A medicine stability test box control and data tracing device

Abstract

The application discloses a kind of drug stability test box control and data tracing device, including prescription fingerprint module, trusted acquisition latch module, offset attribution module, storage evidence gate control module and chain tracing freezing module.Prescription fingerprint module analyzes stability test scheme and generates prescription fingerprint;Trusted acquisition latch module collects the environment quantity in test box, door body state and actuator state, generates trusted environment vector and acquisition abstract;Offset attribution module couples environment target, tolerance boundary, door body state, actuator state and last sampling period control instruction, and outputs offset type and offset intensity;Storage evidence gate control module releases adjustment instruction after tracing pre-sealing is completed;Chain tracing freezing module writes execution receipt and freezes record when checking exception.The device improves the reliability of drug stability test process control and the credibility of data tracing.

Description

Technical Field

[0001] This invention relates to the field of pharmaceutical stability testing equipment technology, specifically to a pharmaceutical stability testing chamber control and data traceability device, and more particularly to a pharmaceutical stability testing chamber control and data traceability device capable of coupling stability testing protocols, environmental data acquisition, offset attribution results, adjustment commands, execution receipts, and traceability records. Background Technology

[0002] Drug stability test chambers are primarily used to conduct long-term stability tests, accelerated stability tests, or influencing factor tests on drugs, active pharmaceutical ingredients (APIs), packaging materials, or related samples under set temperature, humidity, light, and air circulation conditions. The results of stability tests directly affect the drug's shelf life, storage conditions, and quality evaluation. Therefore, during operation, the test chamber must not only maintain stable environmental parameters but also ensure complete recording and traceability of all test data.

[0003] Existing drug stability test chambers typically adjust automatically according to set temperature and humidity targets and can record data such as temperature, humidity, alarm information, and operator information. Some devices also have remote monitoring, access control, electronic recording, audit trail, and data export functions to meet the data management requirements of the drug testing process.

[0004] However, existing technologies still have the following shortcomings: First, the stage objectives, tolerance boundaries, sampling periods, and execution constraints in stability test schemes are usually only used as control parameters and do not form a verifiable data chain with subsequent data acquisition, control actions, and traceability records. Second, sensor-acquired values ​​usually directly participate in control and recording. When sensor drift, communication anomalies, or local sampling anomalies occur, it is difficult to distinguish between real environmental deviations and acquisition anomalies. Third, when environmental deviations occur in the test chamber, existing control methods mostly adjust based on the results of exceeding limits, making it difficult to distinguish the causes of door disturbances, actuator mismatches, and gradual environmental deviations. Fourth, existing audit trails mostly focus on recording "who, when, and what operation," while there is a lack of mandatory binding relationships between the acquisition summary, adjustment instruction summary, execution receipt, and prescription fingerprint during the test process, making it difficult to verify the consistency between control actions and test data. Fifth, in the event of network anomalies, power outage recovery, or recording anomalies, existing systems lack the ability to continue recording, freeze, lock, and externally verify test process data, which can easily affect the integrity and reliability of stability test data.

[0005] Therefore, it is necessary to provide a drug stability test chamber control and data traceability device that enables the stability test plan, reliable data acquisition, offset attribution, regulation control, execution receipt, and chain traceability record to form a closed-loop coupling, thereby improving the reliability of drug stability test process control and the credibility of data traceability. Summary of the Invention

[0006] The purpose of this invention is to provide a control and data traceability device for a drug stability test chamber, so as to solve the problems in existing drug stability test chambers where the test plan, data collection, control actions, execution feedback and traceability records are disconnected, making it difficult to accurately identify the cause of environmental deviation, difficult to strongly bind control actions with traceability records, difficult to freeze and lock abnormal records, and difficult to externally verify the consistency of test process data.

[0007] This invention achieves a closed-loop control and traceability system by coupling data between the prescription fingerprint module, the trusted acquisition and latching module, the offset attribution module, the evidence storage gating control module, and the chain-based traceability and freezing module. This system ensures that environmental data acquisition, offset attribution, adjustment instructions, execution receipts, and traceability hashes are generated in a fixed order within each sampling period, thereby realizing a closed-loop control and traceability system of "pre-sealing, releasing control, writing receipts, and freezing upon anomaly".

[0008] To achieve the aforementioned objective, this invention provides a drug stability test chamber control and data traceability device, comprising a prescription fingerprint module, a trusted acquisition and latching module, an offset attribution module, an evidence storage gating control module, and a chain-based traceability freezing module. The prescription fingerprint module is used to parse the stability test scheme, obtaining the stage sequence, environmental target, tolerance boundary, sampling period, execution constraints, and prescription fingerprint. The trusted acquisition and latching module is used to acquire environmental quantities, gate status, and actuator status within the test chamber in each sampling period, generating a trusted environment vector and acquisition summary, and latching the original sampling data. The offset attribution module is used to couple the trusted environment vector with the environmental target, tolerance boundary, gate status, actuator status, and control command from the previous sampling period, and output... The system outputs the offset type and offset strength. The evidence gating control module generates an adjustment instruction summary based on the offset type, offset strength, and execution constraints. After the chain-based traceability freezing module completes pre-sealing with the current acquisition summary, adjustment instruction summary, prescription fingerprint, and traceability hash value from the previous sampling period as input, it releases the corresponding adjustment instruction. The chain-based traceability freezing module receives the execution receipt of the adjustment instruction from the executor, writes the execution receipt to the current traceability record, and feeds the execution receipt back to the offset attribution module as the executor status input for the next sampling period. When any of the hash recalculation result, adjustment instruction summary consistency check result, or execution receipt consistency check result is abnormal, the corresponding traceability record is frozen, and overwriting the corresponding traceability record is prohibited.

[0009] Compared with the prior art, the present invention has at least the following beneficial effects:

[0010] 1. This invention uses a prescription fingerprint module to parse the stability test plan into a phase sequence, environmental target, tolerance boundary, sampling period, execution constraint, and prescription fingerprint. This allows the test plan to no longer exist merely as a control parameter, but as a basic field for subsequent data collection summary, adjustment instruction summary, execution receipt, and traceability hash generation, thereby improving the consistency between the test plan and process data.

[0011] 2. This invention uses a trusted acquisition latch module to acquire environmental quantities, gate status, and actuator status in each sampling cycle, and generates a trusted environment vector and acquisition summary. At the same time, it latches the original sampling data, so that subsequent control can be adjusted based on the trusted environment vector, while retaining the original sampling data as a traceability basis, reducing the impact of sensor anomalies, communication anomalies, or local sampling anomalies on control results and recording results.

[0012] 3. This invention couples the reliable environment vector with the environment target, tolerance boundary, gate state, actuator state and control command of the previous sampling period through the offset attribution module. It can distinguish between environmental gradual offset, gate disturbance offset, actuator mismatch offset and composite offset. Compared with simple over-limit alarm, it can more accurately identify the cause of the environmental anomaly in the stability test chamber.

[0013] 4. Before releasing the adjustment command, the present invention generates an adjustment command summary through the evidence storage gating control module, and the chain traceability freezing module completes the pre-sealing with the collection summary, adjustment command summary, prescription fingerprint and traceability hash value of the previous sampling period as input, forming a "first evidence storage, then control" gating mechanism, avoiding the data inconsistency problem caused by the control action being executed first and the traceability record being written later.

[0014] 5. This invention uses a chain-type traceability freezing module to write the actuator's execution receipt of the adjustment command into the current traceability record, and feeds the execution receipt back to the offset attribution module as the actuator status input for the next sampling cycle. This allows the actual response result of the actuator to participate in the offset attribution of the next cycle, thereby improving the accuracy of subsequent control judgments.

[0015] 6. When any verification of hash recalculation, adjustment instruction digest, or execution receipt is abnormal, the present invention can freeze the corresponding traceability record and prohibit overwrite of the corresponding traceability record, thereby preventing the existing record from being changed by overwrite after the anomaly occurs, and improving the integrity and auditability of the drug stability test process data.

[0016] 7. This invention generates a traceable chain that can be recalculated, verified, and frozen by associating prescription fingerprints, collection summaries, adjustment instruction summaries, execution receipts, and traceability hashes during the stability test, thereby improving the external verification capabilities of verifying the consistency of test process data, control actions, and traceability records.

[0017] 8. This invention is applicable to drug testing scenarios that require continuous control and long-term data retention, such as long-term stability testing, accelerated stability testing, and influencing factor testing. It can improve the reliability of drug stability testing chamber operation control and the credibility of data traceability without changing the basic environmental regulation function of the test chamber. Attached Figure Description

[0018] Figure 1 This is a schematic diagram of the structure of the drug stability test chamber provided by the present invention.

[0019] Figure 2 This is a block diagram of the control and data traceability device for the drug stability test chamber provided by the present invention. Detailed Implementation

[0020] The specific embodiments of the present invention will be further described below.

[0021] like Figure 1 As shown, this embodiment provides a drug stability test chamber, which includes a chamber body 1, a door 2, an observation window 3, a door sealing structure 4, an inner liner 5, a sample shelf 6, a sample tray 7, a first sensor 8, a second sensor 9, a third sensor 10, a circulating fan 11, an air duct assembly 12, a temperature control assembly 13, a humidity control assembly 14, an auxiliary processing assembly 15, a lighting assembly 16, a door detection assembly 17, a control panel 18, an equipment compartment door 19, and casters 20.

[0022] The chamber 1 constitutes the main outer shell and supporting structure of the drug stability test chamber. Inside the chamber 1 is an inner liner 5, which forms a test chamber for containing drug samples. A door 2 is located on the front side of the chamber 1, hinged to the chamber 1, allowing it to open and close relative to the chamber 1. An observation window 3 is provided on the door 2, allowing observation of the sample condition within the test chamber without opening the door 2. A door seal structure 4 is provided on the inner periphery of the door 2. When the door 2 is closed, the door seal structure 4 seals against the periphery of the front opening of the chamber 1, reducing the influence of the external environment on the temperature, humidity, light, and air circulation within the test chamber.

[0023] The inner liner 5 contains multiple layers of sample shelves 6, which are spaced apart vertically to hold drug samples of different batches or specifications. Sample trays 7 are mounted on the sample shelves 6 for holding drug samples, sample vials, or test containers. The sample shelves 6 have a perforated or mesh structure to allow circulating airflow to pass through them and flow evenly within the test chamber.

[0024] The test chamber is equipped with a first sensor 8, a second sensor 9, and a third sensor 10. These sensors are positioned at different locations within the test chamber to collect environmental parameters. These environmental parameters include one or more of temperature, humidity, light intensity, and air circulation status. The multi-point data collection by the first sensor 8, second sensor 9, and third sensor 10 provides multi-source sampling data for test chamber control and data traceability.

[0025] A circulating fan 11 and an air duct assembly 12 are installed at the rear of the test chamber. The circulating fan 11 drives the air circulation within the test chamber, and the air duct assembly 12 guides the circulating airflow along the test chamber. The circulating fan 11 and the air duct assembly 12 work together to maintain a uniform distribution of temperature, humidity, and airflow within the test chamber, thereby reducing environmental differences between different areas of the sample shelf 6.

[0026] The side of the chamber 1 is equipped with a temperature control component 13, a humidity control component 14, and an auxiliary processing component 15. The temperature control component 13 is used to heat or cool the test chamber; the humidity control component 14 is used to humidify or dehumidify the test chamber; the auxiliary processing component 15 is used in conjunction with the temperature control component 13 and the humidity control component 14 for media transportation, condensation treatment, drainage, or other auxiliary treatments. The temperature control component 13, the humidity control component 14, and the auxiliary processing component 15 work in conjunction with the air duct component 12 to allow the regulated air to enter the test chamber and participate in the circulation.

[0027] A lighting component 16 is installed at the top of the test chamber to provide stable illumination into the test chamber. The lighting component 16 can output a preset light intensity or light cycle according to the stability test plan to meet the lighting requirements of drug stability tests or influencing factor tests.

[0028] A door detection component 17 is installed near the connection point between the door 2 and the housing 1. The door detection component 17 is used to detect the open state, closed state, and opening duration of the door 2. The door state collected by the door detection component 17 can serve as a data source for the control device to identify door disturbances and offsets.

[0029] A control panel 18 is located at the lower front of the chamber 1. The control panel 18 displays the test plan, test chamber environmental parameters, actuator operating status, alarm information, and data traceability information, and receives control commands or stability test plans input by the operator. An equipment door 19 is also located at the lower front of the chamber 1. The equipment door 19 conceals and protects the lower control components, electrical components, or actuator status acquisition components, and facilitates inspection and maintenance. Casters 20 are located at the bottom of the chamber 1, supporting the chamber 1 and facilitating movement of the drug stability test chamber.

[0030] During drug stability testing, the control panel 18 receives or calls up the stability test plan. The first sensor 8, the second sensor 9, and the third sensor 10 collect environmental parameters within the test chamber. The door detection component 17 collects the door status of the chamber door 2. The temperature control component 13, humidity control component 14, auxiliary processing component 15, lighting component 16, circulating fan 11, and air duct component 12 adjust the temperature, humidity, lighting, and circulating air status within the test chamber according to control commands. Through the coordination of the above structures, the drug stability test chamber can provide a stable, controllable, and monitorable test environment for drug samples, and provide environmental data, door status data, and actuator status data for the control and data traceability devices.

[0031] like Figure 2 As shown, this embodiment provides a drug stability test chamber control and data traceability device, which includes a prescription fingerprint module, a trusted acquisition and latching module, an offset attribution module, an evidence storage gating control module, and a chain traceability freezing module.

[0032] The prescription fingerprint module is used to analyze the stability test scheme and obtain the stage sequence, environmental target, tolerance boundary, sampling period, execution constraint and prescription fingerprint.

[0033] The trusted acquisition and latching module is used to acquire environmental quantities, door status, and actuator status within the test chamber in each sampling period, generate a trusted environment vector and acquisition summary, and latch the original sampling data.

[0034] The offset attribution module is used to couple the reliable environment vector with the environment target, tolerance boundary, gate state, actuator state and control command of the previous sampling period, and output the offset type and offset intensity.

[0035] The evidence storage gating control module is used to generate an adjustment instruction summary based on the offset type, offset intensity, and execution constraints, and release the corresponding adjustment instruction after the chain-style traceability freezing module completes the pre-sealing with the current collection summary, adjustment instruction summary, prescription fingerprint, and traceability hash value of the previous sampling period as input.

[0036] The chain-based traceability freezing module is used to receive the execution receipt of the actuator for the adjustment command, write the execution receipt into the current traceability record, and feed the execution receipt back to the offset attribution module as the actuator status input for the next sampling period. When any of the hash recalculation result, adjustment command digest consistency check result or execution receipt consistency check result is abnormal, the corresponding traceability record is frozen and overwrite writing to the corresponding traceability record is prohibited.

[0037] Furthermore, in this embodiment, the reliable environmental quantities of temperature, humidity, illumination, and circulating air acquired by the reliable acquisition latch module correspond to the temperature adjustment, humidity adjustment, illumination adjustment, and circulating air adjustment quantities output by the evidence-gated gate control module, respectively. The gate status serves as the basis for judging gate disturbance offset, and the actuator status and execution receipt serve as the basis for judging execution mismatch offset, and are fed back to participate in the generation of adjustment quantities for the next sampling cycle. Thus, a closed-loop coupling relationship is formed between the measured quantities, offset attribution results, control quantities, and execution feedback of the test chamber.

[0038] In this embodiment, the data coupling relationship between the modules is as follows: the prescription fingerprint output by the prescription fingerprint module serves as the basic field for traceability records in each sampling period; the trusted environment vector output by the trusted acquisition latch module is used for offset attribution, and the output acquisition summary is used for chained traceability; the offset type and offset intensity output by the offset attribution module are used to generate adjustment instruction summaries; the adjustment instruction summaries generated by the evidence gating control module are first pre-sealed by the chained traceability freezing module before the corresponding adjustment instruction is released; after receiving the execution receipt, the chained traceability freezing module generates the current traceability hash value and feeds the execution receipt back to the offset attribution module as the actuator state input for the next sampling period. This forms a closed loop from the experimental plan to control execution and then to traceability freezing.

[0039] After receiving the stability test plan, the prescription fingerprint module first identifies the test name, sample batch number, test type, test stage, stage sequence, stage start and end conditions, temperature target, humidity target, light target, circulating air target, tolerance boundary, sampling period, and execution constraints in the test plan. Among them, the test type can include long-term stability test, accelerated stability test, and influencing factor test.

[0040] The prescription fingerprint module converts each trial phase into structured phase data. This structured phase data includes the first... The environmental target vector, tolerance boundary vector, sampling period, and execution constraint vector are defined for each experimental phase. The environmental target vector may include temperature target, humidity target, illumination target, and circulating air target; the tolerance boundary vector may include temperature tolerance boundary, humidity tolerance boundary, illumination tolerance boundary, and circulating air tolerance boundary; the execution constraint vector may include the maximum adjustment range, minimum adjustment interval, prohibited adjustment conditions, and linkage adjustment sequence for each actuator.

[0041] The prescription fingerprint module generates prescription fingerprints according to the following formula: , in, For the first Prescription fingerprints from the trial phase, Numbering for the experimental phase, This is a one-way hash operation function. This is a summary of the stability test protocol. For the first Environmental target vectors for each experimental phase For the first The tolerance boundary vector for each experimental phase. For the first The sampling period for each experimental phase, For the first The execution constraint vector for each experimental phase. This indicates that the fields will be concatenated according to a preset field order.

[0042] In the above formula, and All with the Each experimental phase corresponds to; when The difference indicates that the environmental objectives and tolerance boundaries differ at different stages of the experiment. The prescription fingerprint module will... The data is sent to the trusted acquisition latch module, the evidence storage gating control module, and the chain-based traceability freezing module, so that the data processing of each sampling cycle is bound to the current test phase.

[0043] Furthermore, the prescription fingerprint module can also be configured with a stage switching buffer window. This buffer window is used to buffer and release adjustments to temperature, humidity, light intensity, and circulating air when environmental targets change in adjacent test stages. Specifically, when the stage switching buffer window... The first experimental phase has been switched to the second. During the trial phase, the prescription fingerprint module generates a phase switching identifier and sends it to the evidence storage gating control module. The evidence storage gating control module limits the adjustment range of each actuator within the phase switching buffer window, ensuring a smooth transition of environmental variables from the previous phase target to the next. This method avoids environmental overshoot or frequent alarms caused by excessive changes in target values ​​during phase switching.

[0044] The purpose of this phase switching buffer window is to create a smooth coupling between prescription changes and control actions. Its beneficial effect is that it can reduce temperature and humidity overshoot, sudden changes in light intensity, and circulating wind disturbances caused by phase switching in long-term stability tests, thereby improving the continuity of the test process.

[0045] The reliable data acquisition and latching module collects environmental parameters, door status, and actuator status within the test chamber during each sampling period. The environmental parameters include temperature, humidity, light intensity, and circulating airflow. The door status includes door open status, door closed status, opening start time, closing time, opening duration, and number of openings. The actuator status includes temperature regulation actuator status, humidity regulation actuator status, light intensity regulation actuator status, and circulating airflow regulation actuator status.

[0046] The trusted acquisition latch module latches the raw sampled values ​​from each sensor. During latching, the trusted acquisition latch module generates raw sampled data records according to the sampling period number, sensor identifier, sampling time, and raw sampled value. The raw sampled data records are stored in an append-only manner, and overwriting or modification in the original location is not allowed. If it is necessary to explain abnormal sampled values, only append-generated explanation records are allowed.

[0047] When the same type of environmental quantity corresponds to multiple sensors, the trusted acquisition latch module generates the trusted environmental quantity in the trusted environmental vector according to the following formula: , in, For the first Environmental quantity in the first The reliable environmental quantity for each sampling period Environmental quantity category number, Numbering the sampling period For the first The number of sensors corresponding to each type of environmental quantity. Number the sensor. For the first The first type of environmental quantity The sensor at the first The confidence weight for each sampling period For the first The first type of environmental quantity The sensor at the first The original sampled values ​​for each sampling period.

[0048] in, These can represent temperature, humidity, light intensity, or circulating air, respectively; when When expressing temperature, Indicates the first The reliable environmental temperature quantity for each sampling period; when When indicating humidity, Indicates the first Confidential environmental humidity values ​​for each sampling period.

[0049] The trusted acquisition latch module updates the trusted weights based on the consistency residuals of similar sensors, the continuity residuals of adjacent samples, the calibration validity status, and the communication validity status. Specifically, the consistency residuals of similar sensors are used to represent the weights of the first sample in the same category of environmental quantities. The difference between the original sampled value of a sensor and the sampled value of other sensors of the same type; the adjacent sampling continuity residual is used to represent the change range between the current sampled value of the same sensor and the sampled value of the previous sampling period; the calibration validity status is used to indicate whether the sensor is within the valid calibration period; the communication validity status is used to indicate whether the sensor communication is continuous and complete.

[0050] Furthermore, the reliable acquisition and latching module can also be equipped with a sensor spatial thermal inertia correction unit. This unit records the installation position of different sensors within the test chamber, their distance from the air supply vent, the return air vent, the door, and their shelf level. It also performs spatial consistency verification on the original sampled values ​​based on the sensor's installation position. For example, if a temperature sensor near the door experiences temperature fluctuations shortly after the door is opened, the reliable acquisition and latching module does not directly classify it as a sensor malfunction. Instead, it generates a spatial thermal inertia description based on the door's condition and the sensor's position. Similarly, if a humidity sensor near the air supply vent changes its humidity before other sensors after humidification, the reliable acquisition and latching module identifies this change as a difference in positional response, rather than immediately classifying it as an overall environmental shift.

[0051] The function of this sensor spatial thermal inertia correction unit is to couple the sensor's spatial installation position with the trusted weight update process. Its beneficial effect is that it can reduce misjudgments caused by differences in sensor position, airflow path, and door disturbances, thereby improving the accuracy of the trusted environment vector.

[0052] When the confidence weight of a sensor falls below a preset confidence threshold, the confidence acquisition latch module removes the corresponding sensor's original sampled value from the confidence environment quantity calculation and writes the removed original sampled value, confidence weight, removal reason, removal time, and sensor identifier into the acquisition summary. The acquisition summary includes at least the sampling period number, confidence environment vector, sensor identifiers involved in the calculation, sensor identifiers removed, gate status, actuator status, original sampled data latch location, and sensor spatial thermal inertia description.

[0053] The role of the trusted acquisition latch module is to generate a trusted environment vector without losing the original sampling data. Its benefits include ensuring that the data used for control and judgment is more reliable, and that complete original data and the basis for exclusion can be viewed during subsequent tracing.

[0054] The offset attribution module reads the reliable environment vector, environment target, tolerance boundary, gate state, actuator state, and control command from the previous sampling period, and calculates the offset intensity. The offset attribution module does not simply determine whether a certain environmental quantity exceeds the tolerance boundary, but rather determines the correlation between environmental offset and gate disturbance, actuator response, and control action from the previous sampling period.

[0055] The offset attribution module calculates the offset strength according to the following formula: , in, For the first The offset intensity per sampling period Numbering the sampling period This represents the total number of environmental quantity categories. Environmental quantity category number, For the first Offset weights of class environment quantities For the first Environmental quantity in the first The reliable environmental quantity for each sampling period For the first Environmental quantity in the first Environmental objectives for each experimental phase Numbering for the experimental phase, For the first Environmental quantity in the first The tolerance boundaries for each experimental phase. The gate perturbation weight, For the first Gate disturbance amount per sampling period To implement mismatch weights, For the first The execution mismatch amount per sampling period.

[0056] in, Indicates the first The effective deviation of the environmental quantity after it exceeds the tolerance boundary; when the first When the environmental quantity does not exceed the tolerance boundary, this item is set to zero. According to the The door opening status, opening duration, and number of openings are generated within each sampling period. According to the The matching relationship between the actuator state and the control command of the previous sampling period is generated within each sampling period.

[0057] The offset attribution module determines the offset type as environmentally gradual offset, door disturbance offset, execution mismatch offset, or composite offset based on the offset intensity and the proportion of door disturbance and execution mismatch. If the offset intensity is mainly caused by the continuous deviation of the trusted environmental quantity from the environmental target, and neither the door disturbance nor the execution mismatch reaches the corresponding threshold, it is determined to be an environmentally gradual offset; if the offset intensity is mainly related to the duration or number of door openings, it is determined to be a door disturbance offset; if the offset intensity is mainly related to the actuator not responding, insufficient response, or inconsistent response direction, it is determined to be an execution mismatch offset; if at least two of the above factors simultaneously reach the corresponding threshold, it is determined to be a composite offset.

[0058] Furthermore, the offset attribution module can also be configured with a door disturbance recovery profile unit. This unit records the environmental recovery process after each door opening, including the opening duration, temperature recovery time after closing, humidity recovery time, light recovery time, and air circulation recovery time, as well as whether an adjustment command is triggered during the recovery process. The offset attribution module generates a door disturbance recovery profile based on multiple door disturbance recovery processes. This profile is used to determine whether the current door disturbance is a naturally recoverable disturbance or whether the door control module needs to intervene for adjustment.

[0059] For example, when the door opening duration is short and the historical door disturbance recovery profile shows that such disturbances can usually recover on their own within a preset recovery time, the offset attribution module can determine the current offset type as door disturbance offset and output a low-intensity recovery suggestion to the evidence-based door control module; when the door opening duration is long and the historical recovery profile shows that similar disturbances are prone to causing long-term humidity deviations, the offset attribution module can determine the current offset type as composite offset and increase the offset intensity.

[0060] The function of this door disturbance recovery profiling unit is to couple the recovery process after historical door opening disturbances with the current offset attribution process. Its beneficial effect is that it can avoid excessive control immediately after a short period of door opening, and can also promptly identify environmental offsets requiring active adjustment after a long period of door opening.

[0061] In this embodiment, the offset attribution module can be implemented using rule-based calculation, or an auxiliary attribution model can be introduced based on rule-based calculation. If an auxiliary attribution model is introduced, the training method is as follows: training samples are extracted from historical stability test records. Each training sample includes a reliable environment vector, an environment target, a tolerance boundary, a gate state, an actuator state, the control command of the previous sampling period, a gate disturbance recovery profile, and a manually confirmed offset type. The manually confirmed offset type is used as the training label to train the auxiliary attribution model, ensuring that the offset type output by the auxiliary attribution model is consistent with the training label. After training, the auxiliary attribution model only serves as an auxiliary judgment result for the offset attribution module; the final offset type is still determined by combining the offset intensity, the gate disturbance amount, and the execution mismatch amount. This method can improve the recognition accuracy in complex offset scenarios while maintaining interpretability.

[0062] The evidence storage gating control module reads the offset type and offset intensity output by the offset attribution module, and reads the execution constraints output by the prescription fingerprint module. Based on the offset type, the evidence storage gating control module determines the linkage adjustment objects for temperature, humidity, light, and circulating air; based on the offset intensity, it determines the adjustment range for each linkage adjustment object; and based on the execution constraints, it limits the adjustment range.

[0063] When the offset type is a gradual environmental change offset, the evidence-gated control module generates a progressive adjustment command to gradually return the environmental quantity to the environmental target. When the offset type is a door disturbance offset, the evidence-gated control module determines whether to wait for natural recovery or generate a low-amplitude recovery command based on the door disturbance recovery profile. When the offset type is an execution mismatch offset, the evidence-gated control module generates a protective adjustment command and writes the execution mismatch information into the adjustment command summary. When the offset type is a composite offset, the evidence-gated control module performs linkage limiting control on temperature, humidity, light intensity, and circulating air according to the execution constraints.

[0064] Furthermore, the evidence-gated control module can also be configured with a shadow correction recording unit. This shadow correction recording unit generates at least one unreleased candidate correction scheme before the formal release of the adjustment command, and calculates the expected recovery direction, expected recovery time, and expected actuator load corresponding to each candidate correction scheme. The evidence-gated control module selects the scheme from the candidate correction schemes that satisfies the execution constraints and has the most stable adjustment amplitude to generate an adjustment command summary. Unreleased candidate correction schemes do not act on the actuator; they are only written as extended fields of the traceability record as shadow correction records.

[0065] The purpose of this shadow correction recording unit is to record alternative adjustment schemes without affecting the actual control process. Its advantage is that when it is necessary to review the rationality of a control action later, the candidate correction schemes that were not released at that time can be viewed, thereby improving the interpretability of control decisions.

[0066] After generating the adjustment instruction summary, the evidence storage gating control module does not immediately release the adjustment instruction. Instead, it sends the collection summary, adjustment instruction summary, and prescription fingerprint to the chain-based traceability freezing module. The adjustment instruction summary includes at least the sampling period number, offset type, offset intensity, linked adjustment object, adjustment range, execution constraint reference identifier, adjustment instruction generation time, stage switching buffer window status, and shadow correction record identifier.

[0067] After the chain-based traceability freezing module completes the pre-sealing, it returns a pre-sealing success flag to the evidence storage gating control module. The evidence storage gating control module only releases the adjustment command corresponding to the adjustment command summary upon receiving the pre-sealing success flag. If it does not receive the pre-sealing success flag, the evidence storage gating control module does not release the formal adjustment command and generates a record of the reason for non-release.

[0068] The function of the evidence storage gating control module is to bind the generation, summarization, pre-sealing, and release of control commands in sequence. Its advantage lies in avoiding data inconsistencies caused by controlling actions being executed first and then retroactively recorded.

[0069] The chain-based traceability freeze module receives prescription fingerprints, collection summaries, adjustment instruction summaries, trusted timestamps, execution receipts, and exception identifiers, and generates pre-sealed hash values ​​and current traceability hash values. The trusted timestamp represents the trusted time information formed by the corresponding record within the sampling period.

[0070] The chain-based traceability freeze module generates the pre-sealed hash value and the current traceability hash value according to the following formula: , , in, For the first The pre-sealed hash value for each sampling period, For the first The current trace hash value for each sampling period. Numbering the sampling period This is a one-way hash operation function. For the first The trace hash value for each sampling period, For the first Prescription fingerprints from the trial phase, Numbering for the experimental phase, For the first Summary of data collection for each sampling period For the first Summary of adjustment instructions for each sampling period For the first A reliable timestamp for each sampling period. For the first Summary of execution receipts for each sampling period For the first Anomaly identifier for each sampling period, This indicates that the fields will be concatenated according to a preset field order.

[0071] in, From the trusted acquisition latch module, From the evidence storage gating control module, Feedback from the actuator on the actual execution of the control command. Anomaly verification results from the chain-based traceability freeze module. It is a pre-sealed result generated before the adjustment command is released. It is the complete traceability result generated after the execution receipt is written.

[0072] The execution receipt may include the actuator identifier, the time of receiving the adjustment command, the execution start time, the execution end time, the execution status, the actual output value, and the execution result identifier. After the chain-based traceability freeze module writes the execution receipt into the current traceability record, it feeds the execution receipt back to the offset attribution module. The offset attribution module reads the execution receipt in the next sampling period and uses it as the actuator status input to calculate the execution mismatch.

[0073] Furthermore, the chain-based traceability freezing module can also be equipped with a traceability risk coloring unit. This unit generates risk identifiers for the traceability records of each sampling period based on hash recalculation results, adjustment instruction digest consistency, execution receipt consistency, sensor rejection counts, gate disturbance frequency, and execution mismatch counts. Risk identifiers can include normal identifiers, warning identifiers, pending review identifiers, and frozen identifiers. The traceability risk coloring unit does not alter the original traceability record content; instead, it appends risk identifier records after the original traceability record.

[0074] The purpose of this traceability risk coloring unit is to stratify the risk of a large number of traceability records in long-term stability testing. Its benefit is that external auditors or reviewers can quickly identify high-risk sampling periods without having to review every single record.

[0075] When the hash recalculation result of any trace record is inconsistent with the stored trace hash value, or the adjustment instruction digest is inconsistent with the actual released adjustment instruction, or the execution receipt is inconsistent with the released adjustment instruction, the chained trace freezing module marks the corresponding trace record as a frozen record, marks the trace records after the frozen record that are related to the same test phase as records to be reviewed, and prohibits overwriting the frozen record and the records to be reviewed. When generating a correction description record, only append-only correction description records are allowed, and the correction description record is bound to the trace hash value of the frozen record.

[0076] At the end of each trial phase, the chain-linked traceability freeze module generates a phase sealing record. This record includes a phase identifier, phase start and end times, prescription fingerprint, first traceability hash value, last traceability hash value, anomaly identifier set, and operator identifier. The chain-linked traceability freeze module uses the sealing hash value of the phase sealing record as the preceding hash value of the first traceability record in the next trial phase.

[0077] Furthermore, the phase sealing record may also include a phase switching buffer window record. The phase switching buffer window record is used to record changes in environmental targets, adjustment limits, buffer duration, and actual environmental recovery status during phase switching. When the... The first experimental phase has entered the [stage name missing]. During the first trial phase, the chain-based traceability freezing module will... Phase sealing records of the first test phase and the first The first traceability record of each trial phase is linked together. This way, the traceability chain will not be broken even if the trial phase changes.

[0078] The purpose of this phase-based sealing and cross-phase transition method is to ensure data continuity between different experimental phases. Its beneficial effect is that it prevents the hiding of abnormal records or the reconstruction of the traceability chain through phase switching.

[0079] The chain-based traceability freeze module includes an offline append-only buffer. This offline append-only buffer is used to cache unuploaded collection summaries, adjustment instruction summaries, execution receipts, and trusted timestamps in the order of the sampling period after a communication failure or power outage recovery.

[0080] When communication between the device and the external audit server fails, the chain-based traceability freezing module still generates pre-sealed hash values ​​and the current traceability hash value locally, and writes the unuploaded records to the offline append-only cache. After communication is restored, the chain-based traceability freezing module starts from the traceability hash value of the previous sampling period, continues to generate traceability hash values ​​in the order of the sampling periods, and uploads the corresponding records.

[0081] When the device resumes operation after a power outage, the chain-based traceability freeze module reads the last valid traceability hash value before the power outage and uses this traceability hash value as the preceding hash value for the traceability record of the first sampling period after recovery. If there are records that have been pre-sealed but have not yet been written with execution receipts before the power outage, the chain-based traceability freeze module marks these records as records to be reviewed and generates a power outage recovery description record.

[0082] Furthermore, offline append-only buffers can also have buffer segment boundary markers set. These markers are used to mark the start time of communication failure, the start time of communication recovery, the last record before power failure, and the first record after recovery. These boundary markers participate in the generation of subsequent external verification packets, enabling external verification endpoints to determine whether the offline buffer segments are sequentially continuated.

[0083] The purpose of this offline, append-only buffer is to maintain the continuity of the traceability chain in the event of communication failures or power outages. Its benefit lies in reducing the risk of data loss due to network anomalies or power outages during long-term stability testing.

[0084] The chain-based traceability freezing module also includes an external verification package generation unit. This unit generates external verification packages, which include a trial protocol summary, prescription fingerprint sequence, traceability hash value sequence, stage sealing records, freeze records, and hash recalculation rules.

[0085] After receiving the external verification packet, the external verification terminal recalculates the prescription fingerprint, collection summary, adjustment instruction summary, trusted timestamp, execution receipt summary, and anomaly identifier according to the hash recalculation rules, and compares the recalculation results with the traceability hash value sequence. When the recalculation results are consistent, it indicates that there is consistency between the test process data, control actions, and traceability records in the corresponding sampling period; when the recalculation results are inconsistent, it indicates that there are records in the corresponding sampling period that need to be verified.

[0086] Furthermore, the external validation package generation unit can also generate cross-chamber comparative validation data. This cross-chamber comparative validation data is used to compare the prescription fingerprints, stage switching buffer window records, door disturbance recovery profiles, offset type distributions, and the number of frozen records between different test chambers when performing the same or similar stability test protocols on multiple drug stability test chambers. The cross-chamber comparative validation data does not directly change the control actions of a single test chamber; it is only used to identify abnormal differences between different test chambers for the same test protocol.

[0087] The purpose of this cross-chamber comparison verification data is to conduct a cross-check of the stability testing process across multiple test chambers. Its beneficial effect is that it can detect long-term drift, actuator performance degradation, or abnormal sensor placement that are not easily apparent in a single test chamber, thereby improving the management reliability of batch stability testing.

[0088] The drug stability test chamber control and data traceability device of this embodiment can be operated according to the following steps:

[0089] S1, the prescription fingerprint module receives the stability test plan and parses it to obtain the stage sequence, environmental objectives, tolerance boundaries, sampling period, execution constraints, and prescription fingerprint. The purpose of this step is to convert the stability test plan into a traceable prescription fingerprint, which has the benefit of binding subsequent data collection, control, and traceability records to specific test stages.

[0090] S2, the prescription fingerprint module determines whether a phase switch exists. If a phase switch exists, a phase switch buffer window record is generated and sent to the evidence storage gating control module and the chain traceability freeze module. The purpose of this step is to smoothly handle the target changes between adjacent test phases, and its beneficial effect is to reduce the risk of environmental overshoot and traceability breakpoints during phase switches.

[0091] S3, the reliable data acquisition and latching module acquires temperature, humidity, light intensity, circulating air, door status, and actuator status according to the sampling period, and latches the original sampled data. The purpose of this step is to form a complete original data foundation, which provides irreplaceable original evidence for subsequent verification.

[0092] S4, the trusted acquisition latch module generates a trusted environment vector based on the sensor's trusted weights and combines it with the sensor's spatial thermal inertia correction results to generate an acquisition summary. This step aims to create trusted data for offset attribution and tracing records, effectively reducing the impact of sensor position differences and local anomalies on the control results.

[0093] S5, the offset attribution module reads the reliable environment vector, environment target, tolerance boundary, gate state, actuator state, and control command from the previous sampling period, calculates the offset intensity, and determines the offset type. The purpose of this step is to identify the cause of environmental offset, and its beneficial effect is that subsequent control can adopt different adjustment strategies for different offset types.

[0094] S6, the offset attribution module combines the door disturbance recovery profile to determine whether the current offset can be naturally recovered. If it can be naturally recovered, a low-intensity recovery suggestion is output; if it cannot be naturally recovered, an active adjustment suggestion is output. The purpose of this step is to avoid over-control caused by short-term door opening disturbances, and its beneficial effect is to improve control stability.

[0095] In step S7, the evidence-gated control module generates a summary of adjustment instructions based on the offset type, offset intensity, and execution constraints, and generates a shadow correction record. This step serves to form formal adjustment instructions and candidate correction schemes, and its beneficial effect is to improve the interpretability of control actions.

[0096] In step S8, the evidence storage gating control module sends the collection summary, adjustment instruction summary, and prescription fingerprint to the chain traceability freezing module. The chain traceability freezing module generates a pre-sealed hash value and returns a pre-sealing success identifier. The purpose of this step is to complete the pre-sealing before the adjustment instruction is released, and its beneficial effect is to achieve "evidence storage first, control later".

[0097] S9: Upon receiving the pre-sealing success flag, the evidence-gated control module releases the corresponding adjustment command. The temperature, humidity, light, and circulating air actuators then execute their respective actions according to the command. This step completes environmental adjustment, and its beneficial effect is to provide clear data collection, attribution, and pre-sealing evidence for the adjustment actions.

[0098] S10, the chain-based traceability freeze module receives the execution receipt, writes the execution receipt into the current traceability record, and generates the current traceability hash value. The purpose of this step is to write the execution result into the traceability chain, which makes the adjustment instructions and execution feedback recalculated and verifiable.

[0099] In step S11, the chain-tracking freeze module feeds back the execution receipt to the offset attribution module as the actuator status input for the next sampling cycle. This step aims to use the actual actuator response to inform the offset judgment in the next cycle, thereby improving the accuracy of subsequent execution mismatch identification.

[0100] In step S12, the chain-based traceability freeze module performs hash recalculation, adjustment instruction digest consistency verification, and execution receipt consistency verification. If any verification fails, a freeze record and a traceability risk coloring record are generated. The purpose of this step is to discover and freeze abnormal records, which helps prevent historical records from being overwritten or modified.

[0101] S13, When the test phase ends, the chain-based traceability freeze module generates a phase sealing record and uses the sealing hash value of the phase sealing record as the preceding hash value of the first traceability record in the next test phase. The purpose of this step is to maintain traceability continuity between test phases, and its beneficial effect is to prevent the traceability chain from being rebuilt or replaced through phase switching.

[0102] S14, when communication fails or power outages are recovered, the chain-based tracing freeze module continues the tracing records offline by only appending them to the buffer, and generates an external verification package after communication is restored. This step ensures data continuity in abnormal scenarios, and its beneficial effect is reducing the risk of data chain disruptions during long-term stability testing.

[0103] This implementation method primarily achieves its purpose through rule computation, hash recalculation, and consistency verification, without relying on a black-box model that must be trained. If an auxiliary attribution model or candidate correction evaluation model is introduced in the specific implementation, the following training method can be adopted.

[0104] The training samples for the auxiliary attribution model are derived from historical stability test records. Each training sample includes a reliable environment vector, environment target, tolerance boundary, gate state, actuator state, control command from the previous sampling period, gate disturbance recovery profile, and manually confirmed offset type. The training label is the manually confirmed offset type. The training objective is to ensure that the offset type output by the auxiliary attribution model is consistent with the training label. After training, the auxiliary attribution model only outputs the auxiliary offset type; the final offset type is still determined by the offset attribution module based on the offset intensity, gate disturbance amount, and execution mismatch.

[0105] The training samples for the candidate correction evaluation model are derived from historical adjustment records. Each training sample includes the offset type, offset intensity, execution constraint, actual adjustment command, execution receipt, post-adjustment recovery time, and whether overshoot occurred. The training label is the adjustment effect level. The training objective is to enable the candidate correction evaluation model to evaluate the expected recovery effect of candidate correction schemes. After training, the candidate correction evaluation model is only used by the shadow correction record unit to generate candidate correction schemes and does not directly bypass the evidence gating control module to release adjustment commands.

[0106] Through the above training method, the model output is limited to the auxiliary attribution and candidate evaluation stages, without disrupting the interpretable data chain between prescription fingerprints, trusted collection, evidence gating, and chain-based traceability freezing.

[0107] This implementation uses a prescription fingerprint module to incorporate the stage sequence, environmental objectives, tolerance boundaries, sampling period, and execution constraints in the stability test plan into the generation of subsequent traceability records; a trusted acquisition latch module to latch the original sampled data and generate a trusted environmental vector and acquisition summary; a sensor spatial thermal inertia correction unit to incorporate sensor installation position, air supply path, and door distance into trusted weighting judgment; an offset attribution module and a door disturbance recovery profiling unit to distinguish between environmental gradual offset, door disturbance offset, execution mismatch offset, and composite offset; and an evidence-based gate control module... The system ensures that adjustment instructions are released only after successful pre-sealing; the shadow correction recording unit allows unreleased candidate solutions to be recorded and used for subsequent review; the chain-based traceability freeze module ensures that execution receipts, anomaly markers, and stage sealing records enter the same traceability chain; the traceability risk coloring unit allows a large number of long-term test records to be stratified according to risk level; the offline append-only buffer and external verification package generation unit ensures traceability continuity in scenarios such as communication anomalies, power outage recovery, and external audits; and cross-compartment comparison verification data allows long-term offsets and execution anomalies between multiple test cases to be detected laterally.

[0108] This implementation method can achieve the following overall beneficial effects: The present invention uses the prescription fingerprint module to parse the stability test plan into a stage sequence, environmental target, tolerance boundary, sampling period, execution constraint and prescription fingerprint, so that the test plan no longer exists only as a control parameter, but also as a basic field for subsequent collection summary, adjustment instruction summary, execution receipt and traceability hash generation to participate in traceability, thereby improving the consistency between the test plan and process data.

[0109] This invention uses a trusted acquisition latch module to acquire environmental quantities, gate status, and actuator status in each sampling cycle, and generates a trusted environment vector and acquisition summary. At the same time, it latches the original sampled data, so that subsequent control can be adjusted based on the trusted environment vector, while retaining the original sampled data as a traceability basis, reducing the impact of sensor anomalies, communication anomalies, or local sampling anomalies on control results and recording results.

[0110] This invention couples the reliable environment vector with the environmental target, tolerance boundary, gate state, actuator state and control command of the previous sampling period through the offset attribution module. It can distinguish between environmental gradual offset, gate disturbance offset, actuator mismatch offset and composite offset. Compared with simple over-limit alarm, it can more accurately identify the cause of environmental anomalies in the stability test chamber.

[0111] This invention uses a gating control module to generate a summary of the adjustment instruction before releasing the adjustment instruction. The chain-like traceability freezing module then pre-seales the summary, the adjustment instruction summary, the prescription fingerprint, and the traceability hash value of the previous sampling period as input, forming a "storage first, control later" gating mechanism. This avoids the data inconsistency caused by the control action being executed first and the traceability record being written later.

[0112] This invention uses a chain-style traceability freezing module to write the actuator's execution receipt of the adjustment command into the current traceability record, and feeds the execution receipt back to the offset attribution module as the actuator status input for the next sampling cycle. This allows the actual response result of the actuator to participate in the offset attribution of the next cycle, thereby improving the accuracy of subsequent control judgments.

[0113] When any verification of hash recalculation, adjustment instruction digest, or execution receipt is abnormal, the present invention can freeze the corresponding traceability record and prohibit overwrite operations on the corresponding traceability record, thereby preventing the existing record from being changed by overwrite after the anomaly occurs, and improving the integrity and auditability of the data in the drug stability test process.

[0114] This invention generates a traceability chain that is recalculated, verifiable, and freezeable by associating prescription fingerprints, collection summaries, adjustment instruction summaries, execution receipts, and traceability hashes during the stability test. This improves the external verification capabilities of verifying the consistency of test process data, control actions, and traceability records.

[0115] This invention is applicable to drug testing scenarios that require continuous control and long-term data retention, such as long-term stability testing, accelerated stability testing, and influencing factor testing. It can improve the reliability of drug stability testing chamber operation control and the credibility of data traceability without changing the basic environmental regulation function of the test chamber.

[0116] The above specific embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit the scope of protection of the present invention. Without departing from the technical concept of the present invention, those skilled in the art can make equivalent substitutions or conventional adjustments to the module deployment location, number of sensors, actuator type, field concatenation order, hash operation function, cache medium, and external verification method; any technical solution that uses the data coupling relationship between prescription fingerprinting, trusted acquisition latching, offset attribution, evidence gating control, and chain-based traceability freezing to achieve control and data traceability of the drug stability test chamber should fall within the scope of protection of the present invention.

Claims

1. A control and data traceability device for a drug stability test chamber, characterized in that, The system includes a prescription fingerprint module, a trusted acquisition and latching module, an offset attribution module, an evidence storage gating control module, and a chain-based traceability and freezing module. The prescription fingerprint module is used to parse the stability test plan, obtaining the stage sequence, environmental target, tolerance boundary, sampling period, execution constraints, and prescription fingerprint. The trusted acquisition and latching module is used to acquire environmental quantities, door status, and actuator status within the test chamber in each sampling period, generating a trusted environment vector and acquisition summary, and latching the original sampling data. The offset attribution module is used to correlate the trusted environment vector with the environmental target, tolerance boundary, door status, actuator status, and... The control command of the previous sampling period is coupled to output the offset type and offset intensity; the evidence gating control module is used to generate an adjustment command summary based on the offset type, offset intensity and execution constraints, and release the corresponding adjustment command after the chain traceability freezing module completes the pre-sealing with the current acquisition summary, adjustment command summary, prescription fingerprint and traceability hash value of the previous sampling period as input; the chain traceability freezing module is used to receive the execution receipt of the adjustment command from the actuator, write the execution receipt into the current traceability record, and feed the execution receipt back to the offset attribution module as the actuator status input for the next sampling period; If any of the hash recalculation result, adjustment instruction digest consistency check result, or execution receipt consistency check result is abnormal, the corresponding trace record is frozen, and overwrite writes to the corresponding trace record are prohibited.

2. The drug stability test chamber control and data traceability device according to claim 1, characterized in that, The prescription fingerprint module generates the prescription fingerprint according to the following formula: ， in, For the first Prescription fingerprints from the trial phase, Numbering for the experimental phase, This is a one-way hash operation function. This is a summary of the stability test protocol. For the first Environmental target vectors for each experimental phase For the first The tolerance boundary vector for each experimental phase. For the first The sampling period for each experimental phase, For the first The execution constraint vector for each experimental phase. This indicates that the fields will be concatenated according to a preset field order.

3. The drug stability test chamber control and data traceability device according to claim 2, characterized in that, The trusted acquisition latch module generates the trusted environment quantity in the trusted environment vector according to the following formula: ， in, For the first Environmental quantity in the first The reliable environmental quantity for each sampling period Environmental quantity category number, Numbering the sampling period For the first The number of sensors corresponding to each type of environmental quantity. Number the sensor. For the first The first type of environmental quantity The sensor at the first The confidence weight for each sampling period For the first The first type of environmental quantity The sensor at the first The original sampled values ​​for each sampling period; the reliable environment vector includes reliable environmental quantities for temperature, humidity, illumination, circulating air, door status, and actuator status.

4. The drug stability test chamber control and data traceability device according to claim 3, characterized in that, The trusted acquisition latch module updates the trusted weight based on the consistency residual of similar sensors, the continuity residual of adjacent samples, the calibration validity status, and the communication validity status. When the trusted weight is lower than the preset trusted threshold, the trusted acquisition latch module removes the original sampled value of the corresponding sensor from the trusted environmental quantity calculation and writes the removed original sampled value, trusted weight, removal reason, removal time, and sensor identifier into the acquisition summary.

5. The drug stability test chamber control and data traceability device according to claim 3, characterized in that, The offset attribution module calculates the offset intensity according to the following formula: ， in, For the first The offset intensity per sampling period Numbering the sampling period This represents the total number of environmental quantity categories. Environmental quantity category number, For the first Offset weights of class environment quantities For the first Environmental quantity in the first The reliable environmental quantity for each sampling period For the first Environmental quantity in the first Environmental objectives for each experimental phase Numbering for the experimental phase, For the first Environmental quantity in the first The tolerance boundaries for each experimental phase. The gate perturbation weight, For the first Gate disturbance amount per sampling period To implement mismatch weights, For the first The execution mismatch amount per sampling period, Indicates the first The effective deviation amount after the environmental quantity exceeds the tolerance boundary; the offset attribution module determines the offset type as environmental gradual change offset, door disturbance offset, execution mismatch offset or composite offset based on the offset intensity and the proportion of door disturbance amount and execution mismatch amount.

6. The drug stability test chamber control and data traceability device according to claim 5, characterized in that, The evidence storage gating control module determines the linkage adjustment objects of temperature, humidity, light, and circulating air based on the offset type, determines the adjustment range of each linkage adjustment object based on the offset intensity, and limits the adjustment range based on the execution constraints. After generating the adjustment instruction summary, the evidence storage gating control module first sends the collection summary, adjustment instruction summary, and prescription fingerprint to the chain traceability freezing module. Only after the chain traceability freezing module returns the pre-sealing success mark will it release the adjustment instruction corresponding to the adjustment instruction summary.

7. The drug stability test chamber control and data traceability device according to claim 6, characterized in that, The chain-based traceability freezing module generates the pre-sealed hash value and the current traceability hash value according to the following formula: ， in, For the first The pre-sealed hash value for each sampling period, For the first The current trace hash value for each sampling period. Numbering the sampling period This is a one-way hash operation function. For the first The trace hash value for each sampling period, For the first Prescription fingerprints from the trial phase, Numbering for the experimental phase, For the first Summary of data collection for each sampling period For the first Summary of adjustment instructions for each sampling period For the first A reliable timestamp for each sampling period. For the first Summary of execution receipts for each sampling period For the first Anomaly identifier for each sampling period, This indicates that the fields will be concatenated according to a preset field order.

8. The drug stability test chamber control and data traceability device according to claim 7, characterized in that, At the end of each test phase, the chain-based traceability freezing module generates a phase sealing record by including the phase identifier, phase start and end time, prescription fingerprint, first traceability hash value, last traceability hash value, abnormal identifier set, and operator identifier. The sealing hash value of the phase sealing record is then used as the preceding hash value of the first traceability record in the next test phase.

9. The drug stability test chamber control and data traceability device according to claim 7, characterized in that, When the hash recalculation result of any trace record is inconsistent with the stored trace hash value, or the adjustment instruction digest is inconsistent with the actual released adjustment instruction, or the execution receipt is inconsistent with the released adjustment instruction, the chain trace freezing module marks the corresponding trace record as a frozen record, marks the trace records after the frozen record that are related to the same test phase as records to be reviewed, and prohibits overwrite writing to the frozen record and the records to be reviewed; when generating a correction description record, only the correction description record is allowed to be generated in an append manner, and the correction description record is bound to the trace hash value of the frozen record.

10. The drug stability test chamber control and data traceability device according to claim 1, characterized in that, The chain-based traceability freezing module includes an offline append-only cache area and an external verification package generation unit. The offline append-only cache area is used to cache unuploaded collection summaries, adjustment instruction summaries, execution receipts, and trusted timestamps in the order of sampling period after communication failure or power outage recovery, and to generate traceability hash values ​​starting from the traceability hash value of the previous sampling period. The external verification package generation unit is used to generate an external verification package including a test protocol summary, prescription fingerprint sequence, traceability hash value sequence, stage sealing record, freeze record, and hash recalculation rules, so that the external verification end can verify the consistency between stability test process data, control actions, and traceability records.

Images