A data processing method for monitoring sodium chondroitin sulfate production

Abstract

The application discloses a data processing method for sodium chondroitin sulfate production monitoring, relates to the technical field of sodium chondroitin sulfate preparation, and solves the technical problem that existing sodium chondroitin sulfate detection data in the production process are affected by factors in the production process, so that the detection data is not accurate enough. Step one: obtaining raw material data and control data corresponding to the precipitation stage of the present sodium chondroitin sulfate production; step two: inputting the raw material data and the control data into a precipitation amount estimation model to obtain estimated precipitation amounts of sodium chondroitin sulfate at various set nodes; step three: generating an adhesion influence coefficient according to the estimated precipitation amounts; step four: obtaining setting position data of various detection devices, and generating a parameter fluctuation range according to the setting position and the adhesion influence coefficient; and step five: obtaining detection data detected by various detection devices, and generating a detection result according to the detection data, the control data and the parameter fluctuation range, so that the accuracy of the detection result is ensured.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of sodium chondroitin sulfate preparation, and relates to a sodium chondroitin sulfate preparation technology, in particular to a data processing method for sodium chondroitin sulfate production monitoring. BACKGROUND

[0002] Sodium chondroitin sulfate is rich in biological functions and has a wide range of uses in clinical medicine. In the existing preparation process of sodium chondroitin sulfate, pig laryngeal bone, nasal bone, tracheal cartilage and other cartilage tissues are usually used as raw materials, and the product is obtained through alkaline hydrolysis, salt hydrolysis, enzymatic hydrolysis and other extraction processes. Taking pig cartilage as an example, the specific reaction extraction process mainly includes the leaching stage, the enzymatic hydrolysis stage, the adsorption stage and the precipitation stage. In the leaching stage, NaOH is used to stir, dissolve and filter the pig cartilage residue, and hydrochloric acid is added to neutralize the alkalinity to adjust the pH value to weak alkalinity. In the enzymatic hydrolysis stage, trypsin is added for hydrolysis, and the temperature is controlled, and the hydrolysis time is usually about 7 hours, and NaOH is added at any time to adjust the pH value. In the adsorption stage, after hydrolysis is completed, hydrochloric acid is added to make the pH value neutral, and white clay and activated carbon are added for adsorption, stirring and pH adjustment, and after adsorption is completed, the pH value is adjusted to weak acidity and precipitated and filtered. In the precipitation stage, NaOH is added to the filtrate to adjust the pH value to neutral, and sodium chloride solution is added to dissolve and filter to clarify, and after stirring and standing for more than 8 hours, sodium chondroitin sulfate precipitate product appears.

[0003] The existing sodium chondroitin sulfate needs to monitor the temperature and pH value and other data in the production process to judge the reaction progress and reaction state of the corresponding reaction stage. However, the temperature detection device and the pH value detection device in the existing chondroitin sulfate production equipment are usually arranged inside the reaction device and directly contact with the reaction medium. In the production process of sodium chondroitin sulfate, precipitates are generated, which will adhere to the sensor. Especially in the production of a large amount of sodium chondroitin sulfate at one time, the adhesion of the precipitates is more serious, which causes the temperature and pH value detected by the detection device to fluctuate. This not only reduces the accuracy of the temperature and pH value data, but also further affects the accuracy of the reaction progress and state based on these data. Therefore, a data processing method for sodium chondroitin sulfate production monitoring is needed. SUMMARY

[0004] The present application aims to solve at least one of the technical problems in the prior art. To this end, the present application provides a data processing method for sodium chondroitin sulfate production monitoring, which solves the technical problem that the detection data of the existing sodium chondroitin sulfate in the production process is affected by the factors in the production process, so that the detection data is not accurate.

[0005] To achieve the above object, the first aspect of the present application provides a data processing method for monitoring the production of chondroitin sulfate sodium, comprising the following steps:

[0006] Step one: obtaining the raw material data and control data corresponding to the precipitation stage of this chondroitin sulfate sodium production;

[0007] Step two: inputting the raw material data and control data into the precipitation amount estimation model to obtain the estimated precipitation amount of chondroitin sulfate sodium at each set node;

[0008] Step three: generating an adhesion influence coefficient according to the estimated precipitation amount;

[0009] Step four: obtaining the setting position data of each detection device, and generating a parameter fluctuation range according to the setting position and the adhesion influence coefficient;

[0010] Step five: obtaining the detection data detected by each detection device, and generating a detection result according to the detection data, the control data and the parameter fluctuation range.

[0011] Because the precipitation adhesion is more serious in the production of chondroitin sulfate sodium, the temperature and pH value detected by the detection device will appear fluctuation phenomenon, and these temperature fluctuation phenomenon may be caused by the precipitation adhesion on the detection device, or may be caused by abnormal reaction, which needs to be distinguished in time and handled accordingly.

[0012] The present application obtains the raw material data and control data corresponding to the precipitation stage of this chondroitin sulfate sodium production, obtains the estimated precipitation amount of chondroitin sulfate sodium at each node in the precipitation stage according to the raw material data and control data, generates an adhesion influence coefficient according to the estimated precipitation amount, obtains the setting position data of each detection device, generates a position influence coefficient of the corresponding detection device according to the device position data, adjusts the originally set normal fluctuation range to obtain the corresponding parameter fluctuation range according to the adhesion influence coefficient and the position influence coefficient, generates a normal range of the corresponding node using the parameter fluctuation range, and finally evaluates the detected detection value using the normal range. By considering the precipitation influence within the normal fluctuation range of the detection data, the applicability of the normal fluctuation range is stronger, and the problem of reducing the evaluation accuracy of the detection data caused by the precipitation adhesion is solved.

[0013] Preferably, the precipitation amount estimation model is obtained by training an artificial intelligence model, comprising:

[0014] Obtain a plurality of historical data, extract the raw material data, control data and precipitation amount generation curve of the precipitation stage in the historical data, obtain the set node, obtain the corresponding node precipitation amount on the precipitation amount generation curve according to the set node, integrate the raw material data, control data and a plurality of node precipitation amounts in the plurality of historical data into a plurality of groups of training data and verification data;

[0015] training the artificial intelligence model using the training data; testing the artificial intelligence model after training using the test data to obtain an artificial intelligence model with input of raw material data and control data and output of node deposition amounts; outputting each node deposition amount as an estimated deposition amount of a corresponding set node to obtain a deposition amount estimation model; wherein the artificial intelligence model comprises a BP neural network model and an RBF neural network model.

[0016] Preferably, the step of generating the adhesion influence coefficient according to the estimated deposition amount comprises:

[0017] obtaining the estimated deposition amount of each set node and marking it as YCi, i being the number of the set node; the number of the first set node at the beginning of the deposition stage being 1;

[0018] the adhesion influence coefficient NFi corresponding to the set node numbered i is calculated by the formula ; wherein, a is an adjustment coefficient, and KFC is the adherable deposition amount of the detection equipment.

[0019] Preferably, the step four comprises the following steps:

[0020] S41: obtaining the set position data of each sensor; extracting the equipment height and equipment center distance in the set position data;

[0021] S42: generating a position influence coefficient according to the equipment height and the center distance;

[0022] S43: generating the parameter fluctuation range of each detection equipment of each set node according to the position influence coefficient and the adhesion influence coefficient.

[0023] Preferably, the step of generating the position influence coefficient according to the equipment height and the center distance comprises:

[0024] marking the equipment height as SH and the equipment center distance as SZ;

[0025] the position influence coefficient WY of the corresponding detection equipment is calculated by the formula ; wherein, β1 and β2 are proportional coefficients for adjusting the proportional relationship of the equipment height and the equipment center distance on adhesion; GH is the total height of the reaction equipment; and ZZ is the maximum center distance of the reaction equipment.

[0026] Preferably, the step of generating the parameter fluctuation range of each detection equipment of each set node according to the position influence coefficient and the adhesion influence coefficient comprises:

[0027] obtaining the position influence coefficient WY corresponding to each detection equipment and the adhesion influence coefficient NFi of each set node;

[0028] The parameter fluctuation range JZBi of the setting node numbered i is calculated by formula JZB i = SB + (γ1 x WY + γ2 x NF i ) x KTB; wherein, SB is the normal fluctuation range of the original setting of the detection equipment; γ1 and γ2 are proportional coefficients; KTB is the adjustable fluctuation range of the detection equipment.

[0029] Preferably, the generating the detection result according to the detection data, the control data and the parameter fluctuation range comprises:

[0030] extracting the detection value corresponding to each detection equipment in the detection data, extracting the control value corresponding to each detection equipment in the control data, and extracting the parameter fluctuation range of each setting node corresponding to each detection equipment;

[0031] generating the normal range of each setting node of the corresponding detection equipment according to the control value and the parameter fluctuation range corresponding to the same detection equipment;

[0032] obtaining the normal range of each setting node of each detection equipment, and judging whether the detection value of the current setting node of the detection equipment is within the normal range corresponding to the setting node in sequence; if yes, marking the node detection result of the corresponding detection equipment at the setting node as normal; if no, marking the node detection result of the corresponding detection equipment at the setting node as abnormal;

[0033] obtaining the node detection result corresponding to the current setting node and the previous setting node of each detection equipment;

[0034] when the node detection result of the same equipment type of the detection equipment at the current setting node exceeds the setting number and is abnormal, setting the detection value of the corresponding detection equipment as an abnormal value, and setting the detection result of the corresponding detection equipment as abnormal;

[0035] when the number of the node detection result of each setting node corresponding to any detection equipment is abnormal and exceeds the node abnormality number threshold, setting the detection equipment as an abnormal equipment.

[0036] Preferably, the generating the normal range of each setting node of the corresponding detection equipment according to the control value and the parameter fluctuation range corresponding to the same detection equipment comprises:

[0037] obtaining the control value KS of the setting node corresponding to the detection equipment, and the parameter fluctuation range JZBi corresponding thereto

[0038] The lower limit value of the normal range under the set node corresponding to the detection device is calculated by the formula ZXi=KS-JZBi; the upper limit value of the normal range under the set node corresponding to the detection device is calculated by the formula ZSi=KS+JZBi; and the lower limit value of the normal range and the upper limit value of the normal range are integrated into the normal range of the set node corresponding to the detection device.

[0039] Compared with the prior art, the beneficial effects of the present application are:

[0040] 1. The present application obtains the raw material data and control data corresponding to the precipitation stage of this time chondroitin sulfate sodium production; estimates the precipitation amount of chondroitin sulfate sodium at each node of the precipitation stage according to the raw material data and control data; generates an adhesion influence coefficient according to the estimated precipitation amount; and obtains the setting position data of each detection device, generates a position influence coefficient of the corresponding detection device according to the device position data; adjusts the originally set normal fluctuation range to obtain the corresponding parameter fluctuation range according to the adhesion influence coefficient and the position influence coefficient; generates the normal range of the corresponding node using the parameter fluctuation range, and finally evaluates the detected detection value using the normal range; by considering the precipitation influence within the normal fluctuation range of the detection data, the applicability of the normal fluctuation range is stronger, and the problem of reduced detection data evaluation accuracy caused by precipitation adhesion is solved. BRIEF DESCRIPTION OF DRAWINGS

[0041] 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 embodiments or the prior art description will be briefly introduced as follows. 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 labor.

[0042] Figure 1 A data processing method for generating chondroitin sulfate sodium detection of the present application is shown in the figure;

[0043] Figure 2 A parameter fluctuation range generation process diagram of the present application is shown in the figure. DETAILED DESCRIPTION

[0044] The technical solutions of the present application will be described in detail below in conjunction with the embodiments. Obviously, the described embodiments are only some of the embodiments of the present application, not all. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the present application.

[0045] Please refer to Figure 1 The first aspect of the present application provides a data processing method for monitoring chondroitin sulfate sodium production, comprising the following steps:

[0046] Step one: obtaining raw material data and control data corresponding to the precipitation stage of this chondroitin sulfate sodium production, the raw material data including several main raw materials in the precipitation stage and their corresponding planned raw material consumption, such as planned chondroitin consumption; the control data of the precipitation stage including control temperature and control pH value and other data;

[0047] Step two: inputting the raw material data and control data into the precipitation amount estimation model to obtain the estimated precipitation amount of chondroitin sulfate sodium at each set node; the precipitation amount estimation model is a model trained according to historical data for estimating the precipitation amount of each stage in the precipitation process; the estimated precipitation amount is the estimated precipitation amount at the corresponding node under the raw material and control conditions;

[0048] Step three: generating an adhesion influence coefficient according to the estimated precipitation amount; the adhesion influence coefficient is the influence of the precipitation amount on the adhesion of the precipitation particles on the detection equipment, the larger the adhesion influence coefficient, the greater the probability of adhesion of the corresponding detection equipment to the precipitation particles;

[0049] Step four: obtaining the setting position data of each detection equipment, generating a parameter fluctuation range according to the setting position and the adhesion influence coefficient; the parameter fluctuation range is the normal fluctuation of the detection data of each detection equipment at each set node;

[0050] Step five: obtaining the detection data detected by each detection equipment, generating a detection result according to the detection data, the control data and the parameter fluctuation range; the detection equipment includes temperature detection equipment and pH value detection equipment and the like.

[0051] In the one-time large-scale production of chondroitin sulfate sodium, the precipitation adhesion is more serious, which leads to fluctuations in the temperature and pH values detected by the detection equipment. These temperature fluctuations may be caused by the precipitation adhering to the detection equipment or by abnormal reactions, which need to be distinguished and treated in a timely manner. In this embodiment, the raw material data and control data corresponding to the precipitation stage of the current chondroitin sulfate sodium production are obtained; the estimated precipitation amount of chondroitin sulfate sodium at each node in the precipitation stage is determined according to the raw material data and control data; the adhesion influence coefficient is generated according to the estimated precipitation amount; the setting position data of each detection equipment is obtained, and the position influence coefficient of the corresponding detection equipment is generated according to the equipment position data; the original set normal fluctuation range is adjusted to obtain the corresponding parameter fluctuation range according to the adhesion influence coefficient and the position influence coefficient; the normal range of the corresponding node is generated using the parameter fluctuation range, and finally the normal range is used to evaluate the detected detection value; by considering the precipitation influence in the normal fluctuation range of the detection data, the applicability of the normal fluctuation range is stronger, the problem of reduced detection data evaluation accuracy caused by precipitation adhesion is solved, the abnormal test data can be processed in a timely manner, and the subsequent production efficiency of chondroitin sulfate sodium is improved.

[0052] The precipitation amount estimation model is obtained by training an artificial intelligence model, including: obtaining a plurality of historical data, extracting the raw material data, control data and precipitation amount generation curve of the precipitation stage in the historical data; the precipitation amount production curve is a curve of the change of the precipitation amount of chondroitin sulfate sodium with time, which is integrated from the precipitation amount of chondroitin sulfate sodium at each time node in the chondroitin sulfate sodium production process corresponding to the historical data;

[0053] A set node is obtained, and the corresponding node precipitation amount is obtained on the precipitation amount generation curve according to the set node; the raw material data, control data and a plurality of node precipitation amounts in the plurality of historical data are integrated into a plurality of groups of training data and test data;

[0054] The artificial intelligence model is trained using training data; the trained artificial intelligence model is tested using test data; specifically, the raw material data and the control data in the test data are input into the trained artificial intelligence model to obtain the node deposition amount of each set node; it is judged whether the difference between the node deposition amount of each set node recorded in the test data and the corresponding output node deposition amount of each node is within an acceptable range; if yes, it means that the test data passes the test, and the next group of test data is tested; if no, the related parameters of the artificial intelligence are adjusted, and the test is continued using the test data; until a certain proportion of test data passes the test; at this time, the artificial intelligence model with input of raw material data and control data and output of a plurality of node deposition amounts is obtained; each node deposition amount is output as the estimated deposition amount of the corresponding set node to obtain a deposition amount estimation model; wherein the artificial intelligence model includes a BP neural network model and an RBF neural network model.

[0055] It can be understood that the artificial intelligence model needs to be trained in advance, and the set node can be a node set at a fixed time interval, or a node set according to the raw material data and the control data. For example, the more raw materials, the greater the change in the deposition amount in a short period of time during the deposition process; the more suitable the control data is for the optimal data, the greater the change in the deposition amount in a short period of time; at this time, the time interval between the set nodes needs to be set shorter; so that the situation of each set node generated subsequently being affected by the deposition does not change drastically, thereby reducing the probability of false judgment of the test data in the later stage.

[0056] According to the estimated deposition amount, an adhesion influence coefficient is generated, including: obtaining the estimated deposition amount of each set node, and marking it as YCi, i is the number of the set node; the number of the first set node at the beginning of the deposition stage is 1;

[0057] The adhesion influence coefficient NFi corresponding to the set node numbered i is calculated by the formula ; wherein a is an adjustment coefficient, the specific value is set according to experience, and KFC is the attachable deposition amount of the detection equipment.

[0058] The adhesion influence coefficient corresponding to each set node is calculated by the above formula in this embodiment. When corresponding to the generation of deposition in the reaction equipment, the more deposition generated between each adjacent set node, the more deposition attached to the detection equipment; the greater the fluctuation influence on the subsequent detection value; therefore, the corresponding adhesion influence coefficient is set larger in this embodiment.

[0059] Please refer to Figure 2 , step four includes the following steps:

[0060] S41: Obtain the setting position data of each sensor; extract the equipment height and equipment axis distance in the setting position data; the equipment height is the distance from the equipment bottom end corresponding to the detection equipment setting position distance, and the equipment axis distance is the distance from the equipment internal center axis corresponding to the detection equipment setting position distance. The internal center axis can be a stirring shaft;

[0061] S42: Generate a position influence coefficient according to the equipment height and the axis distance; the position influence coefficient is a correlation between the position set by the detection equipment and the deposition adhesion. The greater the position influence coefficient, the greater the possibility of deposition adhesion and the greater the adhesion amount corresponding to the position set by the detection equipment;

[0062] S43: Generate the parameter fluctuation range of each detection equipment of each setting node according to the position influence coefficient and the adhesion influence coefficient.

[0063] The position influence coefficient is generated according to the equipment height and the axis distance, including: the equipment height is marked as SH; and the equipment axis distance is marked as SZ.

[0064] The position influence coefficient WY of the corresponding detection equipment is calculated by the formula wherein β1 and β2 are proportional coefficients for adjusting the proportional relationship of the equipment height and the equipment axis distance on adhesion, and the specific values are set according to experience; GH is the total height of the reaction equipment, that is, the height of the inside of the reaction container; and ZZ is the maximum axis distance of the reaction equipment, that is, the maximum distance from the inside surface of the reaction container to the axis.

[0065] The installation position of the detection equipment affects the deposition adhesion, which is calculated by the above formula. Since the density of the deposition is generally greater than that of the solution, deposition particles are generated at each position in the solution. However, due to the difference in density, the deposition particles move to the low end of the solution, so that the detection equipment located at the low end of the solution is more likely to have deposition adhesion. In addition, the general reaction device is a cylindrical cylinder with a stirring device, and the stirring device is located at the axis position. At this time, the farther the distance from the cylinder inner wall to the axis, the weaker the scouring ability of the solution to the inner wall, and the easier the deposition to adhere. The corresponding position influence coefficient is set to be greater.

[0066] The parameter fluctuation range of each detection equipment of each setting node is generated according to the position influence coefficient and the adhesion influence coefficient, including: obtaining the position influence coefficient WY corresponding to each detection equipment, and the adhesion influence coefficient NF1 of each setting node.

[0067] The parameter fluctuation range of each detection equipment of each setting node is generated according to the position influence coefficient and the adhesion influence coefficient, including: obtaining the position influence coefficient WY corresponding to each detection equipment, and the adhesion influence coefficient NF1 of each setting node. i = SB + (γ1 × WY + γ2 × NF i)×KTB calculation of the number of i set node parameter fluctuation range JZBi; wherein, SB is the original setting of the corresponding detection equipment normal fluctuation range; the normal fluctuation range is the original setting of the fluctuation range without considering the influence of the sediment adhesion on the detection equipment, at this time the data in the fluctuation range is normal data; γ1 and γ2 are the proportion coefficient, which is used to adjust the position influence coefficient and the adhesion influence coefficient to the same scale, and also used to adjust the proportion relationship of the position influence coefficient and the adhesion influence coefficient, which plays the role of adjustment and weight distribution, the specific value is set according to experience; KTB is the adjustable fluctuation range corresponding to the detection equipment; the parameter fluctuation range is the fluctuation range of the data collected by the corresponding detection equipment, and the data in the fluctuation range is normal data.

[0068] If the normal fluctuation range corresponding to the temperature data is 1, and the set temperature is 50 degrees, the temperature in the range of 49 to 51 degrees is normal temperature; if the influence of sediment adhesion is considered, the parameter fluctuation range is 1.32, and the temperature in the range of 48.68 to 51.32 degrees is normal temperature.

[0069] According to the detection data, the control data and the parameter fluctuation range, the detection result is generated, including: extracting the detection value corresponding to each detection equipment in the detection data, extracting the control value corresponding to each detection equipment in the control data, and the parameter fluctuation range of each set node corresponding to each detection equipment;

[0070] According to the control value and the parameter fluctuation range corresponding to the same detection equipment, the normal range of the corresponding detection equipment at each set node is generated;

[0071] The normal range of each set node of each detection equipment is obtained, and whether the detection value of the current set node of the detection equipment is in the normal range corresponding to the set node is judged in sequence; if yes, the node detection result of the corresponding detection equipment at the set node is marked as normal; if no, the node detection result of the corresponding detection equipment at the set node is marked as abnormal;

[0072] The node detection result corresponding to the current set node and the previous set node of each detection equipment is obtained;

[0073] When the node detection result of more than a set number of detection equipment of the same device type at the current set node is abnormal, the detection value of the corresponding detection equipment is set to an abnormal value, and the detection result of the corresponding detection equipment is set to abnormal; for example, the control value of pH value at a certain set node is 7, and the corresponding normal range is 6.6 to 7.4; at this time, 6 of the detection results corresponding to 10 pH value detection equipment on the reaction equipment have detection values not in the normal range, and the set number is 3, so the detection result corresponding to the pH value detection of the corresponding node is abnormal.

[0074] When the number of abnormal node detection results of any detection device corresponding to each set node exceeds the node abnormality threshold, the abnormality threshold is set according to experience; at this time, the detection device is set as an abnormal device; when only an individual device continuously appears detection data fluctuation abnormality, it indicates that the detection device may have a problem; in the embodiment, when the detection data fluctuation abnormality of part of the detection devices is caused by their own reasons, the detection data of the abnormal device is not considered in the subsequent detection data processing process, so as to reduce the calculation amount and improve the efficiency of data processing.

[0075] According to the control value corresponding to the same detection device and the parameter fluctuation range, the normal range of the corresponding detection device at each set node is generated, including:

[0076] The control value of the detection device corresponding to the set node is marked as KS, and the corresponding parameter fluctuation range is JZBi

[0077] The lower limit value of the normal range of the detection device corresponding to the set node is calculated by the formula ZXi=KS-JZBi; the upper limit value of the normal range of the detection device corresponding to the set node is calculated by the formula ZSi=KS+JZBi; the lower limit and the upper limit of the normal range are integrated into the normal range of the detection device corresponding to the set node.

[0078] Part of the data in the above formula is removed dimension and its numerical value is calculated, and the formula is obtained by software simulation of a large amount of collected data to obtain a formula closest to the real situation; the preset parameters and the preset threshold in the formula are set by the person skilled in the art according to the actual situation or obtained by a large amount of data simulation.

[0079] The working principle of the present application is:

[0080] The present application obtains the raw material data and control data corresponding to the precipitation stage of the present sodium chondroitin sulfate production; according to the raw material data and control data, the estimated precipitation amount of sodium chondroitin sulfate at each node of the precipitation stage is obtained; according to the estimated precipitation amount, an adhesion influence coefficient is generated; and the setting position data of each detection device is obtained, and the position influence coefficient of the corresponding detection device is generated according to the device position data; according to the adhesion influence coefficient and the position influence coefficient, the originally set normal fluctuation range is adjusted to obtain the corresponding parameter fluctuation range; the normal range of the corresponding node is generated by using the parameter fluctuation range, and finally the detection value obtained by detection is evaluated by using the normal range; by considering the precipitation influence in the normal fluctuation range of the detection data, the applicability of the normal fluctuation range is stronger, and the problem of reducing the evaluation accuracy of the detection data caused by the precipitation adhesion is solved.

[0081] The above examples are used to illustrate the technical method of the present application and are not limiting. Although the present application has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical method of the present application can be modified or equivalently replaced without departing from the spirit and scope of the technical method of the present application.

Claims

1. A data processing method for sodium chondroitin sulfate production monitoring, characterized by, The method comprises the following steps: Step one: obtaining the raw material data and control data corresponding to the precipitation stage of the current sodium chondroitin sulfate production; Step two: inputting the raw material data and control data into the precipitation amount estimation model to obtain the estimated precipitation amount of sodium chondroitin sulfate at each set node; Step three: generating an adhesion influence coefficient according to the estimated precipitation amount; Step four: obtaining the setting position data of each detection device, and generating a parameter fluctuation range according to the setting position and the adhesion influence coefficient; comprising the following steps: S41: obtaining the setting position data of each sensor; extracting the device height and device axis distance in the setting position data; S42: generating a position influence coefficient according to the device height and axis distance; S43: generating the parameter fluctuation range of each detection device at each set node according to the position influence coefficient and the adhesion influence coefficient; Step five: obtaining the detection data detected by each detection device, and generating a detection result according to the detection data, the control data and the parameter fluctuation range, comprising: extracting the detection values of each detection device in the detection data, extracting the control values of each detection device in the control data, and extracting the parameter fluctuation range of each detection device at each set node; generating the normal range of the same detection device at each set node according to the control value and the parameter fluctuation range of the same detection device; comprising: obtaining the control value KS of the detection device at the set node, and the corresponding parameter fluctuation range JZBi calculating the lower limit value of the normal range of the detection device at the set node by the formula ZXi=KS-JZBi; calculating the upper limit value of the normal range of the detection device at the set node by the formula ZSi=KS+JZBi; integrating the lower limit and upper limit of the normal range into the normal range of the detection device at the set node; obtaining the normal range of each detection device at each set node, and sequentially judging whether the detection value of the detection device at the current set node is within the normal range corresponding to the set node; if yes, marking the node detection result of the corresponding detection device at the set node as normal; if no, marking the node detection result of the corresponding detection device at the set node as abnormal; obtaining the node detection result corresponding to the current set node and the previous set node of each detection device; when the node detection result of more than a set number of detection devices of the same device type at the current set node is abnormal, setting the detection value of the corresponding detection device as an abnormal value, and setting the detection result of the corresponding detection device as abnormal; when the number of node detection results of any detection device at each set node is more than a node abnormality threshold, the detection device is set as an abnormal device.

2. A data processing method for monitoring the production of sodium chondroitin sulfate according to claim 1, characterized by, The precipitation amount estimation model is obtained by training an artificial intelligence model, comprising: Obtain a plurality of historical data, extract raw material data, control data and sedimentation amount generation curve of the sedimentation stage in the historical data; obtain a set node, and obtain a corresponding node sedimentation amount on the sedimentation amount generation curve according to the set node; integrate the raw material data, the control data and the plurality of node sedimentation amounts in the plurality of historical data into a plurality of groups of training data and test data; Train the artificial intelligence model using the training data; test the artificial intelligence model after training using the test data; obtain an artificial intelligence model with input of raw material data and control data and output of a plurality of node sedimentation amounts; output each node sedimentation amount as an estimated sedimentation amount of the corresponding set node to obtain a sedimentation amount estimation model; wherein the artificial intelligence model comprises a BP neural network model and an RBF neural network model.

3. A data processing method for monitoring the production of sodium chondroitin sulfate according to claim 1, characterized by, The adhesion influence coefficient is generated according to the estimated sedimentation amount, and comprises a pH value Obtain the estimated sedimentation amount of each set node and mark it as YCi, i is the number of the set node; the number of the first set node at the beginning of the sedimentation stage is 1; The adhesion influence coefficient NFi corresponding to the setting node numbered i is calculated by the formula ; wherein, a is an adjustment coefficient, and KFC is the attachable deposit amount of the detection device.

4. A data processing method for monitoring the production of sodium chondroitin sulfate according to claim 1, characterized by, The position influence coefficient is generated according to the equipment height and the shaft center distance, and comprises: Mark the equipment height as SH; and mark the equipment shaft center distance as SZ; The position influence coefficient WY corresponding to the detection device is calculated by the formula ; wherein, β1 and β2 are proportional coefficients; GH is the total height of the reaction device; and ZZ is the maximum axial distance of the reaction device.

5. A data processing method for monitoring the production of sodium chondroitin sulfate according to claim 1, characterized by, The parameter fluctuation range of each detection equipment of each set node is generated according to the position influence coefficient and the adhesion influence coefficient, and comprises: Obtain the position influence coefficient WY corresponding to each detection equipment, and the adhesion influence coefficient NFi of each set node; The parameter fluctuation range JZBi of the setting node numbered i is calculated by the formula ; wherein SB is the normal fluctuation range of the original setting of the detection device; γ1 and γ2 are proportional coefficients; and KTB is the adjustable fluctuation range of the detection device.

Images