A method and system for adjusting the internal environment of a mobile laboratory cabin

Abstract

This invention discloses a method and system for adjusting the internal environment of a mobile laboratory cabin, specifically relating to the field of environmental control technology. It addresses the problem that existing environmental adjustment methods, which use fixed control targets, cannot adapt to the differentiated environmental requirements of mobile laboratories in multi-task scenarios. The method involves acquiring the current task identifier of the mobile laboratory, identifying the corresponding experimental task execution stage, and dynamically generating a matrix of collaborative constraint relationships between parameters based on a fuzzy inference rule base. It also involves collecting multi-dimensional environmental parameter measurements and combining them with pre-stored expected environmental parameter values ​​to generate a control demand vector. Based on the matrix and vector, it calculates the control conflict risk entropy value and performs risk assessment. When the risk tolerance is exceeded, it generates a roundabout control strategy as an optimization guidance parameter. Finally, based on the optimization guidance parameter, it coordinates and generates environmental control commands and sends them to the actuators. This achieves dynamic adaptation between the environmental parameter control strategy and the experimental task, improving the accuracy and reliability of multi-parameter collaborative control.

Description

Technical Field

[0001] This invention relates to the field of environmental control technology, and in particular to a method and system for adjusting the internal environment of a mobile laboratory cabin. Background Technology

[0002] The interior of a mobile laboratory cabin needs to maintain stable environmental parameters such as temperature, humidity, air pressure, and cleanliness to ensure the accuracy of experimental results and operational safety. Existing environmental adjustment methods typically employ pre-set fixed control targets, using sensor monitoring and feedback control mechanisms to adjust these parameters independently or in a simple, coordinated manner. This approach treats the target values ​​and tolerance ranges of environmental parameters as static settings, failing to consider the varying environmental requirements of mobile laboratories performing different scientific tasks. In practical applications, the fixed control mode struggles to adapt to changes in requirements caused by task switching, resulting in a lack of flexibility in the allocation of control resources.

[0003] The main problem with existing technologies is that environmental adjustment systems cannot adaptively reconfigure control objectives and parameter priorities according to dynamic task requirements. When the type of laboratory task changes, fixed control strategies struggle to adjust the control emphasis and tolerance range of different environmental parameters in a timely manner, easily leading to reduced energy utilization or insufficient control of key parameters, thus affecting the reliability and adaptability of the experimental process. Summary of the Invention

[0004] This invention addresses the technical problems existing in the prior art by providing a method and system for adjusting the internal environment of a mobile laboratory cabin.

[0005] The technical solution of the present invention to solve the above-mentioned technical problems is as follows:

[0006] This invention provides the following technical solution:

[0007] A method for adjusting the internal environment of a mobile laboratory cabin includes:

[0008] S1: Get the current task identifier of the mobile lab;

[0009] S2: Identify the experimental task execution stage corresponding to the current task identifier, and dynamically generate a parameter co-constraint relationship matrix that matches the experimental task execution stage based on the fuzzy inference rule base;

[0010] S3: Collect multi-dimensional environmental parameter measurements inside the mobile laboratory cabin, and calculate and generate a control demand vector based on the multi-dimensional environmental parameter measurements and the pre-stored expected environmental parameter values;

[0011] S4: Calculate the entropy value of regulation conflict risk based on the parameter coordination constraint relationship matrix and regulation demand vector, and determine whether the entropy value of regulation conflict risk exceeds the preset risk tolerance.

[0012] S5: When the preset risk tolerance is exceeded, the control conflict node is identified based on the parameter coordination constraint relationship matrix, and a roundabout control strategy is generated as the optimization guidance parameter according to the control demand vector.

[0013] S6: Based on the optimized guidance parameters and following the collaborative constraint relationship matrix between parameters, coordinate and generate environmental control commands and send them to the actuators.

[0014] Furthermore, obtain the current task identifier of the mobile lab, including:

[0015] Obtain the task selection instruction input by the user through the human-computer interaction interface. The task selection instruction contains a current task identifier used to uniquely distinguish different experimental task types.

[0016] Receive task scheduling instructions from the mobile laboratory's central control system. The task scheduling instructions contain the current task identifier generated according to the experiment plan.

[0017] Furthermore, the experimental task execution stage corresponding to the current task identifier is identified, and a parameter collaborative constraint relationship matrix matching the experimental task execution stage is dynamically generated based on the fuzzy inference rule base, including:

[0018] The task-stage mapping table is queried based on the current task identifier to determine the execution stage of the experimental task;

[0019] The experimental task execution phase is used as the input to the fuzzy inference rule base, and the collaborative constraint strength weights between various environmental parameters are obtained through fuzzy inference.

[0020] A matrix of collaborative constraint relationships between parameters is constructed based on the collaborative constraint strength weights. The rows and columns of the matrix represent different environmental parameters, and the values ​​of the matrix elements represent the strength of the constraint relationship between the corresponding parameters.

[0021] Furthermore, the experimental task execution stage is used as input to the fuzzy inference rule base. The collaborative constraint strength weights between various environmental parameters are obtained through fuzzy inference. This includes: inputting the experimental task execution stage into a pre-established fuzzy inference rule base, which contains corresponding rules for the collaborative constraint relationships between different experimental task execution stages and environmental parameters; processing the input information through the fuzzy inference mechanism and outputting collaborative constraint strength weights that represent the degree of mutual influence between environmental parameters such as temperature, humidity, air pressure, and cleanliness; and using the collaborative constraint strength weights to construct a collaborative constraint relationship matrix between parameters.

[0022] Furthermore, multi-dimensional environmental parameter measurements are collected within the mobile laboratory cabin, and a control demand vector is generated based on these measurements and pre-stored expected environmental parameter values, including:

[0023] Multi-dimensional environmental parameter measurements inside the mobile laboratory cabin are collected simultaneously using temperature sensors, humidity sensors, air pressure sensors, and particulate matter sensors.

[0024] Calculate the parameter deviation between the measured values ​​of multi-dimensional environmental parameters and the expected values ​​of pre-stored environmental parameters;

[0025] The parameter deviations of each environmental parameter are combined in a preset order to form a control demand vector, wherein the dimension of the control demand vector is consistent with the number of environmental parameters.

[0026] Furthermore, based on the parameter coordination constraint relationship matrix and the regulation demand vector, the regulation conflict risk entropy value is calculated, and it is determined whether the regulation conflict risk entropy value exceeds the preset risk tolerance, including:

[0027] The conflict probability distribution of each environmental parameter is obtained by performing matrix multiplication on the regulation demand vector and the collaborative constraint relationship matrix between parameters.

[0028] The information entropy value is calculated based on the conflict probability distribution and used as the entropy value for regulating conflict risk.

[0029] The risk entropy value of the regulatory conflict is compared with the preset risk tolerance. When the risk entropy value of the regulatory conflict is greater than the preset risk tolerance, it is determined that the preset risk tolerance has been exceeded.

[0030] Furthermore, the information entropy value is calculated based on the conflict probability distribution, and the information entropy value is used as the regulation conflict risk entropy value. This includes: calculating the information entropy value using the information entropy calculation formula according to the distribution of conflict probabilities of each environmental parameter in the conflict probability distribution; and using the information entropy value as the regulation conflict risk entropy value to characterize the degree of regulation conflict risk.

[0031] Furthermore, when the preset risk tolerance is exceeded, control conflict nodes are identified based on the parameter coordination constraint relationship matrix, and a roundabout control strategy is generated as an optimization guidance parameter according to the control demand vector, including:

[0032] Analyze the matrix elements in the parameter co-constraint relationship matrix whose constraint strength exceeds the strength threshold, and determine the corresponding environmental parameters as control conflict nodes.

[0033] Based on the magnitude of the parameter deviation values ​​of each environmental parameter in the regulation demand vector, the regulation conflict nodes are prioritized.

[0034] A roundabout control strategy is generated based on the priority ranking result. The roundabout control strategy includes a sequence of instructions that adjust the environmental parameters corresponding to each control conflict node in order of priority.

[0035] The roundabout control strategy containing the instruction sequence is output as the optimization guidance parameter.

[0036] Furthermore, based on the optimized guidance parameters and following the collaborative constraint matrix between parameters, environmental control commands are coordinated and generated and sent to the actuators, including:

[0037] The instruction sequence contained in the optimization guidance parameters is analyzed to obtain the adjustment priority order of each environmental parameter;

[0038] Based on the strength of the constraint relationships in the parameter coordination constraint matrix, determine the combination of environmental parameters to be adjusted simultaneously and their adjustment range;

[0039] Environmental control instructions are generated according to the adjustment priority and adjustment range. The environmental control instructions contain specific environmental parameter control quantities.

[0040] The environmental control commands are sent to the corresponding actuators in the temperature control mechanism, humidity control mechanism, air pressure control mechanism, and air purification mechanism.

[0041] On the other hand, the present invention provides a mobile laboratory cabin internal environment adjustment system, comprising:

[0042] The identifier acquisition module is used to acquire the current task identifier of the mobile laboratory;

[0043] The matrix generation module is used to identify the experimental task execution stage corresponding to the current task identifier, and dynamically generate a parameter co-constraint relationship matrix that matches the experimental task execution stage based on the fuzzy inference rule base.

[0044] The vector generation module is used to collect multi-dimensional environmental parameter measurements inside the mobile laboratory cabin and calculate and generate a control demand vector based on the multi-dimensional environmental parameter measurements and the pre-stored expected environmental parameter values.

[0045] The risk assessment module is used to calculate the risk entropy value of the regulation conflict based on the parameter coordination constraint relationship matrix and the regulation demand vector, and to determine whether the risk entropy value of the regulation conflict exceeds the preset risk tolerance.

[0046] The strategy generation module is used to identify control conflict nodes based on the parameter coordination constraint relationship matrix when the preset risk tolerance is exceeded, and to generate a roundabout control strategy as an optimization guidance parameter according to the control demand vector.

[0047] The instruction execution module is used to coordinate and generate environmental control instructions based on the optimized guidance parameters and in accordance with the matrix of collaborative constraints between parameters, and then send them to the actuator.

[0048] The beneficial effects of this invention are:

[0049] 1. By introducing a task-oriented dynamic environmental control mechanism, the adaptability and accuracy of environmental control in mobile laboratories are significantly improved. By identifying experimental task identifiers and their corresponding execution stages, a matrix of collaborative constraint relationships between parameters is dynamically generated based on a fuzzy inference rule base. This achieves deep coupling between environmental parameter control strategies and experimental requirements. It can adaptively adjust the control priority and tolerance range of parameters such as temperature, humidity, air pressure, and cleanliness according to different scientific task characteristics, ensuring that key experimental parameters receive optimal control resource allocation. At the same time, by establishing collaborative constraint relationships between multiple parameters, mutual conflicts during parameter adjustment are effectively avoided, improving the overall coordination and stability of environmental control.

[0050] 2. By calculating and assessing the entropy value of conflict risk, intelligent early warning and optimized decision-making in the environmental control process are realized. It can promptly identify conflict risks in the parameter control process and generate a roundabout control strategy based on the optimized guiding parameters. This enables efficient utilization of control resources while ensuring environmental stability. It not only improves the robustness of the system, but also ensures the collaborative operation between multiple actuators through an intelligent command coordination mechanism. Ultimately, it achieves high-precision and high-reliability control of the environmental parameters inside the mobile laboratory cabin. Attached Figure Description

[0051] Figure 1 This is a flowchart of a method for adjusting the internal environment of a mobile laboratory cabin according to the present invention;

[0052] Figure 2 This is a schematic diagram of the internal environment adjustment system of a mobile laboratory cabin according to the present invention. Detailed Implementation

[0053] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0054] Example 1: Figure 1 This invention provides a method for adjusting the internal environment of a mobile laboratory cabin, comprising:

[0055] S1: Get the current task identifier of the mobile lab;

[0056] S2: Identify the experimental task execution stage corresponding to the current task identifier, and dynamically generate a parameter co-constraint relationship matrix that matches the experimental task execution stage based on the fuzzy inference rule base;

[0057] S3: Collect multi-dimensional environmental parameter measurements inside the mobile laboratory cabin, and calculate and generate a control demand vector based on the multi-dimensional environmental parameter measurements and the pre-stored expected environmental parameter values;

[0058] S4: Calculate the entropy value of regulation conflict risk based on the parameter coordination constraint relationship matrix and regulation demand vector, and determine whether the entropy value of regulation conflict risk exceeds the preset risk tolerance.

[0059] S5: When the preset risk tolerance is exceeded, the control conflict node is identified based on the parameter coordination constraint relationship matrix, and a roundabout control strategy is generated as the optimization guidance parameter according to the control demand vector.

[0060] S6: Based on the optimized guidance parameters and following the collaborative constraint relationship matrix between parameters, coordinate and generate environmental control commands and send them to the actuators.

[0061] S1: Obtain the current task identifier of the mobile lab, implemented as follows:

[0062] The acquisition of the current task identifier in the mobile laboratory is achieved through two parallel methods to ensure redundancy and reliability in the task identification process. The first method involves obtaining user-input task selection commands through a human-computer interaction interface (HCI). This HCI uses a touchscreen design and displays a menu with four experimental task types: cell culture, sample centrifugation, nucleic acid extraction, and pathological sectioning. Users generate task selection commands by clicking the corresponding menu item. These commands contain a string encoding in the format of "CF-01", "CE-02", "NE-03", or "PS-04" as the current task identifier, where the first two letters represent the experimental type abbreviation and the last two digits represent the task sequence number. The second method involves the mobile laboratory's central control system automatically generating task scheduling commands. This module automatically generates scheduling commands containing the corresponding task identifiers based on a pre-entered experimental schedule when predetermined time nodes are reached. The current task identifier uses the same encoding rules as manually entered identifiers, ensuring a unified resolution standard for identifiers from both sources.

[0063] The task selection function of the human-computer interaction interface is implemented through a graphical user interface. The interface layout is divided into a task selection area and a parameter display area. The task selection area uses four function buttons arranged in a grid, each labeled with the corresponding experimental task name and icon. When the user clicks a specific button, the interface program executes the button response function, extracts the corresponding string code from the pre-stored identifier-task mapping, and encapsulates it into a task selection instruction data packet. This data packet uses JSON format and contains instruction type, timestamp, and task identifier fields. It is sent to the instruction parsing module of the environmental control system via the internal communication bus. The JSON format data packet structure has three levels: the first level is the instruction header information, the second level is the timestamp data, and the third level is the task identifier content. The fields are stored in key-value pairs.

[0064] The task scheduling instruction generation process of the mobile laboratory central control system is based on a timed trigger mechanism. The system maintains an experiment plan queue, which stores experiment task plans sorted by time. Each plan item contains three basic fields: task start time, duration, and task type code. The system clock module checks the plan queue every 60 seconds. When the difference between the current time and the start time of a plan item is less than or equal to a tolerance value of 120 seconds, the corresponding task scheduling instruction is automatically generated. The task scheduling instructions use the same communication protocol and data format as manual instructions, ensuring consistent treatment in subsequent processing stages. The time tolerance value is set based on the minimum preparation time for the experiment task, typically 120 seconds, and this value can be adjusted through the system configuration interface.

[0065] The current task identifier encoding follows a hierarchical naming principle. The first level uses two uppercase letters to represent the major experimental task category: CF for cell culture, CE for sample centrifugation, NE for nucleic acid extraction, and PS for pathological sections. The second level uses two digits to represent the instance number of this task type, sequentially numbered starting from 01. This encoding method ensures the uniqueness of the identifier and allows for direct retrieval of task type information through code parsing, providing foundational data for subsequent task execution phase identification. The encoding rules are stored in the system configuration file and loaded into memory during system startup for use by various modules.

[0066] The transmission of task selection instructions employs a reliable transmission mechanism based on the TCP protocol, with a checksum field included in the data packet to ensure data integrity. Upon receiving the task selection instruction, the environmental control system first performs data verification. If the verification passes, the task identifier field is extracted and stored in the current task register. Simultaneously, the system sends confirmation feedback to the human-machine interface, illuminating the indicator light for the current task status in the interface parameter display area. Data verification uses the CRC32 algorithm, generating a 4-byte checksum appended to the end of the data packet. The receiving end recalculates the checksum and compares it with the received checksum for verification.

[0067] The task scheduling instruction generation process includes a conflict detection mechanism. When the system detects multiple overlapping schedules, it selects which task's scheduling instruction to generate based on the scheduler's priority settings. Each scheduler is assigned a priority flag during entry, with priorities ranging from 0 to 3, where higher numbers indicate higher priority. When a time conflict occurs, the system selects the scheduler with the highest priority to generate the scheduling instruction, simultaneously recording the conflict information in the system log and sending a notification to the operator. Priority settings are based on the importance and urgency of the experimental task; for example, culture experiments are typically set to priority 3, while centrifugation experiments are set to priority 2.

[0068] The current task identifier is stored using a dual-backup mechanism: a copy is stored in the main controller's non-volatile memory and another copy is stored in the environment control module's cache. This design ensures rapid recovery of the task state when some system modules restart. The system persistently stores the current task identifier every 5 minutes to prevent loss of task information due to unexpected power outages. The storage format uses both binary and text formats: binary for fast reading and text for manual viewing and debugging.

[0069] The task identifier validation process includes two steps: format checking and validity checking. Format checking verifies that the string length is at least 5 characters, that the first two digits are a predefined letter combination, and that the last two digits are numbers. Validity checking queries the task type code table to confirm that the code is within the system's supported range. Only valid task identifiers are accepted by the system and proceed to subsequent processing. Invalid identifiers trigger an error handling procedure, returning an error code and explanatory information to the user. Error codes use a three-digit numeric encoding; for example, 101 indicates a format error, and 102 indicates an invalid task type code.

[0070] S2: Identify the experimental task execution stage corresponding to the current task identifier, and dynamically generate a parameter co-constraint relationship matrix that matches the experimental task execution stage based on the fuzzy inference rule base. This is implemented as follows:

[0071] The process of identifying the experimental task execution stage corresponding to the current task identifier begins with parsing the current task identifier. The mobile laboratory's environmental control system first reads the string content of the current task identifier, parsing out the task type code represented by the first two letters and the task instance number represented by the last two numbers. Then, it queries the task-stage mapping table stored in the system's non-volatile memory. This mapping table exists in the form of a relational database table, containing three main columns: task type code, task instance number, and experimental task execution stage. During the query, the task type code and task instance number are used as a composite primary key for exact matching, retrieving the corresponding experimental task execution stage value from the mapping table. The experimental task execution stage is represented in string form; for example, a cell culture task may include different experimental task execution stages such as inoculation, culture, and observation, each stage represented by a unique stage identifier.

[0072] The task-stage mapping table is constructed based on the experimental operation manual, decomposing each experimental task type into several standardized experimental task execution stages. The mapping table maintains a set of experimental task execution stages, each containing attributes such as stage number, stage name, duration, and environmental parameter requirements. Stage numbers employ a three-level coding system: the first level represents the task type, the second level represents the stage sequence, and the third level represents the sub-stage number. For example, CF-01-01 represents the first sub-stage of the first stage of the cell culture task. During system initialization, the complete mapping table data is loaded into memory from the configuration file to improve query efficiency. Updates to the mapping table are implemented through configuration file version management, and each update requires verification of data integrity and consistency.

[0073] The fuzzy inference rule base is constructed based on expert experience and historical experimental data, employing a production rule representation. Each rule consists of a premise and a conclusion. The premise describes the characteristics of the experimental task execution stage, while the conclusion outputs the weights of the collaborative constraints between environmental parameters. The rule base contains multiple rule sets, each targeting a specific experimental task execution stage. For example, the rule set for the cell culture stage contains 5 to 10 specialized rules. The premise of each rule uses fuzzy linguistic variables to describe stage characteristics, such as early culture stage and mid-culture stage. These linguistic variables are converted into numerical representations using membership functions. The membership functions employ triangular or trapezoidal functions, with function parameters set by domain experts according to experimental requirements.

[0074] The fuzzy inference process employs the Mamdani fuzzy inference method, comprising four steps: fuzzification, rule evaluation, aggregation, and defuzzification. First, the input experimental task execution phase information is transformed into a fuzzy set using a membership function, and then matched against the rule premises in the rule base. The matching degree calculation uses the minimum operational rule, taking the minimum matching degree of each sub-condition in the premise as the activation strength of that rule. In the rule evaluation phase, the conclusions of all activated rules are weighted and combined; the output conclusion is a weight vector of the collaborative constraint strengths among environmental parameters. The defuzzification process uses the centroid method to calculate the final precise weight values.

[0075] The weight of the synergistic constraint strength represents the degree of mutual influence between environmental parameters, and its value ranges from 0 to 1 as a real number. The weight is determined based on the physical correlation between the parameters and experimental requirements. For example, the weight between temperature and humidity is typically set to 0.7 to 0.9, indicating a high correlation between the two parameters; while the weight between temperature and cleanliness might be set to 0.3 to 0.5, indicating a weaker correlation. The weight values ​​are determined through a combination of expert scoring and historical data statistical analysis, and are adjusted and optimized every six months based on actual operational data. The adjustment process uses a weighted average algorithm; the new weight equals the original weight multiplied by 0.7 plus the newly calculated weight multiplied by 0.3.

[0076] The construction of the parameter co-constraint matrix is ​​based on an environmental parameter list. The system maintains a standard environmental parameter list containing four basic parameters: temperature, humidity, air pressure, and cleanliness. The rows and columns of the matrix are arranged in this fixed order to ensure the consistency of the matrix structure. The matrix element values ​​are calculated using a weighted product rule. For example, the value of the element in the i-th row and j-th column equals the weight of the co-constraint strength between the temperature parameter and other parameters multiplied by the weight of the co-constraint strength between the humidity parameter and other parameters, and is then adjusted according to the specific relationship type between the parameters. The calculation considers both positive and negative correlations between parameters; positive correlations use positive weights, and negative correlations use negative weights.

[0077] The normalization of matrix element values ​​employs a max-min normalization method, mapping all element values ​​to the range of 0-1. The normalization process first identifies the maximum and minimum values ​​in the matrix, then applies a linear transformation formula to each element value. The normalized matrix exhibits better mathematical properties, facilitating subsequent matrix operations and risk entropy calculations. The system records the original data and normalization parameters for each matrix generation for quality traceability and performance analysis. The normalization parameters include the maximum and minimum values ​​and a scaling factor, which are stored along with the matrix.

[0078] The update and maintenance mechanism of the fuzzy inference rule base includes two approaches: automatic learning and manual review. Automatic learning is based on actual environmental parameter data during the experiment, using clustering algorithms to discover potential association rules between parameters. The clustering algorithm used is the K-means algorithm, and the number of clusters is dynamically determined according to the experimental stage, typically setting 3 to 5 cluster centers. Manual review involves domain experts periodically checking the effectiveness of the rule base and adjusting the rule content and weight settings according to changes in experimental requirements. Each rule base update requires testing and verification to ensure that new rules do not conflict with existing rules and that the accuracy of inference results is improved. Testing and verification uses cross-validation, dividing historical data into training and test sets to verify the generalization ability of the rules.

[0079] The dynamic adjustment function of the collaborative constraint matrix allows the system to fine-tune the matrix element values ​​based on real-time operating status. When the system detects that an environmental parameter continuously deviates from the expected value and is difficult to adjust, it automatically reduces the constraint strength weight of that parameter compared to other parameters, thereby improving the flexibility of control. The adjustment range is limited, with each adjustment not exceeding 10% of the original value, and adjustment logs must be recorded for subsequent analysis. This mechanism ensures that the system maintains stability while possessing a certain degree of self-adaptability. Adjustment decisions are based on the duration and magnitude of parameter deviations; for example, when a parameter deviation persists for more than 30 minutes and the deviation value exceeds 20% of the allowable range, the matrix adjustment mechanism is triggered.

[0080] The anomaly handling mechanism includes rule conflict detection and matrix consistency verification. When contradictory rules appear in the rule base, the system marks these rules and submits them for manual processing. Matrix consistency verification ensures that the generated collaborative constraint matrix meets the mathematical requirements of positive definiteness and symmetry; matrices that do not meet the requirements are automatically corrected. The correction method uses matrix regularization techniques, improving the matrix properties by adding a small identity matrix multiple. All these processes are logged in detail for easy problem tracking and system optimization.

[0081] The synergistic constraints between environmental parameters consider multiple factors, including physical correlations, experimental process requirements, and equipment characteristics. For example, temperature and humidity have a strong physical correlation, so high constraint weights are set for them in most experimental stages. In contrast, the correlation between air pressure and cleanliness is weaker, resulting in relatively lower weights. The specific weight values ​​are determined through extensive experimental data analysis and expert experience, and are continuously optimized and adjusted as experimental experience accumulates. The system maintains an independent weight configuration file for each stage of the experimental task, ensuring that the most appropriate parameter constraint relationships are used for the characteristics of different stages.

[0082] The matrix generation process also considers the uncertainty of parameter measurement and control accuracy. For parameters with low measurement accuracy or slow control response, their constraint weights with other parameters are appropriately reduced to avoid system oscillations caused by over-tuning. The weight adjustment coefficients are dynamically calculated based on the parameter's control performance indicators, including response time, overshoot, and steady-state error. These indicators are obtained through system identification experiments and are updated periodically to ensure accuracy. Through this mechanism, the system can improve the stability and reliability of the control system while ensuring environmental control quality.

[0083] S3: Collect multi-dimensional environmental parameter measurements within the mobile laboratory cabin, and calculate and generate a control demand vector based on the multi-dimensional environmental parameter measurements and pre-stored expected environmental parameter values, implemented as follows:

[0084] The process of collecting multi-dimensional environmental parameter measurements inside the mobile laboratory cabin is achieved through multiple sensors deployed at different locations within the cabin. Temperature sensors utilize PT100 platinum resistance temperature sensors, covering a measurement range of -20℃ to +50℃ with an accuracy of ±0.5℃. These sensors are installed at five measurement points, located at the four corners and the center of the cabin. Humidity sensors employ capacitive polymer film humidity sensors, covering a measurement range of 0%RH to 100%RH with an accuracy of ±3%RH, and are installed alongside the temperature sensors. Barometric pressure sensors use piezoresistive MEMS barometric pressure sensors, measuring a range of 300hPa to 1100hPa with an accuracy of ±1hPa, and are installed at the center of the top of the cabin. Particulate matter sensors employ PM2.5 sensors based on the laser scattering principle, measuring a range of 0μg / m³ to 1000μg / m³ with an accuracy of ±10μg / m³, and are installed at the air inlet and outlet of the air circulation system.

[0085] The sensors employ a synchronous sampling mechanism for data acquisition. All sensors are equipped with a unified clock source, performing a measurement simultaneously every 60 seconds. Measurement data is transmitted to the data acquisition unit via a CAN bus. The data acquisition unit performs preliminary processing on the raw data, including unit standardization, dimension normalization, and outlier removal. Outlier removal uses the 3σ criterion: data with three consecutive measurements exceeding the average value ± 3 times the standard deviation is marked as outlier and removed, and replaced with the previous valid measurement. All sensors are calibrated regularly: temperature and humidity sensors undergo on-site calibration every 30 days, while barometric pressure and particulate matter sensors are returned to the manufacturer for professional calibration every 90 days.

[0086] The acquisition of pre-stored environmental parameter expected values ​​is based on the requirements of each experimental task execution phase. The system maintains an environmental parameter expected value configuration file for each experimental task execution phase. This file contains standard ranges for four parameters: expected temperature, expected humidity, expected air pressure, and expected cleanliness. Expected values ​​are stored in the form of target values ​​and allowable deviation ranges. For example, the expected temperature for the cell culture phase might be set to 37℃ ± 0.5℃, and the expected humidity to be set to 60%RH ± 5%RH. The expected value configuration file is developed according to experimental operating procedures and industry standards, and is stored in the system database after expert review and confirmation.

[0087] The parameter deviation values ​​are calculated using a relative deviation algorithm. For each environmental parameter, the difference between the current measured value and the expected target value is calculated, and then divided by the allowable deviation range to obtain the standardized deviation value. For example, the temperature parameter deviation value is equal to (current temperature measurement value - expected target temperature value) / allowable temperature deviation range. This calculation method makes parameter deviation values ​​of different dimensions comparable, and all parameter deviation values ​​are normalized to the same numerical range, facilitating subsequent vector synthesis and comparative analysis. The importance of parameters is considered during the calculation process; weighted deviation calculations are used for key parameters. The weighting coefficients are set according to experimental requirements; for example, the weight of the temperature parameter is usually set to 1.2, while the weight of the air pressure parameter may be set to 0.8.

[0088] The generation of the control demand vector follows a fixed parameter order. The standard parameter order defined by the system is temperature, humidity, air pressure, and cleanliness, and this order remains consistent throughout the control process. The deviation values ​​of each parameter are arranged sequentially in this order, forming a four-dimensional control demand vector. Each dimension of the vector corresponds to a standardized deviation value of an environmental parameter, typically ranging from -1 to +1. Negative values ​​indicate that the current value is lower than the expected value, and positive values ​​indicate that the current value is higher than the expected value. The vector generation process includes data validity checks to ensure that the values ​​of each dimension are within a reasonable range; values ​​exceeding the range are truncated to boundary values.

[0089] The sensor data preprocessing includes temperature compensation and linearization correction. Temperature sensor readings are spatially averaged based on the installation location; the arithmetic mean of the temperature values ​​from five measurement points is taken as the final temperature measurement. Humidity sensor readings are temperature-compensated, using the simultaneous temperature measurements to correct the humidity readings and eliminate the influence of temperature on humidity measurement. Barometric pressure sensor readings are altitude-compensated, correcting the readings based on the altitude of the mobile laboratory's location. Particulate matter sensor readings are flow-compensated, correcting the particulate matter concentration readings based on the real-time airflow of the air circulation system.

[0090] The calculation of parameter deviation values ​​also considers the influence of the time dimension. The system calculates not only the instantaneous deviation value at the current moment but also the recent deviation trend, such as the rate of change of deviation over the last 10 minutes. The rate of change is calculated using a linear regression method, performing a linear fit on the last 10 measurements (one measurement every 60 seconds), and the slope of the fitted line is taken as the rate of change of deviation. This rate of change information is stored as an additional parameter in the metadata of the control demand vector for subsequent control decision-making processes. The sign of the rate of change indicates the trend of deviation development; a positive value indicates that the deviation is expanding, and a negative value indicates that the deviation is shrinking.

[0091] Standardization of the demand vector ensures comparability between different parameters. Since the physical dimensions of various environmental parameters differ, the original deviation values ​​cannot be directly compared. Therefore, all deviation values ​​need to be normalized to a unified dimensionless scale. The normalization method employs extreme value normalization, mapping all deviation values ​​to the range of -1 to +1. The normalization parameters are dynamically adjusted based on historical data. The system maintains records of the maximum and minimum deviation values ​​for each parameter, updating the normalization parameters every 24 hours. This dynamic normalization method ensures that the demand vector accurately reflects the deviation between the current environmental state and the desired state.

[0092] Multiple quality control measures are implemented during data acquisition. Each sensor is equipped with a self-diagnostic function to monitor its operating status in real time, including indicators such as power supply voltage, signal strength, and noise level. When a sensor anomaly is detected, the system automatically switches to a backup sensor or uses a predicted value as a substitute. The predicted values ​​are generated based on time series analysis of historical data, using an autoregressive integral moving average model to predict environmental parameter values ​​for a future period. All quality control operations are logged in detail, including information such as the time of anomaly occurrence, handling measures, and recovery time. These logs are used for subsequent system performance analysis and optimization.

[0093] The storage and transmission of demand vectors utilize a standardized data format. Vector data is stored as an array of floating-point numbers, with each floating-point number occupying 4 bytes of storage space, for a total vector size of 16 bytes. A data packet header is added during data transmission, containing information such as a timestamp, data version number, and checksum. The checksum is calculated using a cyclic redundancy check algorithm to detect errors during data transmission. Upon receiving the demand vector, the receiving end first verifies the checksum; only after successful verification does it proceed with subsequent processing, ensuring data integrity and reliability. The system retains historical records of all demand vectors from the most recent 24 hours; these records are used for trend analysis and system performance evaluation.

[0094] S4: Calculate the regulation conflict risk entropy value based on the parameter coordination constraint relationship matrix and the regulation demand vector, and determine whether the regulation conflict risk entropy value exceeds the preset risk tolerance limit. The implementation is as follows:

[0095] The process of calculating the entropy value of regulatory conflict risk based on the inter-parameter collaborative constraint relationship matrix and the regulatory demand vector begins with matrix multiplication. When performing matrix multiplication on the regulatory demand vector and the inter-parameter collaborative constraint relationship matrix, the regulatory demand vector is treated as a row vector, and the inter-parameter collaborative constraint relationship matrix is ​​treated as a 4×4 square matrix. The operation is performed according to the rules of matrix multiplication. During the multiplication process, each element of the regulatory demand vector is multiplied pairwise with the corresponding element of the inter-parameter collaborative constraint relationship matrix, and then all the product results are added together to obtain a new vector element. This operation produces a new vector called the conflict probability distribution vector, whose dimension is consistent with the number of environmental parameters. For example, in the case of four environmental parameters, the conflict probability distribution vector is also a four-dimensional vector.

[0096] Each element of the conflict probability distribution vector represents an estimated probability of conflict for the corresponding environmental parameter under the current regulatory requirements. These probability values ​​need to be normalized so that the sum of all probability values ​​equals 1. Normalization is performed by dividing each element by the sum of all elements, ensuring that the resulting conflict probability distribution satisfies the basic properties of probability distributions. For example, assuming the original calculation result is [0.8, 0.6, 0.4, 0.2], normalization yields [0.4, 0.3, 0.2, 0.1]. The normalized conflict probability distribution vector is used for subsequent information entropy calculations; this distribution reflects the probability distribution of conflicts among various environmental parameters during the regulatory process.

[0097] The process of calculating information entropy based on the conflict probability distribution employs the entropy calculation formula from information theory. The information entropy value is obtained by multiplying the conflict probability of each environmental parameter by the logarithm of that probability, summing the results, and taking the negative value. Specifically, for each probability value in the conflict probability distribution, the product of the probability value and the logarithm to the base 2 is calculated first, then all products are summed, and finally, the negative number is taken. This calculation process quantifies the degree of uncertainty in the conflict probability distribution; a larger information entropy value indicates a more uniform conflict distribution and a higher risk of conflict control. The information entropy value ranges from 0 to log2(n), where n is the number of environmental parameters. For example, when n=4, the maximum entropy value is 2.

[0098] When using information entropy as the risk entropy value for regulatory conflict, the calculated information entropy value needs to be mapped to a standardized risk metric. The risk entropy value for regulatory conflict is a dimensionless numerical value, typically ranging from 0 to 1; a higher value indicates a higher risk of regulatory conflict. This mapping process uses a linear scaling method, dividing the original information entropy value by the maximum possible entropy value. For example, with four environmental parameters, dividing the information entropy value by 2 yields the standardized risk entropy value for regulatory conflict. The standardized risk entropy value facilitates comparison with preset risk tolerances and also facilitates comparison of risk levels between systems of different sizes.

[0099] The preset risk tolerance is determined based on historical operational data and expert experience. The system collects records of conflict events and corresponding information entropy values ​​from historical control processes, and determines a suitable risk threshold through statistical analysis. For example, by analyzing the past 1000 control operations, it was found that when the information entropy value exceeds 0.7, there is a 90% probability of a control conflict occurring. Therefore, the preset risk tolerance can be set to 0.7. This threshold can be dynamically adjusted according to actual operating conditions, initially using a more conservative value and gradually optimizing it as system operating experience accumulates. Adjustments to the risk tolerance require a safety assessment to ensure that improper threshold settings will not lead to uncontrolled system risks.

[0100] When comparing the risk entropy value of the regulatory conflict with the preset risk tolerance, a numerical comparison method is used. When the risk entropy value exceeds the preset risk tolerance, it is determined that the risk has exceeded the preset risk tolerance, and the system needs to activate the conflict resolution mechanism. The comparison operation is performed every 60 seconds to ensure timely detection of changes in risk status. The comparison results are divided into three levels: low risk (entropy value less than 0.5 times the tolerance), medium risk (entropy value between 0.5 and 1 times the tolerance), and high risk (entropy value exceeding the tolerance). The system adopts corresponding processing strategies according to different risk levels: normal regulation continues in low-risk situations, monitoring is strengthened in medium-risk situations, and the conflict resolution procedure is activated in high-risk situations.

[0101] The calculation of the conflict probability distribution includes an outlier handling mechanism. When the calculated conflict probability of a certain environmental parameter is abnormally high or low, the system will initiate a verification procedure. The outlier judgment criteria are based on historical distribution; for example, if the conflict probability of a parameter exceeds twice the standard deviation of the historical average for three consecutive periods, it is considered an outlier. For outliers, the system uses a moving average method for smoothing, replacing the current value with the average of the most recent 10 periods to ensure the stability of the probability distribution. Abnormal events are also recorded for subsequent analysis.

[0102] The information entropy calculation process considers boundary case handling for probability values. When the conflict probability of a certain environmental parameter is 0, 0 is directly used in the calculation to avoid logarithmic calculation errors. When all probability values ​​are 0, the system defaults to a uniform distribution as a substitute; for example, with four parameters, each parameter has a probability of 0.25. This handling ensures that the information entropy calculation yields valid results under any circumstances, avoiding calculation interruptions due to data anomalies. The system records these special handling cases and marks the corresponding calculation period in the log.

[0103] The calculation of conflict risk entropy also includes trend analysis functionality. The system not only calculates the current risk entropy value but also analyzes its changing trend. Trend analysis employs time series analysis, performing linear regression on the risk entropy values ​​over the most recent 10 periods to calculate the slope and acceleration of change. These trend indicators serve as auxiliary criteria for risk warning; when the risk entropy value shows a rapid upward trend, the system will issue an early warning signal even if the current value has not yet exceeded the risk tolerance limit. Trend analysis helps the system achieve more proactive risk management, improving its robustness and reliability.

[0104] The risk assessment results are output in a standardized data format. The results include the current risk entropy value, risk level, flags indicating exceeded tolerance limits, and trend indicators. This data is encapsulated in a risk report data structure and transmitted to relevant processing modules via network protocols. Risk reports are generated every 60 seconds, including timestamps and sequence numbers to ensure timeliness and completeness. The system stores all risk reports from the last 24 hours; this historical data is used for system performance analysis and risk pattern identification, providing data support for optimizing and adjusting risk tolerance limits.

[0105] Numerical stability safeguards are implemented during matrix multiplication. When the condition number of the co-constraint matrix between parameters is detected to be too large, the system automatically performs regularization on the matrix by adding a small identity matrix multiple to improve its numerical properties. This process ensures the numerical stability of matrix multiplication and avoids distortion of calculation results due to ill-conditioned matrices. The regularization parameter is dynamically determined based on matrix eigenvalue analysis and is typically set to 0.001 times the largest eigenvalue of the matrix.

[0106] Information entropy calculation employs a high-precision numerical method. The system uses 64-bit floating-point numbers for all calculations to ensure required accuracy. Logarithmic calculations utilize an optimized approximation algorithm, improving computational efficiency while maintaining accuracy. Rounding error control is implemented during calculations, employing a rounding rule of rounding to even numbers (four for six, five for five) to prevent error accumulation from affecting the final result. All calculation operations have an error detection mechanism; when a calculation error is detected, the system automatically recalculates or uses a backup calculation method.

[0107] The dynamic adjustment mechanism for risk tolerance employs an incremental learning algorithm. The system continuously optimizes the risk tolerance value based on recent feedback on regulatory effects, using gradient descent to find the optimal value. The adjustment range is limited, with each adjustment not exceeding 10% of the current value, and the adjusted value must remain within a pre-defined safety range, such as 0.5 to 0.9. The adjustment process requires multiple cycles of verification to ensure that the adjusted tolerance value effectively identifies genuine regulatory conflict risks. The system records all tolerance adjustment history; this data is used for subsequent performance analysis and optimization.

[0108] S5: When the preset risk tolerance is exceeded, the control conflict node is identified based on the parameter coordination constraint relationship matrix, and a roundabout control strategy is generated as the optimization guidance parameter according to the control demand vector, which is implemented as follows:

[0109] When the entropy value of the regulatory conflict risk exceeds the preset risk tolerance, the system initiates the regulatory conflict node identification process. When analyzing matrix elements in the parameter co-constraint relationship matrix where the constraint strength exceeds the strength threshold, the system traverses all off-diagonal elements of the matrix. These elements represent the constraint strength between different environmental parameters. The strength threshold is dynamically determined based on historical operating data, typically set to 1.5 to 2 times the average value of the matrix elements. For example, if the temperature-humidity constraint strength is 0.8 and the average matrix constraint strength is 0.4, and the threshold coefficient is set to 1.5, the threshold becomes 0.6, and this element will be identified as exceeding the strength threshold. For each matrix element exceeding the strength threshold, the system records its row and column numbers; the environmental parameters represented by the corresponding row and column are identified as regulatory conflict nodes. The dynamic adjustment of the strength threshold is based on the system's operating status. When the number of identified nodes is too high or too low, the system automatically adjusts the threshold coefficient, with the adjustment not exceeding 10% of the current value, ensuring the threshold remains within a reasonable range of 0.3 to 0.8.

[0110] After identifying the conflict nodes, the system prioritizes environmental parameters based on their deviation values ​​in the control demand vector. The ranking principle is based on the absolute value of the parameter deviation; parameters with larger deviations have higher priority. For example, when the temperature parameter deviation is +0.8 and the humidity parameter deviation is -0.5, the temperature parameter is given higher priority. The ranking algorithm uses a quicksort method to sort the environmental parameter deviation values ​​corresponding to all conflict nodes in descending order. The ranking process considers the importance weight of the parameters, which is pre-set according to experimental requirements. For example, the importance weight for temperature is 1.2, and for humidity it is 1.0. The actual ranking value is the parameter deviation value multiplied by the importance weight. When the weighted deviation difference of multiple parameters is less than 0.05, the system refers to historical adjustment records and prioritizes parameters with fewer recent adjustments.

[0111] When generating a roundabout control strategy based on priority ranking, the system processes each control conflict node sequentially from highest to lowest priority. For each control conflict node, a corresponding environmental parameter adjustment instruction is generated, containing information such as the target parameter, adjustment direction, and adjustment magnitude. The adjustment direction is determined by the sign of the parameter deviation value; a positive deviation indicates that the parameter value needs to be decreased, and a negative deviation indicates that the parameter value needs to be increased. The adjustment magnitude is calculated based on the magnitude of the parameter deviation value, typically set as a certain percentage of the absolute value of the deviation, such as 50% to 80%. The instruction sequence also includes execution time information; higher-priority instructions are scheduled for earlier execution, and sufficient time intervals are maintained between adjacent instructions to ensure effective execution. The time interval is set according to the response characteristics of the actuators; temperature control mechanisms require a 120-second response time, and humidity control mechanisms require a 60-second response time.

[0112] The roundabout control strategy considers the constraints between parameters to avoid triggering new conflicts when adjusting one parameter. For parameter pairs with strong constraints, the system employs a coordinated adjustment strategy, adjusting both parameters simultaneously but with different adjustment magnitudes. The adjustment magnitude is determined based on the strength of the constraint relationship; the stronger the constraint, the closer the adjustment magnitudes of the two parameters. For example, if there is a strong constraint relationship between temperature and humidity, when temperature needs to be adjusted, humidity is adjusted simultaneously by a smaller magnitude to maintain the relative relationship between the two parameters. The proportion of coordinated adjustment is calculated using the parameter coordination constraint matrix, and the adjusted parameter values ​​must satisfy the constraints defined in the matrix. This coordinated adjustment ensures that no new control conflicts are generated while resolving existing ones.

[0113] When outputting a roundabout control strategy containing instruction sequences as optimization guidance parameters, the system encapsulates the instruction sequences into a standardized data structure. This data structure comprises three main parts: the number of instructions, the instruction list, and the execution schedule. Each instruction in the instruction list includes four fields: parameter type, target value, execution duration, and priority. The execution schedule details the start and end times of each instruction, ensuring the orderly execution of instructions. The optimization guidance parameters are encapsulated in JSON format, containing metadata such as version number 1.0, generation timestamp, and validity period. Network transmission uses the TCP protocol, with data packets enhanced with CRC32 checksums to ensure reliable transmission. The system also stores the optimization guidance parameters in a local database for 30 days for subsequent analysis and optimization.

[0114] The dynamic adjustment mechanism for the intensity threshold is based on adaptive adjustment according to the system's operating status. The system monitors the identification effect of conflict nodes and automatically adjusts the intensity threshold when too many or too few nodes are identified. The adjustment method uses a gradient descent algorithm, with the conflict resolution success rate as the objective function, to progressively optimize the threshold parameters. The system records the success rate data of the last 100 control operations, and initiates the threshold adjustment procedure when the success rate falls below 85%. During the adjustment process, the system tests multiple candidate thresholds and selects the threshold that results in the highest control success rate as the new setpoint. The entire adjustment process requires at least 5 control cycles to complete, ensuring the stability of the adjustment.

[0115] Stability safeguards are implemented during the priority ranking process. When the deviations of multiple environmental parameters are very close, the system employs additional ranking conditions to ensure the stability of the ranking results. These additional conditions include factors such as parameter importance weights, historical adjustment frequency, and adjustment difficulty. For example, when the deviations of two parameters differ by less than 0.05, the parameter with the higher importance weight is prioritized. Historical adjustment frequency is calculated by recording the number of adjustments each parameter has made in the past 24 hours; parameters with higher adjustment frequencies are given lower priority. Adjustment difficulty is assessed based on the response speed and accuracy of the actuators; parameters corresponding to actuators with slow response speeds and low accuracy are assigned lower priority. This multi-condition ranking mechanism avoids frequent changes in ranking results due to small numerical fluctuations, thus improving system stability.

[0116] The generation of the roundabout control strategy includes a risk assessment step. The system simulates the execution of the generated instruction sequence and predicts the potential new conflict risks that may arise after execution. The prediction method is based on the parameter co-constraint matrix, calculating the strength of the constraint relationships between the adjusted parameter values. When a new conflict is predicted, the system automatically adjusts the instruction sequence, such as changing the execution order or adjusting the execution magnitude. The risk assessment uses the Monte Carlo simulation method, randomly generating multiple sets of possible parameter changes and statistically analyzing the probability of conflict occurrence. When the predicted conflict probability exceeds 0.3, the system regenerates the control strategy. During the simulation execution, the response characteristics of the actuator, including response delay and overshoot, are considered to ensure the accuracy of the prediction results.

[0117] The output format of the optimization guidance parameters adopts a standardized design. The data format includes two parts: header information and body information. The header information includes metadata such as version number, generation time, and validity period. The body information contains the specific instruction sequence. The instruction sequence is described in XML format, with each instruction as an independent element containing complete parameter information. Semantic versioning is used for version numbers; major version numbers indicate major updates, minor version numbers indicate feature additions, and revision numbers indicate bug fixes. The validity period is set according to the urgency of the control task, typically ranging from 10 to 60 minutes. Optimization guidance parameters exceeding their validity period automatically expire. This standardized format ensures compatibility between different systems and facilitates subsequent processing and analysis.

[0118] The scheduling of command execution time takes into account the response characteristics of the actuators. Different actuators have different response speeds; for example, temperature control mechanisms respond slowly, while humidity control mechanisms respond quickly. The system calculates the average response time of each actuator based on historical data and schedules the command execution time interval accordingly. The average response time for temperature control mechanisms is 120 seconds, for humidity control mechanisms it is 60 seconds, and for air pressure control mechanisms it is 90 seconds. Longer execution times are reserved for slower-responding actuators, while faster-responding actuators can have more compact execution times. The execution time scheduling also considers the constraints between parameters; parameters with strong constraints should have adjustment times as close as possible to avoid prolonged periods of incoordination. This differentiated timing ensures the accuracy and reliability of command execution.

[0119] The optimization process of the control strategy employs an iterative improvement method. The system records the execution effect of each control strategy, including indicators such as the degree of conflict resolution and execution time. Based on this historical data, the system continuously optimizes the algorithm for generating control strategies, such as adjusting the command amplitude calculation parameters or optimizing the execution time schedule. The optimization algorithm uses reinforcement learning, with the control effect as a reward signal to gradually improve the quality of strategy generation. The reward signal includes a weight of 0.6 for the degree of conflict resolution, a weight of 0.3 for execution time, and a weight of 0.1 for energy consumption indicators. The optimization process is subject to safety constraints to ensure that any adjustments do not exceed the safe operating range. The parameter adjustment range is limited to ±20% of the expected value, and the execution time interval must not be less than 30 seconds. The system conducts a comprehensive evaluation of the optimization algorithm weekly and adjusts the algorithm parameters based on the evaluation results.

[0120] S6: Based on the optimized guidance parameters and following the inter-parameter coordination constraint matrix, coordinate and generate environmental control commands and send them to the actuators, implemented as follows:

[0121] The process of coordinating and generating environmental control instructions based on optimization guidance parameters and following the inter-parameter coordination constraint matrix begins with the parsing of these parameters. The system first reads the optimization guidance parameter data package, verifying its integrity and validity, including version number compatibility, timestamp validity, and digital signature verification. The parsing process employs a hierarchical parsing method. First, it extracts metadata information from the instruction sequence, including the total number of instructions, execution time range, and priority distribution. Then, it parses the detailed content of each instruction one by one to obtain the adjustment priority order of each environmental parameter. The adjustment priority order is represented numerically, with smaller values ​​indicating higher priority; for example, 1 represents the highest priority, and 4 represents the lowest priority. Data validation is performed during parsing to ensure that priority values ​​are within a valid range and that there are no duplicate values. Simultaneously, the logical consistency between instructions is checked to avoid contradictory instruction content.

[0122] When determining the combination of environmental parameters to be adjusted simultaneously based on the strength of constraints in the parameter co-constraint matrix, the system analyzes the strength of constraints on all parameter pairs in the matrix. A constraint strength threshold of 0.6 is set; when the constraint strength between two parameters exceeds this threshold, these two parameters are included in the same adjustment combination. The adjustment combination is determined using graph theory, treating parameters as nodes and constraints as edges, and finding the largest possible set of parameters that can be adjusted simultaneously by searching a complete subgraph. The adjustment magnitude for each combination is calculated based on the parameter co-constraint matrix; parameters with strong constraints are adjusted with similar ratios, while parameters with weak constraints can be adjusted independently. The adjustment magnitude calculation considers the difference between the current parameter value and the target value, while also referencing historical adjustment effect data to ensure that the adjusted parameter values ​​meet the experimental requirements.

[0123] When generating environmental control instructions according to adjustment priority and magnitude, the system first processes each adjustment combination in descending order of priority. For each adjustment combination, a corresponding environmental control instruction is generated, which includes a list of target parameters, target values, adjustment rate, and execution time window. The adjustment rate is determined based on parameter characteristics and equipment performance; for example, the adjustment rate for temperature parameters is limited to within 0.5°C per minute, and the adjustment rate for humidity parameters is limited to within 5%RH per minute. The execution time window is set according to the adjustment priority, with higher-priority instructions receiving earlier execution times, and sufficient time intervals maintained between adjacent instructions. Conflict detection is implemented during instruction generation to ensure that newly generated instructions do not conflict with currently executing instructions, while also considering equipment workload to avoid simultaneous startup of multiple high-power devices.

[0124] Environmental control commands contain specific environmental parameter control quantities, calculated based on the difference between target and current values. The control quantity calculation employs a PID control algorithm, with the proportional gain, integral time, and derivative time set according to the parameter characteristics. For example, the proportional gain for temperature control is set to 2.0, the integral time to 300 seconds, and the derivative time to 60 seconds. Before output, the control quantity undergoes a limiting process to ensure the output value remains within the equipment's permissible operating range; for example, the heater's output power is limited to between 0% and 100%. Simultaneously, a rate of change limitation is implemented to prevent sudden changes in the control quantity from impacting the system; for example, the temperature rate of change is limited to between 0.1°C and 1.0°C per minute. Each control command is accompanied by a timestamp and sequence number to ensure traceability.

[0125] When environmental control commands are sent to the corresponding actuators in the temperature control, humidity control, air pressure control, and air purification systems, the system selects the appropriate actuator based on the parameter type in the command. Temperature control commands are sent to the heating / cooling unit, humidity control commands to the humidification / dehumidification unit, air pressure control commands to the pressure regulating unit, and cleanliness control commands to the air purification unit. Command transmission uses the Modbus RTU protocol with a transmission rate of 19200bps and a data format of 8 data bits, no parity check, and 1 stop bit. After each command transmission, the system waits for an acknowledgment response from the actuator, with a timeout of 5 seconds and a maximum of 3 retries. Data encryption is implemented during transmission to ensure command security.

[0126] A status monitoring mechanism is implemented during command transmission, with the system monitoring the operational status of each actuator in real time. Status monitoring includes information such as equipment readiness status, current output values, and fault codes. When an actuator malfunction is detected, the system automatically activates backup equipment or adjusts command parameters. For example, if the main heating unit fails, it automatically switches to the backup heating unit, simultaneously adjusting the heating power and temperature rise rate. Status information is updated every 10 seconds to ensure the system can respond promptly to changes in equipment status. Monitoring data is saved to the system log for subsequent fault analysis and preventative maintenance.

[0127] The effectiveness of control commands is verified through sensor feedback. The system compares the actual parameter values ​​after command execution with the target values. When the difference exceeds the allowable range, compensation adjustment is initiated. Compensation adjustment uses an incremental method, with the adjustment increment being a certain percentage of the difference, such as 50%. The number of compensation adjustments is limited; if the target value cannot be reached after three consecutive compensation adjustments, the system will re-evaluate the control strategy and generate a new command sequence. During verification, sensor measurement errors are considered, and a weighted averaging method is used to process measurement data from multiple sensors to improve data reliability.

[0128] Energy consumption optimization is considered during instruction execution. The system selects the lowest energy-consuming adjustment path, such as prioritizing natural heat exchange for temperature control and energy-efficient dehumidification technology for humidity control. Energy optimization is based on equipment power characteristics and operating cost data, using a multi-objective optimization algorithm to find the optimal balance between energy consumption and performance. During optimization, environmental parameters are maintained within permissible ranges, ensuring that environmental control quality is not sacrificed for energy saving. Energy consumption data is displayed in real-time on the human-machine interface, helping operators understand the system's operating efficiency.

[0129] All control command execution records are saved to the system database, including command content, sending time, execution result, and energy consumption data. This data is used for subsequent performance analysis and system optimization, helping to improve command generation algorithms and control accuracy. Data is retained for 90 days; data exceeding this period is automatically archived to long-term storage. The system generates weekly execution reports, statistically analyzing metrics such as command execution success rate, average execution time, and energy efficiency. The reports are presented in charts and graphs, allowing administrators to intuitively understand the system's operational status. Based on these analyses, the system automatically adjusts control parameters and optimization strategies, continuously improving the quality and efficiency of environmental control.

[0130] The system also establishes an anomaly handling mechanism to promptly address any abnormal situations that occur during command execution. These anomalies include equipment failure, communication interruption, and parameter anomalies, with different handling strategies implemented based on the type and severity of the anomaly. For example, upon detecting an abnormal temperature rise, the system immediately stops the heating command and initiates the cooling program, while simultaneously issuing an alarm to notify the operator. The anomaly handling process has been thoroughly tested to ensure the system's safe and stable operation under various abnormal conditions. Each anomaly event generates a detailed report, including the time of occurrence, the handling process, and the result assessment. These reports are used to refine the anomaly handling strategy and improve system reliability.

[0131] Example 2: Figure 2 A schematic diagram of the internal environment adjustment system of a mobile laboratory cabin according to the present invention is provided. The mobile laboratory cabin internal environment adjustment system includes:

[0132] The identifier acquisition module is used to acquire the current task identifier of the mobile laboratory;

[0133] The matrix generation module is used to identify the experimental task execution stage corresponding to the current task identifier, and dynamically generate a parameter co-constraint relationship matrix that matches the experimental task execution stage based on the fuzzy inference rule base.

[0134] The vector generation module is used to collect multi-dimensional environmental parameter measurements inside the mobile laboratory cabin and calculate and generate a control demand vector based on the multi-dimensional environmental parameter measurements and the pre-stored expected environmental parameter values.

[0135] The risk assessment module is used to calculate the risk entropy value of the regulation conflict based on the parameter coordination constraint relationship matrix and the regulation demand vector, and to determine whether the risk entropy value of the regulation conflict exceeds the preset risk tolerance.

[0136] The strategy generation module is used to identify control conflict nodes based on the parameter coordination constraint relationship matrix when the preset risk tolerance is exceeded, and to generate a roundabout control strategy as an optimization guidance parameter according to the control demand vector.

[0137] The instruction execution module is used to coordinate and generate environmental control instructions based on the optimized guidance parameters and in accordance with the matrix of collaborative constraints between parameters, and then send them to the actuator.

[0138] All calculations involved in the embodiments are dimensionless numerical calculations, and the preset parameters and thresholds in the calculations are set by those skilled in the art according to the actual situation.

[0139] It should be noted that this invention can be deployed on the device itself to realize embedded applications, or it can run on a PC or other terminal with a user interface, thereby meeting various hardware environments and usage requirements.

[0140] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented, in whole or in part, as a computer program product. A computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, all or part of the processes or functions according to the embodiments of this application are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. Computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wireless or wired transmission; wired transmission methods include optical fiber, twisted pair, coaxial cable, etc.; wireless transmission includes infrared, microwave, etc. Computer-readable storage media can be any available medium that a computer can access or a data storage device such as a server or data center that contains one or more sets of available media. Available media can be magnetic media (e.g., floppy disks, hard disks, magnetic tapes), optical media (e.g., DVDs), or semiconductor media. Semiconductor media can be solid-state drives.

[0141] Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and modules described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0142] In the several embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of modules is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple modules 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 apparatuses or modules may be electrical, mechanical, or other forms.

[0143] The modules described as separate components may or may not be physically separate. The components shown as modules may or may not be physical modules; they may be located in one place or distributed across multiple network modules. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0144] In addition, the functional modules in the various embodiments of this application can be integrated into one processing module, or each module can exist physically separately, or two or more modules can be integrated into one module.

[0145] If a function is implemented as a software module and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a portion 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 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 in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0146] The above are merely specific embodiments of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

[0147] In conclusion, the above are merely preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for adjusting the internal environment of a mobile laboratory cabin, characterized in that, include: S1: Get the current task identifier of the mobile lab; S2: Identify the experimental task execution stage corresponding to the current task identifier, and dynamically generate a parameter co-constraint relationship matrix matching the experimental task execution stage based on the fuzzy inference rule base, including: The task-stage mapping table is queried based on the current task identifier to determine the execution stage of the experimental task; The experimental task execution phase is used as the input to the fuzzy inference rule base, and the collaborative constraint strength weights between various environmental parameters are obtained through fuzzy inference. A matrix of collaborative constraint relationships between parameters is constructed based on the collaborative constraint strength weights. The rows and columns of the matrix represent different environmental parameters, and the values ​​of the matrix elements represent the strength of the constraint relationship between the corresponding parameters. S3: Collect multi-dimensional environmental parameter measurements inside the mobile laboratory cabin, and calculate and generate a control demand vector based on the multi-dimensional environmental parameter measurements and the pre-stored expected environmental parameter values; S4: Calculate the entropy value of regulation conflict risk based on the parameter coordination constraint relationship matrix and regulation demand vector, and determine whether the entropy value of regulation conflict risk exceeds the preset risk tolerance. The calculation of the regulation conflict risk entropy value includes: performing matrix multiplication on the regulation demand vector and the collaborative constraint relationship matrix between parameters to obtain the conflict probability distribution of each environmental parameter; calculating the information entropy value using the information entropy calculation formula based on the distribution of conflict probabilities of each environmental parameter in the conflict probability distribution; and using the information entropy value as the regulation conflict risk entropy value characterizing the degree of regulation conflict risk. S5: When the preset risk tolerance is exceeded, the control conflict node is identified based on the parameter coordination constraint relationship matrix, and a roundabout control strategy is generated as the optimization guidance parameter according to the control demand vector. S6: Based on the optimized guidance parameters and following the collaborative constraint relationship matrix between parameters, coordinate and generate environmental control commands and send them to the actuators.

2. The method for adjusting the internal environment of a mobile laboratory cabin according to claim 1, characterized in that, Obtain the current task identifier of the mobile lab, including: Obtain the task selection instruction input by the user through the human-computer interaction interface. The task selection instruction contains a current task identifier used to uniquely distinguish different experimental task types. Receive task scheduling instructions from the mobile laboratory's central control system. The task scheduling instructions contain the current task identifier generated according to the experiment plan.

3. The method for adjusting the internal environment of a mobile laboratory cabin according to claim 1, characterized in that, The experimental task execution stage is used as input to the fuzzy inference rule base. The collaborative constraint strength weights between various environmental parameters are obtained through fuzzy inference. This includes: inputting the experimental task execution stage into a pre-established fuzzy inference rule base, which contains corresponding rules for the collaborative constraint relationships between different experimental task execution stages and environmental parameters; processing the input information through the fuzzy inference mechanism and outputting collaborative constraint strength weights that represent the degree of mutual influence between environmental parameters such as temperature, humidity, air pressure, and cleanliness; and using the collaborative constraint strength weights to construct a collaborative constraint relationship matrix between parameters.

4. The method for adjusting the internal environment of a mobile laboratory cabin according to claim 1, characterized in that, The system collects multi-dimensional environmental parameter measurements within the mobile laboratory chamber and calculates and generates a control demand vector based on these measurements and pre-stored expected environmental parameter values. This vector includes: Multi-dimensional environmental parameter measurements inside the mobile laboratory cabin are collected simultaneously using temperature sensors, humidity sensors, air pressure sensors, and particulate matter sensors. Calculate the parameter deviation between the measured values ​​of multi-dimensional environmental parameters and the expected values ​​of pre-stored environmental parameters; The parameter deviations of each environmental parameter are combined in a preset order to form a control demand vector, wherein the dimension of the control demand vector is consistent with the number of environmental parameters.

5. The method for adjusting the internal environment of a mobile laboratory cabin according to claim 1, characterized in that, Determining whether the risk entropy value of regulatory conflict exceeds the preset risk tolerance includes: The risk entropy value of the regulatory conflict is compared with the preset risk tolerance. When the risk entropy value of the regulatory conflict is greater than the preset risk tolerance, it is determined that the preset risk tolerance has been exceeded.

6. The method for adjusting the internal environment of a mobile laboratory cabin according to claim 1, characterized in that, When the preset risk tolerance is exceeded, the control conflict nodes are identified based on the parameter coordination constraint relationship matrix, and a roundabout control strategy is generated as the optimization guidance parameter according to the control demand vector, including: Analyze the matrix elements in the parameter co-constraint relationship matrix whose constraint strength exceeds the strength threshold, and determine the corresponding environmental parameters as control conflict nodes. Based on the magnitude of the parameter deviation values ​​of each environmental parameter in the regulation demand vector, the regulation conflict nodes are prioritized. A roundabout control strategy is generated based on the priority ranking result. The roundabout control strategy includes a sequence of instructions that adjust the environmental parameters corresponding to each control conflict node in order of priority. The roundabout control strategy containing the instruction sequence is output as the optimization guidance parameter.

7. The method for adjusting the internal environment of a mobile laboratory cabin according to claim 1, characterized in that, Based on optimized guidance parameters and following the inter-parameter coordination constraint matrix, environmental control commands are generated and sent to the actuators, including: The instruction sequence contained in the optimization guidance parameters is analyzed to obtain the adjustment priority order of each environmental parameter; Based on the strength of the constraint relationships in the parameter coordination constraint matrix, determine the combination of environmental parameters to be adjusted simultaneously and their adjustment range; Environmental control instructions are generated according to the adjustment priority and adjustment range. The environmental control instructions contain specific environmental parameter control quantities. The environmental control commands are sent to the corresponding actuators in the temperature control mechanism, humidity control mechanism, air pressure control mechanism, and air purification mechanism.

8. A mobile laboratory cabin interior environment adjustment system, used to implement the mobile laboratory cabin interior environment adjustment method according to any one of claims 1-7, characterized in that, include: The identifier acquisition module is used to acquire the current task identifier of the mobile laboratory; The matrix generation module is used to identify the experimental task execution stage corresponding to the current task identifier, and dynamically generate a parameter co-constraint relationship matrix that matches the experimental task execution stage based on the fuzzy inference rule base. The vector generation module is used to collect multi-dimensional environmental parameter measurements inside the mobile laboratory cabin and calculate and generate a control demand vector based on the multi-dimensional environmental parameter measurements and the pre-stored expected environmental parameter values. The risk assessment module is used to calculate the risk entropy value of the regulation conflict based on the parameter coordination constraint relationship matrix and the regulation demand vector, and to determine whether the risk entropy value of the regulation conflict exceeds the preset risk tolerance. The strategy generation module is used to identify control conflict nodes based on the parameter coordination constraint relationship matrix when the preset risk tolerance is exceeded, and to generate a roundabout control strategy as an optimization guidance parameter according to the control demand vector. The instruction execution module is used to coordinate and generate environmental control instructions based on the optimized guidance parameters and in accordance with the matrix of collaborative constraints between parameters, and then send them to the actuator.

Images