An intelligent data collection method and system for a liquid drug preparation production line

Abstract

The present invention relates to the field of data processing, and in particular to an intelligent data collection method and system for a liquid drug preparation production line. The method first uses sensors to collect detection data of different production parameters, obtains the number of failures of liquid drug preparation production equipment, collects the detection data of all sensors of the same production parameter at the same time for integration, obtains the comprehensive data of each production parameter, obtains the collection demand coefficient of the production parameter according to the change of the comprehensive data of the production parameter at each time, the correlation between the target production parameter and other production parameters except the target production parameter, and the number of failures, adjusts the parameters of each sensor according to the difference of the detection data of each sensor collecting the same production parameter and the collection demand coefficient of the production parameter, and uses the adjusted sensor for data collection. The present invention can adaptively adjust the collection accuracy of the sensor and improve the effect of data collection.

Description

Technical Field

[0001] The present invention relates to the field of data processing, and in particular to an intelligent data acquisition method and system for a liquid drug preparation production line. Background Art

[0002] In the actual production process of liquid drug preparations, accurate collection of data on various production parameters such as temperature, pressure or concentration can help to promptly identify problems in the production process, ensure the stability and consistency of the quality of the preparations, and improve the production efficiency of liquid drug preparations. It can also predict production equipment failures, perform maintenance in advance, and reduce the risk of production interruptions.

[0003] In the related art, sensors with default or fixed parameter values ​​are usually used to collect data on various production parameters in the production process of liquid drug preparations. However, since there are many production parameters involved in the production process of liquid drug preparations, and the changes and importance of different production parameters in the production process are different, the sensors with default or fixed parameter values ​​cannot guarantee the data collection accuracy of different production parameters, thereby reducing the effect of data collection. Summary of the invention

[0004] In order to solve the technical problem that sensors using default or fixed parameter values ​​cannot guarantee the data collection accuracy of different production parameters, thereby reducing the effect of data collection, the purpose of the present invention is to provide an intelligent data collection method and system for a liquid drug preparation production line. The technical solution adopted is as follows:

[0005] The present invention proposes an intelligent data collection method for a liquid drug preparation production line, the method comprising:

[0006] The sensors are used to collect the detection data of different production parameters of the production link to be tested at each moment within a preset time period, and the number of failures that occurred in the liquid drug preparation production equipment within the preset time period is obtained; the sensors that collect the same production parameters are divided into the same sensor set, and the detection data collected by all sensors in each sensor set at the same moment are integrated to obtain the integrated data of each production parameter at each moment;

[0007] Taking any production parameter as the target production parameter, obtaining the acquisition requirement coefficient of the target production parameter according to the change of the comprehensive data of the target production parameter at each moment, the correlation between the target production parameter and other production parameters except the target production parameter, and the number of failures;

[0008] The sensor set where the sensor for collecting the target production parameter is located is taken as the target sensor set. If there is only one sensor in the target sensor set, the acquisition requirement coefficient is taken as the adjustment coefficient of the sensor in the target sensor set. If there are multiple sensors in the target sensor set, any one sensor in the target sensor set is taken as the target sensor, and the adjustment coefficient of the target sensor is obtained according to the difference in detection data collected at the same time between the target sensor and other sensors in the target sensor set except the target sensor, the detection data collected by the target sensor, and the acquisition requirement coefficient.

[0009] Based on the adjustment coefficient of each sensor, the parameters in the sensor are adjusted to obtain an adjusted sensor, and the adjusted sensor is used to collect data.

[0010] Furthermore, the acquisition requirement coefficient for obtaining the target production parameter includes:

[0011] According to the changes of the comprehensive data of the target production parameters at each moment, the data change degree of the target production parameters is obtained;

[0012] Analyze the correlation between the target production parameter and other production parameters except the target production parameter to obtain the criticality of the target production parameter;

[0013] The data variation degree, the criticality and the number of failures are integrated and normalized to obtain a collection requirement coefficient of a target production parameter.

[0014] Furthermore, the degree of data variation of the target production parameter obtained includes:

[0015] Mapping the comprehensive data of the target production parameter at each moment into a two-dimensional coordinate system to obtain multiple data points of the target production parameter;

[0016] Extract extreme value points from all data points of the target production parameters, take any two adjacent extreme value points as extreme value point groups, and obtain the mutation degree of each extreme value point group according to the absolute value of the difference of the comprehensive data between the two extreme value points in each extreme value point group and the absolute value of the difference at the moment, wherein the mutation degree is positively correlated with the absolute value of the difference of the comprehensive data between the two extreme value points in each extreme value point group, and the mutation degree is negatively correlated with the absolute value of the difference at the moment between the two extreme value points in each extreme value point group;

[0017] Analyzing the overall level of the mutation degree of all extreme value point groups of the target production parameter to obtain the fluctuation degree of the target production parameter;

[0018] The fluctuation degree and the number of extreme value points of the target production parameter are integrated and normalized to obtain the data variation degree of the target production parameter.

[0019] Furthermore, the criticality of obtaining the target production parameters includes:

[0020] The absolute value of the Pearson correlation coefficient of the comprehensive data at each moment between the target production parameter and each other production parameter except the target production parameter is taken as the correlation coefficient between the target production parameter and each other production parameter.

[0021] The overall level of the correlation coefficients between the target production parameter and all other production parameters is analyzed to obtain the criticality of the target production parameter.

[0022] Further, obtaining the adjustment coefficient of the target sensor according to the difference between the detection data collected at the same time between the target sensor and other sensors in the target sensor set except the target sensor, the detection data collected by the target sensor, and the collection requirement coefficient includes:

[0023] Obtaining an overall difference value of the target sensor according to a difference in detection data collected at the same time between the target sensor and other sensors in the target sensor set except the target sensor;

[0024] Analyze the overall level of the detection data collected by the target sensor at all times to obtain the overall data value of the target sensor;

[0025] The overall difference value and the overall data value of the target sensor and the acquisition requirement coefficient of the target production parameter are integrated and normalized to obtain the adjustment coefficient of the target sensor.

[0026] Further, obtaining the overall difference value of the target sensor includes:

[0027] An arbitrary moment is taken as a target moment, and the absolute value of the difference between the detection data collected by the target sensor and each other sensor at the target moment is taken as the initial data difference value between the target sensor and each other sensor at the target moment;

[0028] Taking the accumulated value of the initial data difference values ​​between the target sensor and all other sensors at the target time as the comprehensive data difference value of the target sensor at the target time;

[0029] The overall level of the comprehensive data difference value of the target sensor at all times is analyzed to obtain the overall difference value of the target sensor.

[0030] Further, analyzing the overall level of the comprehensive data difference value of the target sensor at all times to obtain the overall difference value of the target sensor includes:

[0031] The average value of the comprehensive data difference values ​​of the target sensor at all times is taken as the overall difference value of the target sensor.

[0032] Further, adjusting the parameters in the sensor based on the adjustment coefficient of each sensor to obtain the adjusted sensor includes:

[0033] For any sensor, if the adjustment coefficient of the sensor is greater than a preset adjustment threshold, the product value of the adjustment coefficient of the sensor and the original value of the parameter of the sensor is used as the adjustment value of the parameter of the sensor; the sum of the original value of the parameter of the sensor and the adjustment value is used as the adjustment value of the parameter of the sensor, and the value used by the parameter of the sensor is adjusted to the adjustment value;

[0034] If the adjustment coefficient of the sensor is not greater than the preset adjustment threshold, the parameters in the sensor are not adjusted.

[0035] Furthermore, the obtaining of comprehensive data of each production parameter at each moment includes:

[0036] The average value of the detection data collected by all sensors in each sensor set at the same time is used as the comprehensive data of each production parameter at each time.

[0037] The present invention also proposes an intelligent data acquisition system for a liquid drug preparation production line, the system comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, the processor implementing any one of the steps of an intelligent data acquisition method for a liquid drug preparation production line when executing the computer program.

[0038] The present invention has the following beneficial effects:

[0039] The present invention takes into account that the existing sensors using default or fixed parameter values ​​cannot guarantee the data collection accuracy for different production parameters, thereby reducing the effect of data collection. Therefore, the sensor is first used to collect the detection data of the production parameters of the production link to be tested at each moment within a preset time period, and the number of failures that occurred in the liquid drug preparation production equipment within the preset time period is obtained. Considering that multiple sensors of the same type are required for data collection for certain production parameters, the sensors that collect the same production parameter are divided into the same sensor set, and then the comprehensive data of each production parameter at each moment is obtained. For production parameters with more drastic data changes and relatively important production parameters, a higher-precision sensor is required to collect the production parameter, and the production equipment When the number of failures is high, higher-precision sensors are also needed to collect production parameters. Therefore, the collection requirement coefficient can be used to reflect the degree of accuracy required by the sensor for data collection of the target production parameters. At the same time, when there is only one sensor for collecting the target production parameters, the collection requirement coefficient of the target production parameters can be directly used as the adjustment coefficient of the sensor. When there are multiple sensors for collecting the target production parameters, the requirements of each sensor for data collection accuracy are still different. Therefore, the adjustment coefficient of each sensor can be further obtained, and then the adjustment coefficient can be used to adjust the parameters in the sensor, so that the sensor can meet the data collection requirements and improve the accuracy of data collection for different production parameters, thereby improving the final data collection effect. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0041] Figure 1 A flow chart of an intelligent data collection method for a liquid drug preparation production line provided by one embodiment of the present invention;

[0042] Figure 2 A flow chart of a method for obtaining a collection requirement coefficient of a target production parameter provided by an embodiment of the present invention;

[0043] Figure 3 A flow chart of a method for obtaining an adjustment coefficient of a target sensor provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0044] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the following is a detailed description of the intelligent data collection method and system for a liquid drug preparation production line proposed by the present invention, its specific implementation, structure, characteristics and effects, in combination with the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" does not necessarily refer to the same embodiment. In addition, specific features, structures or characteristics in one or more embodiments may be combined in any suitable form.

[0045] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.

[0046] The specific scheme of the intelligent data collection method and system for a liquid drug preparation production line provided by the present invention is described in detail below with reference to the accompanying drawings.

[0047] See also Figure 1 , which shows a flow chart of an intelligent data collection method for a liquid drug preparation production line provided by an embodiment of the present invention, the method comprising:

[0048] Step S1: Use sensors to collect detection data of different production parameters of the production link to be tested at each moment within a preset time period, and obtain the number of failures that occur in the liquid drug preparation production equipment within the preset time period; divide the sensors that collect the same production parameter into the same sensor set, and integrate the detection data collected by all sensors in each sensor set at the same moment to obtain the comprehensive data of each production parameter at each moment.

[0049] The production process of liquid pharmaceutical preparations usually includes multiple production links, such as extraction, preparation or filtration. In each production link, it is usually necessary to use various sensors installed on the production equipment to collect data on various production parameters such as temperature, pressure, pH value and concentration in the production process to analyze the production progress and situation of liquid pharmaceutical preparations.

[0050] The embodiment of the present invention first uses various sensors arranged on the liquid drug preparation production equipment to collect the detection data of the production parameters of the production link to be tested at each moment within a preset time period, wherein the production environment to be tested can be selected from the preparation link of the preparation, and the preset time period is set to 2 hours. The specific value of the preset time period can also be set by the implementer according to the specific implementation scenario, which is not limited here.

[0051] It should be noted that the sampling frequencies of the sensors are the same. In one embodiment of the present invention, the sampling frequency of the sensor is set to 100 Hz. The specific value of the sampling frequency can also be set by the implementer according to the specific implementation scenario and is not limited here.

[0052] At the same time, the production log of the liquid drug preparation production equipment usually records the number of failures that occur during the operation of the equipment in real time. Therefore, the embodiment of the present invention also needs to obtain the number of failures that occur in the production equipment within a preset time period from the log. Subsequently, the data collection accuracy of the sensor can be adjusted based on the number of failures and the detection data collected by the sensor to improve the data collection effect.

[0053] In the production process of liquid pharmaceutical preparations, for certain production parameters, such as temperature, it is necessary to arrange multiple temperature sensors at different positions of the production equipment to collect temperature data so as to analyze the temperature changes at different positions. That is to say, there are multiple sensors of the same type used to collect data for a certain production parameter, and there are one sensor used to collect data for a certain production parameter. Therefore, in an embodiment of the present invention, sensors that collect the same production parameter are first divided into the same set to obtain multiple sensor sets. At this time, a specific sensor set corresponds to a specific production parameter. When there is only one sensor in a sensor set, it means that only one sensor is used for data collection for the production parameter. When there are multiple sensors in a sensor set, it means that multiple sensors of the same type are used for data collection for the production parameter. Then, the detection data collected by all sensors in each sensor set at the same time are integrated to obtain the integrated data of each production parameter at each time. Subsequently, based on the integrated data, the data collection requirements of each production parameter can be analyzed to adjust the sensor collection accuracy.

[0054] Preferably, in one embodiment of the present invention, the method for acquiring comprehensive data of each production parameter at each moment specifically includes:

[0055] The average value of the detection data collected by all sensors in each sensor set at the same time is used as the comprehensive data of each production parameter at each time.

[0056] It should be noted that, for a sensor set including only one sensor, the comprehensive data of the production parameters corresponding to the sensor set at each moment is the detection data collected by the sensor included in the sensor set at each moment.

[0057] Step S2: Take any production parameter as the target production parameter, and obtain the acquisition requirement coefficient of the target production parameter according to the changes in the comprehensive data of the target production parameter at each moment, the correlation between the target production parameter and other production parameters except the target production parameter, and the number of failures.

[0058] Since specific production parameters require the use of specific sensors for data collection, such as temperature parameters require the use of temperature sensors to collect data, and pressure parameters require the use of pressure sensors to collect data. In the production process of liquid pharmaceutical preparations, due to the different importance of different production parameters and different data changes, the data collection accuracy of sensors for different production parameters should also be different. The data collection accuracy of sensors is usually reflected through parameters such as resolution or sensitivity. For production parameters with more drastic data changes and relatively important production parameters, higher-precision sensors are required to collect the production parameters in order to capture more subtle data changes of the production parameters, providing a more accurate basis for data analysis and processing. At the same time, if the liquid pharmaceutical preparation production equipment fails frequently, higher-precision sensors are also required to collect production parameters, which helps to more accurately capture and analyze the operating status of the equipment, predict production equipment failures in advance, and thus improve the reliability and stability of the equipment.

[0059] Therefore, the embodiment of the present invention first analyzes any production parameter, takes any production parameter as the target production parameter, and analyzes the changes in the comprehensive data of the target production parameter at each moment, the correlation between the target production parameter and other production parameters except the target production parameter, and the number of failures of the production equipment. The acquired acquisition requirement coefficient reflects the degree of accuracy requirement of the sensor for data acquisition of the target production parameter. The larger the acquisition requirement coefficient, the more a sensor with higher accuracy is required for data acquisition of the target production parameter. The data acquisition accuracy of the sensor can be adjusted based on the acquisition requirement coefficient to improve the data acquisition effect of the sensor on the target production parameter.

[0060] Preferably, in one embodiment of the present invention, the method for obtaining the acquisition requirement coefficient of the target production parameter specifically includes:

[0061] See also Figure 2 , which shows a flow chart of a method for obtaining a collection requirement coefficient of a target production parameter provided by an embodiment of the present invention.

[0062] Step S201: Obtain the data change degree of the target production parameter according to the change of the comprehensive data of the target production parameter at each moment.

[0063] From the above analysis, it can be seen that the more drastic the change in the comprehensive data of the target production parameters, the more a sensor with higher accuracy is needed to collect data on the target production parameters. Therefore, the change in the comprehensive data of the target production parameters at each moment can be analyzed, and the degree of change in the comprehensive data of the target production parameters can be reflected by the degree of change in the acquired data. The greater the degree of data change, the more drastic the change in the comprehensive data of the target production parameters. Subsequently, the collection demand coefficient of the target production parameters can be accurately calculated based on the degree of data change.

[0064] Preferably, in one embodiment of the present invention, the method for obtaining the data variation degree of the target production parameter specifically includes:

[0065] First, the comprehensive data of the target production parameter at each moment is mapped into a two-dimensional coordinate system to obtain multiple data points of the target production parameter, wherein the horizontal axis of the two-dimensional coordinate system represents the moment and the vertical axis represents the comprehensive data of the target production parameter;

[0066] By using the existing Newton method, extreme points are extracted from all data points of the target production parameters, including maximum points and minimum points. Other existing methods can also be used to extract extreme points, which are not limited here. Any two adjacent extreme points are taken as an extreme point group. The greater the difference in the comprehensive data of the two extreme points in the extreme point group, and the smaller the difference in time, the more obvious the data change between the two extreme points. Therefore, the mutation degree of each extreme point group can be obtained according to the absolute value of the difference in the comprehensive data between the two extreme points in each extreme point group and the absolute value of the difference in time. The mutation degree is positively correlated with the absolute value of the difference in the comprehensive data between the two extreme points in each extreme point group, and the mutation degree is negatively correlated with the absolute value of the difference in time between the two extreme points in each extreme point group. The greater the mutation degree, the more obvious the change in the comprehensive data of each data point between the two extreme points in the extreme point group.

[0067] Then, the overall level of mutation degree of all extreme point groups of the target production parameters is analyzed to obtain the fluctuation degree of the target production parameters. The greater the fluctuation degree, the more obvious the fluctuation change characteristics of the comprehensive data of the target production parameters.

[0068] In an embodiment of the present invention, the average value or median of the mutation degree of all extreme point groups of the target production parameter can be used as the fluctuation degree of the target production parameter to achieve the overall level analysis of the mutation degree of all extreme point groups of the target production parameter, which is not limited here.

[0069] At the same time, the more extreme points there are, the more obvious the change characteristics of the comprehensive data of the target production parameters are. Therefore, the fluctuation degree of the target production parameters and the number of extreme points are integrated and normalized to obtain the data change degree of the target production parameters.

[0070] In the embodiment of the present invention, the combination of the fluctuation degree of the target production parameter and the number of extreme value points can be achieved by calculating the sum or product of the two, which is not limited here.

[0071] In one embodiment of the present invention, the normalization processing can be specifically, for example, maximum and minimum value normalization processing, and the normalization in subsequent steps can all adopt maximum and minimum value normalization processing. In other embodiments of the present invention, other normalization methods can be selected according to the specific range of numerical values, which will not be described in detail.

[0072] As an example, in one embodiment of the present invention, the expression of the data variation degree of the target production parameter may be specifically, for example, as follows:

[0073]

[0074] in, Indicates the degree of data change of target production parameters; Indicates the number of extreme points of the target production parameter; and Respectively represent Comprehensive data of two extreme points in a group of extreme points; and Respectively represent The moment of two extreme points in a group of extreme points; Indicates the number of extreme point groups; represents the normalization function; Indicates The degree of mutation of each extreme point group; Indicates the degree of fluctuation of the target production parameter.

[0075] It should be noted that in other embodiments of the present invention, the positive and negative correlations may be represented by other basic mathematical operations, which will not be elaborated herein.

[0076] Step S202: Analyze the correlation between the target production parameter and other production parameters except the target production parameter to obtain the criticality of the target production parameter.

[0077] The stronger the correlation between the target production parameter and other production parameters, the more data changes of the target production parameter are usually accompanied by data changes of other production parameters, which further indicates that the target production parameter needs more attention in the production process of liquid drug preparations, and the more precise sensors are needed to collect data on the target production parameter. Therefore, the correlation between the target production parameter and other production parameters except the target production parameter can be analyzed, and the importance of the target production parameter can be reflected by the criticality obtained. Subsequently, the collection requirement coefficient of the target production parameter can be calculated based on the criticality.

[0078] Preferably, in one embodiment of the present invention, the method for obtaining the criticality of the target production parameter specifically includes:

[0079] The absolute value of the Pearson correlation coefficient of the comprehensive data at each moment between the target production parameter and each other production parameter except the target production parameter is used as the correlation coefficient between the target production parameter and each other production parameter. The Pearson correlation coefficient is a technical means well known to those skilled in the art and will not be elaborated here. Then, the overall level of the correlation coefficient between the target production parameter and all other production parameters is analyzed to obtain the criticality of the target production parameter.

[0080] In an embodiment of the present invention, the average value or median of the correlation coefficient between the target production parameter and all other production parameters may be used as the criticality of the target production parameter to achieve an overall level analysis of the correlation coefficient between the target production parameter and all other production parameters, which is not limited here.

[0081] As an example, in one embodiment of the present invention, the expression of the criticality of the target production parameter may be specifically, for example, as follows:

[0082]

[0083] in, Indicates the criticality of the target production parameter; Indicates the target production parameters and the Pearson correlation coefficients between other production parameters; Indicates target production parameters and the second Correlation coefficients between other production parameters; Indicates the number of production parameters other than the target production parameters.

[0084] Step S203: The data variation degree, criticality and number of failures are integrated and normalized to obtain the acquisition requirement coefficient of the target production parameter.

[0085] The greater the data variation and criticality of the target production parameters, and the more failures the production equipment has, the more accurate sensors are needed to collect data on the target production parameters. Therefore, the data variation, criticality and failure times can be combined for analysis to obtain the collection requirement coefficient of the target production parameters.

[0086] In the embodiment of the present invention, the integration of the three can be achieved by calculating the sum or product of the data change degree and criticality of the target production parameter and the number of failures of the production equipment, which is not limited here.

[0087] As an example, in one embodiment of the present invention, the expression of the acquisition requirement coefficient of the target production parameter may be specifically, for example, as follows:

[0088]

[0089] in, represents the acquisition requirement coefficient of the target production parameter; Indicates the degree of data change of target production parameters; Indicates the criticality of the target production parameter; Indicates the number of failures that occurred in the liquid drug preparation production equipment within a preset time period; Represents the normalization function.

[0090] The collection requirement coefficient of each production parameter can be obtained by the same method.

[0091] Step S3: The sensor set including the sensor for collecting the target production parameters is taken as the target sensor set. If there is only one sensor in the target sensor set, the acquisition requirement coefficient is taken as the adjustment coefficient of the sensors in the target sensor set. If there are multiple sensors in the target sensor set, any one of the sensors in the target sensor set is taken as the target sensor. The adjustment coefficient of the target sensor is obtained according to the difference in detection data collected at the same time between the target sensor and other sensors in the target sensor set except the target sensor, the detection data collected by the target sensor and the acquisition requirement coefficient.

[0092] In the production process of liquid pharmaceutical preparations, for certain production parameters such as temperature and pressure, due to the different temperatures or pressures in different areas of the production equipment, usually multiple sensors of the same type are set in different areas of the production equipment to collect data. For example, it is necessary to set multiple temperature sensors in different areas of the production equipment to collect temperature data. That is to say, for the target production parameter, there may be multiple sensors to collect data from it, or there may be only one sensor to collect data from it. When there are multiple sensors to collect data for the target production parameter, the collection accuracy of different sensors still needs to be adjusted to different degrees. Therefore, it is necessary to discuss it in two situations.

[0093] First, the sensor set where the sensors that collect the target production parameters are located is taken as the target sensor set. In the first case, when there is only one sensor in the target sensor set, it means that only one sensor collects data for the target production parameters. At this time, the collection requirement coefficient of the target production parameters can be directly used as the adjustment coefficient of the sensors in the target sensor set. The larger the adjustment coefficient, the greater the degree of subsequent adjustment of the collection accuracy of the sensor.

[0094] As an example, in one embodiment of the present invention, the expression of the adjustment coefficient of the sensor in the target sensor set may be specifically, for example, as follows:

[0095]

[0096] in, represents the adjustment coefficient of the sensor in the target sensor set; Represents the acquisition requirement coefficient of the target production parameter.

[0097] In the second case, when there are multiple sensors in the target sensor set, it means that there are multiple sensors of the same type for collecting data on the target production parameters. In this case, it means that the target production parameters are special production parameters such as temperature or pressure. For example, if the target production parameter is a temperature parameter, then among the multiple temperature sensors for collecting temperature parameters, the closer a certain temperature sensor is to the heat source, the greater the difference in detection data collected at the same time between the temperature sensor and other temperature sensors, and the larger the detection data collected by the temperature sensor, the temperature sensor requires higher data collection accuracy. Therefore, any sensor in the target sensor set can be first used as the target sensor, and the distance between the target sensor and the heat source or pressure source is analyzed according to the difference in detection data collected at the same time between the target sensor and other sensors in the target sensor set except the target sensor, and the detection data collected by the target sensor, and the adjustment coefficient of the target sensor is obtained in combination with the collection requirement coefficient of the target production parameter.

[0098] Preferably, in one embodiment of the present invention, the method for obtaining the adjustment coefficient of the target sensor in the target sensor set specifically includes:

[0099] See also Figure 3 , which shows a flow chart of a method for obtaining an adjustment coefficient of a target sensor provided by an embodiment of the present invention.

[0100] Step S301: obtaining an overall difference value of the target sensor according to the difference between the detection data collected at the same time by the target sensor and other sensors in the target sensor set except the target sensor.

[0101] The greater the difference in detection data collected at the same time between the target sensor and other sensors in the target sensor set except the target sensor, the closer the target sensor is to the heat source or pressure source, which means that the data collection accuracy of the target sensor needs to be adjusted to a greater extent in the subsequent process. Therefore, the overall difference value obtained can be used to reflect the degree of difference in detection data collected between the target sensor and other sensors in the target sensor set, and the adjustment coefficient of the target sensor can be calculated and analyzed based on the overall difference value.

[0102] Preferably, in one embodiment of the present invention, the method for obtaining the overall difference value of the target sensor specifically includes:

[0103] Take any moment as the target moment, take the absolute value of the difference between the detection data collected at the target moment between the target sensor and each other sensor as the initial data difference value between the target sensor and each other sensor at the target moment, take the accumulated value of the initial data difference values ​​between the target sensor and all other sensors at the target moment as the comprehensive data difference value of the target sensor at the target moment. The larger the comprehensive data difference value, the greater the difference in the detection data collected at the target moment between the target sensor and other sensors. The comprehensive data difference value of the target sensor at each moment can be obtained by the same method, and then the overall level of the comprehensive data difference value of the target sensor at all moments is analyzed to obtain the overall difference value of the target sensor.

[0104] In the embodiment of the present invention, the average value or median of the comprehensive data difference values ​​of the target sensor at all times may be used as the overall difference value of the target sensor to implement the overall level analysis of the comprehensive data difference values ​​of the target sensor at all times, which is not limited here.

[0105] As an example, in one embodiment of the present invention, the expression of the overall difference value of the target sensor may be specifically, for example, as follows:

[0106]

[0107]

[0108] in, Represents the overall difference value of the target sensor; Indicates that the target sensor is The comprehensive data difference value at the moment; Indicates the number of detection data collected by the target sensor, that is, the number of time moments; Indicates the comprehensive data difference value of the target sensor at the target time; Represents the detection data collected by the target sensor at the target time; Indicates the first sensor in the target sensor set except the target sensor Detection data collected by other sensors at the target time; Indicates the number of other sensors in the target sensor set except the target sensor; Indicates the target sensor and the first sensor in the target sensor set other than the target sensor. The initial data difference value between the other sensors at the target time.

[0109] Step S302: Analyze the overall level of the detection data collected by the target sensor at all times to obtain the overall data value of the target sensor.

[0110] The larger the detection data collected by the target sensor at each moment, the closer the target sensor is to the heat source or pressure source, which means that the data collection accuracy of the target sensor needs to be adjusted to a greater extent in the subsequent process. Therefore, the overall level of the detection data collected by the target sensor at all moments can be analyzed to obtain the overall data value of the target sensor.

[0111] In an embodiment of the present invention, the average value or median of the detection data collected by the target sensor at all times may be used as the overall data value of the target sensor to implement overall level analysis of the detection data collected by the target sensor at all times, which is not limited here.

[0112] As an example, in one embodiment of the present invention, the expression of the overall data value of the target sensor may be specifically, for example, as follows:

[0113]

[0114] in, Represents the overall data value of the target sensor; Indicates the number of detection data collected by the target sensor, that is, the number of time moments; Indicates the first The detection data at a certain moment.

[0115] Step S303: The overall difference value and the overall data value of the target sensor and the acquisition requirement coefficient of the target production parameter are integrated and normalized to obtain the adjustment coefficient of the target sensor.

[0116] The larger the overall difference value and the overall data value of the target sensor, the closer the target sensor is to the heat source or pressure source, and the larger the collection requirement coefficient of the target production parameter collected by the target sensor, the higher the collection accuracy of the target sensor needs, and the greater the degree of subsequent adjustment of the collection accuracy of the target sensor. Therefore, the overall difference value and the overall data value of the target sensor and the collection requirement coefficient of the target production parameter can be combined and normalized to obtain the adjustment coefficient of the target sensor.

[0117] In the embodiment of the present invention, the integration of the three can be achieved by calculating the sum or product of the overall difference value and the overall data value of the target sensor and the acquisition requirement coefficient of the target production parameter, which is not limited here.

[0118] As an example, in one embodiment of the present invention, the expression of the adjustment coefficient of the target sensor may be specifically, for example, as follows:

[0119]

[0120] in, represents the adjustment coefficient of the target sensor in the target sensor set; Represents the overall difference value of the target sensor; Represents the overall data value of the target sensor; represents the acquisition requirement coefficient of the target production parameter; Represents the normalization function.

[0121] By using the same method as above, the adjustment coefficient of each sensor in the target sensor set can be obtained, and further, for any target sensor set, the adjustment coefficient of each sensor included therein can be obtained.

[0122] Step S4: Based on the adjustment coefficient of each sensor, adjust the parameters in the sensor to obtain an adjusted sensor, and use the adjusted sensor to collect data.

[0123] The larger the adjustment coefficient of a sensor, the greater the acquisition accuracy required for the sensor to collect data, and the greater the degree of adjustment of the sensor's data acquisition accuracy. The data acquisition accuracy of the sensor is usually reflected by parameters such as resolution or sensitivity. The higher the resolution and sensitivity of the sensor, the higher the acquisition accuracy of the sensor. Therefore, based on the adjustment coefficient of each sensor, the parameters in the sensor can be adjusted to obtain the adjusted sensor, so that the acquisition accuracy of the adjusted sensor is more in line with the data acquisition requirements of the corresponding production parameters, thereby improving the sensor's data collection effect on each production parameter.

[0124] Preferably, in one embodiment of the present invention, the method for obtaining the adjusted sensor specifically includes:

[0125] For any sensor, if the adjustment coefficient of the sensor is greater than the preset adjustment threshold, it means that the acquisition accuracy of the sensor needs to be improved. Therefore, the product value of the adjustment coefficient of the sensor and the original value of the sensor parameter can be used as the adjustment amount of the sensor parameter; the sum of the original value of the sensor parameter and the adjustment amount is used as the adjustment value of the sensor parameter, and the value used by the sensor parameter is adjusted to the adjustment value.

[0126] As an example, in one embodiment of the present invention, the expression of the adjustment value of the parameter of the sensor may be specifically, for example, as follows:

[0127]

[0128] in, Indicates the adjustment value of the sensor parameters, where the sensor parameters are parameters related to acquisition accuracy, such as resolution and sensitivity; Represents the raw value of the sensor parameter; Indicates the adjustment amount of the sensor parameters.

[0129] If the adjustment coefficient of the sensor is not greater than the preset adjustment threshold, it means that the demand for improving the acquisition accuracy of the sensor is low. At this time, the parameters in the sensor can be left unchanged to avoid excessively increasing the parameters of the sensor, which will reduce the final data acquisition effect. The value range of the preset adjustment threshold is In one embodiment of the present invention, the preset adjustment threshold is set to 0.5. The specific value of the preset adjustment threshold can also be set by the implementer according to the specific implementation scenario, which is not limited here.

[0130] After completing the parameter adjustment of each sensor on the liquid drug preparation production equipment, the adjusted sensors can be used to collect data on various production parameters in the liquid drug preparation production process, thereby improving the final data collection effect.

[0131] An embodiment of the present invention provides an intelligent data acquisition system for a liquid drug preparation production line, the system comprising a memory, a processor and a computer program, wherein the memory is used to store the corresponding computer program, the processor is used to run the corresponding computer program, and when the computer program runs in the processor, the method described in steps S1 to S4 can be implemented.

[0132] It should be noted that the sequence of the above embodiments of the present invention is for description only and does not represent the advantages and disadvantages of the embodiments. The processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0133] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referenced to each other, and each embodiment focuses on the differences from other embodiments.

Claims

1. An intelligent data collection method for a liquid drug preparation production line, characterized in that: The method comprises: The sensors are used to collect the detection data of different production parameters of the production link to be tested at each moment within a preset time period, and the number of failures that occurred in the liquid drug preparation production equipment within the preset time period is obtained; the sensors that collect the same production parameters are divided into the same sensor set, and the detection data collected by all sensors in each sensor set at the same moment are integrated to obtain the integrated data of each production parameter at each moment; Taking any production parameter as the target production parameter, obtaining the acquisition requirement coefficient of the target production parameter according to the change of the comprehensive data of the target production parameter at each moment, the correlation between the target production parameter and other production parameters except the target production parameter, and the number of failures; The sensor set where the sensor for collecting the target production parameter is located is taken as the target sensor set. If there is one sensor in the target sensor set, the acquisition requirement coefficient is taken as the adjustment coefficient of the sensor in the target sensor set. If there are multiple sensors in the target sensor set, any one sensor in the target sensor set is taken as the target sensor, and the adjustment coefficient of the target sensor is obtained according to the difference in detection data collected at the same time between the target sensor and other sensors in the target sensor set except the target sensor, the detection data collected by the target sensor, and the acquisition requirement coefficient; Based on the adjustment coefficient of each sensor, the parameters in the sensor are adjusted to obtain an adjusted sensor, and the adjusted sensor is used to collect data.

2. The intelligent data collection method for a liquid drug preparation production line according to claim 1, characterized in that: The acquisition requirement coefficients for obtaining the target production parameters include: According to the changes of the comprehensive data of the target production parameters at each moment, the data change degree of the target production parameters is obtained; Analyze the correlation between the target production parameter and other production parameters except the target production parameter to obtain the criticality of the target production parameter; The data variation degree, the criticality and the number of failures are integrated and normalized to obtain a collection requirement coefficient of a target production parameter.

3. The intelligent data collection method for a liquid drug preparation production line according to claim 2, characterized in that: The degree of change in the data of obtaining the target production parameters includes: Mapping the comprehensive data of the target production parameter at each moment into a two-dimensional coordinate system to obtain multiple data points of the target production parameter; Extract extreme value points from all data points of the target production parameters, take any two adjacent extreme value points as extreme value point groups, and obtain the mutation degree of each extreme value point group according to the absolute value of the difference of the comprehensive data between the two extreme value points in each extreme value point group and the absolute value of the difference at the moment, wherein the mutation degree is positively correlated with the absolute value of the difference of the comprehensive data between the two extreme value points in each extreme value point group, and the mutation degree is negatively correlated with the absolute value of the difference at the moment between the two extreme value points in each extreme value point group; Analyzing the overall level of the mutation degree of all extreme value point groups of the target production parameter to obtain the fluctuation degree of the target production parameter; The fluctuation degree and the number of extreme value points of the target production parameter are integrated and normalized to obtain the data variation degree of the target production parameter.

4. The intelligent data collection method for a liquid drug preparation production line according to claim 2, characterized in that: The criticality of obtaining the target production parameters includes: The absolute value of the Pearson correlation coefficient of the comprehensive data at each moment between the target production parameter and each other production parameter except the target production parameter is taken as the correlation coefficient between the target production parameter and each other production parameter; The overall level of the correlation coefficients between the target production parameter and all other production parameters is analyzed to obtain the criticality of the target production parameter.

5. The intelligent data collection method for a liquid drug preparation production line according to claim 1, characterized in that: The step of obtaining the adjustment coefficient of the target sensor according to the difference between the detection data collected at the same time by the target sensor and other sensors in the target sensor set except the target sensor, the detection data collected by the target sensor, and the collection requirement coefficient includes: Obtaining an overall difference value of the target sensor according to a difference in detection data collected at the same time between the target sensor and other sensors in the target sensor set except the target sensor; Analyze the overall level of the detection data collected by the target sensor at all times to obtain the overall data value of the target sensor; The overall difference value and the overall data value of the target sensor and the acquisition requirement coefficient of the target production parameter are integrated and normalized to obtain the adjustment coefficient of the target sensor.

6. The intelligent data collection method for a liquid drug preparation production line according to claim 5, characterized in that: The obtaining of the overall difference value of the target sensor comprises: An arbitrary moment is taken as a target moment, and the absolute value of the difference between the detection data collected by the target sensor and each other sensor at the target moment is taken as the initial data difference value between the target sensor and each other sensor at the target moment; Taking the accumulated value of the initial data difference values ​​between the target sensor and all other sensors at the target time as the comprehensive data difference value of the target sensor at the target time; The overall level of the comprehensive data difference value of the target sensor at all times is analyzed to obtain the overall difference value of the target sensor.

7. The intelligent data collection method for a liquid drug preparation production line according to claim 6, characterized in that: Analyzing the overall level of the comprehensive data difference value of the target sensor at all times to obtain the overall difference value of the target sensor includes: The average value of the comprehensive data difference values ​​of the target sensor at all times is taken as the overall difference value of the target sensor.

8. The intelligent data collection method for a liquid drug preparation production line according to claim 1, characterized in that: The step of adjusting the parameters in the sensor based on the adjustment coefficient of each sensor to obtain the adjusted sensor comprises: For any sensor, if the adjustment coefficient of the sensor is greater than a preset adjustment threshold, the product value of the adjustment coefficient of the sensor and the original value of the parameter of the sensor is used as the adjustment value of the parameter of the sensor; the sum of the original value of the parameter of the sensor and the adjustment value is used as the adjustment value of the parameter of the sensor, and the value used by the parameter of the sensor is adjusted to the adjustment value; If the adjustment coefficient of the sensor is not greater than the preset adjustment threshold, the parameters in the sensor are not adjusted.

9. The intelligent data collection method for a liquid drug preparation production line according to claim 1, characterized in that: The obtaining of comprehensive data of each production parameter at each moment includes: The average value of the detection data collected by all sensors in each sensor set at the same time is used as the comprehensive data of each production parameter at each time.

10. An intelligent data acquisition system for a liquid drug preparation production line, the system comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 9 are implemented.

Images