A monitoring and analysis system for the preparation of marine peptides

Abstract

This invention relates to the field of data processing technology, and in particular to a monitoring and analysis system for marine peptide preparation. The system includes: collecting raw production parameter data during the production process; intelligently processing the raw production parameter data to obtain intelligently processed production parameter data; intelligently analyzing the intelligently processed production parameter data to obtain state analysis results; adaptively evaluating the state analysis results using a multi-dimensional dynamic state adaptive evaluation algorithm for marine peptide preparation to generate state evaluation results; generating adjustment suggestions based on the state evaluation results; and converting the adjustment suggestions into control commands to automatically regulate the marine peptide preparation equipment. This system solves the technical problems of traditional marine peptide preparation processes, such as the difficulty in accurately processing and analyzing multi-dimensional production parameter data, and the low precision and efficiency of production control of marine peptide preparation equipment.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and in particular to a monitoring and analysis system for the preparation of marine peptides. Background Art

[0002] Marine peptides are a class of bioactive peptides derived from marine organisms. They have multiple biological functions such as antioxidant, anti-inflammatory, and antibacterial properties. Therefore, they have broad application prospects in the fields of medicine, health products, and food. However, the extraction and preparation process of marine peptides is relatively complex, involving multiple production links and multidimensional parameters such as temperature, pH value, flow rate, pressure, etc. These parameters directly affect the yield and quality of marine peptides. In the actual production process, how to ensure the precise control of these key parameters without affecting production efficiency has become an important factor restricting the industrialization development of marine peptides.

[0003] The traditional marine peptide preparation process mostly relies on manual operation and experience-based adjustment. The monitoring and adjustment of production parameters usually rely on the operator's judgment based on experience, and it is difficult to accurately obtain multi-dimensional production parameter data in real time. Due to the complex interactions and dynamic changes between various production parameters, manual control often cannot respond in a timely manner, which can easily lead to problems such as inconsistent product quality and low production efficiency. In addition, random fluctuations, noisy data, and occasional abnormal phenomena in the production environment also bring great uncertainty to the production process. Traditional production parameter control methods are difficult to effectively handle these complex situations.

[0004] In summary, the traditional marine peptide preparation process has at least the following technical problems: it is difficult to accurately process and analyze multi-dimensional production parameter data, and the accuracy and efficiency of production regulation of marine peptide preparation equipment are low. Summary of the invention

[0005] The present invention provides a monitoring and analysis system for the preparation of marine peptides to solve the technical problems that the multi-dimensional production parameter data existing in the process of marine peptide preparation is difficult to accurately process and analyze, and the accuracy and efficiency of the production regulation of marine peptide preparation equipment are low.

[0006] A monitoring and analysis method for the preparation of marine peptides, comprising the following steps:

[0007] S1. Collect the original production parameter data in the production process, perform intelligent processing on the original production parameter data, and obtain the production parameter data after intelligent processing; perform intelligent analysis on the production parameter data after intelligent processing to obtain the status analysis results;

[0008] S2. Perform adaptive evaluation on the state analysis results through the multi-dimensional dynamic state adaptive evaluation algorithm of marine peptide preparation to generate state evaluation results; generate adjustment suggestions based on the state evaluation results; convert the adjustment suggestions into control instructions to automatically regulate the marine peptide preparation equipment.

[0009] Preferably, the S1 specifically includes:

[0010] In the process of implementing intelligent processing, the original production parameter data is denoised; the outlier detection algorithm with adaptive threshold is used to detect and eliminate outliers in the denoised production parameter data; the production parameter data after outlier detection and elimination is standardized to obtain the production parameter data after intelligent processing.

[0011] Preferably, the S1 specifically includes:

[0012] The dynamic hierarchical feature mapping and time domain offset transformation analysis method are used to perform intelligent analysis on the production parameter data after intelligent processing to obtain the state analysis results.

[0013] Preferably, the S1 specifically includes:

[0014] The specific implementation process of dynamic hierarchical feature mapping and time domain offset transformation analysis method is as follows:

[0015] Through the mapping matrix, the production parameter data after intelligent processing is dynamically hierarchically mapped to obtain the mapping results of each feature layer; the mapping results of each feature layer are transformed in the time domain; the feature cross term is introduced to calculate the feature interaction strength between different feature layers; the feature layers are weighted according to the feature interaction strength between the feature layers, and the production state deviation is calculated based on the weights of different feature layers; the mapping formula is defined as:

[0016]

[0017] in, Representative The mapping result of feature layers; is the dynamic mapping matrix; It's time The intelligently processed production parameter data set;

[0018] The time domain migration transformation formula is as follows:

[0019]

[0020] in, Indicates The production characteristics of each feature layer after time domain migration transformation; is the time offset coefficient; It is The rate of change of production characteristics of each characteristic layer over time;

[0021] The characteristic interaction strength is calculated by the following formula:

[0022]

[0023] in, It is Layer and Layer feature interaction strength between layers, ; Used to prevent the denominator from being zero; It is The production features of each feature layer after time domain migration transformation.

[0024] Preferably, the S1 specifically includes:

[0025] According to the production state deviation, a state correction model is introduced to generate correction values ​​of production conditions; based on the correction values ​​of production conditions, state analysis results are formed.

[0026] Preferably, the S2 specifically includes:

[0027] In the process of realizing the adaptive evaluation algorithm of multi-dimensional dynamic state of marine peptide preparation, the state analysis results are dynamically nonlinearly combined by introducing time transformation and dynamic weights to obtain the state analysis results after dynamic nonlinear combination.

[0028] Preferably, the S2 specifically includes:

[0029] In the process of implementing the adaptive evaluation algorithm for the multi-dimensional dynamic state of marine peptide preparation, the state analysis results after dynamic nonlinear combination are adaptively evaluated to obtain the production state deviation evaluation value.

[0030] Preferably, the S2 specifically includes:

[0031] By comparing the production status deviation evaluation value with the preset threshold, it is evaluated whether there is an abnormality in the production process; and based on the difference between the production status deviation evaluation value and the preset threshold, a status evaluation result is generated.

[0032] The monitoring and analysis system for the preparation of marine peptides of the present invention specifically includes the following technical solutions:

[0033] A monitoring and analysis system for the preparation of marine peptides, comprising the following parts:

[0034] Data acquisition module, data intelligent processing module, intelligent analysis module, adaptive evaluation module, automatic control module, data storage module;

[0035] The data acquisition module collects the original production parameter data in the production process in real time and sends the original production parameter data to the data intelligent processing module;

[0036] The data intelligent processing module performs intelligent processing on the original production parameter data to obtain the intelligently processed production parameter data, and sends the intelligently processed production parameter data to the intelligent analysis module and the data storage module;

[0037] The intelligent analysis module uses dynamic hierarchical feature mapping and time domain offset transformation analysis methods, combined with the production characteristics after the historical time domain offset transformation of the data storage module, to perform intelligent analysis on the production parameter data after intelligent processing, and obtains the mapped feature data, the production characteristics after the time domain offset transformation, the production state deviation value, and the correction value of the production condition. Based on the correction value of the production condition, a state analysis result is formed, and the state analysis result is sent to the adaptive evaluation module, and the production characteristics after the time domain offset transformation and the state analysis result are sent to the data storage module;

[0038] An adaptive evaluation module performs dynamic nonlinear combination on the state analysis results to obtain a state analysis result after the dynamic nonlinear combination, performs adaptive evaluation on the state analysis result after the dynamic nonlinear combination to obtain a state evaluation result, and sends the state evaluation result to the automatic control module, and sends the state analysis result after the dynamic nonlinear combination and the state evaluation result to the data storage module;

[0039] An automatic control module generates adjustment suggestions based on the status evaluation results, generates control instructions based on the adjustment suggestions, and automatically adjusts the production conditions according to the control instructions; and sends the adjustment suggestions to the data storage module;

[0040] The data storage module is used to store production parameter data after intelligent processing in the production process, production characteristics after time domain offset transformation, state analysis results, state assessment results, state analysis results after dynamic nonlinear combination, and adjustment suggestions, and is used as historical data.

[0041] The beneficial effects of the technical solution of the present invention are:

[0042] 1. Introduce adaptive thresholds for anomaly detection. By combining the mean, standard deviation and skewness to dynamically adjust the threshold, it can adapt to changes under different production conditions and avoid the misjudgment phenomenon that is prone to occur in the traditional fixed threshold method. At the same time, by using skewness and time attenuation coefficients, it can detect symmetrically distributed outliers, cope with scenarios with uneven data distribution, and flexibly adjust the detection sensitivity of outliers to adapt to data characteristics at different stages. It is especially suitable for complex marine peptide production processes, which helps to improve the accuracy of anomaly detection and optimize production control.

[0043] 2. Dynamic processing of different production parameter data through dynamic hierarchical feature mapping and time domain offset transformation can better capture subtle changes in production parameter data, especially the complex relationship between various production parameters in a multi-dimensional dynamic environment. Dynamic hierarchical feature mapping projects different production parameter data into an independent feature space, which can more clearly separate the specific impact of each production parameter on the production process; and time domain offset transformation can better track the changing trend of production parameter data through the time delay effect, thereby forming a more accurate assessment of the current production status.

[0044] 3. Feature interaction analysis not only evaluates the independent impact of each production parameter, but also analyzes the interaction between different production parameters through feature cross terms, so as to more comprehensively understand the correlation and synergy between various production parameters. By calculating the intensity of feature interaction, it is possible to give higher weights to production parameters that have a greater impact on the production process, ensuring that these production parameters are focused on during optimization and adjustment, thereby effectively improving the accuracy and efficiency of production control. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] Figure 1 This is a structural diagram of a monitoring and analysis system for preparing a marine peptide according to the present invention;

[0046] Figure 2 The present invention is a flow chart of the monitoring and analysis method for preparing a marine peptide. DETAILED DESCRIPTION

[0047] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the technical scheme in the embodiment of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiment of the present invention. Obviously, the described embodiment is only a part of the embodiment of the present invention, not all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0048] 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.

[0049] The specific scheme of the monitoring and analysis system for the preparation of marine peptides provided by the present invention is described in detail below with reference to the accompanying drawings.

[0050] See attached Figure 1 , which shows a structure diagram of a monitoring and analysis system for preparing marine peptides provided by one embodiment of the present invention, the system includes the following parts:

[0051] Data acquisition module, data intelligent processing module, intelligent analysis module, adaptive evaluation module, automatic control module, data storage module;

[0052] The data acquisition module uses sensors deployed in the marine peptide preparation equipment to collect the original production parameter data in the production process in real time, such as temperature, pH value, flow rate, etc., and sends the original production parameter data to the data intelligent processing module;

[0053] The sensors include temperature sensors, pH sensors, pressure sensors, flow meters, spectrometers, redox potential sensors, dissolved oxygen sensors, etc.

[0054] The data intelligent processing module performs intelligent processing on the original production parameter data to obtain the intelligently processed production parameter data, and sends the intelligently processed production parameter data to the intelligent analysis module and the data storage module;

[0055] The intelligent analysis module uses dynamic hierarchical feature mapping and time domain offset transformation analysis methods, combined with the production characteristics after the historical time domain offset transformation of the data storage module, to perform intelligent analysis on the production parameter data after intelligent processing, and obtains the mapped feature data, the production characteristics after the time domain offset transformation, the production state deviation value, and the correction value of the production condition. Based on the correction value of the production condition, a state analysis result is formed, and the state analysis result is sent to the adaptive evaluation module, and the production characteristics after the time domain offset transformation and the state analysis result are sent to the data storage module;

[0056] An adaptive evaluation module performs dynamic nonlinear combination on the state analysis results to obtain a state analysis result after the dynamic nonlinear combination, performs adaptive evaluation on the state analysis result after the dynamic nonlinear combination to obtain a state evaluation result, and sends the state evaluation result to the automatic control module, and sends the state analysis result after the dynamic nonlinear combination and the state evaluation result to the data storage module;

[0057] The automatic control module generates adjustment suggestions according to the status evaluation results, generates control instructions according to the adjustment suggestions, and automatically adjusts the production conditions according to the control instructions; and sends the adjustment suggestions to the data storage module;

[0058] The data storage module is used to store production parameter data after intelligent processing in the production process, production characteristics after time domain offset transformation, state analysis results, state assessment results, state analysis results after dynamic nonlinear combination, and adjustment suggestions, and is used as historical data.

[0059] See attached Figure 2 , which shows a flow chart of a monitoring and analysis method for preparing a marine peptide provided by an embodiment of the present invention, the method comprising the following steps:

[0060] S1. Collect the original production parameter data in the production process, perform intelligent processing on the original production parameter data, and obtain the production parameter data after intelligent processing; perform intelligent analysis on the production parameter data after intelligent processing to obtain the status analysis results;

[0061] In the marine peptide preparation equipment, the original production parameter data in the production process is collected in real time based on a variety of sensors deployed by professional technicians. The sensors include temperature sensors, pH sensors, pressure sensors, flow meters, spectrometers, redox potential sensors, dissolved oxygen sensors, etc.;

[0062] The original production parameter data is intelligently processed. The specific process is as follows:

[0063] The original production parameter data is subjected to Gaussian filtering to remove random noise in the original production parameter data, and the denoised production parameter data is obtained. Then, the denoised production parameter data is subjected to an outlier detection algorithm with an adaptive threshold to remove extreme data that may appear in the production process, thereby realizing outlier detection and removal. The adaptive threshold is defined as follows:

[0064]

[0065] in, Is an adaptive threshold used to determine the time Whether the production parameter data after denoising is an abnormal value, changes over time, and is dynamically adjusted; It's time The mean of the denoised production parameter data at the moment reflects the central trend of the denoised production parameter data at the moment; is the weight coefficient of the standard deviation, which controls the influence of the standard deviation on the threshold; It's time The standard deviation of the denoised production parameter data at the time instant indicates the degree of fluctuation of the denoised production parameter data. The larger the standard deviation, the more drastic the fluctuation of the production parameter data. is the skewness adjustment coefficient, which is used to control the effect of skewness on the threshold correction and increase the sensitivity of asymmetric data distribution; It's time The skewness coefficient at time is used to describe the symmetry of the data distribution. , It is the expected value operator, which is used to calculate the third-order moment of the data and reflect the distribution form of the data; It's time The denoised production parameter data at each moment; is the time decay coefficient, which changes with time As increases, the adjustment amount for anomaly detection is gradually reduced.

[0066] When the difference between the denoised production parameter data and the mean of the denoised production parameter data is greater than the adaptive threshold, the denoised production parameter data is an abnormal value and is removed.

[0067] The production parameter data that has been detected and eliminated for outliers are standardized to obtain production parameter data of the same magnitude, thereby realizing intelligent processing of the production parameter data and obtaining production parameter data after intelligent processing.

[0068] Furthermore, the dynamic hierarchical feature mapping and time domain offset transformation analysis method are combined with the production characteristics after historical time domain offset transformation to perform intelligent analysis on the production parameter data after intelligent processing to obtain the state analysis results. The specific implementation process is as follows:

[0069] The dynamic hierarchical feature mapping and time domain offset transformation analysis method can capture subtle changes in production parameter data after intelligent processing in a multi-dimensional dynamic environment through dynamic hierarchical feature mapping, time domain offset transformation and feature interaction analysis, form an accurate assessment of the current production status, and provide a data basis for adaptive adjustment suggestions.

[0070] The production parameter data after intelligent processing is processed by dynamic hierarchical feature mapping. The purpose of dynamic hierarchical feature mapping is to map the matrix , projecting different production parameter data into independent feature spaces. Each feature space independently reflects the internal characteristics of a type of production parameter, which helps to further separate the impact of various production parameters on the production process. The mapping formula is defined as:

[0071]

[0072] in, Representative The mapping result of the feature layer reflects the moment Production features after dynamic hierarchical feature mapping; It is a dynamic mapping matrix that converts the current intelligently processed production parameter data set Mapping to a specific feature space; It's time The intelligently processed production parameter data set contains production parameter data in multiple dimensions;

[0073]

[0074] is the dynamic mapping matrix The Line The elements of the column represent the time No. How to map the production parameter data to the feature dimensions, and transform the intelligently processed production parameter data into data in the feature space through a mapping matrix for subsequent analysis and optimization; It is The standard deviation of the intelligently processed production parameter data is used to standardize the fluctuation of the production parameter data; It is The production parameters after intelligent processing are The value of the moment; It is feature dimension in time The expected value (mean) at the moment; is the periodic adjustment coefficient, used to control the The time periodic fluctuation of production parameter data after intelligent processing;

[0075] In order to further capture the time-varying characteristics of the data, the time domain offset transformation is introduced. By performing the time domain offset transformation operation on the mapped production features, a new feature space is constructed. The time domain offset transformation formula is as follows:

[0076]

[0077] in, Indicates The production characteristics of each feature layer after time domain migration transformation; is the time shift coefficient, which controls the delay effect of the time domain shift transform; It is The rate of change of production characteristics of a feature layer over time.

[0078] Furthermore, after dynamic hierarchical feature mapping and time domain shift transformation, feature interaction analysis is performed on data from different feature layers to evaluate the intrinsic connection between data in each dimension. By introducing feature cross terms, the feature interaction strength between different feature layers is calculated. The feature interaction strength is calculated using the following formula:

[0079]

[0080] in, It is Layer and The feature interaction strength between the feature layers indicates the feature interaction strength between the data of two feature layers. ; is the characteristic cross term; is a small positive number used to prevent the denominator from being zero; It is The production features of each feature layer after time domain migration transformation.

[0081] After obtaining the feature interaction strength between feature layers, in order to give priority to the feature layers that have a greater impact on production and ensure that important production parameters can be focused on during analysis, weights are assigned to different feature layers according to the feature interaction strength. The allocation formula is as follows:

[0082]

[0083] in, Indicates the total number of feature layers;

[0084] Furthermore, the weighted feature results of different feature layers are comprehensively evaluated to calculate the production state deviation. The production state deviation is used to evaluate the difference between the current production state and the historical best production state. The formula is as follows:

[0085]

[0086] in, It is the current moment The production state deviation indicates the degree of deviation between the current production state and the historical optimal production state. change; It is the current moment No. Layer characteristic layer's production characteristics after historical best time domain migration transformation; is the weighted feature result;

[0087] After calculating the production state deviation, the production conditions are adjusted according to the production state deviation. The state correction model is introduced to adjust the production conditions in real time, and multi-dimensional nonlinear adjustment items are also introduced to make the state correction model more sensitive to abnormal situations. The mathematical formula of the state correction model is as follows:

[0088]

[0089] in, It is the correction value of the production condition, indicating how to adjust the first Data on production parameters over time Change, the characteristics of each layer correspond to a production parameter; It is a multidimensional nonlinear adjustment term; It is the correction sensitivity coefficient that controls the response intensity to production status deviations.

[0090] Finally, the correction values ​​of all production conditions are combined to form the state analysis result , including all Provide data basis for adaptive assessment.

[0091] S2. Perform adaptive evaluation on the state analysis results through the multi-dimensional dynamic state adaptive evaluation algorithm of marine peptide preparation to generate state evaluation results; generate adjustment suggestions based on the state evaluation results; convert the adjustment suggestions into control instructions to automatically regulate the marine peptide preparation equipment.

[0092] The state analysis results are adaptively evaluated using the multi-dimensional dynamic state adaptive evaluation algorithm for marine peptide preparation to evaluate whether there are abnormalities in the production process and obtain the state evaluation results; the specific implementation process is as follows:

[0093] Firstly, the state analysis results are dynamically nonlinearly combined by introducing time transformation and dynamic weight to capture the dynamic performance of the state analysis results in different time periods and under different production conditions. The specific formula is:

[0094]

[0095] in, It is the state analysis result after dynamic nonlinear combination, and the correction value of each production condition is The complex nonlinear behaviors of the production process are combined together and the cumulative effect over time is added to describe the dynamic changes of the entire production process. is the nonlinear regulation coefficient, which is used to control the intensity of the sinusoidal regulation function; It is a nonlinear trigonometric function, reflecting the correction value of production conditions Dynamic behavior of is the smoothing factor, which is used to control the sensitivity of this term to small deviations; is the amplification factor of the correction rate of change of production conditions; is the historical cumulative weight factor, which is used to control the contribution of the correction value of production conditions in the historical cumulative effect; is the corrected rate of change of production conditions.

[0096] Furthermore, the state analysis results after dynamic nonlinear combination are adaptively evaluated to detect whether there are abnormalities in the production process and to evaluate whether the current production state needs to be adjusted. State analysis results after combining with historical best dynamic nonlinearity By comparison, the degree of deviation of the current production status can be evaluated. The adaptive evaluation formula is:

[0097]

[0098] in, It is the production state deviation evaluation value, which is used to measure the deviation degree of the state analysis result after the current dynamic nonlinear combination relative to the state analysis result after the historical best dynamic nonlinear combination, and integrates the cumulative effect over time and the nonlinear behavior of the current production state; is an integral variable, representing the time point between the start of production and the current moment; is the adaptive adjustment coefficient, which controls the nonlinear adjustment sensitivity to the current production status and is used to balance the contribution of the sinusoidal term in the overall production status evaluation; It is the nonlinear adjustment coefficient, which controls the response amplitude of the sine function to the current production status and is used to balance the impact of nonlinear adjustment on production status evaluation.

[0099] Finally, the production status deviation assessment value is compared with the threshold value preset according to the expert experience method. If the production status deviation assessment value exceeds the threshold value, the monitoring and analysis system for the preparation of marine peptides determines that there is an abnormality in the current production process. According to the difference between the production status deviation assessment value and the preset threshold value, a status assessment result is generated. The status assessment results are divided into slight deviations, moderate deviations and severe deviations. Adjustment suggestions are generated based on the status assessment results. For example, for slight deviations, it is recommended to lower or raise the temperature by 0.1-0.5 degrees through fine adjustment to keep it within the optimal range. It is recommended to use trace amounts of acid and alkali solutions for slow adjustment to gradually adjust the pH value to the optimal range, and the adjustment range should not exceed 0.1. It is recommended to moderately adjust the pump speed or valve opening to adjust the flow rate to the predetermined optimal value. For moderate deviations, it is recommended to quickly adjust the temperature to the set target range, with an adjustment range of about 0.5-2 degrees, which varies appropriately according to the current environment and process requirements. It is recommended to quickly add acid and alkali solutions to quickly restore the pH value to a reasonable range, but it is necessary to avoid adding too much solution at one time to prevent over-adjustment and keep it within the range of 0.1-0.3. It is recommended to adjust the pump speed or increase or decrease the valve opening according to production needs, and the adjustment range should be within the range of 10%-20% to ensure that the material flow rate adapts to the current production status. For severe deviations, it is recommended to immediately force the temperature to return to a safe range. At this time, it is recommended to suspend some production operations and continue production after the temperature returns to normal; it is recommended to take emergency measures immediately to quickly neutralize excess acid or alkali, and accelerate uniform mixing through stirring to avoid excessive concentration in local areas.

[0100] After the adjustment suggestions are generated, the above suggestions are converted into specific control instructions to achieve automatic regulation of the marine peptide preparation equipment.

[0101] The production parameter data after intelligent processing in the production process, the production characteristics after time domain offset transformation, the state analysis results, the state assessment results, the state analysis results after dynamic nonlinear combination, and the adjustment suggestions are stored and used as historical data.

[0102] In summary, a monitoring and analysis system for the preparation of marine peptides was completed.

[0103] The order of the embodiments of the invention is for description only and does not represent the advantages and disadvantages of the embodiments. The processes depicted in the 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.

[0104] 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.

[0105] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that the technical solutions described in the above embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention, and should be included in the protection scope of the present invention.

Claims

1. A monitoring and analysis method for the preparation of marine peptides, characterized in that: The following steps are involved: S1. Collect the original production parameter data in the production process, perform intelligent processing on the original production parameter data, and perform denoising on the original production parameter data during the implementation of intelligent processing; use an outlier detection algorithm with an adaptive threshold to detect and eliminate outliers on the denoised production parameter data, and the adaptive threshold is defined as follows: ； in, is the adaptive threshold; It's time The mean value of the denoised production parameter data at the moment; is the weight coefficient of the standard deviation; It's time The standard deviation of the production parameter data after denoising at each moment; is the skewness adjustment coefficient; It's time The skewness coefficient at time, , is the expected value operator; It's time The denoised production parameter data at each moment; is the time decay coefficient; The production parameter data after outlier detection and elimination is standardized to obtain intelligently processed production parameter data; the production parameter data after intelligent processing is intelligently analyzed using dynamic hierarchical feature mapping and time domain offset transformation analysis to obtain state analysis results; The specific implementation process of dynamic hierarchical feature mapping and time domain offset transformation analysis method is as follows: Through the mapping matrix, the production parameter data after intelligent processing is dynamically layered and mapped to obtain the mapping results of each feature layer. The mapping formula is defined as: ； in, Representative The mapping result of feature layers; is the dynamic mapping matrix; It's time The intelligently processed production parameter data set; The mapping results of each feature layer are subjected to time domain offset transformation. The time domain offset transformation formula is as follows: ； in, Indicates The production characteristics of each feature layer after time domain migration transformation; is the time offset coefficient; It is The rate of change of production characteristics of each characteristic layer over time; The feature cross term is introduced to calculate the feature interaction strength between different feature layers. The feature interaction strength is calculated by the following formula: ； in, It is Layer and Layer feature interaction strength between layers, ; Used to prevent the denominator from being zero; It is The production characteristics of each feature layer after time domain migration transformation; The feature layers are weighted according to the strength of feature interaction between them. The allocation formula is as follows: ； in, Indicates the total number of feature layers; Based on the weights of different feature layers, the production state deviation is calculated as follows: ； in, It is the current moment Deviation of production status; It is the current moment No. Layer characteristic layer's production characteristics after historical best time domain migration transformation; is the weighted feature result; According to the production state deviation, the state correction model is introduced to generate the correction value of the production conditions. The specific implementation formula is: ； in, is the correction value of production conditions; It is a multidimensional nonlinear adjustment term; is the corrected sensitivity coefficient; Based on the correction value of production conditions, the status analysis results are formed; S2. The state analysis results are adaptively evaluated by the multi-dimensional dynamic state adaptive evaluation algorithm for the preparation of marine peptides. In the process of implementing the multi-dimensional dynamic state adaptive evaluation algorithm for the preparation of marine peptides, the state analysis results are dynamically nonlinearly combined by introducing time transformation and dynamic weights to obtain the state analysis results after dynamic nonlinear combination. The specific formula is: ； in, It is the state analysis result after dynamic nonlinear combination; is the nonlinear adjustment coefficient; is a nonlinear trigonometric function; is the smoothing factor; is the amplification factor of the correction rate of change of production conditions; is the historical cumulative weight factor; is the modified rate of change of production conditions; The state analysis results after dynamic nonlinear combination are adaptively evaluated to obtain the production state deviation evaluation value. The adaptive evaluation formula is: ； in, is the production status deviation assessment value; is the integration variable; It is the state analysis result after dynamic nonlinear combination; It is the state analysis result after the best historical dynamic nonlinear combination; is the adaptive adjustment coefficient; is the nonlinear adjustment coefficient; By comparing the production status deviation assessment value with the preset threshold, it is evaluated whether there is an abnormality in the production process; and based on the difference between the production status deviation assessment value and the preset threshold, a status assessment result is generated; based on the status assessment result, an adjustment suggestion is generated; the adjustment suggestion is converted into a control instruction to automatically adjust the marine peptide preparation equipment.

2. A monitoring and analysis system for the preparation of marine peptides, applied to the monitoring and analysis method for the preparation of marine peptides according to claim 1, characterized in that: Includes the following sections: Data acquisition module, data intelligent processing module, intelligent analysis module, adaptive evaluation module, automatic control module, data storage module; The data acquisition module collects the original production parameter data in the production process in real time and sends the original production parameter data to the data intelligent processing module; The data intelligent processing module performs intelligent processing on the original production parameter data to obtain the intelligently processed production parameter data, and sends the intelligently processed production parameter data to the intelligent analysis module and the data storage module; The intelligent analysis module uses dynamic hierarchical feature mapping and time domain offset transformation analysis methods, combined with the production characteristics after the historical time domain offset transformation of the data storage module, to perform intelligent analysis on the production parameter data after intelligent processing, and obtains the mapped feature data, the production characteristics after the time domain offset transformation, the production state deviation value, and the correction value of the production condition. Based on the correction value of the production condition, a state analysis result is formed, and the state analysis result is sent to the adaptive evaluation module, and the production characteristics after the time domain offset transformation and the state analysis result are sent to the data storage module; An adaptive evaluation module performs dynamic nonlinear combination on the state analysis results to obtain a state analysis result after the dynamic nonlinear combination, performs adaptive evaluation on the state analysis result after the dynamic nonlinear combination to obtain a state evaluation result, and sends the state evaluation result to the automatic control module, and sends the state analysis result after the dynamic nonlinear combination and the state evaluation result to the data storage module; An automatic control module generates adjustment suggestions based on the status evaluation results, generates control instructions based on the adjustment suggestions, and automatically adjusts the production conditions according to the control instructions; and sends the adjustment suggestions to the data storage module; The data storage module is used to store production parameter data after intelligent processing in the production process, production characteristics after time domain offset transformation, state analysis results, state assessment results, state analysis results after dynamic nonlinear combination, and adjustment suggestions, and is used as historical data.

Images