabilistic model, trained using historical operational data, fault records of the laboratory equipment, and preset operational specification data. Its input consists of real-time collected equipment and environmental data from the laboratory equipment, and its output is the probability of an anomaly in the actual value corresponding to each parameter type, or the probability of an anomaly in the laboratory equipment. Specifically, the probabilities output by the anomaly assessment model can be scored (e.g., multiplying the output probability by 100) to obtain an anomaly score. For example, if the probability output by the anomaly assessment model is 0.9, the anomaly score can be 90.

[0110] Alternatively, in a specific implementation, the anomaly assessment model can be the threshold prediction model described above. The difference between the threshold range corresponding to each parameter type in the preset parameter thresholds predicted by the threshold prediction model and the actual value can be used to determine the anomaly score. Alternatively, the anomaly score can be calculated based on the difference between the threshold range corresponding to each parameter type in the preset parameter thresholds predicted by the threshold prediction model and the actual value; for example, the standardized value of the difference can be used as the anomaly score.

[0111] In one possible example, the step of inputting the device data and the environmental data into a pre-trained anomaly assessment model for identification to obtain an anomaly score includes: inputting the device data and the environmental data into the anomaly assessment model to calculate the reconstruction error; and normalizing the reconstruction error to obtain the anomaly score.

[0112] Specifically, the anomaly assessment model can be an autoencoder model, which can be trained using environmental and equipment data and corresponding experimental batch information (such as reagent formulations and operators) from historical operational data of similar laboratory equipment during normal operation.

[0113] In practice, after equipment data and environmental data are input into the anomaly assessment model, the model reconstructs the data to obtain reconstructed values ​​for each parameter type. The model then calculates the absolute or mean square error between the actual and reconstructed values ​​of each parameter type in the equipment and environmental data, thus obtaining the reconstruction error. Based on this, the reconstruction errors for each parameter type can be normalized using the Min-Max normalization method, determined by the following formula:

[0114]

[0115] Where X is the calculated reconstruction error, X min X refers to the minimum reconstruction error in a normally functioning sample set. max X refers to the maximum reconstruction error in a normally functioning sample set. normThis refers to the normalized reconstruction error. By normalizing the reconstruction errors corresponding to each parameter type, the magnitude differences in reconstruction errors for different parameter types can be unified and transformed to the same scale standard, i.e., transformed to the [0,1] interval. The value corresponding to the normalized reconstruction error can be directly determined as the anomaly score. Alternatively, a score can be calculated on the normalized reconstruction error (e.g., multiplying the value of the normalized reconstruction error by 100) to obtain the anomaly score. For example, if the value of the normalized reconstruction error is 0.9, the anomaly score can be 90.

[0116] As can be seen in this example, by inputting the collected equipment data and environmental data into the anomaly assessment model for identification and calculating the anomaly score, the anomaly assessment model can determine the credibility of the laboratory equipment anomaly based on the currently collected equipment data and environmental data. Combining this credibility with the judgment helps to improve the accuracy of anomaly identification.

[0117] S340, Based on the comparison results and the anomaly score, determine the abnormal event and the corresponding early warning strategy.

[0118] In practice, abnormal events can be identified based on the comparison results, and then the early warning strategy can be determined by combining the comparison results and the abnormality score. Alternatively, the abnormal events and their corresponding early warning strategies can be determined by combining the comparison results and the abnormality score.

[0119] Specifically, when determining an abnormal event based on the comparison results, and then determining an early warning strategy by combining the comparison results and the abnormal score, in a possible example, determining the abnormal event and the corresponding early warning strategy based on the comparison results and the abnormal score includes: determining at least one target parameter type that exceeds the preset parameter threshold based on the comparison results, wherein the parameter type is an environmental parameter or a device parameter; determining the abnormal event based on the at least one target parameter type; comparing the magnitude relationship between the abnormal score and multiple preset score thresholds to obtain a score level; and determining the early warning strategy corresponding to the abnormal event based on the abnormal event and the score level.

[0120] The comparison results can represent the relationship between each parameter type and the corresponding threshold range in the preset parameter threshold.

[0121] In practice, if the comparison results determine that no target parameter type exceeds the preset parameter threshold, it indicates that the laboratory equipment is operating normally. If the comparison results determine that at least one target parameter type exceeds the threshold range corresponding to the preset parameter threshold, it indicates that the data for these target parameter types is abnormal, and the laboratory equipment may be experiencing an abnormal event. For example, if the laboratory equipment is a centrifuge, and the target parameter type includes the equipment speed, then the possible abnormal event for the laboratory equipment can be determined to be "abnormal centrifuge speed due to bearing wear".

[0122] In practice, multiple preset scoring thresholds can be set, which can be used to assess the severity of the abnormal events corresponding to abnormal scores. For example, if the preset scoring thresholds include a first scoring threshold and a second scoring threshold, when there is only one target parameter type, if the abnormal score corresponding to the target parameter type is greater than or equal to the first scoring threshold, the scoring level can be determined as Level 1; if the abnormal score corresponding to the target parameter type is less than the first scoring threshold but greater than the second scoring threshold, the scoring level can be determined as Level 2; if the abnormal score corresponding to the target parameter type is less than or equal to the second scoring threshold, the scoring level can be determined as Level 3. In this example, the server pre-stores warning strategies for different scoring levels corresponding to different abnormal events. After determining the abnormal event and its corresponding scoring level, the appropriate pre-stored warning strategy can be retrieved from the database as the target warning strategy for possible abnormal events of the current laboratory equipment.

[0123] Specifically, when there are multiple target parameter types, the abnormal scores corresponding to multiple target parameter types can be weighted and summed, and the weighted sum result can be compared with multiple preset scoring thresholds to obtain the scoring level. Specifically, the weight ratio of multiple parameter types associated with laboratory equipment can be obtained based on the experimental batch. When weighting and summing the abnormal scores corresponding to multiple target parameter types, the weighted summation can be calculated based on the weight ratio. For example, if the multiple parameter types associated with laboratory equipment include equipment temperature, equipment speed, and ambient temperature, and their corresponding ratio is 2:3:1, and if the target parameter types are equipment temperature and equipment speed, and their corresponding abnormal scores are 80 and 90 respectively, then the weighted summation result is 80*0.4 + 90*0.6 = 86.

[0124] As can be seen, in this example, determining the abnormal time by comparing the results, and then determining the corresponding early warning strategy for the abnormal event by combining the rating level corresponding to the abnormal score with the abnormal event, helps to improve the accuracy of the early warning strategy adopted, thereby enabling precise prevention and control of abnormal events.

[0125] When determining abnormal events and corresponding early warning strategies by combining comparison results and anomaly scores, the parameter types that exceed the corresponding threshold range in the preset parameter thresholds can be identified based on the comparison results. The probability of data anomalies corresponding to each parameter type can also be determined by comparing the anomaly scores of each parameter type with preset values. Multiple preset values ​​can be set. Specifically, the presence of anomalies in the data corresponding to a parameter type can be determined by combining the parameter types that exceed the corresponding threshold range in the preset parameter thresholds with the anomaly scores of that parameter type. For example, when the anomaly score is expressed as a percentage, multiple preset values ​​can include 60 and 90. If the actual value corresponding to a parameter type exceeds the threshold range and the anomaly score is greater than or equal to 60, it indicates that the data for that parameter type is abnormal; if the actual value corresponding to a parameter type does not exceed the threshold range but the anomaly score is greater than or equal to 90, it indicates that the data for that parameter type is abnormal; otherwise, it indicates that the data for that parameter type is not abnormal. Specifically, the possible abnormal events currently occurring in laboratory equipment can be determined based on the target parameter type that shows anomalies. For example, if the laboratory equipment is a centrifuge, and the parameter types with data anomalies include equipment speed, then the current abnormal event can be determined to be "centrifuge speed abnormal due to bearing wear". Each abnormal event corresponds to a pre-warning strategy. After an abnormal event is identified, its corresponding pre-warning strategy can be directly retrieved, thereby combining the comparison results and the abnormal score to determine the abnormal event and the corresponding pre-warning strategy. The pre-warning strategy includes the pre-warning device that executes the pre-warning strategy.

[0126] S350, send a warning message to the target warning device to instruct the target warning device to execute the warning strategy.

[0127] Among them, target early warning equipment includes linkage equipment (such as exhaust systems, cooling systems, or fire extinguishing systems) and audible and visual alarm devices deployed in the laboratory. Alternatively, target alarm equipment can also be external mobile devices.

[0128] In practical implementation, the server can determine the target early warning device to execute the early warning strategy based on the strategy, and send early warning information carrying the strategy to the target device. For example, if the abnormal event is an excessive concentration of toxic or harmful gases in the laboratory, the early warning strategy could be to activate the laboratory's exhaust system through a linkage control mechanism, in which case the target early warning device is the exhaust system. Alternatively, if the abnormal event is an abnormal rotation speed of laboratory equipment, the early warning strategy could be to shut down the laboratory equipment through linkage control, in which case the target early warning device is the laboratory equipment itself. Furthermore, if the early warning strategy includes integrating the JPush SDK into the user terminal, the target early warning device could also be a mobile device (i.e., the user terminal) that communicates with the server.

[0129] As can be seen, in this application, the server acquires environmental and equipment data corresponding to each laboratory device. This environmental and equipment data consists of data collected by the detection sensors corresponding to the laboratory devices. At least one laboratory device is deployed within the laboratory, and each device is configured with at least one detection sensor. These sensors are detachably connected to the server's communication network. The server compares the environmental and equipment data with preset parameter thresholds to obtain comparison results. It then inputs the equipment and environmental data into a pre-trained anomaly assessment model for identification, obtaining an anomaly score. Based on the comparison results and anomaly scores, it determines the abnormal event and its corresponding early warning strategy. Finally, it sends an early warning message to the target early warning device to instruct it to execute the early warning strategy. Therefore, this application's multi-dimensional detection and analysis of the laboratory by combining environmental and equipment data is beneficial for improving the accuracy of anomaly event detection. Furthermore, this application determines abnormal events and corresponding early warning strategies through preset parameter thresholds and anomaly scores, and sends early warning information to the target early warning device to instruct it to execute the early warning strategy. This not only improves detection accuracy but also enables timely and accurate response to abnormal events, shortens response time, and eliminates the impact of abnormal events on the experiment, thereby improving experimental safety and the reliability of experimental results. Meanwhile, this application can simultaneously support multiple protocols such as Modbus, MQTT, and CoAP, achieving multi-protocol compatibility.

[0130] In one possible example, the security management platform could be equipped with a visual configuration platform. This platform allows for customized configuration of alarm rules and priorities by configuring preset scoring thresholds and pre-defined operational specifications. For instance, the update nodes of the threshold prediction model could be configured seasonally to enable different monitoring modes at different times. Developing a visual configuration platform allows non-technical personnel to complete routine maintenance tasks through a drag-and-drop interface, thus reducing professional maintenance costs.

[0131] In one possible example, the detection sensor is detachably connected to the server's communication network via magnetic mounting. This magnetic mounting method improves the ease of installation and removal of the detection sensor.

[0132] In one possible example, the method further includes: when a target device is detected to be connected, obtaining device information of the target device, wherein the target device is a detection sensor or an early warning device; obtaining a driver program for the target device based on the device information; loading the driver program; and driving the target device based on the driver program.

[0133] The target device can be a detection sensor or early warning device of the communication network of the newly connected server.

[0134] In practice, when the server detects that a target device has been connected, it can obtain the device information of the connected target device (such as vendor ID, device ID, etc.), and then search for a matching driver in the driver library based on the device information, and record the driver to realize driver control of the connected target device.

[0135] Specifically, the driver loading process can support the dynamic injection of JSON format configuration parameters to flexibly adapt to the personalized needs of different devices (such as port number, function switch, etc.), so that device behavior can be adjusted without modifying the driver code.

[0136] As can be seen, in this example, by searching for and loading the corresponding driver through the device information of the newly connected target device, the remote installation of the device driver can be achieved. Compared with the situation where manual on-site installation of the driver is required and work is interrupted, this example can achieve the installation of the driver for the newly connected device by the administrator entering the corresponding driver for the device information in the system background. This is beneficial for adapting to the newly connected target device and can achieve plug-and-play functionality for the device.

[0137] In one possible example, after loading the driver and driving the target device according to the driver, the method further includes: when an upgrade instruction for the target device is detected, obtaining a first firmware package carried by the upgrade instruction; obtaining a second firmware package currently used by the target device; comparing the first firmware package and the second firmware package to determine the firmware to be upgraded; and updating the driver of the target device according to the firmware to be upgraded.

[0138] In practice, the server periodically checks and downloads the latest driver (i.e., the first firmware package) applicable to the target device. That is, the server periodically detects upgrade commands. Alternatively, each time the server receives a first firmware package for upgrade from the target device, it detects a corresponding upgrade command for that first firmware package. Upon detecting an upgrade command, the server retrieves the first firmware package corresponding to the upgrade command and compares it with the driver currently used by the target device (i.e., the second firmware package) to determine the firmware to be upgraded within the first firmware package, which differs from the second firmware package. The server then loads this firmware onto the target device to update its driver.

[0139] As can be seen in this example, by comparing the first firmware package used for upgrading the target device with the currently used second firmware package when an upgrade command is detected, the firmware to be upgraded is obtained. The driver of the target device is then updated according to the firmware to be upgraded. Remote driver updates can be achieved through differential upgrade technology, which helps to reduce bandwidth consumption.

[0140] In one possible example, updating the driver of the target device according to the firmware to be upgraded includes: updating the driver of the target device stored in a first partition according to the firmware to be upgraded, wherein the driver image of the target device is stored in the first partition and a second partition; after the driver stored in the first partition is updated, updating the driver of the target device stored in the second partition according to the firmware to be upgraded.

[0141] In practice, the driver programs and application data of the target device accessing the server are stored in a dual-mirror format (i.e., mirrored storage in a first partition and a second partition). Specifically, when upgrading a driver, the firmware to be updated can be loaded and updated first for the driver in the first partition, and then the firmware to be updated for the driver in the second partition can be loaded and updated after the first partition has been updated. In this way, while the driver in the first partition is being upgraded, the driver in the second partition ensures the normal operation of the target device and prevents downtime due to the update. By upgrading the driver in the first partition in ascending order and then further upgrading the driver in the second partition, a complete update of the target device can be achieved, which is beneficial for subsequent use.

[0142] As can be seen, in this example, updating the drivers for the first and second partitions in sequence helps ensure the security of the upgrade process, eliminates the need for lab downtime deployment, improves the scalability of lab equipment, and effectively enhances the convenience and efficiency of deploying new functional equipment in the lab.

[0143] This application can divide the server into functional units based on the above method examples. For example, each function can be divided into its own functional unit, or two or more functions can be integrated into one processing unit. The integrated unit can be implemented in hardware or as a software functional unit. It should be noted that the unit division in this application embodiment is illustrative and only represents one logical functional division; other division methods may be used in actual implementation.

[0144] For embodiments consistent with those shown above, please refer to... Figure 4 , Figure 4 This is a block diagram of the functional units of a security management system provided in an embodiment of this application. The security management system is the aforementioned server or a part thereof. The security management system 40 includes:

[0145] The acquisition unit 410 is used to acquire environmental data and equipment data corresponding to each of the laboratory devices. The environmental data and equipment data are data collected by the detection sensors corresponding to the laboratory devices. At least one of the laboratory devices is deployed in the laboratory. Each laboratory device is configured with at least one detection sensor. The detection sensor can be detachably installed and connected to the communication network of the server.

[0146] The comparison unit 420 is used to compare the environmental data and the device data with a preset parameter threshold to obtain a comparison result;

[0147] The identification unit 430 is used to input the device data and the environmental data into a pre-trained anomaly assessment model for identification and to obtain an anomaly score.

[0148] The determining unit 440 is used to determine the abnormal event and the corresponding early warning strategy based on the comparison result and the abnormality score.

[0149] The sending unit 450 is used to send warning information to the target warning device to instruct the target warning device to execute the warning strategy.

[0150] In one possible example, in determining the abnormal event and the corresponding early warning strategy based on the comparison result and the abnormal score, the determining unit is specifically configured to: determine at least one target parameter type that exceeds the preset parameter threshold based on the comparison result, wherein the parameter type is an environmental parameter or a device parameter; determine the abnormal event based on the at least one target parameter type; compare the magnitude relationship between the abnormal score and multiple preset score thresholds to obtain a score level; and determine the early warning strategy corresponding to the abnormal event based on the abnormal event and the score level.

[0151] In one possible example, the safety management system further includes a threshold generation unit, which is configured to: acquire historical operating data of the laboratory equipment, fault records of the laboratory equipment, and preset operating specification data before comparing the environmental data and the equipment data with preset parameter thresholds to obtain the comparison results; train an initial prediction model based on the historical operating data, fault records, and preset operating specification data of the laboratory equipment to obtain a threshold prediction model; acquire target environmental data and target equipment data of the laboratory equipment within a target time period; and input the target environmental data and target equipment data into the threshold prediction model for calculation to generate the preset parameter threshold.

[0152] In one possible example, regarding the step of inputting the target environment data and the target device data into the threshold prediction model for calculation to generate the preset parameter threshold, the threshold generation unit is further configured to: acquire current time data; determine whether the model update condition is met based on the environment data and the time data; when the model update condition is met, acquire the target weight based on the time data; update the weight parameters of the threshold prediction model to the target weight, and perform update training to obtain the updated threshold prediction model; input the target environment data and the target device data into the updated threshold prediction model for calculation to generate the preset parameter threshold.

[0153] In one possible example, in the process of inputting the device data and the environmental data into a pre-trained anomaly assessment model for identification and obtaining an anomaly score, the identification unit is specifically used to: input the device data and the environmental data into the anomaly assessment model, calculate the reconstruction error, and normalize the reconstruction error to obtain the anomaly score.

[0154] In one possible example, in acquiring the environmental data and equipment data corresponding to each of the laboratory devices, the acquisition unit is specifically configured to: acquire signal data collected by the target sensor and environmental data or equipment data; compare the signal data with a fingerprint database to determine the target laboratory device corresponding to the target sensor; and mark the environmental data or equipment data as data corresponding to the target laboratory device.

[0155] In one possible example, the detection sensor is detachably connected to the server's communication network via magnetic mounting; the security management system further includes a communication management unit, which is used to: when a target device is detected to be connected, obtain device information of the target device, wherein the target device is a detection sensor or an early warning device; obtain the driver program of the target device based on the device information; load the driver program; and drive the target device according to the driver program.

[0156] In one possible example, the security management system further includes an upgrade unit, which is configured to: after loading the driver and driving the target device according to the driver, when an upgrade instruction for the target device is detected, obtain a first firmware package carried by the upgrade instruction; obtain a second firmware package currently used by the target device; compare the first firmware package and the second firmware package to determine the firmware to be upgraded; and update the driver of the target device according to the firmware to be upgraded.

[0157] In one possible example, regarding the update of the target device's driver according to the firmware to be upgraded, the upgrade unit is specifically configured to: update the driver of the target device stored in a first partition according to the firmware to be upgraded, wherein the driver image of the target device is stored in the first partition and a second partition; after the driver stored in the first partition is updated, update the driver of the target device stored in the second partition according to the firmware to be upgraded.

[0158] It is understood that since the method embodiments and the device embodiments are different presentations of the same technical concept, the content of the method embodiment section in this application should be adapted to the device embodiment section in a synchronous manner, and will not be repeated here.

[0159] In the case of using integrated units, the functional unit composition block diagram of another security management system provided in this application embodiment is as follows: Figure 5 As shown. In Figure 5 In this document, the security management system 40 includes a processing module 520 and a communication module 510. The processing module 520 controls and manages the actions of the security management system 40, including, for example, the steps performed by the acquisition unit 410, comparison unit 420, identification unit 430, determination unit 440, and sending unit 450, and / or other processes for performing the techniques described herein. The communication module 510 supports interaction between the security management system 40 and other devices. Figure 5 As shown, the security management system 40 may also include a storage module 530, which is used to store the program code and data of the data acquisition security management system 40.

[0160] The processing module 520 can be a processor or controller, such as a central processing unit (CPU), a general-purpose processor, a digital signal processor (DSP), an ASIC, an FPGA, or other programmable logic devices, transistor logic devices, hardware components, or any combination thereof. It can implement or execute various exemplary logic blocks, modules, and circuits described in conjunction with the embodiments of this application. The processor can also be a combination that implements computing functions, such as a combination of one or more microprocessors, a combination of a DSP and a microprocessor, etc. The communication module 510 can be a transceiver, RF circuitry, or a communication interface, etc. The storage module 530 can be a memory.

[0161] All relevant content in each scenario involved in the above method embodiments can be referenced from the functional descriptions of the corresponding functional modules, and will not be repeated here. The above security management system 40 can execute the above... Figure 3 The safety warning method shown.

[0162] Figure 6 This is a structural block diagram of an electronic device provided in an embodiment of this application. Figure 6 As shown, electronic device 60 can be a server in the aforementioned security management system. This electronic device may include a processor 610, a memory 620, a communication interface 630, and one or more programs 621. The processor 610, memory 620, and communication interface 630 are interconnected and perform communication with each other. The one or more programs 621 are stored in the memory 620 and configured to be executed by the processor 610. The one or more programs 621 include instructions for performing any step in the above method embodiments.

[0163] The communication interface 630 is used to support communication between the electronic device 60 and other devices. The processor 610 may be, for example, a central processing unit (CPU), a general-purpose processor, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or other programmable logic devices, transistor logic devices, hardware components, or any combination thereof. It can implement or execute various exemplary logic blocks, units, and circuits described in conjunction with the embodiments of this application. The processor may also be a combination that implements computing functions, such as a combination of one or more microprocessors, a combination of a DSP and a microprocessor, etc.

[0164] The memory 620 can be volatile memory or non-volatile memory, or may include both. The non-volatile memory can be read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), or flash memory. The volatile memory can be random access memory (RAM), which is used as an external cache. By way of example, but not limitation, many forms of random access memory are available, such as static random access memory (SRAM), dynamic random access memory (DRAM), synchronous dynamic random access memory (SDRAM), double data rate synchronous dynamic random access memory (DDR SDRAM), enhanced synchronous dynamic random access memory (ESDRAM), synchronous linked dynamic random access memory (SynchLink DRAM, SLDRAM), and direct memory bus RAM (DRRAM).

[0165] In a specific implementation, the processor 610 is used to execute any step in the above method embodiments, and when performing data transmission such as sending, it can choose to call the communication interface 630 to complete the corresponding operation.

[0166] It should be noted that the above schematic diagram of the electronic device 60 is only an example, and the actual number of components included may be more or less, and no single limitation is made here.

[0167] This application can divide electronic devices into functional units based on the above method examples. For example, each function can be divided into its own functional unit, or two or more functions can be integrated into one processing unit. The integrated unit can be implemented in hardware or as a software functional unit. It should be noted that the unit division in this application embodiment is illustrative and only represents one logical functional division; other division methods may be used in actual implementation.

[0168] This application also provides a computer-readable storage medium storing a computer program for electronic data interchange, which causes a computer to perform some or all of the steps of any of the methods described in the above method embodiments, wherein the computer includes a server.

[0169] This application also provides a computer program product, which includes a non-transitory computer-readable storage medium storing a computer program operable to cause a computer to perform some or all of the steps of any of the security warning methods described in the above method embodiments. The computer program product can be a software installation package.

[0170] It should be noted that, for the sake of simplicity, the foregoing method embodiments are all described as a series of actions. However, those skilled in the art should understand that this application is not limited to the described order of actions, as some steps may be performed in other orders or simultaneously according to this application. Furthermore, those skilled in the art should also understand that the embodiments described in the specification are preferred embodiments, and the actions and modules involved are not necessarily essential to this application.

[0171] In the above embodiments, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions in other embodiments.

[0172] In the several embodiments provided in this application, it should be understood that the disclosed apparatus can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of the units described above is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between devices or units may be electrical or other forms.

[0173] The units described above as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0174] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0175] If the aforementioned integrated units are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage device (CMD). Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a memory and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned memory includes various media capable of storing program code, such as USB flash drives, read-only memory (ROM), random access memory (RAM), portable hard drives, magnetic disks, or optical disks.

[0176] Those skilled in the art will understand that all or part of the steps in the various methods of the above embodiments can be implemented by a program instructing related hardware. The program can be stored in a computer-readable storage device, which may include a flash drive, ROM, RAM, disk, or optical disk, etc.

[0177] The embodiments of this application have been described in detail above. Specific examples have been used to illustrate the principles and implementation methods of this application. The description of the above embodiments is only for the purpose of helping to understand the method and core ideas of this application. At the same time, for those skilled in the art, there will be changes in the specific implementation methods and application scope based on the ideas of this application. Therefore, the content of this specification should not be construed as a limitation of this application.

Claims

1. A safety early warning method, characterized in that, The security alert method, applied to a server in a security management platform, includes: The system acquires environmental data and equipment data corresponding to each laboratory device. The environmental data and equipment data are data collected by the detection sensors corresponding to the laboratory devices. At least one laboratory device is deployed in the laboratory, and each laboratory device is configured with at least one detection sensor. The detection sensor can be detachably installed and connected to the communication network of the server. The environmental data and the device data are compared with preset parameter thresholds to obtain the comparison results; The device data and the environmental data are input into a pre-trained anomaly assessment model for identification, and an anomaly score is obtained. Based on the comparison results and the anomaly score, determine the abnormal events and the corresponding early warning strategies for the abnormal events; Send warning information to the target warning device to instruct the target warning device to execute the warning strategy.

2. The method according to claim 1, characterized in that, The step of determining the abnormal event and the corresponding early warning strategy based on the comparison result and the abnormality score includes: Based on the comparison results, at least one target parameter type exceeding the preset parameter threshold is determined, wherein the parameter type is an environmental parameter or a device parameter; Based on the type of at least one target parameter, determine the abnormal event; By comparing the abnormal scores with multiple preset score thresholds, a score level is obtained; The early warning strategy corresponding to the abnormal event is determined based on the abnormal event and the rating level.

3. The method according to claim 1, characterized in that, Before comparing the environmental data and the device data with preset parameter thresholds to obtain the comparison result, the method further includes: Acquire historical operating data, fault records, and preset operating specification data of the laboratory equipment; An initial prediction model is trained based on the historical operating data of the laboratory equipment, the fault records of the laboratory equipment, and the preset operating specification data to obtain a threshold prediction model; Acquire target environmental data and target equipment data of the laboratory equipment within a target time period; The target environment data and the target device data are input into the threshold prediction model for calculation to generate the preset parameter threshold.

4. The method according to claim 3, characterized in that, The step of inputting the target environment data and the target device data into the threshold prediction model for calculation to generate the preset parameter threshold includes: Get the current time data; Determine whether the model update conditions are met based on the environmental data and the time data; When the model update conditions are met, the target weight is obtained based on the time data; The weight parameters of the threshold prediction model are updated to the target weights, and the updated training is performed to obtain the updated threshold prediction model. The target environment data and the target device data are input into the updated threshold prediction model for calculation to generate the preset parameter threshold.

5. The method according to claim 1, characterized in that, The step of inputting the device data and the environmental data into a pre-trained anomaly assessment model for identification and obtaining an anomaly score includes: The device data and the environmental data are input into the anomaly assessment model to calculate the reconstruction error. The reconstruction error is normalized to obtain the anomaly score.

6. The method according to claim 1, characterized in that, The acquisition of environmental and equipment data corresponding to each laboratory device includes: Acquire signal data collected by the target sensor, as well as environmental or equipment data; By comparing the signal data with the fingerprint database, the target laboratory equipment corresponding to the target sensor is determined; The environmental data or the equipment data is labeled as the data corresponding to the target laboratory equipment.

7. The method according to claim 1, characterized in that, The detection sensor is detachably connected to the server's communication network via magnetic mounting; the method further includes: When a target device is detected to be connected, the device information of the target device is obtained, wherein the target device is the detection sensor or the early warning device; Obtain the driver program for the target device based on the device information; Load the driver and drive the target device according to the driver.

8. The method according to claim 7, characterized in that, After loading the driver and driving the target device according to the driver, the method further includes: When an upgrade instruction for the target device is detected, the first firmware package carried by the upgrade instruction is obtained; Obtain the second firmware package currently used by the target device; By comparing the first firmware package and the second firmware package, the firmware to be upgraded is determined. Update the driver for the target device according to the firmware to be upgraded.

9. The method according to claim 8, characterized in that, The step of updating the driver for the target device according to the firmware to be upgraded includes: The driver for the target device stored in the first partition is updated according to the firmware to be upgraded, and the driver image of the target device is stored in the first partition and the second partition; After the driver stored in the first partition is updated, the driver stored in the second partition of the target device is updated according to the firmware to be upgraded.

10. A security management system, characterized in that, A server used in a security management platform, wherein the security management system includes: The acquisition unit is used to acquire environmental data and equipment data corresponding to each laboratory device. The environmental data and equipment data are data collected by the detection sensors corresponding to the laboratory devices. At least one laboratory device is deployed in the laboratory, and each laboratory device is configured with at least one detection sensor. The detection sensor can be detachably installed and connected to the communication network of the server. The comparison unit is used to compare the environmental data and the device data with a preset parameter threshold to obtain the comparison result; The identification unit is used to input the device data and the environmental data into a pre-trained anomaly assessment model for identification and to obtain an anomaly score. The determining unit is used to determine the abnormal event and the corresponding early warning strategy based on the comparison result and the abnormality score. The sending unit is used to send warning information to the target warning device to instruct the target warning device to execute the warning strategy.

Images