Laboratory intelligent management and control method based on Internet of Things and Internet of Things system

Abstract

The invention relates to a laboratory intelligent management and control method based on the Internet of Things and an Internet of Things system, the method is used for a management and control unit of Internet of Things equipment in a laboratory, and the method comprises the following steps: collecting multi-source parameter data SD in real time, transmitting the data to a central processing unit, and generating an abnormal parameter set AP; generating an abnormal degree grade based on the abnormal parameter set AP, and generating an alarm grade and an instruction set in combination with a preset alarm grading rule; and generating a resource allocation scheme based on the alarm level and the instruction set, generating adjusted environmental parameter data PD according to the resource allocation scheme, generating a real-time change trend, and dynamically monitoring the real-time change trend. According to the invention, intelligent monitoring, rapid anomaly detection, accurate resource allocation and adaptive regulation and control of the laboratory are realized, and safe and stable operation of the laboratory is effectively guaranteed.

Description

Technical Field

[0001] The present invention relates to the field of Internet of Things devices, and in particular to an intelligent management and control method for a laboratory based on the Internet of Things and an Internet of Things system. Background Art

[0002] As the core place for scientific research and technological innovation, the safety and operation efficiency of a laboratory are directly related to the reliability of scientific research results and the life and property safety of personnel. With the rapid development of Internet of Things technology, intelligent management and control of laboratories has become a key direction for improving safety management levels. By real-time monitoring of data such as environmental parameters and equipment status, Internet of Things technology provides new management possibilities for laboratories.

[0003] In the prior art, a single operating parameter of a laboratory is monitored and an alarm is given based on a set threshold. However, the monitoring of a single parameter and simple threshold alarm lack the comprehensive analysis ability for abnormal parameters of multiple types and degrees, and the alarm mechanism and the distribution of processing instructions lack intelligence and flexibility, making it difficult to fully meet the safety requirements in complex scenarios.

[0004] In summary, the existing solutions lack pertinence, and it is difficult to achieve rapid response and precise response in the face of emergencies in the laboratory, and potential risks still exist. Summary of the Invention

[0005] The technical problem to be solved by the present invention is: how to design an intelligent management and control solution for a laboratory based on the Internet of Things, which can effectively ensure the safe and stable operation of the laboratory.

[0006] The inventor has found through research that with the increase in the investment budget for laboratory fixed assets in China, the scale of laboratories is getting larger and larger. For example, provincial and municipal engineering centers with an area of more than 1,000 square meters. The Internet of Things (IoT) intelligent transformation of large laboratories is significantly different from most laboratories in terms of solutions. Here, most laboratories are referred to as "conventional laboratories". There are many problems in the process of migrating the IoT system applied to conventional laboratories to large laboratories, and the solution needs to be adjusted and optimized. The leaders of this adjustment and optimization, as laboratory managers, are concerned about the safety status of the laboratory and require technicians to design a targeted solution. Based on this, the present invention proposes an IoT-based intelligent management and control method for laboratories. The method is used for the control unit of IoT devices in the laboratory, and the method includes: S1, collecting multi-source parameter data SD in real time and transmitting it to the central processing unit to generate an abnormal parameter set AP; S2, generating an abnormal degree level based on the abnormal parameter set AP, and then generating an alarm level and an instruction set in combination with a preset alarm classification rule; S3, generating a resource allocation plan based on the alarm level and the instruction set, generating adjusted environmental parameter data PD according to the resource allocation plan, and then dynamically monitoring the real-time change trend to achieve intelligent management and control.

[0007] A further technical solution thereof is that the steps of S1 include: S101, collecting multi-source parameter data SD in real time through IoT devices and a sensor network, and transmitting it to the central processing unit to generate temperature, humidity, and time series data; S102, comparing each parameter with a preset threshold range according to the temperature, humidity, and time series data. If the parameter exceeds the threshold range, it is marked as an abnormal parameter, and an abnormal parameter set AP is generated.

[0008] A further technical solution thereof is that the steps of S2 include: S103, extracting the characteristic values of each abnormal parameter from the abnormal parameter set AP, classifying the abnormal types through the K-means clustering algorithm to generate an abnormal type distribution; S104, obtaining the historical data and current characteristic values of each type of abnormal parameter according to the abnormal type distribution, analyzing the mutual influence and change trend between parameters through the LSTM model to generate an abnormal degree level; S105, generating an alarm level and an instruction set based on the abnormal degree level according to the preset alarm classification rule.

[0009] A further technical solution thereof is that the steps of S3 include: S106, generating a resource allocation plan according to the alarm level and the instruction set; S107, obtaining the control interface of the target device according to the resource allocation plan, adjusting the device operation state through the processing instructions in the instruction set, and generating adjusted environmental parameter data PD; S108, extracting the real-time change trend from the adjusted environmental parameter data PD, comparing the change trend with the preset recovery target. If the recovery target is reached, continue to compare after waiting for the preset interval time. If the recovery target is not reached, update the instruction set through a feedback loop to generate an updated instruction set; S109, obtaining the latest status of the abnormal parameters and the system logs generated during the adjustment process according to the updated instruction set, and verifying the adjustment effect through a sliding window statistical method to achieve intelligent control.

[0010] A further technical solution thereof is that the laboratory includes a main experimental hall, the main experimental hall is provided with Internet of Things devices, and the control unit of the Internet of Things devices is connected to multiple sensors and corresponds to a sensor network; the steps of S101 include: S1011, collecting multi-source parameter data SD in real time through the Internet of Things devices and the sensor network to generate an original data stream; S1012, extracting time series features from the original data stream to construct multi-dimensional time series data including timestamps; S1013, performing format standardization processing on the multi-dimensional time series data including timestamps to generate standardized multi-dimensional time series data; S1014, clustering and grouping the standardized multi-dimensional time series data according to parameter types to generate grouped multi-dimensional time series data; S1015, verifying the integrity of the grouped multi-dimensional time series data to generate temperature and humidity time series data.

[0011] A further technical solution thereof is that the laboratory further includes a temperature rise experimental room, a humidity state experimental room, a fire prevention experimental room, and other experimental rooms. The main experimental hall and the rooms are physically isolated by walls; the control unit is respectively connected to multiple temperature sensors, multiple humidity sensors, and multiple gas concentration sensors; at least half of the temperature sensors are arranged in the temperature rise experimental room, at least half of the humidity sensors are arranged in the humidity state experimental room, and at least half of the gas concentration sensors are arranged in the fire prevention experimental room; the steps of S102 include: S1021, comparing the temperature and humidity time series data with the preset threshold range of the temperature parameter. If the temperature exceeds the threshold range, it is marked as a temperature abnormal parameter TP; S1022, continuing to compare the temperature and humidity time series data with the preset threshold range of the humidity parameter. If the humidity exceeds the threshold range, it is marked as a humidity abnormal parameter HP; S1023, continuing to compare the temperature and humidity time series data with the preset threshold range of the gas concentration parameter. If the gas concentration exceeds the threshold range, it is marked as a gas concentration abnormal parameter GP, and combining the temperature abnormal parameter TP and the humidity abnormal parameter HP to generate an abnormal parameter set AP.

[0012] Its further technical solution is that the steps of S108 include: S1081, extracting the real-time change trend from the adjusted environmental parameter data, determining the trend direction by time series analysis, and obtaining a change trend set; S1082, comparing the change trend set with a preset recovery target. If the recovery target is not reached, a primary adjustment signal is generated according to the deviation value to obtain adjustment signal data; S1083, updating the current instruction set according to the adjustment signal data, and generating a primary optimization instruction set by dividing the instruction priorities through a preset rule; S1084, adjusting the environmental parameters through the primary optimization instruction set to obtain real-time change data and obtaining a parameter change set; S1085, analyzing the change trend according to the parameter change set. If the recovery target is not achieved, a secondary adjustment signal is generated based on the latest parameter change, and the steps of updating the current instruction set according to the adjustment signal data are executed again until the recovery target is achieved; S1086, if the recovery target is achieved after analyzing the change trend, an updated instruction set is generated.

[0013] Its further technical solution is that the steps of S109 include: S1091, extracting the device priority sequence from the resource allocation plan, and generating an abnormal parameter tracking identifier in combination with real-time monitoring data; S1092, according to the abnormal parameter tracking identifier, obtaining the parameter change sequence during the adjustment process from the system log to generate a log time series data set; S1093, performing a sliding window mean filtering process on the log time series data set, calculating the parameter change rate and variance, and generating a stability evaluation index; S1094, if the stability evaluation index does not reach the preset threshold, generating a dynamic adjustment coefficient according to the parameter change rate and updating the execution frequency of the control instructions in the instruction set; S1095, re-obtaining the real-time monitoring data through the updated instruction set. If the parameter value of the real-time monitoring data continuously exceeds the safety range, a reconstruction request for the resource allocation plan is triggered, the device sorting is re-determined and a new resource allocation plan is generated; if the parameter value of the real-time monitoring data returns to the safety range, the process is terminated, thereby realizing intelligent control.

[0014] In a second aspect, the present application proposes an Internet of Things system. The Internet of Things devices of the Internet of Things system are arranged in a laboratory, and the control unit of the Internet of Things devices is used to implement the laboratory intelligent control method as described in the first aspect.

[0015] In summary, the implementation plan of this application for large laboratories includes a main experimental hall, as well as a temperature rise experimental room, a humidity state experimental room, a fire prevention experimental room, and other experimental rooms. The main experimental hall and the rooms are physically isolated by walls. Laboratory managers cannot, through patrols, understand the safety status of the laboratory in real time. The effect of setting up full-time patrol posts is not ideal enough, which is a "long-term hidden danger" for laboratory managers. The time investment in laboratory safety also takes up a large amount of time of laboratory managers, affecting the long-term development of large laboratories. Based on this, the inventor proposes a laboratory intelligent control method and an Internet of Things system. The method collects multi-source parameter data SD in real time through Internet of Things devices and generates temperature, humidity, and time series data. For abnormal parameters, eigenvalue extraction and classification using the K-means clustering algorithm are performed, and the mutual influence between parameters is analyzed in combination with the LSTM model to generate an abnormal degree level. According to the abnormal degree, corresponding alarm levels and processing instructions are triggered, and the device operation status is adjusted according to the processing instructions to achieve intelligent control. The present invention adjusts the device operation status through a control interface, monitors the change trend of environmental parameters in real time, adopts a feedback loop to update the instruction set, and finally verifies the adjustment effect through a sliding window statistical method. This application realizes intelligent monitoring, rapid anomaly detection, precise resource allocation, and adaptive regulation of the laboratory, effectively ensuring the safe and stable operation of the laboratory. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] The accompanying drawings herein are incorporated into the specification and form a part of the specification, showing embodiments consistent with the present invention and, together with the specification, are used to explain the principles of the present invention.

[0017] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or in the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, for those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0018] Figure 1 It is a flowchart of the laboratory intelligent control method based on the Internet of Things provided by the present invention.

[0019] Figure 2 It is another flowchart of the laboratory intelligent control method based on the Internet of Things provided by the present invention.

[0020] Figure 3 It is the alarm classification rule of the laboratory intelligent control method based on the Internet of Things provided by the present invention.

[0021] Figure 4 It is a framework schematic diagram of the electronic device provided by the embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0022] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0023] It should be understood that when used in this specification and the appended claims, the terms "comprising" and "including" indicate the presence of the described features, wholes, steps, operations, elements, and / or components, but do not exclude the presence or addition of one or other features, wholes, steps, operations, elements, components, and / or their combinations.

[0024] It should also be understood that the terms used in the specification of the present invention are merely for the purpose of describing specific embodiments and are not intended to limit the present invention. As used in the specification of the present invention and the appended claims, unless the context clearly indicates otherwise, the singular forms "a", "an", and "the" are intended to include the plural forms.

[0025] It should be further understood that the term " / and" used in the specification of the present invention and the appended claims refers to any one or any combination of the related listed items and all possible combinations, and includes these combinations.

[0026] As used in this specification and the appended claims, the term "if" may be interpreted as "when...", "once", "in response to determining", or "in response to detecting" according to the context. Similarly, the phrase "if determined" or "if detecting [the described condition or event]" may be interpreted as meaning "once determined", "in response to determining", "once detecting [the described condition or event]", or "in response to detecting [the described condition or event]" according to the context.

[0027] In this specification and the appended claims, there may be multiple expression forms for the same technical feature or professional term, such as using upper-level generalization, lower-level limitation, or synonymous substitution and other different expression forms; those skilled in the art can clearly understand the substantially same technical meaning pointed to by different expression forms based on their professional knowledge and in combination with the overall content of the specification and the accompanying drawings; the differences in different expression forms are only reflected in the diversity of the literal level, do not constitute a substantial modification or limitation to the technical solution, and do not affect the certainty of the protection scope of the patent claim and the full disclosure of the technical content of the specification.

[0028] Embodiment 1 Please refer to Figures 1 to 3 as shown, where the Figures 1 to 2An embodiment of the present invention provides an intelligent control method for a laboratory based on the Internet of Things. The method is used for the control unit of Internet of Things devices in a laboratory, and the method includes: S1, collecting multi-source parameter data SD in real time and transmitting it to a central processing unit to generate an abnormal parameter set AP; S2, generating an abnormal degree level based on the abnormal parameter set AP, and then generating an alarm level and an instruction set in combination with a preset alarm classification rule; S3, generating a resource allocation plan based on the alarm level and the instruction set, generating adjusted environmental parameter data PD according to the resource allocation plan, and then dynamically monitoring the real-time change trend to achieve intelligent control. In the above solution, the realization of intelligent control includes confirming whether the state of the laboratory has returned to the safe range.

[0029] In one embodiment, the steps of S1 include: S101, collecting multi-source parameter data SD in real time through Internet of Things devices and a sensor network, and transmitting it to a central processing unit to generate temperature, humidity, and time series data; S102, comparing each parameter with a preset threshold range according to the temperature, humidity, and time series data. If the parameter exceeds the threshold range, it is marked as an abnormal parameter to generate an abnormal parameter set AP.

[0030] In one embodiment, the steps of S2 include: S103, extracting the characteristic values of each abnormal parameter from the abnormal parameter set AP, classifying the abnormal types through the K-means clustering algorithm to generate an abnormal type distribution; S104, obtaining the historical data and current characteristic values of each type of abnormal parameter according to the abnormal type distribution, analyzing the mutual influence and change trend between parameters through the LSTM model to generate an abnormal degree level; S105, according to the abnormal degree level, based on the preset alarm classification rule, generating an alarm level and an instruction set. Among them, S105 may include, according to the abnormal degree level, based on the preset alarm classification rule, if the abnormal degree is low, assigning a low-level alarm and recording a log, if it is medium, assigning a medium-level alarm and generating a preliminary instruction, if it is high, triggering a high-level alarm and generating a detailed processing instruction, generating an alarm level and an instruction set. Among them, for the preset alarm classification rule, please refer to Figure 3 as shown.

[0031] In one embodiment, S103 includes first normalizing the raw data in the abnormal parameter set, mapping parameter values of different magnitudes to a unified numerical interval by eliminating the dimension difference, so that the temperature, humidity, and gas concentration are comparable; subsequently classifying the abnormal types through the K-means clustering algorithm (i.e., unsupervised classification, determining the cluster centers of abnormal categories with similar change patterns through iterative optimization), and each abnormal category cluster center corresponds to a prototype of an abnormal pattern, such as a composite abnormality like "rapid temperature rise accompanied by gas leakage" or "slow humidity change superimposed with device offline"; the system divides the abnormal types according to the feature similarity between the samples and the abnormal category cluster centers, statistically calculates the distribution density of each type in the space-time dimension, and finally generates the abnormal type distribution.

[0032] In one embodiment, the steps of S3 include: S106, generating a resource allocation plan according to the alarm level and the instruction set; S107, obtaining the control interface of the target device according to the resource allocation plan, adjusting the device operation state through the processing instructions in the instruction set, and generating the adjusted environmental parameter data PD; S108, extracting the real-time change trend from the adjusted environmental parameter data PD, comparing the change trend with the preset recovery target, if the recovery target is reached, continue to compare after waiting for the preset interval time, if the recovery target is not reached, update the instruction set through a feedback loop to generate an updated instruction set; S109, obtaining the latest status of the abnormal parameters and the system log generated during the adjustment process according to the updated instruction set, and verifying the adjustment effect through the sliding window statistical method to achieve intelligent control.

[0033] In one embodiment, the laboratory includes a main experimental hall, and the main experimental hall is provided with Internet of Things devices, and the management and control unit of the Internet of Things devices is connected to multiple sensors and corresponds to a sensor network; the steps of S101 include: S1011, collecting multi-source parameter data SD in real time through the Internet of Things devices and the sensor network to generate a raw data stream; S1012, extracting time series features from the raw data stream to construct multi-dimensional time series data including timestamps; S1013, performing format normalization processing on the multi-dimensional time series data including timestamps to generate normalized multi-dimensional time series data; S1014, clustering and grouping the normalized multi-dimensional time series data according to the parameter type to generate grouped multi-dimensional time series data; S1015, verifying the integrity of the grouped multi-dimensional time series data to generate temperature, humidity, and gas time series data.

[0034] In the above solution, first, regarding the problem of timestamp alignment, according to the differences in data acquisition frequencies of different sensors, an interpolation method is used to fill in the parameter values at missing time points. The interpolation method includes, for high-frequency acquisition parameters (such as temperature data per second), retaining the original timestamps and supplementing the interpolation data at adjacent time points; for low-frequency parameters (such as humidity data per minute), virtual data points with second-level timestamps are generated based on the linear interpolation algorithm to ensure that all parameters have continuous observed values on a unified time basis. Subsequently, dimensional unification processing is performed on the parameter values. The original measurement values of each sensor are mapped to a dimensionless relative interval through a standardization formula to eliminate the influence of different physical quantity units on data analysis. The dimensional unification processing is to convert data with different units, such as temperature or humidity, into a unified numerical range to eliminate the influence of unit differences such as degrees Celsius and percentages on data analysis. During the standardization process, abnormal timestamp data caused by communication delays (such as future timestamps earlier than the current system time or error data significantly deviating from the device clock synchronization range) are detected and removed synchronously. After completing spatio-temporal alignment and dimensional unification, the system further verifies the continuity of the time series of each parameter. For long data missing segments caused by equipment failures, the statistical characteristics of historical data of similar sensors are used for filling, and finally, standardized multi-dimensional time series data are generated.

[0035] In one embodiment, the laboratory further includes a temperature rise test chamber, a dampness test chamber, a fire protection test chamber, and other test chambers. The main test hall and the chambers are physically isolated by walls and are provided with access doors; the laboratory is a large-scale laboratory, and the building area can be more than one thousand square meters. Most of the experimental equipment is located in the main test hall, and the number of other test chambers can be between two and twenty, depending on the function of the laboratory; the control unit is respectively connected to a plurality of temperature sensors, a plurality of humidity sensors, and a plurality of gas concentration sensors; at least half of the temperature sensors are located in the temperature rise test chamber, at least half of the humidity sensors are located in the dampness test chamber, and at least half of the gas concentration sensors are located in the fire protection test chamber; the steps of S102 include: S1021, comparing the temperature and humidity time series data with the preset threshold range of the temperature parameter. If the temperature exceeds the threshold range, it is marked as the temperature abnormal parameter TP; S1022, continuing to compare the temperature and humidity time series data with the preset threshold range of the humidity parameter. If the humidity exceeds the threshold range, it is marked as the humidity abnormal parameter HP; S1023, continuing to compare the temperature and humidity time series data with the preset threshold range of the gas concentration parameter. If the gas concentration exceeds the threshold range, it is marked as the gas concentration abnormal parameter GP, and combining the temperature abnormal parameter TP and the humidity abnormal parameter HP, an abnormal parameter set AP is generated. Optionally, the gas concentration abnormal parameter GP, the temperature abnormal parameter TP, and the humidity abnormal parameter HP are superimposed to generate the abnormal parameter set AP; the abnormal parameter set AP generated after the superposition, that is, the gas concentration abnormal parameter GP, the temperature abnormal parameter TP, and the humidity abnormal parameter HP are simply combined to generate the abnormal parameter set AP, that is, the abnormal parameter set AP is composed of the above three parts.

[0036] In one embodiment, after the step S103, the steps of S104 include: S1041, extracting the feature vectors of each clustering category from the abnormal type distribution to generate a clustering identification set; S1042, according to the clustering identification set, matching the full-cycle time series data of the same type of abnormal events from the historical database to generate a historical data set; S1043, extracting the real-time feature values of each abnormal parameter from the current multi-dimensional time series data to generate a current feature set; S1044, performing timestamp alignment processing on the historical data set and the current feature set, constructing an LSTM input sequence, and generating a standardized time series input set; S1045, inputting the standardized time series input set into a pre-trained LSTM model to output an inter-parameter correlation weight matrix and a trend prediction value; S1046, calculating a parameter influence factor according to the correlation weight matrix, and combining the deviation value between the trend prediction value and the real-time feature value to generate an abnormal degree level. Among them, in the step of inputting the standardized time series input set into the pre-trained LSTM model, the pre-training specifically includes: in the first step, extracting the complete time series of the full-cycle data of the same type of abnormal events (temperature / humidity / gas concentration) from the historical database, after performing data cleaning on it, based on the clustering identification, matching similar historical events (such as temperature anomalies, etc.), and screening out the same type of data set; for the same type of data set, performing timestamp alignment, dividing it into fixed-duration windows, and standardizing the temperature, humidity, and gas concentration values within the fixed-duration windows to obtain standardized data; in the second step, defining a supervision task: based on the standardized data, setting the task that the model needs to predict future parameter values and output parameter correlation weights; in the third step, constructing a dual-branch LSTM: for the supervision task, designing a dual-branch LSTM that inputs standardized data, outputs prediction values, and correlation weights; in the fourth step, training the dual-branch LSTM: training the dual-branch LSTM with the standardized data, jointly optimizing the prediction and correlation tasks until the model converges, and thus completing the pre-training.

[0037] In the above solution, similar historical events (such as temperature anomalies, etc.) are matched based on clustering identifiers, and similar data sets are filtered out. That is, the complete time series of each historical anomaly event will be extracted as a high-dimensional vector containing fluctuation patterns, parameter correlations, and spatio-temporal distribution characteristics, and dynamically matched with the feature vector of the current clustering identifier. The matching rule uses a composite condition judgment: First, the basic feature similarity is required (for example, the deviation of the rising slope of the temperature anomaly from the historical case does not exceed twice its standard deviation). Second, the parameter correlation pattern is verified (for example, whether the humidity anomaly is accompanied by a change in the operating state of a specific device). Finally, the morphological similarity of the time evolution trend is checked, where the sequence morphological distance calculated by the dynamic time warping algorithm is less than a preset threshold. For historical data that meets all conditions, the system classifies it into a similar data set. Among them, the pre-trained model adopts a three-layer LSTM network structure, and the number of neurons in each layer is dynamically configured according to the anomaly type (for example, a dense neuron is set in the temperature rise anomaly layer to capture mutation features). The initial learning rate is set to an adaptive adjustment mode. When the decline amplitude of the training loss for three consecutive cycles is less than the threshold, the learning rate is automatically reduced. At the same time, an attention mechanism is introduced in the parameter correlation weight branch to strengthen the modeling of the mutual influence of key parameters. This screening mechanism ensures that the model training data highly conforms to the current anomaly pattern, enabling the prediction result to accurately reflect the actual environmental evolution law.

[0038] In one embodiment, the steps of S106 include: According to the alarm level and instruction set, for the abnormal parameters corresponding to the high-level alarm, obtain the device status and availability in the real-time monitored resource pool, and allocate device resources through a priority-weighted greedy algorithm to generate a resource allocation plan. During the process of generating the resource allocation plan, the system dynamically allocates device resources through a priority-weighted greedy algorithm. This method is based on the alarm levels of the gas concentration abnormal parameter GP, temperature abnormal parameter TP, and humidity abnormal parameter HP in the abnormal parameter set AP, and calculates a comprehensive priority score for each abnormal task. The priority score consists of three parts: the basic weight of the abnormal parameter type (GP is given the highest weight, TP is the second, and HP is the lowest), the dynamic correction coefficient of the current abnormal degree (for example, the coefficient of the high-level alarm is 1.5 times that of the medium-level alarm), and the spatial matching degree of the device resources (preferably allocate devices directly associated with the abnormal area). The greedy algorithm is based on making the optimal choice (the most favorable choice) in each step of the selection, traversing all pending abnormal tasks, generating a task queue in descending order of the priority score, and sequentially allocating the currently available device resources to each task. When multiple tasks compete for the same device, the task with the highest priority score is selected to exclusively occupy the device resources, and the remaining tasks are automatically downgraded to the standby device pool or enter the waiting queue. In this way, the "resource allocation plan" can ensure that each device only executes one core task at the same time.

[0039] Step S108: Extract the real-time change trend from the adjusted environmental parameter data, compare the change trend with the preset recovery target. If the recovery target is not reached, update the instruction set through a feedback loop to generate an updated instruction set.

[0040] In one embodiment, the steps of S108 include: S1081, extract the real-time change trend from the adjusted environmental parameter data, use time series analysis to determine the trend direction, and obtain a set of change trends; S1082, compare the set of change trends with the preset recovery target. If the recovery target is not reached, generate a primary adjustment signal according to the deviation value to obtain adjustment signal data; S1083, update the current instruction set according to the adjustment signal data, and generate a primary optimized instruction set by dividing the instruction priorities through a preset rule; S1084, adjust the environmental parameters through the primary optimized instruction set to obtain real-time change data, and obtain a set of parameter changes; S1085, analyze the change trend according to the set of parameter changes. If the recovery target is not achieved, generate a secondary adjustment signal based on the latest parameter changes, and execute again according to the secondary adjustment signal. Update the steps of the current instruction set according to the adjustment signal data until the recovery target is achieved; S1086, if the recovery target is achieved after analyzing the change trend, generate an updated instruction set. The following is an example to illustrate the relationship between the adjustment signal and the instruction set: First, monitor the environmental parameters (such as temperature) in real time to generate change trend data, compare it with the preset recovery target (such as the temperature is stable at 25 degrees). If there is a deviation, generate an adjustment signal (such as lowering the temperature); the adjustment signal includes specific control requirements (such as the temperature adjustment range, the rate of temperature adjustment range), which are mapped to the instruction set (such as turning on the air conditioner, starting ventilation, etc., which can be multiple), and sorted by priority, which is equivalent to clarifying which instruction is executed first.

[0041] In the above solution, the primary optimized instruction set is generated by dividing the instruction priorities through a preset rule, that is, a decision-making process for ranking the importance of the environmental control instructions to be executed according to a preset multi-dimensional judgment criterion. For example, the system will comprehensively evaluate the safety threat level, the breadth of the influence range, the requirement for the recovery time limit, and the control cost of various abnormal parameters in the current laboratory environment as key factors; when both the gas concentration exceeds the standard and the local temperature is abnormal, since the safety threat level dimension in the preset rule is given the highest weight, the system will give priority to dealing with the gas leakage problem with an explosion risk and generate a primary optimized instruction set corresponding to the gas leakage problem.

[0042] In one embodiment, the steps of S109 include: S1091, extracting the device priority sequence from the resource allocation scheme and generating an abnormal parameter tracking identifier in combination with real-time monitoring data; S1092, obtaining the parameter change sequence during the adjustment process from the system log according to the abnormal parameter tracking identifier and generating a log time series data set; S1093, performing sliding window mean filtering on the log time series data set, calculating the parameter change rate and variance, and generating a stability evaluation index; S1094, if the stability evaluation index does not reach the preset threshold, generating a dynamic adjustment coefficient according to the parameter change rate and updating the execution frequency of the control instructions in the instruction set; S1095, re-obtaining the real-time monitoring data through the updated instruction set. If the parameter value of the real-time monitoring data continuously exceeds the safety range, a reconstruction request for the resource allocation scheme is triggered, the device sorting is re-determined, and a new resource allocation scheme is generated; if the parameter value of the real-time monitoring data returns to the safety range, the process is terminated, thereby realizing intelligent control. Among them, re-determining the device sorting and regenerating the resource allocation scheme means returning to the steps of S106 to regenerate a resource allocation scheme to replace the original preliminary resource allocation scheme. Since the objective environment has changed, the regenerated resource allocation scheme is different from the original version. Among them, the generation of the stability evaluation includes a change rate threshold and a variance threshold. The former focuses on the change slope within the interval, and the latter focuses on the degree of fluctuation; the change rate threshold is used to measure the smoothness of the parameter recovery process, and the maximum allowable change gradient is set for different regulation targets. When it is detected that the parameter changes by more than the preset standard within a unit time, the system determines that the current adjustment strategy fails to effectively control the environmental state; the variance threshold reflects the allowable range of parameter fluctuations, which is set based on the characteristics of the laboratory functional areas. For example, the fluctuation tolerance is strict in areas sensitive to gas concentration, while a relatively loose fluctuation range is allowed for temperature monitoring.

[0043] In summary, the implementation solution of this application for large laboratories includes a main experimental hall, as well as a temperature rise experimental room, a damp state experimental room, a fire prevention experimental room, and other experimental rooms. The main experimental hall and the rooms are physically isolated by walls. Laboratory managers cannot, through patrols, obtain a real-time understanding of the safety status of the laboratory. The effect of setting up full-time patrol posts is also not ideal, which is a "long-term hidden danger" for laboratory managers and also takes up a large amount of the time of laboratory managers in terms of the time investment in laboratory safety, affecting the long-term development of large laboratories. Based on this, the inventor proposes a laboratory intelligent management and control method and an Internet of Things system. The method collects multi-source parameter data SD in real time through Internet of Things devices and generates temperature, humidity, and time series data. For abnormal parameters, eigenvalue extraction and classification using the K-means clustering algorithm are performed, and the mutual influence between parameters is analyzed in combination with the LSTM model to generate an abnormal degree level. Corresponding alarm levels and processing instructions are triggered according to the abnormal degree, and the device operating state is adjusted according to the processing instructions to achieve intelligent management and control. The present invention adjusts the device operating state through a control interface, monitors the change trend of environmental parameters in real time, updates the instruction set using a feedback loop, and finally verifies the adjustment effect through a sliding window statistical method. This application realizes intelligent monitoring, rapid anomaly detection, precise resource allocation, and adaptive regulation of the laboratory, effectively ensuring the safe and stable operation of the laboratory.

[0044] Embodiment 2 Please refer to Figure 4 , Figure 4 which is a block diagram of an electronic device provided by the present invention. The electronic device can be a terminal or a server. Among them, the terminal can be an electronic device with a communication function such as a smart phone, a tablet computer, a notebook computer, a desktop computer, a personal digital assistant, and a wearable device. It includes a processor 111, a communication interface 112, a memory 113, and a communication bus 114. Among them, the processor 111, the communication interface 112, and the memory 113 complete mutual communication through the communication bus 114.

[0045] The memory 113 is used to store computer programs.

[0046] In an embodiment of the present invention, when the processor 111 executes the program stored on the memory 113, it implements the method provided by any one of the foregoing method embodiments.

[0047] It should be understood that in the embodiments of the present application, the processor 111 may be a central processing unit (CPU), and the processor 502 may also be other general-purpose processors, digital signal processors (DSPs), application specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. Among them, the general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc.

[0048] Those of ordinary skill in the art can understand that all or part of the processes in the methods of the above embodiments can be completed by instructing relevant hardware through a computer program. The computer program can be stored in a storage medium, and the storage medium is a computer-readable storage medium. The computer program is executed by at least one processor in the computer system to implement the process steps of the embodiments of the above methods.

[0049] Those of ordinary skill in the art can realize that the units and algorithm steps of the examples described in combination with the embodiments disclosed herein can be implemented by electronic hardware, computer software, or a combination of the two. To clearly illustrate the interchangeability of hardware and software, the composition and steps of the examples have been generally described according to functions in the above description. Whether these functions are executed in a hardware or software manner depends on the specific application and design constraints of the technical solution. Professional technicians can use different methods to implement the described functions, but such implementation should not be considered to exceed the scope of the present invention.

[0050] In several embodiments provided by the present invention, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For example, the division of each unit is only a logical function division, and there may be other division methods in actual implementation. For example, a unit or component can be combined or integrated into another system, or some features can be ignored or not executed.

[0051] The steps in the method embodiments of the present invention can be adjusted, combined, and deleted according to actual needs. The units in the device embodiments of the present invention can be combined, divided, and deleted according to actual needs. In addition, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically alone, or two or more units can be integrated into one unit.

[0052] When the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a storage medium. Based on such an understanding, the technical solution of the present invention, 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 storage medium and includes several instructions for causing a computer device (which may be a personal computer, a terminal, or a network device, etc.) to execute all or part of the steps of the methods described in various embodiments of the present invention.

[0053] In the above embodiments, the descriptions of the various embodiments have their own emphases. For the parts not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0054] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, provided that these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalent technologies, the present invention is also intended to include these changes and modifications.

[0055] As described above, the above are only the specific embodiments of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present invention can easily think of various equivalent modifications or substitutions, and these modifications or substitutions should all be covered within the protection scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.

Claims

1. A laboratory intelligent management and control method based on the Internet of Things, characterized in that: The method is used for a control unit of an Internet of Things device in a laboratory, and the method comprises: S1, collect multi-source parameter data SD in real time and transmit it to the central processing unit to generate an abnormal parameter set AP; S2, generating an abnormality level based on the abnormal parameter set AP, and then generating an alarm level and instruction set in combination with a preset alarm classification rule; S3, generates a resource allocation plan based on the alarm level and instruction set, generates adjusted environmental parameter data PD according to the resource allocation plan, and then dynamically monitors the real-time change trend to achieve intelligent management and control.

2. The method for intelligent laboratory management and control based on the Internet of Things according to claim 1, characterized in that: The steps of S1 include: S101, collect multi-source parameter data SD in real time through IoT devices and sensor networks, and transmit them to the central processing unit to generate temperature and humidity time series data; S102: Compare the preset threshold range of each parameter according to the temperature and humidity time series data. If the parameter exceeds the threshold range, it is marked as an abnormal parameter, and an abnormal parameter set AP is generated.

3. The method for intelligent laboratory management and control based on the Internet of Things according to claim 2, characterized in that: The steps of S2 include: S103, extracting the characteristic value of each abnormal parameter from the abnormal parameter set AP, classifying the abnormal type by using the K-means clustering algorithm, and generating an abnormal type distribution; S104, according to the abnormal type distribution, obtain the historical data and current characteristic value of each type of abnormal parameter, analyze the mutual influence and change trend between the parameters through the LSTM model, and generate the abnormal degree level; S105, according to the abnormality level, according to the preset alarm classification rules, and based on the alarm classification rules, generate an alarm level and an instruction set.

4. The method for intelligent laboratory management and control based on the Internet of Things according to claim 3 is characterized in that: The steps of S3 include: S106, generating a resource allocation plan according to the alarm level and the instruction set; S107, according to the resource allocation plan, obtain the control interface of the target device, adjust the device operation state through the processing instructions in the instruction set, and generate adjusted environmental parameter data PD; S108, extracting the real-time change trend from the adjusted environmental parameter data PD, comparing the change trend with the preset recovery target, and if the recovery target is reached, continuing the comparison after waiting for a preset interval time, and if the recovery target is not reached, updating the instruction set through a feedback loop to generate an updated instruction set; S109, according to the updated instruction set, obtain the latest status of the abnormal parameters and the system log generated during the adjustment process, verify the adjustment effect through the sliding window statistical method, and realize intelligent management and control.

5. The method for intelligent laboratory management and control based on the Internet of Things according to claim 4 is characterized in that: The laboratory includes a main laboratory hall, the main laboratory hall is equipped with an Internet of Things device, and the control unit of the Internet of Things device is connected to multiple sensors and corresponds to a sensor network; The step of S101 includes: Collect multi-source parameter data SD in real time through IoT devices and sensor networks to generate raw data streams; Extract time series features from the original data stream and construct multidimensional time series data including timestamps; Perform format standardization on multidimensional time series data containing timestamps to generate standardized multidimensional time series data; Cluster and group the standardized multidimensional time series data according to parameter types to generate grouped multidimensional time series data; Verify the integrity of the grouped multidimensional time series data and generate temperature and humidity time series data.

6. The method for intelligent laboratory management and control based on the Internet of Things according to claim 5 is characterized in that: The laboratory also includes a temperature rise test room, a moisture test room, a fire test room, and other test rooms. The main test room and the rooms are physically isolated by walls; the control unit is respectively connected to multiple temperature sensors, multiple humidity sensors, and multiple gas concentration sensors; at least half of the temperature sensors are located in the temperature rise test room, at least half of the humidity sensors are located in the moisture test room, and at least half of the gas concentration sensors are located in the fire test room; the step of S102 includes: According to the temperature and humidity time series data, it is compared with the preset threshold range of the temperature parameter. If the temperature exceeds the threshold range, it is marked as the temperature anomaly parameter TP; Continue to compare the temperature and humidity time series data with the preset threshold range of the humidity parameter. If the humidity exceeds the threshold range, it is marked as a humidity anomaly parameter HP; Continue to compare the temperature and humidity time series data with the preset threshold range of the gas concentration parameter. If the gas concentration exceeds the threshold range, it is marked as a gas concentration anomaly parameter GP. Combined with the temperature anomaly parameter TP and the humidity anomaly parameter HP, an anomaly parameter set AP is generated.

7. The method for intelligent laboratory management and control based on the Internet of Things according to claim 6, characterized in that: The steps of S108 include: Extract the real-time change trend from the adjusted environmental parameter data, determine the trend direction using time series analysis, and obtain a set of change trends; Compare the change trend set with the preset recovery target. If the recovery target is not reached, generate a primary adjustment signal according to the deviation value to obtain adjustment signal data; Update the current instruction set according to the adjustment signal data, and generate a primary optimized instruction set by dividing the instruction priorities according to preset rules; Adjust environmental parameters through primary optimization instruction sets, obtain real-time change data, and obtain parameter change sets; Analyze the change trend according to the parameter change set. If the recovery target is not achieved, generate a secondary adjustment signal based on the latest parameter change, execute again according to the secondary adjustment signal, and update the current instruction set according to the adjustment signal data until the recovery target is achieved; If the recovery goal is achieved after analyzing the change trend, an updated instruction set is generated.

8. The method for intelligent laboratory management and control based on the Internet of Things according to claim 7, characterized in that: The step of S109 includes: Extract the equipment priority sequence from the resource allocation plan and generate abnormal parameter tracking identification based on real-time monitoring data; According to the abnormal parameter tracking mark, the parameter change sequence during the adjustment process is obtained from the system log to generate a log time series data set; Perform sliding window mean filtering on the log time series data set, calculate the parameter change rate and variance, and generate stability evaluation indicators; If the stability evaluation index does not reach the preset threshold, a dynamic adjustment coefficient is generated according to the parameter change rate to update the execution frequency of the control instructions in the instruction set; Real-time monitoring data is reacquired through the updated instruction set. If the parameter value of the real-time monitoring data continues to exceed the safe range, a reconstruction request for the resource allocation plan is triggered, the equipment sorting is redetermined, and the resource allocation plan is regenerated; if the parameter value of the real-time monitoring data returns to the safe range, the process is terminated, thereby realizing intelligent management and control.

9. An Internet of Things system, characterized in that: The Internet of Things device of the Internet of Things system is arranged in a laboratory, and the control unit of the Internet of Things device is used to implement the laboratory intelligent control method according to any one of claims 1 to 8.

Images