Manufacturing process of chicory compound preparation

Abstract

The invention relates to the technical field of production process monitoring, in particular to a manufacturing process of a chicory compound preparation, which comprises a data acquisition module, a data processing module, a data processing module and a processing module, obtaining to-be-analyzed pH data, a characteristic temperature window, a characteristic pH window, a reference temperature window and a reference pH window corresponding to the tth monitoring moment; the drift anomaly identification and correction module is used for carrying out drift anomaly identification and correction on the to-be-analyzed pH data corresponding to the tth monitoring moment to obtain target pH data corresponding to the tth monitoring moment; and the pH monitoring and adjusting module is used for monitoring and adjusting the pH in the chicory compound preparation production and manufacturing container at the t monitoring moment according to the target pH data corresponding to the t monitoring moment. And the pH value in the endive compound preparation production and manufacturing container at the t monitoring moment is monitored and adjusted according to the corrected pH value data, so that the adjusting effect can be improved or guaranteed.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of production process monitoring, in particular to a manufacturing process of a chicory composite preparation. BACKGROUND

[0002] The chicory composite preparation is a composite product made of chicory as the main raw material and mixed with other plant extracts or functional ingredients. Since many functional ingredients in chicory are greatly affected by pH, changes in pH during the manufacturing process not only affect the flavor of the preparation, but also may cause changes in the structure of some ingredients and thus affect the quality of the product. Therefore, it is crucial to monitor and adjust the pH value in real time during the mixing process of the raw materials in the manufacturing process of the chicory composite preparation.

[0003] The existing pH monitoring and adjustment method in the production and manufacturing process of the chicory composite preparation usually refers to directly adjusting the pH in the production and manufacturing process of the chicory composite preparation according to the real-time monitored pH. That is, the existing method usually uses a pH sensor to monitor the pH in real time, and directly compares the real-time monitored pH value with the preset pH interval. When the real-time monitored pH value is found to be out of the preset pH interval, the pH value is adjusted by adding acid or alkali to keep the pH value as stable as possible within a reasonable interval. However, during the production and manufacturing process, the chicory raw juice may contain impurities such as deposition of some compound ingredients, plant fibers and colloid residues. These impurities may block the surface of the pH electrode. When the impurities block the surface of the pH electrode, the data monitored by the pH sensor may deviate from the true value, resulting in poor adjustment effect of the pH monitoring and adjustment in the production and manufacturing process of the chicory composite preparation, affecting the stability of the preparation or the quality of the product. For example, the real pH before adjustment may be within the preset pH interval, but the real pH after adjustment may be out of the preset pH interval. Therefore, how to improve or ensure the adjustment effect when monitoring and adjusting the pH in the production and manufacturing process of the chicory composite preparation has become a problem to be solved. SUMMARY

[0004] To solve the above problems, the present application provides a manufacturing process of a chicory composite preparation, and the technical solution adopted is as follows: One embodiment of the present application provides a manufacturing process of a chicory composite preparation, which comprises: A data acquisition module is configured to acquire, during the production and manufacturing process of the chicory composite preparation, to-be-analyzed pH data, a characteristic temperature window, a characteristic pH window, a reference temperature window and a reference pH window corresponding to a tth monitoring time. The drift anomaly identification and correction module is used for obtaining an initial feature value corresponding to the tth monitoring moment according to the similarity between the feature temperature window and the reference temperature window and the difference between the fitting slope of the feature pH window and the fitting slope of the reference pH window, obtaining a target feature value corresponding to the tth monitoring moment according to the difference between the fitting value obtained by fitting the feature pH window and the pH data to be analyzed and the initial feature value, clustering the target feature values corresponding to the tth monitoring moment and the monitoring moment before the tth monitoring moment to obtain a clustering cluster corresponding to the tth monitoring moment, identifying and correcting the drift anomaly of the pH data to be analyzed corresponding to the tth monitoring moment according to the clustering cluster, and obtaining target pH data corresponding to the tth monitoring moment. The pH monitoring and adjustment module is used for monitoring and adjusting the pH in the chicory compound preparation production and manufacturing container at the tth monitoring moment according to the target pH data corresponding to the tth monitoring moment.

[0005] Beneficial effects: The application comprises a data acquisition module used for acquiring pH data to be analyzed corresponding to the tth monitoring moment, a feature temperature window, a feature pH window, a reference temperature window and a reference pH window in the production and manufacturing process of the chicory compound preparation; a drift anomaly identification and correction module used for obtaining an initial feature value corresponding to the tth monitoring moment according to the similarity between the feature temperature window and the reference temperature window and the difference between the fitting slope of the feature pH window and the fitting slope of the reference pH window, obtaining a target feature value corresponding to the tth monitoring moment according to the difference between the fitting value obtained by fitting the feature pH window and the pH data to be analyzed and the initial feature value, clustering the target feature values corresponding to the tth monitoring moment and the monitoring moment before the tth monitoring moment to obtain a clustering cluster corresponding to the tth monitoring moment, identifying and correcting the drift anomaly of the pH data to be analyzed corresponding to the tth monitoring moment according to the clustering cluster, and obtaining target pH data corresponding to the tth monitoring moment; and a pH monitoring and adjustment module used for monitoring and adjusting the pH in the chicory compound preparation production and manufacturing container at the tth monitoring moment according to the target pH data corresponding to the tth monitoring moment. The pH data is adjusted according to the corrected pH data in the chicory compound preparation production and manufacturing container at the tth monitoring moment, which can improve or guarantee the adjustment effect, thereby ensuring the production stability and production quality of the chicory compound preparation. BRIEF DESCRIPTION OF DRAWINGS

[0006] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the drawings needed to be used in the description of the embodiments or the prior art will be briefly introduced. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative effort based on these drawings.

[0007] Figure 1 The structure block diagram of the manufacturing process of the chicory composite preparation of the present application. DETAILED DESCRIPTION

[0008] The technical solutions in the embodiments of the present application will be described clearly and completely with reference to the drawings of the embodiments of the present application. Obviously, the described embodiments are only some embodiments of the present application, not all embodiments of the present application. Based on the embodiments of the present application, all other embodiments obtained by those skilled in the art without creative effort belong to the scope of protection of the embodiments of the present application.

[0009] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which the present application belongs.

[0010] The present embodiment provides a manufacturing process of a chicory composite preparation, which is described in detail as follows: As shown in Figure 1 The present embodiment provides a manufacturing process of a chicory composite preparation, which includes: The data acquisition module 01 is configured to acquire the to-be-analyzed pH data, the characteristic temperature window, the characteristic pH window, the reference temperature window and the reference pH window corresponding to the tth monitoring time in the production and manufacturing process of the chicory composite preparation.

[0011] The purpose of the embodiment is to determine whether the monitored data is affected by impurity components to correct the data, which can also be said to be the identification and correction of the drift anomaly of the monitored pH data. Based on the corrected data, the pH in the production process of the chicory composite preparation is monitored and adjusted. The impurity component in the embodiment refers to the phenomenon that the monitored pH data deviates from the true pH caused by the blockage of impurity components on the surface of the pH electrode, also known as drift anomaly. The drift anomaly refers to the phenomenon that the monitored pH data deviates from the true pH caused by the blockage of impurity components on the surface of the pH electrode. The impurity components mainly refer to fibers, colloids and some compounds that may be deposited in the chicory raw materials. During the pH monitoring process, these impurity components may adhere to the surface of the pH electrode or block the electrode structure, thereby affecting the accuracy of the pH monitoring data or causing the data monitored by the pH sensor to drift. In addition, for ease of analysis and understanding, the subsequent embodiment will be described by taking the pH monitoring and adjustment process in the production process of any batch of chicory composite preparation as an example.

[0012] In order to realize the identification and correction of drift anomaly and improve the adjustment effect, the embodiment needs to collect the temperature and pH in the production and manufacturing process of the chrysanthemum composite preparation in real time, obtain the to-be-de-noised pH data and the sensed temperature data corresponding to each monitoring time, and the specific process of obtaining the to-be-de-noised pH data and the sensed temperature data corresponding to each monitoring time in the production and manufacturing process of the chrysanthemum composite preparation is as follows: a pH sensor and a temperature sensor are installed in a formula mixing tank for manufacturing the chrysanthemum composite preparation, the pH in the mixing tank at different monitoring times is collected in real time by using the pH sensor, the collected pH data is recorded as initial pH data corresponding to the corresponding monitoring time, the mixing tank is a container for producing and manufacturing the chrysanthemum composite preparation, the temperature in the mixing tank at different monitoring times is collected in real time by using the temperature sensor, the collected temperature data is recorded as initial temperature data corresponding to the corresponding monitoring time, the initial pH data is processed by using the existing normalization method or standardization method, the data obtained by processing is recorded as to-be-de-noised pH data, the initial temperature data is processed by using the existing normalization method or standardization method, the data obtained by processing is recorded as sensed temperature data, and the minimum and maximum normalization method or the Z-Score standardization method can be selected for processing; and since the pH value change is sensitive to temperature, the temperature may affect some chemical reactions and may also cause electrode monitoring deviation, in order to distinguish the influence of temperature and impurity components on the pH change or to more accurately identify the drift anomaly data, the subsequent identification and correction of drift anomaly in the embodiment also needs to refer to the temperature data, so the temperature data needs to be collected; in addition, the temperature sensor and the pH sensor in the embodiment are synchronous collection, and the specific collection frequency or the time interval between adjacent monitoring times needs to be set by the implementer according to the actual situation such as monitoring accuracy, for example, the time interval between adjacent monitoring times can be set to 1 s in the embodiment, and the monitoring time is also the data collection time.

[0013] Therefore, the embodiment can obtain the to-be-de-noised pH data and the sensing temperature data corresponding to each monitoring time through the above process in the production process of the chicory composite preparation. However, in the production process of the chicory composite preparation, factors such as suspended impurities and bubbles may cause the pH data collected by the sensor to have sudden noise, and the noise may affect the subsequent identification and correction of the drift anomaly. Therefore, the to-be-de-noised pH data needs to be de-noised before the real-time identification and correction of the drift anomaly, that is, the to-be-de-noised pH data is de-noised in real time after being obtained, and the de-noised data is recorded as the to-be-analyzed pH data corresponding to the corresponding monitoring time. That is, the data obtained by de-noising the to-be-de-noised pH data corresponding to a monitoring time is the to-be-analyzed pH data corresponding to the monitoring time. In the embodiment, the data for identifying and correcting the drift anomaly is the to-be-analyzed pH data. Since the de-noising method of the to-be-de-noised pH data corresponding to each monitoring time is the same, and in order to facilitate understanding and description, the process of de-noising the to-be-de-noised pH data corresponding to any monitoring time r to obtain the to-be-analyzed pH data corresponding to the monitoring time r will be described as an example in the following. The specific process of de-noising the to-be-de-noised pH data corresponding to the monitoring time r to obtain the to-be-analyzed pH data corresponding to the monitoring time r when the chicory composite preparation is produced from the beginning to the monitoring time r is as follows: Firstly, the absolute value of the difference between the to-be-de-noised pH data corresponding to the monitoring time r and the target pH data corresponding to the monitoring time before the monitoring time r is calculated, and is recorded as a first difference value. The first difference value is normalized, and the result of the normalization is recorded as a first evaluation value of the to-be-de-noised pH data corresponding to the monitoring time r. Then, the target pH data corresponding to the monitoring time before the monitoring time r is used to predict the pH data corresponding to the monitoring time r, to obtain predicted pH data corresponding to the monitoring time r. The specific process of obtaining the predicted pH data corresponding to the monitoring time r is as follows: a time sequence constructed by a continuous preset number of monitoring times before the monitoring time r and closest to the monitoring time r in time is recorded as neighboring historical monitoring times corresponding to the monitoring time r. A two-dimensional space is constructed, with time as the horizontal coordinate and pH data as the vertical coordinate. The monitoring times in the neighboring historical monitoring times and the target pH data corresponding to the monitoring times are mapped into the two-dimensional space, to obtain data points corresponding to the monitoring times in the neighboring historical monitoring times. The horizontal coordinate value of the data point corresponding to a monitoring time is the time corresponding to the monitoring time, and the vertical coordinate value of the data point corresponding to the monitoring time is the target pH data corresponding to the monitoring time. The data points corresponding to the monitoring times in the neighboring historical monitoring times are fitted by using the least square method, to obtain a fitting straight line. The data corresponding to the monitoring time t is obtained on the fitting straight line, and is recorded as the predicted pH data corresponding to the monitoring time r. Then, the absolute value of the difference between the predicted pH data corresponding to the monitoring time t and the to-be-de-noised pH data corresponding to the monitoring time t is calculated, and is recorded as a second difference value. The second difference value is normalized, and the result of the normalization is recorded as a second evaluation value of the to-be-de-noised pH data corresponding to the monitoring time r. Finally, the mean value of the first evaluation value and the second evaluation value of the to-be-de-noised pH data corresponding to the monitoring time r is obtained, and is recorded as a comprehensive evaluation value of the to-be-de-noised pH data corresponding to the monitoring time r. In addition, since the de-noising, identification and correction of the drift anomaly are performed in real time in this embodiment, the monitoring times before the current monitoring time are historical monitoring times, and therefore the historical monitoring times have completed the de-noising, identification and correction of the drift anomaly. Therefore, the corrected data corresponding to the historical monitoring times can be obtained, and the corrected data is also referred to as the target pH data in this embodiment. Therefore, when the chicory compound preparation is produced from the beginning to the monitoring time r, the monitoring time r is the current monitoring time, and the target pH data corresponding to the monitoring times before the monitoring time r is the data obtained by de-noising the to-be-de-noised pH data corresponding to the monitoring times before the monitoring time r, and then performing identification and correction of the drift anomaly. In specific applications, the implementer needs to set the value of the preset number according to the actual situation. For example, the preset number can be set to 10 in this embodiment.

[0014] The specific calculation expression of the comprehensive evaluation value of the to-be-de-noised pH data corresponding to the monitoring time r is as follows:

[0015] wherein, is the comprehensive evaluation value of the to-be-de-noised pH data corresponding to the monitoring moment r, Norm() is a normalization function, is the to-be-de-noised pH data corresponding to the monitoring moment r, is the target pH data corresponding to the previous monitoring moment of the monitoring moment r, is the predicted pH data corresponding to the monitoring moment t; and the difference between the mutation noise and the neighborhood data is larger, and the fluctuation range of the mutation noise is larger relative to the trend of the neighborhood data, so when is larger and is larger, or is larger, the possibility of the to-be-de-noised pH data corresponding to the monitoring moment r being mutation noise is larger, and the probability of subsequent de-noising is larger, when is smaller, the possibility of the to-be-de-noised pH data corresponding to the monitoring moment r being mutation noise is smaller, and the probability of subsequent de-noising is smaller.

[0016] Then, it is judged whether the comprehensive evaluation value of the to-be-de-noised pH data corresponding to the monitoring moment r is larger than a preset noise threshold. If it is larger, it is determined that the to-be-de-noised pH data corresponding to the monitoring moment r is mutation noise data, and the to-be-de-noised pH data corresponding to the monitoring moment r needs to be de-noised. The de-noised data is recorded as the to-be-analyzed pH data corresponding to the monitoring moment r. If it is not larger, it is determined that the to-be-de-noised pH data corresponding to the monitoring moment r is not mutation noise data, and the to-be-de-noised pH data corresponding to the monitoring moment r does not need to be de-noised. The to-be-de-noised pH data corresponding to the monitoring moment r is directly recorded as the to-be-analyzed pH data corresponding to the monitoring moment r. If it is judged that the to-be-de-noised pH data corresponding to the monitoring moment r needs to be de-noised, the embodiment can select a predicted value replacement method to complete the de-noising, that is, if it is judged that the to-be-de-noised pH data corresponding to the monitoring moment r needs to be de-noised, the embodiment records the predicted pH data corresponding to the monitoring moment t as the to-be-analyzed pH data corresponding to the monitoring moment r. As other real-time methods, the existing de-noising method can also be selected to de-noise the to-be-de-noised pH data corresponding to the monitoring moment r.

[0017] And in specific applications, the implementer needs to set the preset noise threshold according to the value range of the comprehensive evaluation value and other actual situations. For example, the change of non-mutation noise data is changed according to the real acid-base situation of the raw material, that is, it changes according to a certain trend, so the comprehensive evaluation value of the non-mutation noise data is relatively small, while the change of the mutation noise data is random and has no regularity with other data, so the comprehensive evaluation value of the mutation noise data is relatively large. Therefore, after normalization processing, the mutation noise data and the non-noise data and the mutation noise data have a large difference in the comprehensive evaluation value, that is, they are concentrated at both ends of the corresponding value range. Therefore, the preset noise threshold can be set to 0.5 in this embodiment. In addition, since the probability of mutation noise occurring at the initial monitoring time is extremely low, the pH data to be denoised at the initial monitoring time can be directly recorded as the pH data to be analyzed at the monitoring time, such as the pH data to be denoised at the monitoring time within the first 10 seconds can be directly recorded as the pH data to be analyzed at the monitoring time.

[0018] The method of denoising the pH data to be denoised at any monitoring time to obtain the pH data to be analyzed at the corresponding monitoring time in this embodiment is the same as the method of denoising the pH data to be denoised at the monitoring time r to obtain the pH data to be analyzed at the monitoring time r. The method of identifying and correcting the drift anomaly of the pH data to be analyzed at any monitoring time to obtain the target pH data to be denoised at the corresponding monitoring time in this embodiment is the same as the method of denoising the pH data to be analyzed at the tth monitoring time to obtain the target pH data at the tth monitoring time.

[0019] After denoising is completed, the process of identifying and correcting the drift anomaly of the pH data to be analyzed at any monitoring time to obtain the target pH data at the corresponding monitoring time is described, for example, when the chicory compound preparation is produced from the beginning to the tth monitoring time, the specific process of identifying and correcting the drift anomaly of the pH data to be denoised at the tth monitoring time to obtain the target pH data at the tth monitoring time is described. In this embodiment, the characteristic temperature window, the characteristic pH window, the reference temperature window and the reference pH window corresponding to the tth monitoring time need to be obtained first, that is, the characteristic temperature window, the characteristic pH window, the reference temperature window and the reference pH window corresponding to the tth monitoring time are the data support for identifying and correcting the drift anomaly of the pH data to be analyzed at the tth monitoring time. Then, the specific acquisition process of the characteristic temperature window, the characteristic pH window, the reference temperature window and the reference pH window corresponding to the tth monitoring time is: Firstly, the time period formed from the beginning of the production of the chicory composite preparation to the tth monitoring moment is obtained, and is recorded as the production and manufacturing monitoring time period corresponding to the tth monitoring moment. The last moment in the production and manufacturing monitoring time period corresponding to the tth monitoring moment is the tth monitoring moment. Then, a sliding window is constructed, and the right end of the sliding window is aligned with the right end moment of the production and manufacturing monitoring time period. From the end moment of the production and manufacturing monitoring time period, sliding is performed forward according to a preset step size, until the initial moment of the production and manufacturing monitoring time period is contained in the sliding window, and the sliding stops. All the windows obtained from the beginning of the sliding to the stopping of the sliding are recorded as the time windows corresponding to the tth monitoring moment. Then, in all the time windows corresponding to the tth monitoring moment, the time window containing the tth monitoring moment is selected and recorded as the characteristic time window corresponding to the tth monitoring moment, and all the time windows not containing the tth monitoring moment are selected and recorded as the reference time windows corresponding to the tth monitoring moment. Next, the time sequence window composed of the sensing temperature data corresponding to all the monitoring moments in the characteristic time window corresponding to the tth monitoring moment is recorded as the characteristic temperature window corresponding to the tth monitoring moment, the time sequence window composed of the sensing temperature data corresponding to all the monitoring moments in the fth reference time window corresponding to the tth monitoring moment is recorded as the fth reference temperature window corresponding to the tth monitoring moment, the time sequence window composed of the target pH data corresponding to all the monitoring moments in the fth reference time window corresponding to the tth monitoring moment is recorded as the fth reference pH window corresponding to the tth monitoring moment, and the time sequence window composed of the target pH data corresponding to the monitoring moments other than the tth monitoring moment in the characteristic time window corresponding to the tth monitoring moment and the to-be-analyzed pH data corresponding to the tth monitoring moment is recorded as the characteristic pH window corresponding to the tth monitoring moment. The characteristic pH window and the characteristic temperature window are the characteristic data windows corresponding to the characteristic time window. The target pH data corresponding to the monitoring moments before the tth monitoring moment refers to the data after the identification and correction of the drift anomaly of the to-be-analyzed pH data corresponding to the monitoring moments, and the method for identifying and correcting the drift anomaly of the to-be-analyzed pH data corresponding to the monitoring moments before the tth monitoring moment is the same as the method for identifying and correcting the drift anomaly of the to-be-analyzed pH data corresponding to the tth monitoring moment. In specific applications, the implementer can set the size of the sliding window and the preset step size according to the actual acid-base adjustment frequency and other actual conditions. However, because the liquid mixing stage of the production of the chicory composite preparation usually lasts for 30 minutes to several hours according to different process complexities, in order to minimize the influence of the pH adjustment process on the analysis, the window should not be set too large or too small. For example, in this embodiment, the length of the sliding window can be set to 1 minute, and the preset step size can be set to 1 second.

[0020] Therefore, the embodiment can obtain the pH data to be analyzed, the characteristic temperature window, the characteristic pH window, the reference temperature window and the reference pH window corresponding to the tth monitoring moment through the above process. The embodiment requires that the characteristic temperature window and the reference temperature window are equal in length. If they are not equal in length, padding and truncation processing can be performed. For example, the longer window is truncated to the same length as the shorter window.

[0021] The drift anomaly identification and correction module 02 is configured to obtain an initial feature value corresponding to the tth monitoring moment according to the similarity between the characteristic temperature window and the reference temperature window and the difference between the fitting slope of the characteristic pH window and the fitting slope of the reference pH window, obtain a target feature value corresponding to the tth monitoring moment according to the difference between the fitting value obtained by fitting the characteristic pH window and the pH data to be analyzed and the initial feature value, cluster the target feature values corresponding to the tth monitoring moment and the monitoring moments before the tth monitoring moment to obtain a cluster corresponding to the tth monitoring moment, and identify and correct the drift anomaly of the pH data to be analyzed corresponding to the tth monitoring moment according to the cluster to obtain target pH data corresponding to the tth monitoring moment.

[0022] Since the pH monitoring value is affected by temperature in addition to impurity components when no acid or base is added for adjustment, temperature is one of the main influencing factors. For example, temperature can affect the pH electrode slope or potential, resulting in deviation of the pH monitoring value. Temperature can promote certain chemical reactions at different temperatures, thereby causing changes in the pH value. However, the influence of temperature on the pH value has strong regularity and is reproducible. The drift degree of data caused by impurities is related to factors such as impurity concentration and deposition, and generally has no fixed regularity. Moreover, when the temperature conditions between the windows are similar, the pH trends between the windows are also similar without the influence of impurities. Therefore, the initial feature value corresponding to the tth monitoring moment can be obtained according to the temperature similarity between the windows and the similarity of the pH trends between the windows in the subsequent process of the embodiment. The initial feature value can reflect whether the pH data to be analyzed corresponding to the tth monitoring moment is a drift anomaly. Therefore, the temperature similarity between the characteristic temperature window corresponding to the tth monitoring moment and the reference temperature windows corresponding to the tth monitoring moment is obtained according to the absolute value of the difference between the data in the characteristic temperature window corresponding to the tth monitoring moment and the corresponding position data in the reference temperature windows corresponding to the tth monitoring moment. The temperature similarity is a key data for accurately identifying whether it is a drift anomaly. The specific process of obtaining the temperature similarity between the characteristic temperature window corresponding to the tth monitoring moment and the reference temperature windows corresponding to the tth monitoring moment is as follows. In the feature temperature window corresponding to the tth monitoring moment, starting from the initial data in the feature temperature window, every preset number of data is marked once until the last data position in the feature temperature window is reached, and the marked data in the feature temperature window is recorded as the marked data in the feature temperature window. In the fth reference temperature window, starting from the initial data in the fth reference temperature window, every preset number of data is marked once until the last data in the reference temperature window is reached, and the marked data in the fth reference temperature window is recorded as the marked data in the fth reference temperature window. If the gth data in the feature temperature window is marked data and the gth data in the fth reference temperature window is also marked data, the absolute value of the difference between each marked data in the feature temperature window corresponding to the tth monitoring moment and the marked data at the corresponding position in the fth reference temperature window corresponding to the tth monitoring moment is obtained to obtain a temperature difference value sequence between the feature temperature window corresponding to the tth monitoring moment and the reference temperature window corresponding to the tth monitoring moment. The a th temperature difference value in the temperature difference value sequence is the absolute value of the difference between the a th marked data in the feature temperature window and the a th marked data in the fth reference temperature window. The result of the negative correlation mapping of the cumulative result of all data in the temperature difference value sequence and then the normalization is recorded as the temperature similarity between the feature temperature window corresponding to the tth monitoring moment and the fth reference temperature window corresponding to the tth monitoring moment. The negative correlation mapping and then the normalization here means that after the cumulative result of all data in the temperature difference value sequence is added by a preset constant and then inverted, the normalization function Norm() is processed. The preset constant is to prevent the denominator from being 0. The specific expression of the temperature similarity between the feature temperature window corresponding to the tth monitoring moment and the fth reference temperature window corresponding to the tth monitoring moment is:

[0023] wherein, is the temperature similarity between the feature temperature window corresponding to the tth monitoring moment and the fth reference temperature window corresponding to the tth monitoring moment, Norm() is a normalization function, A is the total number of marked data in the feature temperature window corresponding to the tth monitoring moment or the fth reference temperature window corresponding to the tth monitoring moment, which can also be referred to as the total number of data in the temperature difference value sequence, is the a th marked data in the feature temperature window corresponding to the tth monitoring moment, is the a th marked data in the fth reference temperature window corresponding to the tth monitoring moment, is the a-th temperature difference value in the temperature difference value sequence, c is a preset constant, c can be any constant, for example, c is set to 0.01 in this embodiment. And The smaller the value is, the more similar the temperature between the characteristic temperature window corresponding to the t-th monitoring moment and the f-th reference temperature window corresponding to the t-th monitoring moment is, and vice versa The larger the value is, the more similar the temperature between the characteristic temperature window corresponding to the t-th monitoring moment and the f-th reference temperature window corresponding to the t-th monitoring moment is, and vice versa The smaller the value is, the more similar the temperature between the characteristic temperature window corresponding to the t-th monitoring moment and the f-th reference temperature window corresponding to the t-th monitoring moment is, and vice versa The smaller the value is, the more similar the temperature between the characteristic temperature window corresponding to the t-th monitoring moment and the f-th reference temperature window corresponding to the t-th monitoring moment is, and vice versa

[0024] Since the pH value needs to be continuously adjusted during the liquid mixing stage of the chicory compound preparation to achieve the final effect, there may be some differences in the normal pH value at different times, but the influence trend of the same temperature condition on the pH value has similarity, therefore, after analyzing the similarity between the characteristic temperature window and the reference temperature window, the embodiment will next obtain the difference between the slope of the straight line obtained by fitting all the data in the characteristic pH window corresponding to the t-th monitoring moment and the slope of the straight line obtained by fitting all the data in the reference pH window corresponding to the t-th monitoring moment, to obtain the pH change trend similarity between the characteristic pH window corresponding to the t-th monitoring moment and the reference pH window corresponding to the t-th monitoring moment, and the specific acquisition process of the pH change trend similarity between the characteristic pH window corresponding to the t-th monitoring moment and the f-th reference pH window corresponding to the t-th monitoring moment is: using the least square method to fit the data in the characteristic pH window corresponding to the t-th monitoring moment, and recording the slope of the straight line obtained by fitting as the trend characteristic value of the characteristic pH window, using the least square method to fit the data in the f-th reference pH window corresponding to the t-th monitoring moment, and recording the slope of the straight line obtained by fitting as the trend characteristic value of the f-th reference pH window, calculating the absolute value of the difference between the trend characteristic value of the characteristic pH window and the trend characteristic value of the f-th reference pH window, and then performing negative correlation mapping and normalization, and recording the result as the pH change trend similarity between the characteristic pH window corresponding to the t-th monitoring moment and the f-th reference pH window corresponding to the t-th monitoring moment, the characteristic pH window and the characteristic temperature data belong to the characteristic time window, the f-th reference pH window and the f-th reference temperature window belong to the f-th reference time window corresponding to the t-th monitoring moment, and the negative correlation mapping and normalization here means that after adding a preset constant to the absolute value of the difference, taking the reciprocal, and then processing with the normalization function Norm(), the preset constant here is also to prevent the denominator from being 0.

[0025] The specific calculation expression of the pH change trend similarity between the characteristic pH window corresponding to the t-th monitoring moment and the f-th reference pH window corresponding to the t-th monitoring moment is:

[0026] wherein, the pH change trend similarity between the characteristic pH window corresponding to the tth monitoring moment and the fth reference pH window corresponding to the tth monitoring moment, K1 is the trend representation value of the characteristic pH window, K2 is the trend representation value of the fth reference pH window; and the smaller the absolute value of the difference between the trend representation value of the characteristic pH window and the trend representation value of the fth reference pH window, that is, the greater, the more similar the pH change trend of the characteristic pH window corresponding to the tth monitoring moment to the pH change trend of the fth reference pH window corresponding to the tth monitoring moment, and vice versa, the greater the absolute value of the difference between the trend representation value of the characteristic pH window and the trend representation value of the fth reference pH window, that is, the smaller, the less similar the pH change trend of the characteristic pH window corresponding to the tth monitoring moment to the pH change trend of the fth reference pH window corresponding to the tth monitoring moment.

[0027] Since the closer the window corresponding to the data to the tth monitoring moment, the greater the reference value, therefore, after obtaining the temperature similarity between the characteristic temperature window corresponding to the tth monitoring moment and the reference temperature window corresponding to the tth monitoring moment, the pH change trend similarity between the characteristic pH window corresponding to the tth monitoring moment and the reference pH window corresponding to the tth monitoring moment, the embodiment combines the time interval between the characteristic pH window and the reference pH window to determine the initial characteristic value corresponding to the tth monitoring moment, that is, the embodiment will next obtain the initial characteristic value corresponding to the tth monitoring moment according to the temperature similarity between the characteristic temperature window corresponding to the tth monitoring moment and the reference temperature window corresponding to the tth monitoring moment, the pH change trend similarity between the characteristic pH window corresponding to the tth monitoring moment and the reference pH window corresponding to the tth monitoring moment, and the time interval between the characteristic pH window corresponding to the tth monitoring moment and the reference pH window corresponding to the tth monitoring moment, the initial characteristic value corresponding to the tth monitoring moment is a key parameter for determining the target characteristic value, and the target characteristic value is a key data for identifying whether it is a drift anomaly, then the specific acquisition process of the initial characteristic value corresponding to the tth monitoring moment is: According to the temperature similarity between the feature temperature window corresponding to the tth monitoring moment and each reference temperature window corresponding to the tth monitoring moment, the pH change trend similarity between the feature pH window corresponding to the tth monitoring moment and each reference pH window corresponding to the tth monitoring moment, and the time interval between the feature pH window corresponding to the tth monitoring moment and the time window corresponding to each reference pH window corresponding to the tth monitoring moment, a comprehensive feature value set corresponding to the tth monitoring moment is obtained. The bth comprehensive feature value in the comprehensive feature value set is the multiplication result of the temperature similarity between the feature temperature window and the bth reference temperature window corresponding to the tth monitoring moment, the negative correlation mapping result of the pH change trend similarity between the feature pH window and the bth reference pH window corresponding to the tth monitoring moment, and the time weight factor between the feature pH window and the bth reference pH window corresponding to the tth monitoring moment. The negative correlation mapping result of the pH change trend similarity is the reciprocal of the pH change trend similarity. The time weight factor between the feature pH window and the bth reference pH window corresponding to the tth monitoring moment is the result of normalizing and then negatively correlating the time interval between the time window corresponding to the feature pH window and the time window corresponding to the bth reference pH window corresponding to the tth monitoring moment. The time weight factor between the feature pH window and the bth reference pH window is , the time interval between the time window corresponding to the feature pH window and the time window corresponding to the bth reference pH window corresponding to the tth monitoring moment, the time interval between the time window corresponding to the feature pH window and the time window corresponding to the bth reference pH window corresponding to the tth monitoring moment can be the time interval between the end time in the time window corresponding to the feature pH window and the end time in the time window corresponding to the bth reference pH window, the maximum time interval between the time window corresponding to the feature pH window and the time windows corresponding to all reference pH windows corresponding to the tth monitoring moment, the role is to normalize , the constant 1 is subtracted from in order to negatively correlate the result that the closer to the tth monitoring moment, the more dominant the temperature similarity and the pH change trend similarity are to the initial feature value. All data in the comprehensive feature value set are accumulated, and the accumulated result is taken as the initial feature value corresponding to the tth monitoring moment. The expression of the bth comprehensive feature value in the comprehensive feature value set is , the temperature similarity between the feature temperature window corresponding to the tth monitoring moment and the bth reference temperature window corresponding to the tth monitoring moment, The pH change trend similarity between the feature pH window corresponding to the tth monitoring moment and the bth reference pH window corresponding to the tth monitoring moment.

[0028] Since the closer the pH change trend between the feature pH window and the reference pH window is, the more likely the pH change of the feature pH window is normal change, that is, the more the data in the feature pH window is not data generated by impurity component interference or is not drift abnormal data, and since the closer the reference value of the reference window of the feature pH window or the tth monitoring moment is, the more it can reflect the situation of the pH data of the tth monitoring moment being interfered by impurity components or whether the pH data of the tth monitoring moment is drift abnormal data, therefore, the smaller the time interval between the time window corresponding to the feature pH window and the time window corresponding to the reference pH window is, the stronger the reaction degree of the reference pH window to the authenticity of the feature pH window data is, or in other words, the reference pH window can better reflect whether the pH data of the tth monitoring moment is drift abnormal data. Therefore, when is larger, it indicates that the temperature influence of the feature temperature window and the bth reference temperature window is more consistent, and the reference of the bth reference pH window is greater, or in other words, is more reliable, when is smaller, the reaction degree of the reference pH window to the authenticity of the feature pH window data is stronger, or in other words, the trend similarity between the feature pH window and the reference pH window can better reflect whether the pH data of the tth monitoring moment is drift abnormal data, or in other words, is more reliable, is smaller, it indicates that the drift abnormality performance of the feature pH window is higher, and it indicates that the pH data of the feature pH window or the tth monitoring moment is more likely to be drift abnormal data, that is, when is larger, is smaller, and is smaller, that is, the initial feature value corresponding to the tth monitoring moment is larger, it indicates that the drift abnormality performance of the feature pH window is higher, and it indicates that the pH data of the feature pH window or the tth monitoring moment is more likely to be drift abnormal data.

[0029] But at the beginning time of the drift anomaly, that is, at the initial stage of the fiber, the colloid and the deposited compound adhering to the surface of the pH electrode or blocking the electrode structure, most of the data in the feature data window corresponding to the feature time window at this time is normal data, the drift anomaly performance degree of the feature pH window at this time is relatively low, or the initial feature value calculated at this time is relatively low, so it is difficult to identify the initial drift anomaly only according to the initial feature value, and in order to avoid identifying the initial drift anomaly data as normal data, the subsequent deviation degree of the to-be-analyzed pH data corresponding to the monitoring time from the fitting value on the fitting curve of the feature pH window corresponding to the monitoring time is also needed to obtain the final feature value, that is, the target feature value, which characterizes the drift anomaly of the to-be-analyzed pH data corresponding to the monitoring time, that is, the target feature value corresponding to the tth monitoring time is obtained according to the difference between the fitting value obtained by fitting the feature pH window corresponding to the tth monitoring time and the to-be-analyzed pH data corresponding to the tth monitoring time and the initial feature value corresponding to the tth monitoring time, and the specific obtaining process of the target feature value corresponding to the tth monitoring time is as follows: The straight line obtained by fitting the data in the feature pH window corresponding to the tth monitoring time by using the least square method is denoted as a feature straight line, the value corresponding to the tth monitoring time obtained on the feature straight line is denoted as a fitting pH data corresponding to the tth monitoring time, the absolute value of the difference between the fitting pH data corresponding to the tth monitoring time and the to-be-analyzed pH data corresponding to the tth monitoring time is calculated and denoted as a feature difference value, the ratio of the feature difference value to the fitting pH data corresponding to the tth monitoring time is denoted as a relative deviation degree, and the product of the relative deviation degree and the initial feature value corresponding to the tth monitoring time is calculated and denoted as a target feature value corresponding to the tth monitoring time; and the specific calculation expression of the target feature value corresponding to the tth monitoring time is as follows:

[0030] Wherein, Q is the target feature value corresponding to the tth monitoring time, q is the initial feature value corresponding to the tth monitoring time, is the to-be-analyzed pH data corresponding to the tth monitoring time, is the fitting pH data corresponding to the tth monitoring time; and when q is larger, Q is larger, that is, the drift anomaly performance degree of the feature pH window is higher, and the probability that the feature pH window or the to-be-analyzed pH data corresponding to the tth monitoring time is drift anomaly data is larger, and vice versa, that is, Q is smaller, which indicates that the drift anomaly performance degree of the feature pH window is lower, and the probability that the feature pH window or the to-be-analyzed pH data corresponding to the tth monitoring time is drift anomaly data is smaller.

[0031] Since the target feature value corresponding to normal data or non-drift abnormal data has higher similarity and is lower than that of drift abnormal data, after obtaining the target feature value corresponding to the monitoring time, the DBSCAN algorithm is used to cluster the target feature value corresponding to the tth monitoring time and the target feature values corresponding to all monitoring times before the tth monitoring time, and the clustering result is recorded as the clustering cluster corresponding to the tth monitoring time. The clustering process is known, and the method for obtaining the target feature value corresponding to the tth monitoring time is the same as that for obtaining the target feature value corresponding to the tth monitoring time. Then, according to the clustering cluster corresponding to the tth monitoring time, the drift abnormality of the to-be-analyzed pH data corresponding to the tth monitoring time is identified and corrected, and the target pH data corresponding to the tth monitoring time is obtained. The specific process of identifying and correcting the drift abnormality of the to-be-analyzed pH data corresponding to the tth monitoring time according to the clustering cluster corresponding to the tth monitoring time to obtain the target pH data corresponding to the tth monitoring time is as follows: In the cluster corresponding to the tth monitoring moment, the cluster to which the target characteristic value corresponding to the tth monitoring moment belongs and the cluster with the most data in the cluster are obtained, the cluster with the most data in the cluster is recorded as the maximum cluster, and it is judged whether the cluster to which the target characteristic value corresponding to the tth monitoring moment belongs belongs to the maximum cluster. If it belongs, it is determined that there is no drift anomaly at the tth monitoring moment or that the pH data to be analyzed corresponding to the tth monitoring moment is not drift anomaly data. At this time, the pH data to be analyzed corresponding to the tth monitoring moment does not need to be corrected and can be directly used as the target pH data corresponding to the tth monitoring moment. If it is judged that the cluster to which the target characteristic value corresponding to the tth monitoring moment belongs does not belong to the maximum cluster, it is determined that there is a drift anomaly at the tth monitoring moment or that the pH data to be analyzed corresponding to the tth monitoring moment is drift anomaly data. At this time, the pH data to be analyzed corresponding to the tth monitoring moment needs to be corrected. The correction process is as follows: the pH data corresponding to the tth monitoring moment is predicted according to the target pH data corresponding to the monitoring moment before the tth monitoring moment, the predicted pH data corresponding to the tth monitoring moment is obtained, and the predicted pH data corresponding to the tth monitoring moment is used as the target pH data corresponding to the tth monitoring moment. That is, under the premise that the target pH data corresponding to the monitoring moment before the tth monitoring moment is known, the LSTM algorithm is used to predict the data corresponding to the tth monitoring moment to obtain the predicted pH data corresponding to the tth monitoring moment. The process of data prediction by the LSTM algorithm is known. In addition, as another real-time method, the pH data to be analyzed corresponding to the tth monitoring moment can also be identified for drift anomaly based on a preset anomaly threshold. For example, when the target characteristic value corresponding to the tth monitoring moment is greater than the preset anomaly threshold, it is determined that there is a drift anomaly at the tth monitoring moment or that the pH data to be analyzed corresponding to the tth monitoring moment is drift anomaly data that needs to be corrected. When the target characteristic value corresponding to the tth monitoring moment is not greater than the preset anomaly threshold, it is determined that the pH data to be analyzed corresponding to the tth monitoring moment does not need to be corrected and can be directly used as the target pH data corresponding to the tth monitoring moment. The preset anomaly threshold can be set by the implementer according to the actual situation. In addition, the pH data to be analyzed corresponding to the initial monitoring moment is not identified for drift anomaly and corrected, and the pH data to be analyzed corresponding to the initial monitoring moment can be directly recorded as the target pH data. For example, the pH data to be analyzed corresponding to the monitoring moment within the first 10 seconds can be directly recorded as the target pH data corresponding to the monitoring moment.

[0032] Therefore, the drift anomaly of the pH data to be analyzed corresponding to the tth monitoring moment is identified and corrected through the above process.

[0033] The pH monitoring and adjusting module 03 monitors and adjusts the pH in the container for producing and manufacturing the composite preparation of chicory at the tth monitoring time according to the target pH data corresponding to the tth monitoring time.

[0034] After the identification and correction of the drift anomaly of the pH data to be analyzed corresponding to the tth monitoring time are completed, that is, the target pH data corresponding to the tth monitoring time is obtained, the pH in the container for producing and manufacturing the composite preparation of chicory at the tth monitoring time is monitored and adjusted according to the target pH data corresponding to the tth monitoring time. The pH in the container for producing and manufacturing the composite preparation of chicory at the tth monitoring time is monitored and adjusted, which is to adjust the pH in the raw material mixing process. The process of monitoring and adjusting the pH corresponding to the tth monitoring time based on the pH data is known. For example, when it is determined that the target pH data corresponding to the tth monitoring time exceeds the preset pH interval, it is determined that the pH in the mixing tank needs to be adjusted at this time. When it is determined that the target pH data corresponding to the tth monitoring time does not exceed the preset pH interval, it is determined that the pH in the mixing tank does not need to be adjusted at this time. The process of adjusting the pH in the mixing tank is known, and thus will not be described in detail. The mixing tank is the container for producing and manufacturing the composite preparation of chicory. The preset pH interval is an empirical value.

[0035] At this point, the embodiment completes the monitoring and adjustment of the pH in the production and manufacturing process of the composite preparation of chicory, and can ensure or improve the effect of adjustment.

[0036] To sum up, the embodiment comprises a data acquisition module, configured to acquire pH data to be analyzed, a characteristic temperature window, a characteristic pH window, a reference temperature window and a reference pH window corresponding to a tth monitoring time in the production and manufacturing process of the chicory composite preparation; a drift anomaly identification and correction module, configured to obtain an initial characteristic value corresponding to the tth monitoring time according to the similarity between the characteristic temperature window and the reference temperature window and the difference between the fitting slope of the characteristic pH window and the fitting slope of the reference pH window, obtain a target characteristic value corresponding to the tth monitoring time according to the difference between the fitting value obtained by fitting the characteristic pH window and the pH data to be analyzed and the initial characteristic value, cluster the target characteristic values corresponding to the tth monitoring time and the monitoring time before the tth monitoring time to obtain a cluster corresponding to the tth monitoring time, identify and correct the drift anomaly of the pH data to be analyzed corresponding to the tth monitoring time according to the cluster, and obtain target pH data corresponding to the tth monitoring time; and a pH monitoring and adjustment module, configured to monitor and adjust the pH in the chicory composite preparation production and manufacturing container at the tth monitoring time according to the target pH data corresponding to the tth monitoring time. According to the embodiment, the pH in the chicory composite preparation production and manufacturing container at the tth monitoring time is monitored and adjusted according to the corrected pH data, which can improve or guarantee the adjustment effect, thereby ensuring the production stability of the chicory composite preparation and the production quality of the chicory composite preparation.

[0037] The above-described embodiments are only used to illustrate the technical solutions of the present application, but not limit the same; although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that they can still modify the technical solutions recorded in the foregoing embodiments, or make equivalent replacements to some technical features thereof; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application, and should be included in the protection scope of the present application.

Claims

1. A manufacturing process for a chicory compound formulation, characterized in that, The manufacturing process of the chicory compound preparation includes: The data acquisition module is used to acquire the pH data to be analyzed, the characteristic temperature window, the characteristic pH window, the reference temperature window, and the reference pH window at the t-th monitoring time during the production and manufacturing process of chicory compound formulation. The drift anomaly identification and correction module is used to obtain the initial feature value corresponding to the t-th monitoring time based on the similarity between the feature temperature window and the reference temperature window and the difference between the fitting slope of the feature pH window and the fitting slope of the reference pH window; to obtain the target feature value corresponding to the t-th monitoring time based on the difference between the fitted value obtained by fitting the feature pH window and the pH data to be analyzed and the initial feature value; to cluster the target feature values ​​corresponding to the t-th monitoring time and the monitoring times before the t-th monitoring time to obtain the cluster corresponding to the t-th monitoring time; and to identify and correct the drift anomaly of the pH data to be analyzed corresponding to the t-th monitoring time based on the cluster to obtain the target pH data corresponding to the t-th monitoring time. The pH monitoring and adjustment module is used to monitor and adjust the pH in the chicory compound preparation production container at the t-th monitoring time based on the target pH data corresponding to the t-th monitoring time.

2. The manufacturing process of the chicory compound preparation as described in claim 1, characterized in that, The method for obtaining the characteristic temperature window, characteristic pH window, reference temperature window, and reference pH window includes: Acquire sensor temperature data at various monitoring points during the production and manufacturing process of chicory compound formulations; The time period from the start of production of the chicory compound preparation to the t-th monitoring time is denoted as the production monitoring time period corresponding to the t-th monitoring time. A sliding window is constructed, and the sliding window is slid backward from the end of the production monitoring time period to obtain all time windows corresponding to the t-th monitoring time. Among all time windows, the time window containing the t-th monitoring time is denoted as the characteristic time window corresponding to the t-th monitoring time, and all time windows not containing the t-th monitoring time are denoted as the reference time window corresponding to the t-th monitoring time. The time series window composed of the sensing temperature data corresponding to all monitoring times in the characteristic time window is denoted as the characteristic temperature window corresponding to the t-th monitoring time, and the time series window composed of the sensing temperature data corresponding to all monitoring times in the reference time window is denoted as the t-th monitoring time. A reference temperature window corresponding to each monitoring time is defined as the time window formed by the target pH data corresponding to all monitoring times within the reference time window. The time window formed by the target pH data corresponding to the other monitoring times excluding the t-th monitoring time and the pH data to be analyzed corresponding to the t-th monitoring time is defined as the time window formed by the characteristic pH window. The target pH data corresponding to other monitoring times before the t-th monitoring time are the data after identifying and correcting the drift anomalies of the pH data to be analyzed corresponding to the corresponding monitoring times. The method for identifying and correcting the drift anomalies of the pH data to be analyzed corresponding to other monitoring times before the t-th monitoring time is the same as the method for identifying and correcting the drift anomalies of the pH data to be analyzed corresponding to the t-th monitoring time.

3. The manufacturing process of the chicory compound preparation as described in claim 1, characterized in that, The methods for obtaining the initial feature value corresponding to the t-th monitoring time include: Starting from the initial data in the characteristic temperature window, the data is marked once every preset number of data points, and the marked data in the characteristic temperature window is recorded as the marked data in the characteristic temperature window. Starting from the initial data in the reference temperature window, the data is marked once every preset number of data points, and the marked data in the reference temperature window is recorded as the marked data in the reference temperature window. The temperature similarity between the characteristic temperature window and the reference temperature window is obtained based on the absolute value of the difference between each marked data in the characteristic temperature window and the marked data at the corresponding position in the reference temperature window. The difference between the slope of the straight line obtained by fitting all data in the characteristic pH window and the slope of the straight line obtained by fitting all data in the reference pH window is used to obtain the similarity of pH change trends between the characteristic pH window and the reference pH window. The initial feature value corresponding to the t-th monitoring time is obtained based on the time interval between the feature pH window and the reference pH window, the temperature similarity, and the pH change trend similarity.

4. The manufacturing process of the chicory compound preparation as described in claim 3, characterized in that, The method for obtaining the temperature similarity between the feature temperature window and the reference temperature window includes: The sequence of absolute values ​​of the differences between each labeled data in the feature temperature window and the corresponding labeled data in the reference temperature window is denoted as the temperature difference sequence. The result of negative correlation mapping and normalization of the sum of all data in the temperature difference sequence is denoted as the temperature similarity between the feature temperature window and the reference temperature window.

5. The manufacturing process of the chicory compound preparation as described in claim 3, characterized in that, The similarity of pH change trends between the characteristic pH window and the reference pH window is the result of normalizing the absolute value of the difference between the slope of the straight line obtained by fitting all data in the characteristic pH window and the slope of the straight line obtained by fitting all data in the reference pH window after negative correlation mapping.

6. The manufacturing process of the chicory compound preparation as described in claim 3, characterized in that, A method for obtaining the initial feature value corresponding to the t-th monitoring time based on the time interval between the characteristic pH window and the reference pH window, the temperature similarity, and the pH change trend similarity includes: Obtain the comprehensive characterization value set corresponding to the t-th monitoring time. The b-th comprehensive characterization value in the comprehensive characterization value set is the result of multiplying the temperature similarity between the characteristic temperature window and the b-th reference temperature window corresponding to the t-th monitoring time, the negative correlation mapping result of the pH change trend similarity between the characteristic pH window and the b-th reference pH window corresponding to the t-th monitoring time, and the result of multiplying the time weight factor between the characteristic pH window and the b-th reference pH window. The time weight factor between the characteristic pH window and the b-th reference pH window is the result of normalizing the time interval between the characteristic pH window and the b-th reference pH window and then performing a negative correlation mapping. The summation result of all data in the comprehensive characterization value set is used as the initial feature value corresponding to the t-th monitoring time.

7. The manufacturing process of a chicory compound preparation as described in claim 1, characterized in that, The methods for obtaining the target feature value corresponding to the t-th monitoring time include: The fitted pH data corresponding to the t-th monitoring time is obtained by fitting all the data in the characteristic pH window. The absolute value of the difference between the fitted pH data and the pH data to be analyzed corresponding to the t-th monitoring time is recorded as the characteristic difference. The ratio of the characteristic difference to the fitted pH data is recorded as the relative deviation. The product of the relative deviation and the initial characteristic value is recorded as the target characteristic value corresponding to the t-th monitoring time.

8. The manufacturing process of a chicory compound preparation as described in claim 1, characterized in that, A method for identifying and correcting drift anomalies in the pH data to be analyzed at the t-th monitoring time based on the clusters, and obtaining the target pH data at the t-th monitoring time, includes: Determine whether the cluster to which the target feature value corresponding to the t-th monitoring time belongs belongs to the largest cluster among all clusters corresponding to the t-th monitoring time. If it does, it is determined that there is no drift anomaly at the t-th monitoring time, and the pH data to be analyzed at the t-th monitoring time is directly used as the target pH data at the t-th monitoring time. Otherwise, it is determined that there is a drift anomaly at the t-th monitoring time, and the pH data at the t-th monitoring time is predicted based on the target pH data corresponding to the monitoring times before the t-th monitoring time, and the predicted value is used as the target pH data at the t-th monitoring time.

9. The manufacturing process of a chicory compound preparation as described in claim 1, characterized in that, Methods for acquiring the pH data to be analyzed at any given monitoring time include: For any monitoring moment in the production process of chicory compound preparation, the pH data collected by the sensor at the monitoring moment is normalized and recorded as the pH data to be denoised at the monitoring moment. The pH data to be denoised at the monitoring moment is then denoised to obtain the pH data to be analyzed at the monitoring moment.

10. The manufacturing process of a chicory compound preparation as described in claim 9, characterized in that, A method for denoising the pH data to be analyzed at the monitoring time corresponding to the monitoring time includes: The result of normalizing the absolute value of the difference between the pH data to be denoised at the monitoring time and the target pH data at the previous monitoring time is recorded as the first evaluation value. The pH data at the monitoring time is predicted based on the target pH data at the previous monitoring time to obtain the predicted pH data at the monitoring time. The result of normalizing the absolute value of the difference between the predicted pH data and the pH data to be denoised at the monitoring time is recorded as the second evaluation value. The average of the first and second evaluation values ​​is recorded as the comprehensive evaluation value. It is determined whether the comprehensive evaluation value is greater than a preset noise threshold. If it is greater, the pH data to be denoised at the monitoring time is denoised, and the denoised data is recorded as the pH data to be analyzed at the corresponding monitoring time. If it is not greater, the pH data to be denoised at the monitoring time is directly recorded as the pH data to be analyzed at the corresponding monitoring time.

Images