An intelligent cell storage state monitoring method and system

Abstract

The present invention discloses an intelligent cell storage state monitoring method and system, the method includes data collection, data preprocessing, cell storage control and cell storage state monitoring. The present invention belongs to the field of cell monitoring, specifically refers to an intelligent cell storage state monitoring method and system, the scheme enhances the activation function by introducing difference factors and nonlinear factors, and then predicts the cell state based on the LSTM model; improves the sensitivity to abnormal cell states; by constructing a feedback-response system, different control methods are taken for abnormal states of different degrees; thereby achieving stable and efficient cell storage state monitoring; abnormal storage state is judged by a dynamically adjusted convergence time judgment mechanism, and a flexible convergence rate adjustment mechanism is designed by introducing an adjustment term, an error term, and a nonlinear index; a rigorous abnormal judgment mechanism is designed to reduce the risk of false alarms; thereby achieving the accuracy of cell storage state monitoring.

Description

Technical Field

[0001] The present invention relates to the technical field of cell monitoring, and in particular to an intelligent cell storage state monitoring method and system. Background Art

[0002] The cell storage state monitoring method is a systematic method for continuously observing and evaluating key parameters in the cell storage environment to ensure that the cells are in a stable and appropriate state during storage. This method collects, processes and analyzes a variety of environmental and cell state data, combined with neural network models and control strategies, to achieve dynamic monitoring and abnormality detection of cell storage status. However, general cell storage state monitoring methods are unable to respond quickly to changes in environmental conditions and have poor monitoring stability under fluctuations in environmental factors; general cell storage state monitoring methods are overly sensitive to environmental fluctuations, leading to false alarms, and single monitoring is not suitable for the needs of different stages of cell culture, resulting in poor monitoring accuracy. Summary of the invention

[0003] In view of the above situation, in order to overcome the defects of the prior art, the present invention provides an intelligent cell storage state monitoring method and system. In view of the fact that the general cell storage state monitoring method cannot achieve rapid response to changes in environmental conditions and has poor monitoring stability under environmental factor fluctuations, this solution introduces difference factors and nonlinear factors to enhance the activation function, and then predicts the cell state based on the LSTM model; improves the sensitivity to abnormal cell states; by constructing a feedback-response system, different control methods are taken for abnormal states of different degrees, and rapid feedback is achieved under abnormal conditions to ensure the stability of the cell state; thereby achieving stable and efficient cell storage state monitoring; in view of the fact that the general cell storage state monitoring method is too sensitive to environmental fluctuations and thus leads to false alarms, and a single monitoring is not suitable for the needs of different stages in cell culture, resulting in poor monitoring accuracy, this solution uses a dynamically adjusted convergence time judgment mechanism to judge the storage state abnormally, and introduces an adjustment term, an error term, a nonlinear exponent to design a flexible convergence rate adjustment mechanism to meet the needs of different stages in cell culture; a rigorous abnormal judgment mechanism is designed to reduce the risk of false alarms; thereby achieving the accuracy of cell storage state monitoring.

[0004] The technical solution adopted by the present invention is as follows: The present invention provides an intelligent cell storage state monitoring method, the method comprising the following steps:

[0005] Step S1: data collection;

[0006] Step S2: data preprocessing;

[0007] Step S3: cell storage control;

[0008] Step S4: Cell storage status monitoring.

[0009] Furthermore, in step S1, the data collection is to collect historical cell monitoring data; the historical cell monitoring data includes environmental parameters, cell state parameters, sampling time and cell short-term state; and the cell short-term state is used as a data label.

[0010] Furthermore, in step S2, the data preprocessing is to perform data cleaning, data conversion and standardization on the collected data; the data cleaning is to process missing values ​​and duplicate values; the data conversion is to convert the data into a vector form; and the standardization is to standardize the data based on the maximum and minimum normalization method.

[0011] Furthermore, in step S3, the cell storage control specifically includes the following steps:

[0012] Step S31: prediction model training; for the preprocessed data set and data labels, the cell short-term state prediction model is trained based on the LSTM neural network; in order to adjust the reference points of different cell monitoring parameters, the activation function introduces a difference factor, and introduces a nonlinear factor to deal with temperature and humidity fluctuations and metabolic abnormalities. The activation function is expressed as: ; Where f(·) is the activation function and x is the activation variable; is the gamma function; α is the difference factor; p is the nonlinear factor; is the adjustment coefficient; is the variance parameter; is an odd number;

[0013] Step S32: Establish a feedback-response system; regard the short-term state of the cell as a feedback-response system, where the feedback system represents the predicted short-term state of the cell, and the response system represents the actual monitored short-term state; define the error term as the difference between the predicted state and the actual state, expressed as: , ; Among them, e 1l (·) and e 2l (·) are the environmental parameter error term and the cell state parameter error term respectively; x 1l (·) and y 1l (·) are the environmental parameter vectors predicted by LSTM and actually monitored; x 2l (·) and y 2l (·) are the cell state parameter vectors predicted by LSTM and actually monitored, respectively; t is the sampling time;

[0014] Step S33: Synchronous controller design; when the short-term state of the cell is slightly abnormal, a limited time synchronous controller is used, and the limited time synchronous controller setting is expressed as:

[0015] ;

[0016] When the short-term state of the cell is seriously abnormal, a fixed-time synchronization controller is used. The fixed-time synchronization controller is expressed as:

[0017] ;

[0018] ; Among them, u 1l (·) and u 2l (·) is a synchronous controller, which is used to adjust the environmental parameter vector and the cell state vector, respectively, acting on the error term to achieve a stable state; sign(·) is a sign function; and Used to adjust the controller's response to large error terms; and Used to adjust the sensitivity of the controller to small error terms; and ρ are the nonlinear exponents controlling small errors and large errors, respectively; is the external disturbance term; is the feedback gain, used to suppress error fluctuations; T max is the maximum allowed convergence time; is the preset convergence threshold; and is the control gain, which is used to adjust the error term of the environment and cell state; , , and is the intensity parameter that controls small and large error terms; is the delayed feedback gain; is the feedback delay time; is the adjustment gain parameter; l is the dimension index;

[0019] Step S34: Convergence proof: construct stability function, expressed as: ; where V(·) is the stability function; p is the exponential parameter; and n is the total number of dimensions.

[0020] Further, in step S4, the cell storage state monitoring is to calculate the convergence time. When the convergence time is greater than the preset convergence threshold or the convergence time is error, the cell storage state is determined to be abnormal, and an early warning is issued to relevant personnel. The convergence time T is expressed as:

[0021] ; in, It is the lower bound of the convergence rate; and is the control error gain threshold; c is the adjustment term used to fine-tune the convergence rate; and is the nonlinear index of the control error term.

[0022] The present invention provides an intelligent cell storage state monitoring system, comprising a data acquisition module, a data preprocessing module, a cell storage control module and a cell storage state monitoring module;

[0023] The data acquisition module collects historical cell monitoring data and sends the data to the data preprocessing module;

[0024] The data preprocessing module performs data cleaning, data conversion and standardization on the collected data; and sends the data to the cell storage control module;

[0025] The cell storage control module introduces a difference factor and a nonlinear factor to enhance the activation function, and then predicts the cell state based on the LSTM model; adopts different control methods for different degrees of abnormal states; and sends the data to the cell storage state monitoring module;

[0026] The cell storage status monitoring module realizes cell storage status monitoring on real-time collected data by designing a dynamically adjusted convergence time determination mechanism.

[0027] The beneficial effects achieved by the present invention using the above scheme are as follows:

[0028] (1) In view of the fact that general cell storage state monitoring methods cannot respond quickly to changes in environmental conditions and have poor monitoring stability under fluctuating environmental factors, this scheme introduces difference factors and nonlinear factors to enhance the activation function, and then predicts the cell state based on the LSTM model; improves the sensitivity to abnormal cell states; and constructs a feedback-response system to take different control methods for different degrees of abnormal states, achieve rapid feedback in abnormal situations, and ensure the stability of the cell state; thereby achieving stable and efficient cell storage state monitoring.

[0029] (2) In view of the fact that general cell storage status monitoring methods are overly sensitive to environmental fluctuations, which may lead to false alarms, and single monitoring is not suitable for the needs of different stages of cell culture, resulting in poor monitoring accuracy, this scheme uses a dynamically adjusted convergence time judgment mechanism to determine the abnormality of the storage status. In order to meet the needs of different stages of cell culture, an adjustment term, an error term, and a nonlinear exponent are introduced to design a flexible convergence rate adjustment mechanism; a rigorous abnormal judgment mechanism is designed to reduce the risk of false alarms; and thus the accuracy of cell storage status monitoring is achieved. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] Figure 1 A schematic diagram of a flow chart of an intelligent cell storage status monitoring method provided by the present invention;

[0031] Figure 2 A schematic diagram of an intelligent cell storage status monitoring system provided by the present invention;

[0032] Figure 3 It is a schematic diagram of the process of step S3.

[0033] The accompanying drawings are used to provide further understanding of the present invention and constitute a part of the specification. They are used to explain the present invention together with the embodiments of the present invention and do not constitute a limitation of the present invention. DETAILED DESCRIPTION

[0034] The technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments; based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0035] In the description of the present invention, it is necessary to understand that terms such as “upper”, “lower”, “front”, “back”, “left”, “right”, “top”, “bottom”, “inside” and “outside” indicating directions or positional relationships are based on the directions or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the referred system or element must have a specific direction, be constructed and operated in a specific direction, and therefore cannot be understood as a limitation on the present invention.

[0036] Example 1, see Figure 1 The present invention provides an intelligent cell storage status monitoring method, which comprises the following steps:

[0037] Step S1: data collection, collecting historical cell monitoring data;

[0038] Step S2: Data preprocessing, data cleaning, data conversion and standardization of the collected data;

[0039] Step S3: Cell storage control, introduce difference factors and nonlinear factors to enhance the activation function, and then predict the cell state based on the LSTM model; take different control methods for different degrees of abnormal states;

[0040] Step S4: Cell storage status monitoring, by designing a dynamically adjusted convergence time determination mechanism to monitor the cell storage status of the real-time collected data.

[0041] Example 2, see Figure 1 This embodiment is based on the above embodiment. In step S1, the historical cell monitoring data includes environmental parameters, cell state parameters, sampling time and cell short-term state; the cell short-term state is used as a data label; the environmental parameters include temperature, humidity, carbon dioxide concentration, oxygen concentration and light intensity; the cell state parameters include cell survival rate, metabolite concentration, cell density and pH value; the cell short-term state includes normal, mild abnormality and severe abnormality.

[0042] Example 3, see Figure 1 This embodiment is based on the above embodiment. In step S2, data cleaning is to process missing values ​​and duplicate values; data conversion is to convert data into vector form; and standardization is to standardize data based on the maximum and minimum normalization method.

[0043] Example 4, see Figure 1 and Figure 3 This embodiment is based on the above embodiment. In step S3, the cell storage control specifically includes the following steps:

[0044] Step S31: prediction model training; for the preprocessed data set and data labels, the cell short-term state prediction model is trained based on the LSTM neural network; in order to adjust the reference points of different cell monitoring parameters, the activation function introduces a difference factor, and introduces a nonlinear factor to deal with temperature and humidity fluctuations and metabolic abnormalities. The activation function is expressed as: ; Where f(·) is the activation function and x is the activation variable; is the gamma function; α is the difference factor; p is the nonlinear factor; is the adjustment coefficient; is the variance parameter; is an odd number;

[0045] Step S32: Establish a feedback-response system; regard the short-term state of the cell as a feedback-response system, where the feedback system represents the predicted short-term state of the cell, and the response system represents the actual monitored short-term state; define the error term as the difference between the predicted state and the actual state, expressed as: , ; Among them, e 1l (·) and e 2l (·) are the environmental parameter error term and the cell state parameter error term respectively; x 1l (·) and y 1l (·) are the environmental parameter vectors predicted by LSTM and actually monitored; x 2l (·) and y 2l (·) are the cell state parameter vectors predicted by LSTM and actually monitored, respectively; t is the sampling time;

[0046] Step S33: Synchronous controller design; when the short-term state of the cell is slightly abnormal, a limited time synchronous controller is used, and the limited time synchronous controller setting is expressed as:

[0047] ;

[0048] When the short-term state of the cell is seriously abnormal, a fixed-time synchronization controller is used. The fixed-time synchronization controller is expressed as:

[0049] ;

[0050] ; Among them, u 1l (·) and u 2l (·) is a synchronous controller, which is used to adjust the environmental parameter vector and the cell state vector, respectively, acting on the error term to achieve a stable state; sign(·) is a sign function; and Used to adjust the controller's response to large error terms; and Used to adjust the sensitivity of the controller to small error terms; and ρ are the nonlinear exponents controlling small errors and large errors, respectively; is the external disturbance term; is the feedback gain, used to suppress error fluctuations; T max is the maximum allowed convergence time; is the preset convergence threshold; and is the control gain, which is used to adjust the error term of the environment and cell state; , , and is the intensity parameter that controls small and large error terms; is the delayed feedback gain; is the feedback delay time; is the adjustment gain parameter; l is the dimension index;

[0051] Step S34: Convergence proof: construct stability function, expressed as: ; where V(·) is the stability function; p is the exponential parameter; and n is the total number of dimensions.

[0052] By performing the above operations, in view of the problem that the general cell storage state monitoring method cannot achieve rapid response to changes in environmental conditions and has poor monitoring stability under fluctuations in environmental factors, this scheme introduces difference factors and nonlinear factors to enhance the activation function, and then predicts the cell state based on the LSTM model; improves the sensitivity to abnormal cell states; by constructing a feedback-response system, different control methods are taken for different degrees of abnormal states in a targeted manner, and rapid feedback is achieved in abnormal situations to ensure the stability of the cell state; thereby achieving stable and efficient cell storage state monitoring.

[0053] Example 5, see Figure 1 This embodiment is based on the above embodiment. In step S4, the cell storage state monitoring is to calculate the convergence time. When the convergence time is greater than the preset convergence threshold or the convergence time is error, the cell storage state is determined to be abnormal, and an early warning is issued to relevant personnel. The convergence time T is expressed as:

[0054] ; in, It is the lower bound of the convergence rate; and is the control error gain threshold; c is the adjustment term used to fine-tune the convergence rate; and is the nonlinear index of the control error term.

[0055] By performing the above operations, the general cell storage status monitoring method is too sensitive to environmental fluctuations, which leads to false alarms, and a single monitoring method is not suitable for the needs of different stages of cell culture, resulting in poor monitoring accuracy. This solution uses a dynamically adjusted convergence time judgment mechanism to make abnormal judgments on the storage status. In order to meet the needs of different stages of cell culture, an adjustment term, an error term, and a nonlinear exponent are introduced to design a flexible convergence rate adjustment mechanism; a rigorous abnormal judgment mechanism is designed to reduce the risk of false alarms; and the accuracy of cell storage status monitoring is achieved.

[0056] Example 6, see Figure 2 , this embodiment is based on the above embodiment, and the present invention provides an intelligent cell storage state monitoring system, including a data acquisition module, a data preprocessing module, a cell storage control module and a cell storage state monitoring module;

[0057] The data acquisition module collects historical cell monitoring data and sends the data to the data preprocessing module;

[0058] The data preprocessing module performs data cleaning, data conversion and standardization on the collected data; and sends the data to the cell storage control module;

[0059] The cell storage control module introduces a difference factor and a nonlinear factor to enhance the activation function, and then predicts the cell state based on the LSTM model; adopts different control methods for different degrees of abnormal states; and sends the data to the cell storage state monitoring module;

[0060] The cell storage status monitoring module realizes cell storage status monitoring on real-time collected data by designing a dynamically adjusted convergence time determination mechanism.

[0061] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device.

[0062] While the embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that many changes, modifications, substitutions and variations can be made to the embodiments without departing from the principles and spirit of the invention.

[0063] The present invention and its embodiments are described above, and such description is not restrictive. The drawings show only one embodiment of the present invention, and the actual structure is not limited thereto. In short, if ordinary technicians in the field are inspired by it, without departing from the purpose of the invention, they can design a structure and embodiment similar to the technical solution without creativity, which should belong to the protection scope of the present invention.

Claims

1. An intelligent cell storage status monitoring method, characterized in that: The method comprises the following steps: Step S1: data collection, collecting historical cell monitoring data; Step S2: data preprocessing; Step S3: cell storage control; Step S4: cell storage status monitoring; In step S3, the cell storage control specifically includes the following steps: Step S31: prediction model training; for the preprocessed data set and data labels, the cell short-term state prediction model is trained based on the LSTM neural network; in order to adjust the reference points of different cell monitoring parameters, the activation function introduces a difference factor, and introduces a nonlinear factor to deal with temperature and humidity fluctuations and metabolic abnormalities. The activation function is expressed as: ; Where f(·) is the activation function and x is the activation variable; is the gamma function; α is the difference factor; p is the nonlinear factor; is the adjustment coefficient; is the variance parameter; is an odd number; Step S32: Establish a feedback-response system; regard the short-term state of the cell as a feedback-response system, where the feedback system represents the predicted short-term state of the cell, and the response system represents the actual monitored short-term state; define the error term as the difference between the predicted state and the actual state, expressed as: , ; Among them, e 1l (·) and e 2l (·) are the environmental parameter error term and the cell state parameter error term respectively; x 1l (·) and y 1l (·) are the environmental parameter vectors predicted by LSTM and actually monitored; x 2l (·) and y 2l (·) are the cell state parameter vectors predicted by LSTM and actually monitored, respectively; t is the sampling time; Step S33: Synchronous controller design; when the short-term state of the cell is slightly abnormal, a limited time synchronous controller is used, and the limited time synchronous controller setting is expressed as: ； When the short-term state of the cell is seriously abnormal, a fixed-time synchronization controller is used. The fixed-time synchronization controller is expressed as: ； ； Among them, u 1l (·) and u 2l (·) is a synchronous controller, which is used to adjust the environmental parameter vector and the cell state vector, respectively, acting on the error term to achieve a stable state; sign(·) is a sign function; and Used to adjust the controller's response to large error terms; and Used to adjust the sensitivity of the controller to small error terms; and ρ are the nonlinear exponents controlling small errors and large errors, respectively; is the external disturbance term; is the feedback gain, used to suppress error fluctuations; T max is the maximum allowed convergence time; is the preset convergence threshold; and is the control gain, which is used to adjust the error term of the environment and cell state; , , and is the intensity parameter that controls the small and large error terms; is the delayed feedback gain; is the feedback delay time; is the adjustment gain parameter; l is the dimension index; e 1l (·) and e 2l (·) are the environmental parameter error term and the cell state parameter error term, respectively; t is the sampling time; Step S34: Convergence proof: construct stability function, expressed as: ; where V(·) is the stability function; p is the exponential parameter; n is the total number of dimensions; In step S4, the cell storage status monitoring is to calculate the convergence time. When the convergence time is greater than the preset convergence threshold or the convergence time is error, the cell storage status is determined to be abnormal, and an early warning is issued to relevant personnel. The convergence time T is expressed as: ;in, It is the lower bound of the convergence rate; and is the control error gain threshold; c is the adjustment term used to fine-tune the convergence rate; and is the nonlinear index of the control error term.

2. The intelligent cell storage status monitoring method according to claim 1, characterized in that: In step S1, the historical cell monitoring data includes environmental parameters, cell state parameters, sampling time and cell short-term state; the cell short-term state is used as a data label.

3. The intelligent cell storage status monitoring method according to claim 1, characterized in that: In step S2, the data preprocessing is to clean, convert and standardize the collected data; the data cleaning is to process missing values ​​and duplicate values; the data conversion is to convert the data into a vector form; and the standardization is to standardize the data based on the maximum and minimum normalization method.

4. An intelligent cell storage state monitoring system, used to implement an intelligent cell storage state monitoring method as claimed in any one of claims 1 to 3, characterized in that: It includes a data acquisition module, a data preprocessing module, a cell storage control module and a cell storage status monitoring module; The data acquisition module collects historical cell monitoring data and sends the data to the data preprocessing module; The data preprocessing module performs data cleaning, data conversion and standardization on the collected data; And send the data to the cell storage control module; The cell storage control module introduces a difference factor and a nonlinear factor to enhance the activation function, and then predicts the cell state based on the LSTM model; Different control methods are adopted for different degrees of abnormal conditions; and the data is sent to the cell storage status monitoring module; The cell storage status monitoring module realizes cell storage status monitoring on real-time collected data by designing a dynamically adjusted convergence time determination mechanism.

Images