An Optimization Method for Controlling a Drug Constant Temperature System

Abstract

The present invention discloses an optimization method for controlling a drug constant temperature system, which mainly relates to the technical field of drug constant temperature control. The method includes: obtaining a pre-deployed temperature sensor array in the target drug constant temperature area; obtaining a cluster of temperature data subsequences; obtaining a set of key temperature value sequences corresponding to the temperature data sequence set; constructing a key temperature value topological network sequence; performing iterative correlation analysis on the key temperature value topological network sequence in chronological order to determine iterative correlation temperature characteristics; and optimizing the control of the drug constant temperature system based on the iterative correlation temperature characteristics to obtain control optimization parameters. The present invention solves the technical problem in the prior art that the control optimization of the drug constant temperature system lacks in-depth analysis of temperature, resulting in low control accuracy and slow system response, and achieves the technical effect of effectively reducing the processing complexity of temperature data and improving the accuracy of the temperature control system.

Description

Technical Field

[0001] The present invention relates to the technical field of drug constant temperature control, and particularly relates to an optimization method for controlling a drug constant temperature system. Background Art

[0002] Traditional drug constant temperature control systems usually rely on temperature sensors to monitor the temperature of the drug storage area in real time to ensure temperature stability. During the temperature monitoring process, a large amount of temperature data usually needs to be processed and analyzed to ensure that the drug storage environment meets the specified temperature requirements. However, due to often relying on single real-time temperature measurement, lacking in-depth analysis of temperature change trends, and being unable to perform efficient correlation analysis on temperature data over a period of time, this may lead to slow or inaccurate response of the temperature control system, thus affecting the quality and safety of drugs.

[0003] The prior art has the technical problems that the optimization of the drug constant temperature system control lacks in-depth analysis of temperature, resulting in low control accuracy and slow system response. Summary of the Invention

[0004] The present application provides an optimization method for controlling a drug constant temperature system, which is used to solve the technical problems in the prior art that the optimization of the drug constant temperature system control lacks in-depth analysis of temperature, resulting in low control accuracy and slow system response.

[0005] In view of the above problems, the present application provides an optimization method for controlling a drug constant temperature system, and the method includes:

[0006] Obtain a pre-deployed temperature sensor array in the target drug constant temperature area;

[0007] Traverse the temperature sensor array to extract temperature data within a preset time period to obtain a set of temperature data sequences, and preset a fixed time window, and use the preset fixed time window to perform sequence segmentation processing on the set of temperature data sequences to obtain a cluster of temperature data subsequences, where the cluster of temperature data subsequences includes a set of temperature data subsequences corresponding to each temperature data sequence;

[0008] Respectively screen the key temperature values of each temperature data subsequence in the cluster of temperature data subsequences, and serialize the screening results to obtain a set of key temperature value sequences corresponding to the set of temperature data sequences;

[0009] Based on the positions of the respective temperature sensors in the temperature sensor array, perform simultaneous topological processing on the set of key temperature value sequences to construct a key temperature value topological network sequence;

[0010] Perform iterative correlation analysis on the key temperature value topological network sequence in chronological order to determine the iterative correlation temperature characteristics of the constant temperature area of the target drug within a preset time period;

[0011] Based on the iterative correlation temperature characteristics, optimize the control of the drug constant temperature system to obtain control optimization parameters, where the drug constant temperature system is used to control the temperature of the constant temperature area of the target drug.

[0012] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0013] In this application, by obtaining the pre-deployed temperature sensor array in the constant temperature area of the target drug, then traversing the temperature sensor array to extract temperature data within a preset time period, obtaining a temperature data sequence set, and presetting a fixed time window, using the preset fixed time window to perform sequence segmentation processing on the temperature data sequence set to obtain a temperature data subsequence cluster, where the temperature data subsequence cluster includes a temperature data subsequence set corresponding to each temperature data sequence. Furthermore, respectively screen the key temperature values of each temperature data subsequence in the temperature data subsequence cluster, and serialize the screening results to obtain a key temperature value sequence set corresponding to the temperature data sequence set. Then, based on the positions of the temperature sensors in the temperature sensor array, perform simultaneous topological processing on the key temperature value sequence set to construct a key temperature value topological network sequence. Perform iterative correlation analysis on the key temperature value topological network sequence in chronological order to determine the iterative correlation temperature characteristics of the constant temperature area of the target drug within a preset time period. Furthermore, optimize the control of the drug constant temperature system based on the iterative correlation temperature characteristics to obtain control optimization parameters, where the drug constant temperature system is used to control the temperature of the constant temperature area of the target drug. It achieves the technical effects of effectively reducing the processing complexity of temperature data and improving the accuracy and response speed of the temperature control system through in-depth temperature data analysis. Description of the Drawings

[0014] Att Figure 1 is a schematic flowchart of a method for optimizing the control of a drug constant temperature system provided by an embodiment of the present invention.

[0015] Att Figure 2 is a schematic flowchart of obtaining iterative correlation temperature characteristics in a method for optimizing the control of a drug constant temperature system provided by an embodiment of the present invention. Detailed Embodiments

[0016] The present invention will be further described below in conjunction with specific embodiments. It should be understood that these embodiments are only used to illustrate the present invention and not to limit the scope of the present invention. In addition, it should be understood that after reading the content taught by the present invention, those skilled in the art can make various changes or modifications to the present invention, and these equivalent forms also fall within the scope defined by the appended claims of this application.

[0017] Embodiment, as shown in the appendix Figure 1 The present application provides an optimization method for controlling a drug constant temperature system. Among them, the method includes:

[0018] S100: Obtain a pre-deployed temperature sensor array in the target drug constant temperature area;

[0019] In the embodiment of the present application, the target drug constant temperature area is an environmental area specifically set for storing drugs, and its temperature needs to be maintained within a certain range to ensure the effectiveness and safety of the drugs. This area usually uses temperature control devices (such as air conditioners, heaters, etc.) to maintain a stable temperature. The pre-deployed temperature sensor array is a plurality of temperature sensors pre-planned and installed in the target drug constant temperature area. The sensor array is used to comprehensively monitor the temperature distribution in this area to ensure that the temperature at each position can be accurately monitored. Pre-deployment means that these sensors have been installed before the system is implemented, and their positions have been optimized to ensure that they can effectively cover the entire target area.

[0020] Through the pre-deployed temperature sensor array, the temperature can be monitored in real time at each position in the drug storage area. The data provided by the sensor array will help the system comprehensively understand the temperature distribution in the drug storage environment to ensure that there are no temperature dead spots. It achieves the technical effect of laying a foundation for the subsequent obtained data for analyzing and optimizing the control system.

[0021] S200: Traverse the temperature sensor array to extract temperature data within a preset time period to obtain a temperature data sequence set, and preset a fixed time window. Use the preset fixed time window to perform sequence segmentation processing on the temperature data sequence set to obtain a temperature data subsequence cluster, where the temperature data subsequence cluster includes a temperature data subsequence set corresponding to each temperature data sequence;

[0022] In one embodiment, each sensor in the temperature sensor array is accessed one by one to obtain their respective temperature data. By traversing the entire array, the temperature information at each position in the constant temperature area of the target drug can be comprehensively collected. The preset time period refers to a time range set in advance during the data acquisition process. The temperature data within this time period will be extracted and used for subsequent analysis. The set of temperature data sequences refers to a group of temperature data obtained from each temperature sensor, and these data are arranged in chronological order to form a data sequence. The data sequence of each sensor represents the temperature change of that sensor within the preset time period. The preset fixed time window is a time period with a fixed length, which is used to segment the temperature data sequence. By setting the fixed time window, the entire temperature data sequence can be divided into multiple smaller subsequences according to a specific time step, so as to obtain the cluster of temperature data subsequences, where the cluster of temperature data subsequences includes a set of temperature data subsequences corresponding to each temperature data sequence.

[0023] After each temperature data sequence is segmented according to the fixed time window, the temperature data within each time period constitutes a subsequence, and multiple subsequences are combined into a set of temperature data subsequences corresponding to a temperature data sequence.

[0024] By performing time period processing on the temperature data obtained from the temperature sensor array and further segmenting these data, the temperature change information within the constant temperature area of the target drug is obtained. Thus, the temperature fluctuations within each time period can be analyzed more precisely to discover possible temperature anomalies or trends. It achieves the technical effect of providing a data basis for subsequent screening of key temperature values, topological analysis, etc.

[0025] S300: Screen the key temperature values of each temperature data subsequence in the cluster of temperature data subsequences respectively, and serialize the screening results to obtain the set of key temperature value sequences corresponding to the set of temperature data sequences;

[0026] Furthermore, when screening the key temperature values of each temperature data subsequence in the cluster of temperature data subsequences respectively and serializing the screening results to obtain the set of key temperature value sequences corresponding to the set of temperature data sequences, step S300 of the embodiment of the present application further includes:

[0027] Randomly extract a first temperature data subsequence from the cluster of temperature data subsequences, and construct a screening space based on the first temperature data subsequence to obtain a temperature screening space, where the temperature screening space includes multiple temperature screening points, and each temperature screening point corresponds to a temperature data;

[0028] Extract the central temperature screening point of the temperature screening space. Using the central temperature screening point as an index, perform key temperature value screening in the temperature screening space according to a preset screening bandwidth to obtain a first key temperature value;

[0029] Perform subsequence key temperature value screening on each temperature data subsequence in the temperature data subsequence cluster to obtain a cluster of subsequence key temperature values;

[0030] Perform serialization processing on each subsequence key temperature value set in the cluster of subsequence key temperature values to obtain the set of key temperature value sequences.

[0031] In one embodiment, select the most representative temperature values from each temperature data subsequence. These key temperature values can reflect important information in the temperature data, such as peaks, valleys, change trends, etc., and perform serialization processing on the screening results in chronological order to obtain the set of key temperature value sequences. The purpose of the screening process is to extract representative key temperature values from the temperature data, achieving the technical effects of dimensionality reduction of the data, improving the data analysis efficiency, and further enhancing the system control optimization response rate.

[0032] The set of key temperature value sequences refers to the set formed by arranging the key temperature values screened from the temperature data subsequence cluster in chronological order from front to back. Each temperature data subsequence corresponds to a key temperature value.

[0033] Preferably, randomly extract a first temperature data subsequence from the temperature data subsequence cluster to construct the temperature screening space. Among them, the temperature screening space refers to the space obtained by inputting the selected first temperature data subsequence into a two-dimensional coordinate system. The screening space includes multiple temperature screening points, and each screening point represents a specific temperature value.

[0034] The central temperature screening point refers to a temperature screening point with representativeness or importance in the screening space, and is usually used as an index point for further screening operations. Optionally, the central temperature screening point can be the center point of the space. The preset screening bandwidth refers to a temperature range set during the screening process to ensure that the selected temperature data points have a certain degree of relevance and representativeness. The bandwidth controls the looseness of the screening, that is, the allowable temperature data deviation range during screening.

[0035] Optionally, it is necessary to screen the key temperature values for each temperature data subsequence. Through this screening process, temperature values that are of great significance for system control and optimization are extracted from the original temperature data, and these key values represent important characteristics in temperature changes. By randomly extracting the first temperature data subsequence from the temperature data subsequence cluster and constructing a screening space based on this, the system can more accurately locate and screen out the key temperature points that are crucial for temperature control optimization. The central temperature screening point serves as the starting point of the screening process to help further refine the screening range. Through this method, the most representative temperature characteristics can be effectively found in complex temperature data, and then the constant temperature control system for drugs can be precisely optimized.

[0036] Based on the central temperature screening point of the temperature screening space, key temperature values are screened in the temperature screening space according to a preset screening bandwidth to obtain the first key temperature value. Moreover, based on the same principle of obtaining the first key temperature value, subsequence key temperature values are screened for each temperature data subsequence in the temperature data subsequence cluster to obtain a subsequence key temperature value cluster.

[0037] After screening and serialization processing, the obtained set of key temperature value sequences provides core data support for subsequent temperature analysis and system optimization. This process ensures that the system can be more efficient and accurate when analyzing temperature fluctuations and provides a more precise decision-making basis for system optimization.

[0038] Furthermore, the central temperature screening point of the temperature screening space is extracted. Taking the central temperature screening point as an index, key temperature values are screened in the temperature screening space according to a preset screening bandwidth to obtain the first key temperature value. The steps S300 of the embodiments of the present application further include:

[0039] Taking the straight line passing through the central temperature screening point and parallel to the horizontal axis of the temperature screening space as the initial screening straight line, where the temperature screening space is a two-dimensional space, the horizontal axis is time, and the vertical axis is the temperature value;

[0040] An initial screening neighborhood of the initial screening straight line is constructed based on the preset screening bandwidth, and the initial screening neighborhood density of the initial screening neighborhood is statistically calculated;

[0041] The initial screening straight line is respectively moved upward and downward by the preset screening bandwidth to obtain an upward-moved screening straight line and a downward-moved screening straight line;

[0042] Based on the upward-moved screening straight line and the downward-moved screening straight line, an upward-moved screening neighborhood and a downward-moved screening neighborhood are respectively constructed in combination with the preset screening bandwidth, and the upward-moved screening neighborhood density and the downward-moved screening neighborhood density are statistically calculated;

[0043] When the upward screening neighborhood density and / or the downward screening neighborhood density is greater than or equal to the initial screening neighborhood density, iterate upward along the upward screening line according to the preset screening bandwidth and / or iterate downward along the downward screening line according to the preset screening bandwidth until the preset number of iterations is satisfied, and determine the target upward screening line and / or the target downward screening line;

[0044] Perform key temperature value screening in the temperature screening space based on the target upward screening line and / or the target downward screening line to obtain the first key temperature value.

[0045] Further, when the upward screening neighborhood density and the downward screening neighborhood density are less than the initial screening neighborhood density, use the temperature data corresponding to the central temperature screening point as the first key temperature value.

[0046] Further, performing key temperature value screening in the temperature screening space based on the target upward screening line and / or the target downward screening line to obtain the first key temperature value, step S300 of the embodiment of the present application further includes:

[0047] Add the temperature data corresponding to the target upward screening line and / or the target downward screening line into the key temperature value screening set;

[0048] Use the key temperature value corresponding to the maximum value in the key temperature value screening set as the first key temperature value.

[0049] In one embodiment, a line passing through the central temperature screening point and parallel to the horizontal axis of the temperature screening space is used as the initial screening line. The temperature screening space is a two-dimensional space, the horizontal axis is time, and the vertical axis is the temperature value.

[0050] The initial screening neighborhood refers to the area composed of temperature screening points whose distance to the initial screening line is the preset screening bandwidth. The range of the neighborhood is determined by the preset screening bandwidth. Count the number of temperature screening points in the initial screening neighborhood to obtain the initial screening neighborhood density. The initial screening neighborhood density refers to the number of temperature data in the screening neighborhood, which reflects the distribution of temperature data in this neighborhood. The larger the density, the more concentrated the temperature data in this area, which may mean that the temperature change in this area is more representative.

[0051] Move the initial screening line upward and downward by the preset screening bandwidth respectively to obtain an upward-moved screening line and a downward-moved screening line. Furthermore, based on the same principle as obtaining the initial screening neighborhood, construct neighborhoods according to the preset screening bandwidth and the upward-moved screening line and the downward-moved screening line to obtain the upward-moved screening neighborhood and the downward-moved screening neighborhood. Count the number of temperature screening points in the upward-moved screening neighborhood and the downward-moved screening neighborhood respectively to obtain the upward-moved screening neighborhood density and the downward-moved screening neighborhood density.

[0052] Furthermore, compare the upward-moved screening neighborhood density and the downward-moved screening neighborhood density with the initial screening neighborhood density respectively. When the upward-moved screening neighborhood density and / or the downward-moved screening neighborhood density is greater than or equal to the initial screening neighborhood density, it indicates that the initial screening neighborhood is not the most densely distributed area in the temperature screening space and further iteration is required.

[0053] Furthermore, perform iteration upward according to the preset screening bandwidth based on the upward-moved screening line and / or iteration downward according to the preset screening bandwidth based on the downward-moved screening line until the preset number of iterations is satisfied. Respectively take the lines obtained in the last iteration as the target upward-moved screening line and / or the target downward-moved screening line. Wherein, the preset number of iterations is the maximum number of iterations preset by those skilled in the art. Add the temperature data corresponding to the target upward-moved screening line and / or the target downward-moved screening line into the critical temperature value screening set, and take the critical temperature value corresponding to the maximum value in the critical temperature value screening set as the first critical temperature value. Thus, it is ensured that the obtained first critical temperature value can best represent the temperature value of the first temperature data subsequence.

[0054] Further, when the upward-moved screening neighborhood density and the downward-moved screening neighborhood density are less than the initial screening neighborhood density, it indicates that the temperature value corresponding to the central temperature screening point can best represent the temperature value of the first temperature data subsequence. Therefore, take the temperature data corresponding to the central temperature screening point as the first critical temperature value.

[0055] By dynamically adjusting (moving up or down) the screening lines in the screening space, more representative key temperature values are finely screened out. In this process, the initial screening lines are formed by intersecting the horizontal and vertical coordinates in the temperature screening space, ensuring that the starting points of the screening are located in the core area of the temperature data. Subsequently, the system moves the screening lines up and down to expand or shrink the screening range, and screens the temperature data depending on the screening bandwidth. And during the screening process, the system counts the density of each screening neighborhood, and the size of the density can reflect the density degree of the temperature data in the screening area. By continuously adjusting the position of the screening lines and deciding whether to continue the iteration according to the change of the density, this adaptive screening method can effectively find the key change area of the temperature data. Through this iterative process, the system can select the most representative temperature data points according to the change of the screening neighborhood, so as to obtain the first key temperature value. It achieves the technical effect of obtaining a reliable first key temperature value and providing a basis for the precise control of the drug constant temperature system.

[0056] S400: Based on the positions of the respective temperature sensors in the temperature sensor array, perform simultaneous chronological topology processing on the set of key temperature value sequences to construct a key temperature value topology network sequence;

[0057] In a possible embodiment, the key temperature value topology network sequence is a temperature change network sequence obtained after topology processing that reflects the constant temperature area of the target drug. The topology network sequence combines the spatial and temporal relationships of the temperature data to form a network structure, where each node represents a temperature sensor. This network can display the propagation characteristics of temperature changes in space and the mutual influence in time.

[0058] The position of each sensor represents a spatial coordinate. Combining this coordinate information with the temperature data sequence, the system can construct a topological structure of temperature changes in space. By performing topology processing on these sequences, a network sequence reflecting the temperature change law in the drug constant temperature area can be obtained.

[0059] Optionally, extract the set of key temperature values corresponding to the first time window in the set of key temperature value sequences, and connect each key temperature value according to the position of the corresponding temperature sensor to obtain a key temperature value topology network. Based on the above network construction process, perform simultaneous chronological topology processing on the set of key temperature value sequences to obtain a key temperature value topology network sequence.

[0060] The significance of this simultaneous sequential topology processing lies in that it not only considers the chronological order of temperature changes, but also reveals the temperature distribution relationship in space through the positions of the sensors. This topological network sequence can effectively represent the mutual influence of temperature changes between different positions, achieving the technical effect of providing accurate spatial and temporal data support for further control optimization.

[0061] S500: Perform iterative correlation analysis on the key temperature value topological network sequence in chronological order to determine the iterative correlation temperature characteristics of the constant temperature region of the target drug within a preset time period;

[0062] Further, as shown in the appendix Figure 2 As shown, performing iterative correlation analysis on the key temperature value topological network sequence in chronological order to determine the iterative correlation temperature characteristics of the constant temperature region of the target drug within a preset time period, step S500 of the embodiment of the present application further includes:

[0063] Extract the first key temperature value topological network and the second key temperature value topological network of the key temperature value topological network sequence in the order from front to back in time;

[0064] Calculate the network node similarity of the first key temperature value topological network and the second key temperature value topological network respectively to obtain a set of first network node similarities, where each network node similarity is the similarity degree of the key temperature values of the network nodes at the same position in the first key temperature value topological network and the second key temperature value topological network;

[0065] Perform matrix processing on the set of first network node similarities to obtain a first iterative correlation matrix;

[0066] Perform convolution calculation on the first iterative correlation matrix and the second key temperature value topological network to obtain a first iterative correlation temperature value topological network;

[0067] Extract the third key temperature value topological network of the key temperature value topological network sequence, and perform iterative correlation analysis on it and the first iterative correlation temperature value topological network to obtain a second iterative correlation temperature value topological network;

[0068] Perform multiple iterations of correlation in sequence until the iterative correlation analysis of the key temperature value topological network sequence is completed to obtain a target iterative correlation temperature value topological network, and perform feature extraction on the target iterative correlation temperature value topological network to obtain the iterative correlation temperature characteristics.

[0069] Further, step S500 of the embodiment of the present application further includes:

[0070] Pre-construct a similarity normalization function, where the similarity normalization function is:

[0071]

[0072] where Nor[lim(x i , y i )] is the normalized value corresponding to the i-th first network node similarity in the first network node similarity set, e is the base of the natural logarithm, m is the total number of first network node similarities in the first network node similarity set, and lim(x i , y j ) is the first network node similarity between the i-th key temperature value in the first key temperature value topological network and the i-th key temperature value at the same position in the second key temperature value topological network;

[0073] The first network node similarity set is normalized by using the similarity normalization function, and the processing result is embedded into an initially empty matrix to obtain the first iterative correlation matrix.

[0074] In one embodiment, each key temperature value topological network in the key temperature value topological network sequence reflects the temperature distribution of the constant temperature region of the target drug within a time window. By performing iterative correlation analysis on the key temperature value topological network sequence in chronological order, the temperature trend change between different time windows is mined, and the temperature characteristics of the constant temperature region of the target drug within a preset time period can be obtained, that is, the iterative correlation temperature characteristics. It achieves the technical effect of providing reliable data support for subsequent accurate optimization of the drug constant temperature system control.

[0075] Optionally, the first key temperature value topological network and the second key temperature value topological network of the key temperature value topological network sequence are extracted in chronological order. The network node similarities of the first key temperature value topological network and the second key temperature value topological network are calculated respectively to obtain a first network node similarity set, where each network node similarity is the similarity degree of the key temperature values of the network nodes at the same position in the first key temperature value topological network and the second key temperature value topological network.

[0076] Preferably, the first key temperature value and the second key temperature value at the same position in the first key temperature value topological network and the second key temperature value topological network are extracted, the difference between the first key temperature value and the second key temperature value is calculated, and the reciprocal of the calculation result is used as the first network node similarity. Based on the above principle, the first key temperature values and the second key temperature values at all the same positions in the first key temperature value topological network and the second key temperature value topological network are calculated for similarity to obtain the first network node similarity set.

[0077] Optionally, by performing matrix processing on the first network node similarity set, the correlation relationship between the first critical temperature value topology network and the second critical temperature value topology network is analyzed to obtain the first iterative correlation matrix.

[0078] Preferably, a similarity normalization function is pre-constructed, where the similarity normalization function is:

[0079]

[0080] where Nor[lim(x i , y i )] is the normalized value corresponding to the i-th first network node similarity in the first network node similarity set, e is the base of the natural logarithm, m is the total number of first network node similarities in the first network node similarity set, and lim(x i , y j ) is the first network node similarity between the i-th critical temperature value in the first critical temperature value topology network and the i-th critical temperature value at the same position in the second critical temperature value topology network.

[0081] Among them, the similarity normalization function is used to normalize the first network node similarities in the first network node similarity set and embed the processing results into an initially empty matrix to obtain the first iterative correlation matrix. Among them, the first iterative correlation matrix reflects the correlation relationship between the first critical temperature value topology network and the second critical temperature value topology network.

[0082] Furthermore, a graph convolutional network is used to perform convolutional calculations on the first iterative correlation matrix and the second critical temperature value topology network, and influence analysis is performed on the second critical temperature value topology network in combination with the correlation relationship to identify more detailed temperature change situations, thereby obtaining the first iterative correlation temperature value topology network.

[0083] Preferably, multiple sample iterative correlation matrices, multiple sample critical temperature value topology networks, and corresponding multiple sample iterative correlation temperature value topology networks are obtained as training data, and the training data is used to perform supervised training on a framework constructed based on a convolutional neural network to learn the one-to-two mapping relationship between the iterative correlation matrix and the critical temperature value topology network and the iterative correlation temperature value topology network, and a trained graph convolutional network is obtained.

[0084] Extract the third key temperature value topology network of the key temperature value topology network sequence again, perform iterative correlation analysis on it and the first iterative correlation temperature value topology network, and obtain the second iterative correlation temperature value topology network based on the same principle as obtaining the first iterative correlation temperature value topology network. Perform multiple iterations of correlation in sequence until the iterative correlation analysis of the key temperature value topology network sequence is completed, and obtain the target iterative correlation temperature value topology network. Use a convolutional neural network to extract features from the target iterative correlation temperature value topology network to obtain the iterative correlation temperature feature.

[0085] By performing iterative correlation analysis on the key temperature value topology network sequence, the temperature change characteristics of the constant temperature region of the target drug within a preset time period are revealed. This process can mine the temperature fluctuation law from the temperature data at multiple time nodes and identify potential problems that may exist in the temperature control system, such as abnormal temperature fluctuations and local temperature runaway. It achieves the technical effect of providing data support for subsequent reliable system control optimization.

[0086] S600: Control and optimize the drug constant temperature system based on the iterative correlation temperature feature to obtain control optimization parameters, where the drug constant temperature system is used to control the temperature of the target drug constant temperature region.

[0087] Further, based on the iterative correlation temperature feature to control and optimize the drug constant temperature system to obtain control optimization parameters, step S600 of the embodiment of the present application further includes:

[0088] Obtain an iterative correlation temperature feature sample set and a control optimization parameter sample set;

[0089] Train based on the iterative correlation temperature feature sample set and the control optimization parameter sample set to obtain a control optimization parameter recognizer;

[0090] Use the control optimization parameter recognizer to identify the iterative correlation temperature feature to obtain the control optimization parameter.

[0091] In one embodiment, the iterative correlation temperature feature sample set is composed of iterative correlation temperature features within multiple time periods, and each sample corresponds to the temperature change pattern within a time period. The control optimization parameter sample set contains control optimization parameters corresponding to each iterative correlation temperature feature sample. The control optimization parameters include various adjustment parameters of the temperature control system, such as the working intensity of the refrigeration or heating equipment, the rotation speed of the fan, the power of the heating tube, etc. Wherein, the drug constant temperature system is used to control the temperature of the target drug constant temperature region.

[0092] Use the iteratively associated temperature feature sample set and the control optimization parameter sample set to train machine learning models (such as neural networks, support vector machines, decision trees, etc.). The purpose of training is to enable the model to predict the corresponding control optimization parameters based on the input temperature change features (iteratively associated temperature features). Through training, a mapping relationship between the iteratively associated temperature features and the control optimization parameters is established. This process continuously adjusts the parameters in the model through an optimization algorithm (such as minimizing the error) to ensure that the prediction of the control parameters is as accurate as possible.

[0093] After the model training is completed, input the iteratively associated temperature features, and use the trained control optimization parameter recognizer to identify these features and predict the corresponding control optimization parameters. This process is carried out in real time and can automatically adjust the control parameters of the drug constant temperature system according to the continuously changing temperature data. According to the recognition results, the corresponding control optimization parameters are output, and these parameters will guide the temperature control system to make real-time adjustments. Through this process, the temperature control system can adapt to environmental temperature changes and maintain the temperature stability of the target drug constant temperature area.

[0094] The system feeds back the temperature change according to the result of each control optimization. If the temperature remains within the desired range, the control optimization parameters can continue to be used. If the temperature change exceeds the preset range, the optimization process will be carried out again to adjust the control optimization parameters. That is, by continuously monitoring and adjusting the working state of the drug constant temperature system, the continuous stability of temperature control is ensured.

[0095] By combining machine learning with control optimization strategies, the control parameters of the drug constant temperature system are dynamically optimized based on the iteratively associated temperature features. This method can not only improve the timeliness of system response, but also improve the accuracy and stability of temperature control, which helps to ensure the quality and safety of drugs.

[0096] In summary, the embodiments of the present application have at least the following technical effects:

[0097] 1. Through the analysis of iteratively associated temperature features, the present application accurately identifies the temperature fluctuation pattern, optimizes the control parameters, and achieves the technical effects of improving the response speed and accuracy of the drug constant temperature system and ensuring temperature stability.

[0098] 2. By training and optimizing the control parameters through machine learning, the drug constant temperature system can automatically adjust to different environmental changes, achieving the technical effect of enhancing the system's adaptability to external changes.

[0099] 3. By performing dimensionality reduction processing on the temperature data, the present application achieves the technical effect of improving the data analysis efficiency.

[0100] It should be noted that the above order of the embodiments of the present application is only for description and does not represent the superiority or inferiority of the embodiments. In addition, the specific embodiments of this specification have been described. Further, the processes depicted in the drawings do not necessarily require the particular order or sequential order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0101] The above are only the preferred embodiments of the present application and are not intended to limit the present application. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present application shall be included within the protection scope of the present application.

[0102] This specification and the drawings are only exemplary descriptions of the present application and are considered to have covered any and all modifications, variations, combinations, or equivalents within the scope of the present application. Obviously, those skilled in the art can make various changes and modifications to the present application without departing from the scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the present application and its equivalent technologies, the present application is intended to include these changes and modifications.

Claims

1. A drug constant temperature system control optimization method, characterized in that: The method comprises: Acquire a pre-placed temperature sensor array of a target drug constant temperature area; Traversing the temperature sensor array to extract temperature data within a preset time period to obtain a temperature data sequence set, and presetting a fixed time window, using the preset fixed time window to perform sequence segmentation processing on the temperature data sequence set to obtain a temperature data subsequence cluster, wherein the temperature data subsequence cluster includes a temperature data subsequence set corresponding to each temperature data sequence; Performing subsequence key temperature value screening on each temperature data subsequence in the temperature data subsequence cluster respectively, and performing serialization processing on the screening results to obtain a key temperature value sequence set corresponding to the temperature data sequence set; Based on the position of each temperature sensor in the temperature sensor array, performing simultaneous topological processing on the key temperature value sequence set to construct a key temperature value topological network sequence; Performing iterative correlation analysis on the key temperature value topological network sequence in chronological order to determine iterative correlation temperature characteristics of the constant temperature area of ​​the target drug within a preset time period; Based on the iterative correlation temperature characteristics, the drug constant temperature system is controlled and optimized to obtain control optimization parameters, wherein the drug constant temperature system is used to control the temperature of the target drug constant temperature area; Among them, performing iterative correlation analysis on the key temperature value topological network sequence in chronological order to determine the iterative correlation temperature characteristics of the constant temperature area of ​​the target drug within a preset time period includes: Extracting a first key temperature value topological network and a second key temperature value topological network of the key temperature value topological network sequence in a time-ordered manner; Respectively calculating the network node similarities of the first critical temperature value topology network and the second critical temperature value topology network to obtain a first network node similarity set, wherein each network node similarity is the degree of similarity of the critical temperature values ​​of the network nodes at the same position in the first critical temperature value topology network and the second critical temperature value topology network; Performing matrix processing on the first network node similarity set to obtain a first iterative association matrix; Performing convolution calculation on the first iterative association matrix and the second key temperature value topological network to obtain a first iterative association temperature value topological network; Extracting a third key temperature value topological network of the key temperature value topological network sequence, performing iterative correlation analysis on the third key temperature value topological network and the first iterative correlation temperature value topological network to obtain a second iterative correlation temperature value topological network; Perform multiple iterations of association in sequence until the iteration association analysis of the key temperature value topological network sequence is completed, obtain the target iteration association temperature value topological network, perform feature extraction on the target iteration association temperature value topological network, and obtain the iteration association temperature feature.

2. A drug constant temperature system control optimization method as claimed in claim 1, characterized in that: Performing subsequence key temperature value screening on each temperature data subsequence in the temperature data subsequence cluster respectively, and performing serialization processing on the screening results to obtain a key temperature value sequence set corresponding to the temperature data sequence set, including: Randomly extracting a first temperature data subsequence from the temperature data subsequence cluster, and constructing a screening space based on the first temperature data subsequence to obtain a temperature screening space, wherein the temperature screening space includes a plurality of temperature screening points, and each temperature screening point corresponds to a temperature data; Extracting a central temperature screening point of the temperature screening space, taking the central temperature screening point as an index, screening a key temperature value in the temperature screening space according to a preset screening bandwidth, and obtaining a first key temperature value; Performing subsequence key temperature value screening on each temperature data subsequence in the temperature data subsequence cluster to obtain a subsequence key temperature value cluster; Each subsequence key temperature value set in the subsequence key temperature value cluster is serialized to obtain the key temperature value sequence set.

3. A drug constant temperature system control optimization method as claimed in claim 2, characterized in that: Extracting a central temperature screening point of the temperature screening space, taking the central temperature screening point as an index, screening a key temperature value in the temperature screening space according to a preset screening bandwidth, and obtaining a first key temperature value, including: A straight line passing through the central temperature screening point and parallel to the abscissa axis of the temperature screening space is used as the initial screening straight line, wherein the temperature screening space is a two-dimensional space, the abscissa axis is time, and the ordinate axis is temperature value; Constructing an initial screening neighborhood of the initial screening straight line based on a preset screening bandwidth, and counting the initial screening neighborhood density of the initial screening neighborhood; The initial screening straight line is moved upward and downward respectively by the preset screening bandwidth to obtain an upward screening straight line and a downward screening straight line; Based on the upward screening straight line and the downward screening straight line, respectively, an upward screening neighborhood and a downward screening neighborhood are constructed in combination with the preset screening bandwidth, and the density of the upward screening neighborhood and the density of the downward screening neighborhood are counted; When the upward-moving screening neighborhood density and / or the downward-moving screening neighborhood density is greater than or equal to the initial screening neighborhood density, the upward-moving screening straight line is iterated upward according to the preset screening bandwidth and / or the downward-moving screening straight line is iterated downward according to the preset screening bandwidth until the preset number of iterations is met, and the target upward-moving screening straight line and / or the target downward-moving screening straight line are determined; Based on the target upward screening line and / or the target downward screening line, a key temperature value screening is performed in the temperature screening space to obtain a first key temperature value.

4. A drug constant temperature system control optimization method as claimed in claim 3, characterized in that: When the upward-moving screening neighborhood density and the downward-moving screening neighborhood density are less than the initial screening neighborhood density, the temperature data corresponding to the central temperature screening point is used as the first key temperature value.

5. A method for optimizing the control of a medicine constant temperature system as claimed in claim 3, characterized in that: Performing key temperature value screening in the temperature screening space based on the target upward screening line and / or the target downward screening line to obtain a first key temperature value includes: Adding the temperature data corresponding to the target upward filter line and / or the target downward filter line into the key temperature value filter set; The key temperature value corresponding to the maximum value in the key temperature value screening set is used as the first key temperature value.

6. A drug constant temperature system control optimization method as claimed in claim 1, characterized in that: include: A similarity normalization function is pre-constructed, wherein the similarity normalization function is: Among them, Nor[lim(x i ,y i )] is the normalized value corresponding to the i-th first network node similarity in the first network node similarity set, e is the base of the natural logarithm, m is the total number of first network node similarities in the first network node similarity set, lim(x i ,y i ) is the first network node similarity between the ith key temperature value in the first key temperature value topological network and the ith key temperature value at the same position in the second key temperature value topological network; The first network node similarity set is normalized using the similarity normalization function, and the processing result is embedded in an initially empty matrix to obtain the first iterative association matrix.

7. A drug constant temperature system control optimization method as claimed in claim 1, characterized in that: Based on the iterative correlation temperature characteristics, the drug constant temperature system is controlled and optimized to obtain control optimization parameters, including: Obtaining an iterative correlation temperature feature sample set and a control optimization parameter sample set; Performing training based on the iterative correlation temperature feature sample set and the control optimization parameter sample set to obtain a control optimization parameter identifier; The iterative correlation temperature feature is identified by using the control optimization parameter identifier to obtain the control optimization parameter.

Images