A mass spectrometry detection kit storage environment control method and system

Abstract

The invention discloses a mass spectrometry detection kit storage environment control method and system, belonging to the technical field of storage environment control. The control method has the following specific steps: I. transporting the mass spectrometry detection kit and preliminarily detecting the storage environment; II. classifying and managing the mass spectrometry detection kit and initializing the storage environment; III. allocating a storage area for each kit and dynamically adjusting the storage environment. The invention can quickly and accurately make environmental adjustment decisions, improve the flexibility and adaptability of environmental regulation, avoid the degradation of kit performance due to fluctuations in storage conditions, realize highly automated environmental regulation, reduce the risk of human error, reduce management and maintenance costs, improve management efficiency, greatly improve classification efficiency and accuracy, effectively reduce classification error rate, realize intelligent and adaptive classification processing, enhance classification flexibility, reduce storage space waste due to random placement or improper classification of kits, and improve storage efficiency.

Description

Technical Field

[0001] The present invention relates to the technical field of storage environment control, and in particular to a storage environment control method and system for a mass spectrometry detection kit. Background Art

[0002] Mass spectrometry detection kits have important applications in scientific research and medicine. They are mainly important reagents commonly used in analytical chemistry, biomedicine and other fields. Their quality and stability are directly related to the accuracy and reliability of experimental results. The quality and performance of mass spectrometry detection kits depend to a large extent on the environmental conditions during storage. In order to ensure the activity and effectiveness of mass spectrometry detection kits during storage, their storage environment must be strictly monitored and regulated. However, traditional storage methods often rely only on manual settings and monitoring, and it is difficult to respond to complex and changeable storage environment changes in real time, which may lead to problems such as kit failure and performance degradation. Therefore, the development of a mass spectrometry detection kit storage environment control method and system is of great significance to ensure the activity, stability and use effect of mass spectrometry detection kits.

[0003] The existing mass spectrometry detection kit storage environment control method and system environment control have poor flexibility and adaptability, and the performance of the kit is easily degraded due to fluctuations in storage conditions. The risk of human error is high, and the management and maintenance costs also increase accordingly. In addition, the existing mass spectrometry detection kit storage environment control method and system cannot realize intelligent and adaptive classification processing, which reduces the classification efficiency and accuracy. There is a lot of storage space waste due to random placement or improper classification of the kit. To this end, we propose a mass spectrometry detection kit storage environment control method and system. Summary of the invention

[0004] The purpose of the present invention is to solve the defects in the prior art and to propose a method and system for controlling the storage environment of a mass spectrometry detection kit.

[0005] In order to achieve the above object, the present invention adopts the following technical solutions:

[0006] A method for controlling the storage environment of a mass spectrometry detection kit, the specific steps of the control method are as follows:

[0007] Ⅰ. Transport the mass spectrometry test kit and conduct preliminary testing of the storage environment;

[0008] II. Classify and manage mass spectrometry detection kits and initialize the storage environment;

[0009] III. Allocate storage areas for each test kit and dynamically adjust the storage environment;

[0010] IV. Build a virtual storage environment and optimize and adjust storage paths in real time;

[0011] V. Optimize storage strategies based on real-time storage environment information;

[0012] VI. Analyze and predict various abnormal situations in the storage environment and make adjustments;

[0013] Ⅶ. Regularly feedback storage environment data and perform real-time environment optimization.

[0014] As a further solution of the present invention, the specific steps of classifying and managing the mass spectrometry detection kit in step II are as follows:

[0015] S1.1: Collect the number of mass spectrometry detection kits to be classified, extract the feature vector of each mass spectrometry detection kit, and use a set of quantum bit sequences to represent the status of each kit. ,in Representative kit Belongs to category The probability amplitude of , and randomly assigns the initial quantum state of each quantum bit sequence;

[0016] S1.2: In each round of iteration, the rotation angle is obtained according to the gradient of the fitness function. Based on the rotation angle, the quantum state of each reagent kit is updated using a quantum rotating gate. After the update is completed, the updated amplitude and phase are processed by a normalization method. After the processing is completed, The probability of projecting each kit onto any category;

[0017] S1.3: Collect the projection probabilities of each category and determine the preliminary classification of the test kit according to the maximum probability principle. Collect the preliminary classification results of all test kits to form the classification scheme of the current iteration, calculate the fitness function value of the classification scheme, and compare it with the fitness value of the previous iteration;

[0018] S1.4: If the current fitness value is better than the previous iteration, the best classification scheme is updated. Otherwise, continue to the next iteration and repeat the quantum revolving door update and quantum measurement process. Repeat the iterative update. When the change in the fitness function is less than the set threshold, stop the iteration and output the classification scheme containing the category label to which each test kit finally belongs. Then the transport robot allocates the test kits to the corresponding storage area according to the classification scheme.

[0019] As a further solution of the present invention, the specific calculation formula of the fitness function described in S1.2 is as follows:

[0020] ;

[0021] In the formula, represents the fitness value; Represents the classification result set of all test kits; Represents the kit category; Representative Category The index of Representative Category The number of test kits in represents the index of the kit; Represents the index of another kit, Inconsistency; Representative kit With the kit Similarity value of ; represents the balancing factor that regulates the similarity within and between categories; Represents another category The index of Inconsistency; Representative Category The number of test kits in

[0022] The specific calculation formula for updating the quantum state of each kit by the quantum revolving gate described in S1.2 is as follows:

[0023] ;

[0024] ;

[0025] In the formula, Representative The amplitude update value after iterations; Representative The phase update value after iterations; Representative Iteration 1 Belongs to category Amplitude of Representative Iteration 1 Belongs to category The phase of Represents the rotation angle The sine value of ; Represents the rotation angle The cosine value of .

[0026] As a further solution of the present invention, the specific steps of initializing the storage environment in step II are as follows:

[0027] S2.1: Collect storage environment data, kit information, kit quality assessment data, and usage records of different types of mass spectrometry detection kits from the historical storage database, and preprocess each set of collected data to obtain a kit data set;

[0028] S2.2: Based on the collected historical data, a set of CNN networks is established, the test kit data set is divided into a training set and a test set, and the training set is passed into the CNN network. Each set of training data starts from the input layer of the CNN network through forward propagation, passes through the convolution layer, pooling layer, fully connected layer and Softmax layer in sequence, and outputs the final prediction result through the output layer;

[0029] S2.3: Calculate the difference between the predicted result and the actual value through the MSE error function. Based on the calculated difference value, update the parameters of each layer of the CNN network through back propagation and Adam optimization algorithm, and evaluate the performance indicators of the CNN network after training through the test set. At the same time, adjust the CNN network hyperparameters according to the evaluation results, and repeat the training and testing until the change value of the CNN network performance indicator reaches the preset range;

[0030] S2.4: The CNN network collects data from each group of newly stored mass spectrometry detection kits and predicts the optimal storage environment for each kit. The predicted optimal environmental conditions are compared with the various parameters of the current storage environment collected by the sensor in real time to obtain environmental differences. According to the environmental difference values, corresponding environmental adjustment strategies are formulated to adjust the various environmental parameters of the storage environment to generate an initial storage environment.

[0031] As a further solution of the present invention, the specific steps of dynamically adjusting the storage environment in step III are as follows:

[0032] S3.1: Initialize a set of A population of storage environment parameter combinations is formed, where each individual represents a storage condition. The fitness value of each individual under the current storage environment is calculated to evaluate the impact of the current storage condition on the quality of the kit. The individual with the highest fitness is selected from the initial population as the optimal solution for the current population.

[0033] S3.2: Set a set of coefficient vectors, and generate a random number between 0 and 1 in each iteration. If the generated random number is greater than the preset probability parameter, the coefficient for controlling the position update is calculated based on the coefficient vector. and step length , according to the size of the direction, determine whether the remaining individuals are shrinking around the optimal solution;

[0034] S3.3: If <1, it means that the remaining individuals are surrounded by the optimal solution, and the positions of the remaining individuals are updated according to the optimal solution, that is, the storage condition parameter values ​​of each individual are adjusted. ≥1, it means that the remaining individuals are far away from the optimal solution. Through random search, random positions are selected in the population space, and the positions of the remaining individuals are updated. After each iteration, the coefficient vector is updated based on the linear decreasing rule;

[0035] S3.4: If the generated random number is less than or equal to the preset probability parameter, the distance between the positions of the remaining individuals in the population and the optimal solution position is calculated, and the motion law of each individual along the spiral trajectory approaching the optimal solution is simulated through the spiral motion formula to update the individual position;

[0036] S3.5: Repeatedly select the optimal solution and iteratively update the position until the fitness value converges to a preset threshold range, then compare the fitness values ​​of individuals in each group, output the optimal environmental parameter combination, and adjust the storage environment in the current storage area based on the optimal environmental parameter combination.

[0037] As a further solution of the present invention, the specific steps of location updating described in S3.3 are as follows:

[0038] ;

[0039] ;

[0040] ;

[0041] ;

[0042] In the formula, represents the position update coefficient; and Respectively represent random vectors in the range [0, 1]; represents the coefficient vector, ranging from [0, 2]; represents the step length; Representative Group individuals in The position in the iteration; Representative Group individuals in The position in the iteration; Representative The current optimal solution position at the iteration; Representative Group individuals in The position in the iteration; represents a new position randomly selected in the entire population space;

[0043] The specific calculation formula of the spiral motion formula described in S3.4 is as follows:

[0044] ;

[0045] In the formula, Represents the current individual in The position in the iteration; Represents the distance between the current individual position and the optimal solution position; represents the helical constant, which is used to control the contraction rate of the helical curve; represents a set of random numbers in the range [−1,1], used to simulate the changes in the spiral path; Represents the change in angle along the helical curve.

[0046] A mass spectrometry detection kit storage environment control system, comprising an acquisition processing module, a central database, a history analysis module, an environment prediction module, a detection and evaluation module, an intelligent control module, a storage optimization module, a classification and handling module, a monitoring and scheduling module, a visualization interaction module and a log report module;

[0047] The acquisition and processing module is used to collect environmental data inside the storage environment in real time through a variety of sensors, and pre-process the collected raw environmental data;

[0048] The central database is used to store and manage all collected environmental data and mass spectrometry detection kit information;

[0049] The historical analysis module is used to analyze the historical environmental data stored in the central database to determine the initial storage conditions;

[0050] The environment prediction module is used to predict the future environment state of the storage environment based on historical data and real-time data;

[0051] The detection and evaluation module is used to monitor various parameters of the storage environment in real time, identify abnormal situations, and evaluate the current environmental conditions based on real-time monitoring data and prediction results;

[0052] The intelligent control module is used to automatically adjust environmental conditions according to the information provided by each module;

[0053] The storage optimization module is used to dynamically adjust storage environment parameters according to the evaluation results;

[0054] The classification and transport module is used to classify different types of mass spectrometry detection kits and control the intelligent robot to transport the kits to a designated storage area;

[0055] The monitoring and scheduling module is used to monitor the operating status and control strategy of each module, and perform comprehensive scheduling according to actual conditions;

[0056] The visual interaction module is used to display the storage environment status of the current warehouse, the reagent kit storage information and the system working status, and provide the user with a manual adjustment function;

[0057] The log report module is used to record the system's operating logs and generate reports regularly for user analysis and reference.

[0058] As a further solution of the present invention, the specific steps of automatically adjusting the environmental conditions by the intelligent control module are as follows:

[0059] S4.1: Collect all environmental states of each storage area and construct the state space of different storage areas , where each state ,and , and Respectively represent the current ambient temperature, ambient humidity, light intensity, air circulation rate and vibration amplitude;

[0060] S4.2: Construct the action space of the corresponding storage area according to the control measures that the system can take in different states , where each group has specific adjustment measures for the storage environment , and based on historical data, calculate from the state In action Then transfer to the next state The state transition probability of

[0061] S4.3: Initialize the policy function based on the initial actions taken in each state Then, according to the current strategy, the state value function of each state is obtained, and the strategy function is improved and updated by selecting actions that can obtain greater expected returns in each state, and the strategy evaluation and strategy improvement are repeated until the strategy function converges to the preset threshold to obtain the optimal strategy. ;

[0062] S4.4: Initialize the Q value of each state-action pair according to the current state and Q value by The greedy strategy selects the action to be executed and executes the selected action, while observing the feedback of the current storage environment, updating the Q value based on the feedback, and then updating the current state to the next state, and continuing to execute the next round of decision-making, repeating the iteration until the Q value change value of each state-action pair reaches the preset range;

[0063] S4.5: Obtain various environmental parameters and corresponding state changes in each storage area in real time, combine real-time data with historical data, calculate the latest state transition frequency, integrate new data into the state transition probability, and update the immediate reward function according to the deviation of real-time data. After the real-time data fusion is completed, use the new state transition probability and immediate reward function to update the Q value;

[0064] S4.6: Collect the action information with the largest Q value in each state, and build the corresponding optimal strategy. Then, perform actual storage environment control based on the optimal strategy, monitor the changes in environmental parameters in real time, record the actual effect after executing the action in the current state, and evaluate whether the execution effect of the current strategy has achieved the expected goal. If the environmental parameters deviate from the target value, update the strategy through the real-time data fusion mechanism and adjust the storage conditions.

[0065] Compared with the prior art, the present invention has the following beneficial effects:

[0066] 1. The storage environment control method of the mass spectrometry detection kit collects all environmental states of each storage area and constructs the state space of different storage areas. According to the control measures that the system can take in different states, the action space of the corresponding storage area is constructed, and according to historical data, the state transition probability of transferring from each state to the next state after executing a random action is calculated. Based on the initial action taken in each state, the policy function is initialized, and then according to the current policy, the state value function of each state is obtained, and the policy function is improved and updated by selecting an action that can obtain a greater expected return in each state. The policy evaluation and policy improvement are repeated until the policy function converges to a preset threshold to obtain the optimal policy, and the Q value of each state-action pair is initialized. According to the current state and Q value, The greedy strategy selects the action to be performed and executes the selected action, while observing the feedback of the current storage environment, updating the Q value based on the feedback, and then updating the current state to the next state, and continuing to execute the next round of decision-making, repeating the iteration until the Q value change value of each state-action pair reaches the preset range, and obtaining various environmental parameters and corresponding state changes in each storage area in real time, and combining real-time data with historical data, by calculating the latest state transition frequency, integrating the new data into the state transition probability, and updating the immediate reward function according to the deviation of the real-time data. After the real-time data fusion is completed, the new state transition probability and immediate reward function are used to update the Q value. Collect the action information with the largest Q value in each state, build the corresponding optimal strategy, and then perform actual storage environment control based on the optimal strategy, monitor the changes in environmental parameters in real time, record the actual effect after executing the action in the current state, and evaluate whether the execution effect of the current strategy has achieved the expected goal. If the environmental parameters deviate from the target value, update the strategy through the real-time data fusion mechanism and adjust the storage conditions. It can make environmental adjustment decisions quickly and accurately, improve the flexibility and adaptability of environmental regulation, avoid the performance degradation of the test kit due to fluctuations in storage conditions, realize highly automated environmental regulation, reduce the risk of human error, and reduce management and maintenance costs, thereby improving management efficiency.

[0067] 2. The present invention collects the number of mass spectrometry detection kits to be classified in the warehouse, extracts the characteristic vector of each mass spectrometry detection kit, and uses a group of quantum bit sequences to represent the state of each kit, and randomly assigns the initial quantum state of each quantum bit sequence. In each round of iteration, the rotation angle is obtained according to the gradient of the fitness function, and based on the rotation angle, the quantum state of each kit is updated using a quantum rotating gate. After the update is completed, the updated amplitude and phase are processed by a normalization method. After the processing is completed, each kit is projected onto any category according to a set probability, the projection probability of each category is collected, and the preliminary classification of the kit is determined according to the maximum probability principle. The preliminary classification results of all kits are collected to form a classification method for the current iteration. The scheme is then calculated, and the fitness function value of the classification scheme is compared with the fitness value of the previous iteration. If the current fitness value is better than the previous iteration, the best classification scheme is updated. Otherwise, the next iteration is continued, and the quantum revolving door update and quantum measurement process are repeated. The iterative update is repeated. When the change in the fitness function is less than the set threshold, the iteration is stopped, and a classification scheme containing the category label to which each kit finally belongs is output. After that, the transport robot allocates the kits to the corresponding storage area according to the classification scheme, which greatly improves the classification efficiency and accuracy, effectively reduces the classification error rate, realizes intelligent and adaptive classification processing, improves the flexibility of classification, reduces the waste of storage space caused by random placement or improper classification of kits, and improves storage efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0068] The accompanying drawings are used to provide further understanding of the present invention and constitute a part of the specification. They are used to explain the present invention together with the embodiments of the present invention and do not constitute a limitation of the present invention.

[0069] Figure 1 A flowchart of a storage environment control method for a mass spectrometry detection kit proposed by the present invention;

[0070] Figure 2 This is a system block diagram of a mass spectrometry detection kit storage environment control system proposed by the present invention. DETAILED DESCRIPTION

[0071] The technical solutions in the embodiments of the present invention will be described clearly and completely below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments. Example 1

[0072] Reference Figure 1 This embodiment discloses a method for controlling the storage environment of a mass spectrometry detection kit. The specific steps of the control method are as follows:

[0073] Transport the mass spectrometry test kit and conduct preliminary testing of the storage environment.

[0074] Categorize and manage mass spectrometry detection kits and initialize the storage environment.

[0075] Specifically, the number of mass spectrometry detection kits to be classified in the library is collected, and the feature vector of each mass spectrometry detection kit is extracted, and the state of each kit is represented by a set of quantum bit sequences. ,in Representative kit Belongs to category The probability amplitude is calculated and the initial quantum state of each quantum bit sequence is randomly assigned. In each round of iteration, the rotation angle is obtained according to the gradient of the fitness function. Based on the rotation angle, the quantum state of each kit is updated using a quantum rotating gate. After the update is completed, the updated amplitude and phase are processed by the normalization method. After the processing is completed, The probability of projecting each test kit onto any category is calculated, and the projection probability of each category is collected. The preliminary classification of the test kit is determined according to the maximum probability principle. The preliminary classification results of all test kits are collected to form the classification scheme of the current iteration, and then the fitness function value of the classification scheme is calculated and compared with the fitness value of the previous iteration. If the current fitness value is better than the previous iteration, the best classification scheme is updated. Otherwise, the next iteration is continued, and the quantum revolving door update and quantum measurement process are repeated. The iterative update is repeated. When the change in the fitness function is less than the set threshold, the iteration is stopped, and the classification scheme containing the category label to which each test kit finally belongs is output. After that, the transport robot distributes the test kits to the corresponding storage area according to the classification scheme.

[0076] Specifically, storage environment data, kit information, kit quality assessment data and usage records of different types of mass spectrometry detection kits are collected from the historical storage database, and each set of collected data is preprocessed to obtain a kit data set. Based on the collected historical data, a set of CNN networks is established, and the kit data set is divided into a training set and a test set. The training set is passed into the CNN network. Each set of training data starts from the input layer of the CNN network through forward propagation, passes through the convolution layer, the pooling layer, the fully connected layer and the Softmax layer in sequence, and outputs the final prediction result through the output layer. The difference between the predicted result and the actual value is calculated by the MSE error function, and the calculated difference is used. Outliers, update the parameters of each layer of the CNN network through back propagation and Adam optimization algorithm, and evaluate the performance indicators of the CNN network after training through the test set. At the same time, adjust the CNN network hyperparameters according to the evaluation results, and repeat training and testing until the change value of the CNN network performance indicator reaches the preset range. The CNN network collects each group of data of each group of newly stored mass spectrometry detection kits, and predicts the optimal storage environment for each kit. The predicted optimal environmental conditions are compared with the various parameters of the current storage environment collected by the sensor in real time to obtain the environmental difference. According to the environmental difference value, the corresponding environmental adjustment strategy is formulated to adjust the various environmental parameters of the storage environment to generate the initial storage environment.

[0077] In this embodiment, the specific calculation formula of the fitness function is as follows:

[0078] ;

[0079] In the formula, represents the fitness value; Represents the classification result set of all test kits; Represents the kit category; Representative Category The index of Representative Category The number of test kits in represents the index of the kit; Represents the index of another kit, Inconsistency; Representative kit With kit Similarity value of ; represents the balancing factor that regulates the similarity within and between categories; Represents another category The index of Inconsistency; Representative Category The number of test kits in

[0080] The specific calculation formula for the quantum revolving door to update the quantum state of each kit is as follows:

[0081] ;

[0082] ;

[0083] In the formula, Representative The amplitude update value after iterations; Representative The phase update value after iterations; Representative Iteration 1 Belongs to category Amplitude of Representative Iteration 1 Belongs to category The phase of Represents the rotation angle The sine value of ; Represents the rotation angle The cosine value of .

[0084] Assign storage areas to each reagent kit and dynamically adjust the storage environment.

[0085] Specifically, initialize a set of A population of storage environment parameter combinations is set, where each individual represents a storage condition. The fitness value of each individual under the current storage environment is calculated to evaluate the impact of the current storage condition on the quality of the kit. In the initial population, the individual with the highest fitness is selected as the optimal solution for the current population. A set of coefficient vectors is set. At the same time, a random number between 0 and 1 is generated in each round of iteration. If the generated random number is greater than the preset probability parameter, the coefficient for controlling the position update is calculated based on the coefficient vector. and step length , according to the size of the direction, determine whether the remaining individuals are shrinking around the optimal solution. If <1, it means that the remaining individuals are surrounded by the optimal solution, and the positions of the remaining individuals are updated according to the optimal solution, that is, the storage condition parameter values ​​of each individual are adjusted. ≥1, it means that the remaining individuals are far away from the optimal solution. Through random search, random positions are selected in the population space, and the positions of the remaining individuals are updated. After each iteration, the coefficient vector is updated based on the linear decreasing rule. If the generated random number is less than or equal to the preset probability parameter, the distance between the positions of the remaining individuals in the population and the optimal solution position is calculated, and the spiral motion formula is used to simulate the movement law of each individual approaching the optimal solution along the spiral trajectory, update the individual position, repeatedly select the optimal solution and iteratively update the position until the fitness value converges to the preset threshold range, then compare the fitness values ​​of the individuals in each group, output the optimal environmental parameter combination, and adjust the storage environment in the current storage area based on the optimal environmental parameter combination.

[0086] In addition, it needs to be further explained that the specific steps of location update are as follows:

[0087] ;

[0088] ;

[0089] ;

[0090] ;

[0091] In the formula, represents the position update coefficient; and Respectively represent random vectors in the range [0, 1]; represents the coefficient vector, ranging from [0, 2]; represents the step length; Representative Group individuals in The position in the iteration; Representative Group individuals in The position in the iteration; Representative The current optimal solution position at the iteration; Representative Group individuals in The position in the iteration; represents a new position randomly selected in the entire population space;

[0092] The specific calculation formula of spiral motion formula is as follows:

[0093] ;

[0094] In the formula, Represents the current individual in The position in the iteration; Represents the distance between the current individual position and the optimal solution position; represents the helical constant, which is used to control the contraction rate of the helical curve; represents a set of random numbers in the range [−1,1], used to simulate the changes in the spiral path; Represents the change in angle along the helical curve.

[0095] Build a virtual storage environment and optimize and adjust storage paths in real time.

[0096] Optimize storage strategies based on real-time storage environment information.

[0097] Analyze and predict various abnormal situations in the storage environment and make adjustments.

[0098] Regularly feed back storage environment data and perform real-time environment optimization. Example 2

[0099] Reference Figure 2 , this embodiment discloses a mass spectrometry detection kit storage environment control system, including an acquisition and processing module, a central database, a historical analysis module, an environmental prediction module, a detection and evaluation module, an intelligent control module, a storage optimization module, a classification and handling module, a monitoring and scheduling module, a visual interaction module, and a log report module;

[0100] The acquisition and processing module is used to collect environmental data inside the storage environment in real time through a variety of sensors, and to pre-process the collected raw environmental data; the central database is used to store and manage all collected environmental data and mass spectrometry detection kit information; the historical analysis module is used to analyze the historical environmental data stored in the central database and determine the initial storage conditions; the environmental prediction module is used to predict the future environmental status of the storage environment based on historical data and real-time data; the detection and evaluation module is used to monitor various parameters of the storage environment in real time, identify abnormal situations, and evaluate the current environmental conditions based on real-time monitoring data and prediction results.

[0101] Specifically, all environmental states of each storage area are collected and the state space of different storage areas is constructed. , where each state ,and , and They represent the current ambient temperature, ambient humidity, light intensity, air circulation rate, and vibration amplitude respectively. According to the control measures that the system can take in different states, the action space of the corresponding storage area is constructed. , where each group has specific adjustment measures for the storage environment , and based on historical data, calculate from the state In action Then transfer to the next state The state transition probability is based on the initial actions taken in each state, and the strategy function is initialized Then, according to the current strategy, the state value function of each state is obtained, and the strategy function is improved and updated by selecting actions that can obtain greater expected returns in each state, and the strategy evaluation and strategy improvement are repeated until the strategy function converges to the preset threshold to obtain the optimal strategy. , initialize the Q value of each state-action pair, according to the current state and Q value, through The greedy strategy selects the action to be executed and executes the selected action, while observing the feedback of the current storage environment, updates the Q value based on the feedback, and then updates the current state to the next state, and continues to execute the next round of decision-making, repeating the iteration until the Q value change value of each state-action pair reaches the preset range, and obtains the various environmental parameters and corresponding state changes in each storage area in real time, and combines the real-time data with the historical data. By calculating the latest state transition frequency, the new data is integrated into the state transition probability. At the same time, according to the deviation of the real-time data, the immediate reward function is updated. After the real-time data fusion is completed, the new state transition probability and immediate reward function are used to update the Q value, collect the action information with the largest Q value in each state, and construct the corresponding optimal strategy. Then, the actual storage environment control is performed according to the optimal strategy, the environmental parameter changes are monitored in real time, the actual effect after the action is executed in the current state is recorded, and the execution effect of the current strategy is evaluated to see whether it has achieved the expected goal. If the environmental parameters deviate from the target value, the strategy is updated through the real-time data fusion mechanism, and the storage conditions are adjusted.

[0102] The intelligent control module is used to automatically adjust the environmental conditions according to the information provided by each module; the storage optimization module is used to dynamically adjust the storage environment parameters according to the evaluation results; the classification and transportation module is used to classify different types of mass spectrometry detection kits and control the intelligent robot to transport the kits to the designated storage area; the monitoring and scheduling module is used to monitor the operating status and control strategies of each module, and perform comprehensive scheduling according to actual conditions; the visual interaction module is used to display the current warehouse storage environment status, kit storage information and system working status, and provide users with manual adjustment functions; the log report module is used to record the system operation log and regularly generate reports for user analysis and reference.

Claims

1. A method for controlling the storage environment of a mass spectrometry detection kit, characterized in that: The specific steps of the control method are as follows: Ⅰ. Transport the mass spectrometry test kit and conduct preliminary testing of the storage environment; II. Classify and manage mass spectrometry detection kits and initialize the storage environment; III. Allocate storage areas for each test kit and dynamically adjust the storage environment; IV. Build a virtual storage environment and optimize and adjust storage paths in real time; V. Optimize storage strategies based on real-time storage environment information; VI. Analyze and predict various abnormal situations in the storage environment and make adjustments; Ⅶ. Regularly feed back storage environment data and perform real-time environment optimization; The specific steps of classifying and managing the mass spectrometry detection kit in step II are as follows: S1.1: Collect the number of mass spectrometry detection kits to be classified, extract the feature vector of each mass spectrometry detection kit, and use a set of quantum bit sequences to represent the status of each kit. ,in Representative kit Belongs to category The probability amplitude of , and randomly assigns the initial quantum state of each quantum bit sequence; S1.2: In each round of iteration, the rotation angle is obtained according to the gradient of the fitness function. Based on the rotation angle, the quantum state of each reagent kit is updated using a quantum rotating gate. After the update is completed, the updated amplitude and phase are processed by a normalization method. After the processing is completed, The probability of projecting each kit onto any category; S1.3: Collect the projection probabilities of each category and determine the preliminary classification of the test kit according to the maximum probability principle. Collect the preliminary classification results of all test kits to form the classification scheme of the current iteration, calculate the fitness function value of the classification scheme, and compare it with the fitness value of the previous iteration; S1.4: If the current fitness value is better than the previous iteration, the best classification scheme is updated. Otherwise, the next iteration is continued, and the quantum revolving door update and quantum measurement process are repeated. The iteration update is repeated. When the change in the fitness function is less than the set threshold, the iteration is stopped, and the classification scheme containing the category label to which each test kit finally belongs is output. After that, the transport robot allocates the test kits to the corresponding storage area according to the classification scheme. The specific steps of dynamically adjusting the storage environment described in step III are as follows: S3.1: Initialize a set of A population of storage environment parameter combinations is formed, where each individual represents a storage condition. The fitness value of each individual under the current storage environment is calculated to evaluate the impact of the current storage condition on the quality of the kit. The individual with the highest fitness is selected from the initial population as the optimal solution for the current population. S3.2: Set a set of coefficient vectors, and generate a random number between 0 and 1 in each iteration. If the generated random number is greater than the preset probability parameter, the coefficient for controlling the position update is calculated based on the coefficient vector. and step length , according to the size of the direction, determine whether the remaining individuals are shrinking around the optimal solution; S3.3: If <1, it means that the remaining individuals are surrounded by the optimal solution, and the positions of the remaining individuals are updated according to the optimal solution, that is, the storage condition parameter values ​​of each individual are adjusted. ≥1, it means that the remaining individuals are far away from the optimal solution. Through random search, random positions are selected in the population space, and the positions of the remaining individuals are updated. After each iteration, the coefficient vector is updated based on the linear decreasing rule; S3.4: If the generated random number is less than or equal to the preset probability parameter, the distance between the positions of the remaining individuals in the population and the optimal solution position is calculated, and the motion law of each individual along the spiral trajectory approaching the optimal solution is simulated through the spiral motion formula to update the individual position; S3.5: Repeatedly select the optimal solution and iteratively update the position until the fitness value converges to a preset threshold range, then compare the fitness values ​​of individuals in each group, output the optimal environmental parameter combination, and adjust the storage environment in the current storage area based on the optimal environmental parameter combination.

2. A method for controlling the storage environment of a mass spectrometry detection kit according to claim 1, characterized in that: The specific calculation formula of the fitness function described in S1.2 is as follows: ; In the formula, represents the fitness value; Represents the classification result set of all test kits; Represents the kit category; Representative Category The index of Representative Category The number of test kits in represents the index of the kit; Represents the index of another kit, Inconsistency; Representative kit With the kit Similarity value of ; represents the balancing factor that regulates the similarity within and between categories; Represents another category The index of Inconsistency; Representative Category The number of test kits in The specific calculation formula for updating the quantum state of each kit by the quantum revolving gate described in S1.2 is as follows: ; ; In the formula, Representative The amplitude update value after iterations; Representative The phase update value after iterations; Representative Iteration 1 Belongs to category Amplitude of Representative Iteration 1 Belongs to category The phase of Represents the rotation angle The sine value of ; Represents the rotation angle The cosine value of .

3. The method for controlling the storage environment of a mass spectrometry detection kit according to claim 1, characterized in that: The specific steps for initializing the storage environment in step II are as follows: S2.1: Collect storage environment data, kit information, kit quality assessment data, and usage records of different types of mass spectrometry detection kits from the historical storage database, and preprocess each set of collected data to obtain a kit data set; S2.2: Based on the collected historical data, a set of CNN networks is established, the test kit data set is divided into a training set and a test set, and the training set is passed into the CNN network. Each set of training data starts from the input layer of the CNN network through forward propagation, passes through the convolution layer, pooling layer, fully connected layer and Softmax layer in sequence, and outputs the final prediction result through the output layer; S2.3: Calculate the difference between the predicted result and the actual value through the MSE error function. Based on the calculated difference value, update the parameters of each layer of the CNN network through back propagation and Adam optimization algorithm, and evaluate the performance indicators of the CNN network after training through the test set. At the same time, adjust the CNN network hyperparameters according to the evaluation results, and repeat the training and testing until the change value of the CNN network performance indicator reaches the preset range; S2.4: The CNN network collects data from each group of newly stored mass spectrometry detection kits and predicts the optimal storage environment for each kit. The predicted optimal environmental conditions are compared with the various parameters of the current storage environment collected by the sensor in real time to obtain environmental differences. According to the environmental difference values, corresponding environmental adjustment strategies are formulated to adjust the various environmental parameters of the storage environment to generate an initial storage environment.

4. The method for controlling the storage environment of a mass spectrometry detection kit according to claim 1, characterized in that: The specific steps of location updating described in S3.3 are as follows: ; ; ; ; In the formula, represents the position update coefficient; and Respectively represent random vectors in the range [0, 1]; represents the coefficient vector, ranging from [0, 2]; represents the step length; Representative Group individuals in The position in the iteration; Representative Group individuals in The position in the iteration; Representative The current optimal solution position at the iteration; Representative Group individuals in The position in the iteration; represents a new position randomly selected in the entire population space; The specific calculation formula of the spiral motion formula described in S3.4 is as follows: ; In the formula, Represents the current individual in The position in the iteration; Represents the distance between the current individual position and the optimal solution position; represents the helical constant, which is used to control the contraction rate of the helical curve; represents a set of random numbers in the range [−1,1], used to simulate the changes in the spiral path; Represents the change in angle along the helical curve.

5. A mass spectrometry detection kit storage environment control system, used to implement the method according to any one of claims 1 to 4, characterized in that: It includes collection and processing module, central database, historical analysis module, environmental prediction module, detection and evaluation module, intelligent control module, storage optimization module, classification and handling module, monitoring and scheduling module, visual interaction module and log report module; The acquisition and processing module is used to collect environmental data inside the storage environment in real time through a variety of sensors, and pre-process the collected raw environmental data; The central database is used to store and manage all collected environmental data and mass spectrometry detection kit information; The historical analysis module is used to analyze the historical environmental data stored in the central database to determine the initial storage conditions; The environment prediction module is used to predict the future environment state of the storage environment based on historical data and real-time data; The detection and evaluation module is used to monitor various parameters of the storage environment in real time, identify abnormal situations, and evaluate the current environmental conditions based on real-time monitoring data and prediction results; The intelligent control module is used to automatically adjust environmental conditions according to the information provided by each module; The storage optimization module is used to dynamically adjust storage environment parameters according to the evaluation results; The classification and transport module is used to classify different types of mass spectrometry detection kits and control the intelligent robot to transport the kits to a designated storage area; The monitoring and scheduling module is used to monitor the operating status and control strategy of each module, and perform comprehensive scheduling according to actual conditions; The visual interaction module is used to display the storage environment status of the current warehouse, the reagent kit storage information and the system working status, and provide the user with a manual adjustment function; The log report module is used to record the system's operating logs and generate reports regularly for user analysis and reference.

6. A mass spectrometry detection kit storage environment control system according to claim 5, characterized in that: The specific steps of automatically adjusting the environmental conditions by the intelligent control module are as follows: S4.1: Collect all environmental states of each storage area and construct the state space of different storage areas , where each state ,and , and Respectively represent the current ambient temperature, ambient humidity, light intensity, air circulation rate and vibration amplitude; S4.2: Construct the action space of the corresponding storage area according to the control measures that the system can take in different states , where each group has specific adjustment measures for the storage environment , and based on historical data, calculate from the state In action Then transfer to the next state The state transition probability of S4.3: Initialize the policy function based on the initial actions taken in each state Then, according to the current strategy, the state value function of each state is obtained, and the strategy function is improved and updated by selecting actions that can obtain greater expected returns in each state, and the strategy evaluation and strategy improvement are repeated until the strategy function converges to the preset threshold to obtain the optimal strategy. ; S4.4: Initialize the Q value of each state-action pair according to the current state and Q value by The greedy strategy selects the action to be executed and executes the selected action, while observing the feedback of the current storage environment, updating the Q value based on the feedback, and then updating the current state to the next state, and continuing to execute the next round of decision-making, repeating the iteration until the Q value change value of each state-action pair reaches the preset range; S4.5: Obtain various environmental parameters and corresponding state changes in each storage area in real time, combine real-time data with historical data, calculate the latest state transition frequency, integrate new data into the state transition probability, and update the immediate reward function according to the deviation of real-time data. After the real-time data fusion is completed, use the new state transition probability and immediate reward function to update the Q value; S4.6: Collect the action information with the largest Q value in each state, and build the corresponding optimal strategy. Then, perform actual storage environment control based on the optimal strategy, monitor the changes in environmental parameters in real time, record the actual effect after executing the action in the current state, and evaluate whether the execution effect of the current strategy has achieved the expected goal. If the environmental parameters deviate from the target value, update the strategy through the real-time data fusion mechanism and adjust the storage conditions.

Images