Color titration recognition method, system, and device based on combined prediction model

Abstract

The present application relates to a kind of color titration identification method, system and equipment based on combination prediction model, belong to solution titration identification technical field, its method includes: using hue measured value trains ARIMA model;Judge the difference between the hue measured value after each titration and the hue predicted value of ARIMA model whether greater than first threshold value;If yes, then determine the current titration point as color mutation inflection point, and at least three consecutive hue measured values after color mutation inflection point are used to train linear regression model;Based on the linear regression model after training, iteratively execute three times hue value prediction and titration operation, until the average difference between the predicted value and the measured value for three times is less than second threshold value, determine the first drop point in the first three drop points as color smooth inflection point.The present application uses different models to identify color in different reaction stages, which can adapt to the demand of multi-color titration formula, and improve the accuracy and anti-interference ability of color titration.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of titration reaction, and in particular to a color titration identification method, system and device based on a combined prediction model. BACKGROUND

[0002] Color titration is one of the most widely used analysis methods in chemical analysis. By analyzing the color change of a reaction solution during titration, the chemical composition in the reaction solution can be determined, which is suitable for determining chemical compositions in the fields of petroleum, food, medicine, environment, etc.

[0003] Currently, traditional manual color titration is to add water and an indicator to an unknown concentration solution, and then titrate with a known concentration solution. During titration, when the titration reagent completely reacts with the measured substance, the color of the solution will change obviously, and the color will not recover within 30 seconds, which indicates that the titration end point has been reached. This traditional titration is based on visual judgment by the human eye of the operator, and different personnel have different sensitivities to color judgment, which can cause deviations. This repetitive titration work consumes the operator's energy, and repeated titration operations can cause uncertainty in the experimental results, and the related corrosive titration liquid can also cause damage to the human body, increasing the risk of the titration process.

[0004] In existing automatic color titration identification, the automatic titration equipment does not have universality, and the titration formula or equipment parameters need to be modified or adjusted every time chemical composition analysis is performed. After the titration formula is modified or changed, the equipment usually cannot adapt, and the operator needs to set the related parameters again. At the same time, the titration process conditions are relatively harsh, and the color recognition accuracy is mainly affected by the sensor, which increases the maintenance cost, use cost and manufacturing cost of the automatic titration equipment. SUMMARY

[0005] (I) Technical problems to be solved

[0006] In view of the above-mentioned shortcomings and deficiencies of the prior art, the present application provides a color titration identification method, system and device based on a combined prediction model, which solves the technical problems of poor universality, low anti-interference ability and high operation and maintenance cost of existing automatic color titration identification.

[0007] (II) Technical solutions

[0008] In order to achieve the above-mentioned purposes, the main technical solutions adopted by the present application include:

[0009] In a first aspect, the present application provides a color titration identification method based on a combined prediction model, comprising:

[0010] The color phase measured value of the solution after each titration reagent is added in the titration reaction stage is collected, and when the number of additions reaches the target sample size, the ARIMA model is trained using the obtained color phase measured value;

[0011] The color phase measured value of the solution after each titration reagent is added in the titration reaction stage is collected, and when the number of additions reaches the target sample size, the ARIMA model is trained using the obtained color phase measured value;

[0012] When the comparison difference is greater than the first threshold value, it is determined that the current titration point is a color mutation inflection point, and at least three consecutive color phase measured values after the color mutation inflection point are collected as a dynamic training set to train the preset linear regression model;

[0013] Based on the trained linear regression model, the color phase value prediction and titration reagent addition operation are iteratively performed three times, and when the average difference between the color phase prediction value and the color phase measured value for three consecutive times is less than the second threshold value, the first addition point in the three consecutive addition points is determined as a color stable inflection point.

[0014] Optionally, before the color phase measured value of the solution after each titration reagent is added in the titration reaction stage is collected, and when the number of additions reaches the target sample size, the ARIMA model is trained using the obtained color phase measured value, it further includes:

[0015] In response to the titration operation instruction, the titration system is initialized to include pipeline rinsing, background light source adjustment, and titration recognition positioning;

[0016] The initialized titration system controls the image frame of the solution to be titrated to be collected, and the image frame is converted into an HSV color model;

[0017] After the HSV color model is subjected to Gaussian blur denoising processing, the color phase average value of the pixel points in the region to which the solution to be titrated belongs in the denoised HSV color model is determined as the original color phase value.

[0018] Optionally, before the color phase measured value of the solution after each titration reagent is added in the titration reaction stage is collected, and when the number of additions reaches the target sample size, the ARIMA model is trained using the obtained color phase measured value, it further includes:

[0019] In the titration reaction stage, the image frame after each titration reagent is added is obtained, and when the number of additions reaches the target sample size, the obtained image frame is converted into an HSV color model;

[0020] After all the HSV color models are subjected to Gaussian blur denoising processing, the color phase average value of the pixel points in the region to which the solution to be titrated belongs in the denoised HSV color model is determined as the color phase measured value;

[0021] determining whether the difference between the measured color phase value and the original color phase value of the solution to be titrated is greater than a set color difference threshold value;

[0022] if the difference between the measured color phase value and the original color phase value of the solution to be titrated is greater than a set color difference threshold value, outputting prompt information that the solution to be titrated is contaminated;

[0023] if the difference between the measured color phase value and the original color phase value of the solution to be titrated is not greater than a set color difference threshold value, training a preset ARIMA model using the original color phase value and all measured color phase values.

[0024] Optionally, comparing the color phase measured value after each time of adding the titration reagent in the cyclic titration process with the color phase predicted value obtained by the trained ARIMA model to determine whether the comparison difference is greater than a set first threshold value comprises:

[0025] in the cyclic titration process, obtaining the color phase predicted value of the solution after the first time of adding the titration reagent by the trained ARIMA model;

[0026] comparing the color phase predicted value with the corresponding color phase measured value to determine whether the comparison difference is greater than a set first threshold value;

[0027] if the comparison difference is greater than the set first threshold value, determining that the current titration point is a color mutation inflection point;

[0028] if the comparison difference is not greater than the set first threshold value, supplementing the color phase value of the solution at the current titration point to the training set of the ARIMA model, and iteratively training the ARIMA model by using the updated training set, so that the iteratively trained ARIMA model predicts the color phase predicted value after the next time of adding the titration reagent.

[0029] Optionally, if the comparison difference is greater than the set first threshold value, determining that the current titration point is a color mutation inflection point, and collecting at least three consecutive color phase measured values after the color mutation inflection point as a dynamic training set to train a preset linear regression model comprises:

[0030] in the case that the comparison difference between the color phase predicted value and the color phase measured value is greater than the set first threshold value, determining the current titration point as a color mutation inflection point, and recording the number of times of cyclic titration to reach the color mutation inflection point;

[0031] determining whether the difference amplitude of at least three consecutive color phase measured values after the color mutation inflection point exceeds a set amplitude threshold value;

[0032] if the difference amplitude of at least three consecutive color phase measured values after the color mutation inflection point does not exceed the set amplitude threshold value, outputting a control instruction that the titration is ended;

[0033] When the difference between the measured hue values of at least three consecutive hues after the color mutation inflection point exceeds the set amplitude threshold, the measured hue values after the color mutation inflection point are taken as a dynamic training set to train the preset linear regression model.

[0034] Optionally, based on the trained linear regression model, the color hue value prediction and titration reagent addition operation are iteratively performed three times, and when the average difference between the color hue prediction values and the color hue measured values for three consecutive times is less than the set second threshold, determining that the first titration point in the three consecutive titration points is the color stable inflection point comprises:

[0035] predicting the color hue prediction values of the solution to be titrated after three consecutive titration operations through the trained linear regression model, and synchronously acquiring the color hue measured values of the solution to be titrated;

[0036] determining whether the average difference between the color hue prediction values and the color hue measured values for three consecutive times is less than the set second threshold;

[0037] When the average difference between the color hue prediction values and the color hue measured values for three consecutive times is less than the set second threshold, it is determined that the first titration point in the three consecutive titration points is the color stable inflection point, and the number of cyclic titrations to reach the color stable inflection point and the control instruction to output the titration end are recorded;

[0038] When the average difference between the color hue prediction values and the color hue measured values for three consecutive times is not less than the set second threshold, the three consecutive color hue measured values are replaced with the earliest three consecutive color hue measured values in the dynamic training set, and the linear regression model is iteratively trained through the updated training set, so that the linear regression model after iterative training predicts the color hue prediction values after the next three titration titration reagents.

[0039] In a second aspect, the embodiments of the present application provide a color titration recognition system based on a combined prediction model, comprising:

[0040] An ARIMA model training module is configured to acquire the color hue measured values of the solution after each titration reagent addition in the titration reaction stage, and train the preset ARIMA model using the acquired color hue measured values when the number of titration reaches the target sample size.

[0041] A first titration color judgment module is configured to compare the color hue measured values after each titration reagent addition in the cyclic titration process with the color hue prediction values predicted by the trained ARIMA model, and determine whether the comparison difference is greater than the set first threshold.

[0042] A color mutation inflection point determination module is configured to determine the current titration point as the color mutation inflection point when the comparison difference is greater than the set first threshold, and train the preset linear regression model using at least three consecutive color hue measured values after the color mutation inflection point as a dynamic training set.

[0043] The second titration color judgment module is configured to iteratively perform three times of hue value prediction and titration reagent dropping operation based on the trained linear regression model, and determine whether the average difference between the three consecutive hue prediction values and the measured hue value is less than a second threshold value.

[0044] The color plateau inflection point determination module is configured to determine that the first dropping point in the three consecutive dropping points is a color plateau inflection point when the average difference between the three consecutive hue prediction values and the measured hue value is less than the second threshold value.

[0045] In a third aspect, an embodiment of the present application provides a titration device, comprising:

[0046] The quantitative liquid adding device;

[0047] The industrial camera;

[0048] The motion assembly is connected with the industrial camera, and is configured to drive the industrial camera to a designated position so that the industrial camera collects an image frame of the solution to be titrated;

[0049] The controller is connected with the quantitative liquid adding device, the industrial camera and the motion assembly respectively, and is configured to execute the color titration recognition method based on the combined prediction model based on the collected image frame.

[0050] Optionally, the quantitative liquid adding device comprises a peristaltic pump, a plunger pump and a titration solution warehouse.

[0051] The peristaltic pump is connected with the titration solution warehouse pipeline, and is configured to add titration reagent to the titration solution warehouse.

[0052] The plunger pump is connected with the titration solution warehouse pipeline and electrically connected with the color titration recognition device, and is configured to control the dropping of the corresponding amount of titration reagent into the solution to be titrated according to the control instruction output by the color titration recognition device.

[0053] Optionally, the titration device further comprises a background light source and a light shield.

[0054] The background light source is connected with the controller, and is configured to dynamically adjust the background brightness of the color recognition area according to the original hue value of the solution to be titrated output by the controller.

[0055] (Three) beneficial effects

[0056] The beneficial effects of the present application are: the present application can sensitively perceive and capture the tiny color mutation inflection point by collecting hue value in real time and dynamically training ARIMA model to predict the trend, which avoids the lag and misjudgment risk of traditional subjective visual inspection or fixed threshold method. After detecting the mutation, the measured values near the mutation point are further used to dynamically train the linear regression model, and the prediction and titration verification are iteratively performed, which can quickly and accurately locate the color stable inflection point followed. The present application realizes high-precision closed-loop control of titration endpoint determination through the double-model cooperative detection and iterative verification mechanism, greatly reduces the excessive consumption of reagents and human judgment errors, and is especially suitable for complex titration systems with slow color change or insignificant mutation, thereby improving the automation level and reliability of color titration analysis. BRIEF DESCRIPTION OF DRAWINGS

[0057] Figure 1 A flowchart of a color titration recognition method based on a combined prediction model is provided for an embodiment of the present application.

[0058] Figure 2 A flowchart of the preparation stage before titration is provided for an embodiment of the present application.

[0059] Figure 3 A flowchart of the learning sample collection stage is provided for an embodiment of the present application.

[0060] Figure 4 A flowchart of the solution color mutation inflection point recognition stage is provided for an embodiment of the present application.

[0061] Figure 5 A prediction and measured curve diagram of the solution color mutation inflection point is provided for an embodiment of the present application.

[0062] Figure 6 A flowchart of the solution color smooth inflection point recognition stage is provided for an embodiment of the present application.

[0063] Figure 7 A prediction and measured curve diagram of the solution color smooth inflection point is provided for an embodiment of the present application. DETAILED DESCRIPTION

[0064] In order to better explain the present application, so as to be understood, the present application is described in detail by specific embodiments in combination with the drawings.

[0065] REFERENCES Figures 1 to 7As shown, the color titration recognition method based on the combined prediction model comprises the following steps: collecting the hue measured value of the solution after each time of adding the titration reagent in the titration reaction stage, and training the preset ARIMA model (autoregressive moving average model) by using the acquired hue measured value when the number of adding reaches the target sample amount; comparing the hue measured value after each time of adding the titration reagent in the cyclic titration process with the hue predicted value obtained by the trained ARIMA model, and judging whether the comparison difference is greater than the set first threshold value; when the comparison difference is greater than the set first threshold value, determining that the current titration point is the color mutation inflection point, and collecting at least three continuous hue measured values after the color mutation inflection point as a dynamic training set to train the preset linear regression model; based on the trained linear regression model, iteratively performing the hue value prediction and the titration reagent adding operation for three times, and when the average difference between the hue predicted value and the hue measured value for three times is less than the set second threshold value, determining that the first adding point in the three continuous adding points is the color stable inflection point.

[0066] The embodiment can sensitively perceive and capture the tiny color mutation inflection point by collecting the hue value in real time and dynamically training the ARIMA model to predict the trend, and the lag and misjudgment risk of the traditional subjective visual inspection or fixed threshold method are avoided. After detecting the mutation, the measured value near the mutation point is further used to dynamically train the linear regression model, and the prediction and titration verification are iteratively performed, which can quickly and accurately locate the color stable inflection point following the color mutation inflection point. The embodiment realizes the high-precision closed-loop control of the titration end point determination through the double-model cooperative detection and iterative verification mechanism, greatly reduces the excessive consumption of reagents and the human judgment error, and is especially suitable for the complex titration system with slow color change or insignificant mutation, and improves the automation level and reliability of the color titration analysis.

[0067] In order to better understand the above technical solutions, the exemplary embodiments of the present application will be described in more detail below with reference to the accompanying drawings. Although the exemplary embodiments of the present application are shown in the drawings, it should be understood that the present application can be implemented in various forms and should not be limited by the embodiments described herein. On the contrary, these embodiments are provided so that the present application can be more clearly, thoroughly understood, and the scope of the present application can be completely conveyed to those skilled in the art.

[0068] Specifically, referring to Figure 1 As shown, the color titration recognition method based on the combined prediction model comprises the following steps:

[0069] S100, collecting the hue measured value of the solution after each time of adding the titration reagent in the titration reaction stage, and training the preset ARIMA model by using the acquired hue measured value when the number of adding reaches the target sample amount.

[0070] In this embodiment, before step S100, sub-steps G100 to G300 are further included:

[0071] G100, in response to the titration operation instruction, initializing the titration system including pipeline rinsing, background light source adjustment, and titration recognition positioning.

[0072] Reference Figure 2 As shown in the titration preparation stage, in response to the titration operation instruction, the titration system is initialized to ensure that the whole reaches the standard environment before titration. First, the pipeline of the titration system is rinsed to make the pipeline full of titration reagent liquid, ensuring that one drop of liquid can be added each time to ensure the accuracy of the liquid added later; then, based on the original hue value of the solution to be titrated, adjust the background light source; then, adjust the industrial camera so that the shooting angle of the industrial camera can capture the recognition area of the solution to be titrated; finally, detect whether the recognition area is all the solution to be titrated, if yes, determine that the titration operation can be executed, if not, the solution to be titrated needs to be added until the recognition area is all the solution to be titrated.

[0073] G200, control the initialized titration system to collect image frames of the solution to be titrated, and convert the image frames into an HSV color model.

[0074] G300, after Gaussian blur denoising processing of the HSV color model, the average hue value of the pixel points in the region to which the solution to be titrated belongs in the denoised HSV color model is determined as the original hue value.

[0075] In this embodiment, reference Figure 3 As shown in the titration preparation stage, in response to the titration operation instruction, the titration system is initialized to ensure that the whole reaches the standard environment before titration. First, the pipeline of the titration system is rinsed to make the pipeline full of titration reagent liquid, ensuring that one drop of liquid can be added each time to ensure the accuracy of the liquid added later; then, based on the original hue value of the solution to be titrated, adjust the background light source; then, adjust the industrial camera so that the shooting angle of the industrial camera can capture the recognition area of the solution to be titrated; finally, detect whether the recognition area is all the solution to be titrated, if yes, determine that the titration operation can be executed, if not, the solution to be titrated needs to be added until the recognition area is all the solution to be titrated.

[0076] S110, in the titration reaction stage, acquire image frames after adding titration reagent each time, and convert the acquired image frames into an HSV color model when the number of additions reaches the target sample amount.

[0077] After adding titration reagent each time, the image frame after color stabilization is acquired by the industrial camera, and then the image frame is converted into an HSV color model. In the HSV color model, H represents hue, S represents saturation, and V represents value.

[0078] S120, after Gaussian blur denoising processing of all HSV color models, the average hue value of the pixel points in the region to which the solution to be titrated belongs in the denoised HSV color model is determined as the measured hue value.

[0079] Gaussian blur can be performed on the image to retain color details of the image while reducing noise, and after taking an average of hue values of pixels in the identified region, the average is taken as a training sample, and a target sample amount of training samples is obtained by traversing all HSV color models to perform de-noising and hue value calculation.

[0080] In S130, it is determined whether the difference between the measured hue value and the original hue value of the titration solution to be titrated is greater than a set color difference threshold.

[0081] In S140a, if the difference between the measured hue value and the original hue value of the titration solution to be titrated is greater than the set color difference threshold, a prompt message indicating that the titration solution to be titrated is contaminated is output.

[0082] In S140b, if the difference between the measured hue value and the original hue value of the titration solution to be titrated is not greater than the set color difference threshold, the original hue value and all measured hue values are used to train a preset ARIMA model.

[0083] In this embodiment, the composition of the titration solution before titration is analyzed to determine whether the titration solution is contaminated, so as to ensure the purity and stability of the titration solution. By comparing the color difference between the solution without adding the titration reagent and the solution with the titration reagent but without color transient, the contamination of the titration solution can be quickly determined, and in the case of no contamination, the hue value of the solution with the titration reagent but without color transient can be collected as the training data of the ARIMA model.

[0084] In S200, the measured hue value after adding the titration reagent in each cycle of the titration process is compared with the predicted hue value obtained by the trained ARIMA model, and it is determined whether the comparison difference is greater than a set first threshold.

[0085] In this embodiment, as shown in FIG. 2, step S200 can further include sub-steps S210 to S230: Figure 4

[0086] In S210, the predicted hue value of the solution after adding the titration reagent for the first time in the cycle of the titration process is obtained by the trained ARIMA model.

[0087] In S220, the predicted hue value is compared with the corresponding measured hue value, and it is determined whether the comparison difference is greater than the set first threshold.

[0088] In S230a, when the comparison difference is greater than the set first threshold, it is determined that the current titration point is a color mutation inflection point.

[0089] ​S230b: When the difference is not greater than the set first threshold, the solution hue value at the current titration point is added to the training set of the ARIMA model. The ARIMA model is then iteratively trained using the updated training set so that the iteratively trained ARIMA model can predict the hue value after the next titration reagent is added.

[0090] In this embodiment, during the cyclic titration process, hue recognition and ARIMA model prediction are used to determine the color abrupt change points of the solution. First, based on n training samples from the ARIMA model's training phase, the ARIMA model is trained to predict the hue value after the (n+1)th titration. Next, after the (n+1)th addition of titrant, the actual hue value is obtained by processing the image using an industrial camera. If the difference between the predicted and actual values ​​is not greater than a set first threshold, it indicates that no abrupt change in the titration color has occurred, and the hue value continues to exhibit stable linearity. The actual hue value after the (n+1)th titration is then added to the original training set, and the first training sample is removed to maintain the sample size at n. Then, the ARIMA model is iteratively trained based on the updated training samples to predict the hue value after the (n+2)th titration reagent is added, until the difference between the predicted and actual values ​​exceeds the set first threshold. Finally, when the difference is greater than the first threshold, due to the previously stable linear training set, the prediction value based on the ARIMA model will continue to predict a stable linear hue value. However, the actual solution color change will cause a large deviation between the actual hue value and the predicted value, which verifies that the solution to be titrated has undergone a color change at the current moment, and thus determines the current liquid addition volume as the solution titration color change point.

[0091] In one specific embodiment, if the instant of the first color change is the inflection point, marking the point of complete reaction, the concentration of the unknown stock solution can be determined accordingly. Figure 5 As shown, the ARIMA model is initially trained using n training samples in the initial stage of titration to predict the hue value at the (n+1)th point. After the (n+1)th addition of titration reagent, the actual hue value is obtained by processing the image using an industrial camera, and the difference between the predicted and actual values ​​is calculated in real time. If hue value fluctuations occur during titration but do not change continuously, the difference will not exceed the set threshold, and it will not be identified as a color abrupt change point. When the solution color undergoes a sudden change, the real-time hue value will change accordingly, and the measured hue value curve will show an upward or downward trend. At the same time, since the training value before the abrupt change is still a smooth curve, the predicted value will continue to produce a smooth value prediction during the abrupt change, resulting in a difference greater than the set threshold. The current detection point is considered the instantaneous point of color abrupt change, i.e., the required titration inflection point (inflection point 1). The concentration of the unknown stock solution can be calculated by using the current titration volume, the concentration of the blank sample solution, the concentration of the titration reagent, and the volume of the unknown concentration stock solution.

[0092] S300. When the difference is greater than the set first threshold, the current titration point is determined as the color change inflection point, and at least three consecutive measured hue values ​​after the color change inflection point are collected as a dynamic training set to train the preset linear regression model.

[0093] In this embodiment, step S300 may further include sub-steps S310 to S330:

[0094] S310. If the difference between the predicted hue value and the measured hue value is greater than the set first threshold, the current titration point is determined as the color change inflection point, and the number of cyclic titrations to reach the color change inflection point is recorded.

[0095] S320. Determine whether the difference between at least three consecutive measured hue values ​​after the color abrupt change inflection point exceeds the set amplitude threshold.

[0096] S330a: If the difference between at least three consecutive measured hue values ​​after the color abrupt change inflection point does not exceed the set amplitude threshold, then output a control command to end the titration.

[0097] S330b: When the difference between at least three consecutive measured hue values ​​after the color abrupt change inflection point exceeds a set amplitude threshold, the measured hue values ​​obtained after the color abrupt change inflection point will be used as a dynamic training set to train the preset linear regression model.

[0098] In this embodiment, during the titration reaction, some solutions continue to react after reaching the color inflection point, with further addition of titrants. The color change becomes more pronounced with each titration. During this process, the solution color varies to different degrees depending on the formulation, resulting in fluctuations in the hue value. Eventually, the hue value smooths out, and this smoothing point is defined as the titration endpoint. Therefore, this embodiment uses at least three consecutive measured hue values ​​after the color inflection point as a training set to train the linear regression model. The trained linear regression model then predicts the hue value. As long as the difference between the predicted and measured values ​​is not less than a second threshold, the prediction and titration process is iteratively executed. Specifically, before reaching the titration smoothing point, the measured hue value after each titration is added to the original training set, replacing the earliest measured hue value, forming a dynamic training set. This ensures that the linear regression model is trained before each prediction.

[0099] S400: Based on the trained linear regression model, iteratively execute three hue value prediction and titration reagent addition operations until the average difference between the three consecutive hue prediction values ​​and the measured hue values ​​is less than the set second threshold. Then, determine the first of the three consecutive addition points as the color stability inflection point.

[0100] In this embodiment, referring to Figure 6 As shown in the figure, step S400 can further include sub-steps S410-S430:

[0101] S410, predicting the hue predicted value of the solution to be titrated after three consecutive titration operations by the trained linear regression model, and synchronously acquiring the hue measured value of the solution to be titrated.

[0102] S420, judging whether the average difference between the hue predicted value and the hue measured value of the three consecutive times is less than the set second threshold value.

[0103] S430a, when the average difference between the hue predicted value and the hue measured value of the three consecutive times is less than the set second threshold value, it is determined that the first drop point of the three consecutive drop points is the color stable inflection point, and the cycle titration number reaching the color stable inflection point and the control instruction of titration end are recorded.

[0104] S430b, when the average difference between the hue predicted value and the hue measured value of the three consecutive times is not less than the set second threshold value, the three consecutive hue measured values are replaced with the earliest three consecutive hue measured values in the dynamic training set, and the linear regression model is iteratively trained based on the updated training set, so that the linear regression model after iterative training predicts the hue predicted value after adding the titration reagent for the next three times.

[0105] In this embodiment, the linear regression model is suitable for working as a prediction model for the second half of the titration due to its own characteristics and the trend of the change curve. By using the actual hue value for m times (at least 3 times) as a training set, the next drop hue value is predicted based on the linear regression model, and the actual hue value after adding the reagent is acquired. Since the hue becomes unstable, the process is cycled for 3 times and ensures that the training set is nearly m times, ensuring the accuracy of the prediction. If the average value of the difference between the 3 times of prediction and the actual value is greater than the second threshold value, it means that there is a large fluctuation in the color, and the above iterative training prediction is continued; if the average value of the difference between the 3 times of prediction and the actual value is less than the second threshold value, it means that the predicted curve and the actual curve are fitted, and the color has tended to be stable, and it is determined that the first drop point of the three consecutive drop points is the color stable inflection point.

[0106] In one specific embodiment, as Figure 7As shown, after determining the color mutation point, the actual hue value of the last m times is selected as the training set, and the hue value after adding the titration reagent next time is predicted based on the linear regression model. Then, the actual hue value is obtained for comparison with the predicted value. Due to the characteristics of the linear regression model, the predicted hue value will show a trend of rising or falling according to the change rate. In order to ensure the accuracy of the prediction, the whole process is repeated for 3 times, and the training set is always kept as the data of the last m times. The average value of the difference between the last 3 predictions and the actual value is taken as the judgment condition. When the difference is greater than the second threshold value, it is determined that the color is still fluctuating; when the difference is less than the second threshold value, the predicted curve closely fits the actual hue value, and it is determined that the color has tended to be stable, and the current point is the color smooth inflection point, that is, the inflection point 2 in Figure 7 The concentration value of the stock solution to be titrated can be calculated by the current titration amount, the concentration value of the formula blank sample solution, the concentration of the titration reagent, and the volume of the unknown concentration stock solution.

[0107] In addition, the embodiment also provides a color titration recognition system based on a combined prediction model, which comprises:

[0108] An ARIMA model training module is configured to collect the actual hue value of the solution after adding the titration reagent each time during the titration reaction stage, and train the preset ARIMA model by using the obtained actual hue value when the number of titration reaches the target sample size.

[0109] A first titration color judgment module is configured to compare the actual hue value after adding the titration reagent each time in the cyclic titration process with the predicted hue value predicted by the trained ARIMA model, and judge whether the comparison difference is greater than a set first threshold value.

[0110] A color mutation inflection point judgment module is configured to determine that the current titration point is the color mutation inflection point when the comparison difference is greater than the set first threshold value, and collect at least three consecutive actual hue values after the color mutation inflection point as a dynamic training set to train the preset linear regression model.

[0111] A second titration color judgment module is configured to iteratively execute the hue value prediction and the titration reagent adding operation three times based on the trained linear regression model, and judge whether the average difference between the consecutive three predicted hue values and the actual hue value is less than a set second threshold value.

[0112] A color smooth inflection point judgment module is configured to determine that the first adding point in the consecutive three adding points is the color smooth inflection point when the average difference between the consecutive three predicted hue values and the actual hue value is less than the set second threshold value.

[0113] Finally, the embodiment also provides a titration device, comprising: a quantitative liquid adding device; an industrial camera; a motion assembly connected with the industrial camera, used to drive the industrial camera to a specified position to enable the industrial camera to collect an image frame of a solution to be titrated; and a controller connected with the quantitative liquid adding device, the industrial camera and the motion assembly respectively, used to execute the color titration recognition method based on the combined prediction model based on the collected image frame.

[0114] In the embodiment, the quantitative liquid adding device comprises: a peristaltic pump, a plunger pump and a titration solution warehouse; the peristaltic pump is connected with the titration solution warehouse pipeline, used to add titration reagent to the titration solution warehouse; the plunger pump is connected with the titration solution warehouse pipeline and electrically connected with the color titration recognition device, used to control the drop of a corresponding amount of titration reagent into the solution to be titrated according to the control instruction output by the color titration recognition device.

[0115] In the embodiment, the titration device further comprises: a background light source and a light shield; the background light source is connected with the controller, used to dynamically adjust the background brightness of the color recognition area according to the original hue value of the solution to be titrated output by the controller.

[0116] In summary, the embodiment of the present application provides a color titration recognition method, system and device based on a combined prediction model. First, the color of the first stage of titration is identified based on the autoregressive moving average model (ARIMA model) to find the color mutation inflection point. Then, the color of the second stage of titration is identified based on the linear regression model to find the color smooth inflection point. Finally, the unknown concentration solution is automatically color titrated, and the actual concentration of the stock solution is calculated based on the results of the inflection point obtained by analyzing the formula blank sample and the combined prediction model. The present application can quickly identify the two inflection points of color change by reasonably using the characteristics of ARIMA model and linear regression model, which can be suitable for the needs of various color titration formulas, and different models are used for color titration recognition according to different reaction stages, which can further improve the accuracy of color titration and the anti-interference ability of identification analysis.

[0117] The system / device described in the above embodiments of the present application is used to implement the method of the above embodiments of the present application, so based on the method described in the above embodiments of the present application, those skilled in the art can understand the specific structure and modification of the system / device, and thus it is not repeated here. Any system / device used in the method of the above embodiments of the present application belongs to the scope of protection of the present application.

[0118] Those skilled in the art will appreciate that embodiments of the present application can be devised for a variety of other systems which are currently developed or later developed. Those skilled in the art will appreciate that the application can provide, among other things, a method, a system, or a computer program product. Accordingly, the present application can take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present application can take the form of a computer program product on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROMs, optical storage devices, and the like) embodying computer readable program code.

[0119] The present application is described in reference to the flowchart and / or block diagram illustrations of the method, apparatus (system) and computer program product according to embodiments of the application. It will be understood that each block of the flowchart and / or block diagrams, and combinations of blocks in the flowchart and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general purpose computer, special purpose computer, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions of the flowchart and / or block diagrams.

[0120] It should be noted that the use of the terms "one" and "an" herein does not exclude a plurality, and "a" or "an" means "at least one". The application can be implemented by means of hardware comprising several distinct elements, and by means of a suitably programmed computer. The use of the terms first, second, third, etc., does not imply any ordering, but is used for the purpose of naming. The terms "comprising", "comprise", "comprised of" and "comprising" when used in this specification are taken to specify the presence of stated features, integers, steps or components but do not preclude the presence or addition of one or more other features, integers, steps, components or groups thereof.

[0121] In addition, it should be pointed out that the terms "one embodiment", "some embodiments", "an embodiment", "example", "specific example" or "some examples" and the like, in the description of the specification, mean that the particular feature, structure, material or characteristic being described is included in at least one embodiment or example of the present application. Illustrative representations of the above terms in the specification do not necessarily refer to the same embodiment or example. Moreover, the particular features, structures, materials, or characteristics can be combined in any suitable manner in one or more embodiments or examples. Furthermore, different embodiments or examples described in the specification can be combined and combined with features of other embodiments or examples, without mutual contradiction.

[0122] Although preferred embodiments of the application have been described, those skilled in the art will appreciate that additional modifications and variations to the described embodiments are possible in light of the above teachings.

[0123] It will be apparent to those skilled in the art that various modifications and variations can be made to the present application without departing from the spirit or scope of the application.

Claims

1. A color titration recognition method based on a combination prediction model, characterized in that, The method comprises the following steps: During the titration reaction stage, the color phase measured value of the solution after each drop of titration reagent is collected, and when the number of drops reaches the target sample amount, the preset ARIMA model is trained using the obtained color phase measured value; During the titration reaction stage, the color phase measured value of the solution after each drop of titration reagent is collected, and when the number of drops reaches the target sample amount, the preset ARIMA model is trained using the obtained color phase measured value; During the titration reaction stage, the color phase measured value of the solution after each drop of titration reagent is collected, and when the number of drops reaches the target sample amount, the preset ARIMA model is trained using the obtained color phase measured value; During the titration reaction stage, the color phase measured value of the solution after each drop of titration reagent is collected, and when the number of drops reaches the target sample amount, the preset ARIMA model is trained using the obtained color phase measured value; 2. The method of claim 1, wherein, During the titration reaction stage, the color phase measured value of the solution after each drop of titration reagent is collected, and when the number of drops reaches the target sample amount, the preset ARIMA model is trained using the obtained color phase measured value; During the titration reaction stage, the color phase measured value of the solution after each drop of titration reagent is collected, and when the number of drops reaches the target sample amount, the preset ARIMA model is trained using the obtained color phase measured value; During the titration reaction stage, the color phase measured value of the solution after each drop of titration reagent is collected, and when the number of drops reaches the target sample amount, the preset ARIMA model is trained using the obtained color phase measured value; During the titration reaction stage, the color phase measured value of the solution after each drop of titration reagent is collected, and when the number of drops reaches the target sample amount, the preset ARIMA model is trained using the obtained color phase measured value; 3. The method of claim 1, wherein, During the titration reaction stage, the color phase measured value of the solution after each drop of titration reagent is collected, and when the number of drops reaches the target sample amount, the preset ARIMA model is trained using the obtained color phase measured value; During the titration reaction stage, the color phase measured value of the solution after each drop of titration reagent is collected, and when the number of drops reaches the target sample amount, the preset ARIMA model is trained using the obtained color phase measured value; During the titration reaction stage, the color phase measured value of the solution after each drop of titration reagent is collected, and when the number of drops reaches the target sample amount, the preset ARIMA model is trained using the obtained color phase measured value; During the titration reaction stage, the color phase measured value of the solution after each drop of titration reagent is collected, and when the number of drops reaches the target sample amount, the preset ARIMA model is trained using the obtained color phase measured value; During the titration reaction stage, the color phase measured value of the solution after each drop of titration reagent is collected, and when the number of drops reaches the target sample amount, the preset ARIMA model is trained using the obtained color phase measured value; During the titration reaction stage, the color phase measured value of the solution after each drop of titration reagent is collected, and when the number of drops reaches the target sample amount, the preset ARIMA model is trained using the obtained color phase measured value; 4. The method of claim 1, wherein, During the titration reaction stage, the color phase measured value of the solution after each drop of titration reagent is collected, and when the number of drops reaches the target sample amount, the preset ARIMA model is trained using the obtained color phase measured value; During the titration reaction stage, the color phase measured value of the solution after each drop of titration reagent is collected, and when the number of drops reaches the target sample amount, the preset ARIMA model is trained using the obtained color phase measured value; During the titration reaction stage, the color phase measured value of the solution after each drop of titration reagent is collected, and when the number of drops reaches the target sample amount, the preset ARIMA model is trained using the obtained color phase measured value; During the titration reaction stage, the color phase measured value of the solution after each drop of titration reagent is collected, and when the number of drops reaches the target sample amount, the preset ARIMA model is trained using the obtained color phase measured value; During the titration reaction stage, the color phase measured value of the solution after each drop of titration reagent is collected, and when the number of drops reaches the target sample amount, the preset ARIMA model is trained using the obtained color phase measured value; During the titration reaction stage, the color phase measured value of the solution after each drop of titration reagent is collected, and when the number of drops reaches the target sample amount, the preset ARIMA model is trained using the obtained color phase measured value; During the titration reaction stage, the color phase measured value of the solution after each drop of titration reagent is collected, and when the number of drops reaches the target sample amount, the preset ARIMA model is trained using the obtained color phase measured value; During the titration reaction stage, the color phase measured value of the solution after each drop of titration reagent is collected, and when the number of drops reaches the target sample amount, the preset ARIMA model is trained using the obtained color phase measured value; During the titration reaction stage, the color phase measured value of the solution after each drop of titration reagent is collected, and when the number of drops reaches the target sample amount, the preset ARIMA model is trained using the obtained color phase measured value; During the titration reaction stage, the color phase measured value of the solution after each drop of titration reagent is collected, and when the number of drops reaches the target sample amount, the preset ARIMA model is trained using the obtained color phase measured value; During the titration reaction stage, the color phase measured value of the solution after each drop of titration reagent is collected, and when the number of drops reaches the target sample amount, the preset ARIMA model is trained using the obtained color phase measured value; During the titration reaction stage, the color phase measured value of the solution after each drop of titration reagent is collected, and when the number of drops reaches the target sample amount, the preset ARIMA model is trained using the obtained color phase measured value; During the titration reaction stage, the color phase measured value of the solution after each drop of titration reagent is collected, and when the number of drops reaches the target sample amount, the preset ARIMA model is trained using the obtained color phase measured value; During the titration reaction stage, the color phase measured value of the solution after each drop of titration reagent is collected, and when the number of drops reaches the target sample amount, the preset ARIMA model is trained using the obtained color phase measured value; During the titration reaction stage, the color phase measured value of the solution after each drop of titration reagent is collected, and when the number of drops reaches the target sample amount, the preset ARIMA model is trained using the obtained color phase measured value; During the titration reaction stage, the color phase measured value of the solution after each drop of titration reagent is collected, and when the number of drops reaches the target sample amount, the preset ARIMA model is trained using the obtained color phase measured value; During the titration reaction stage, the color phase measured value of the solution after each drop of titration reagent is collected, and when the number of drops reaches the target sample amount, the preset ARIMA model is trained using the obtained color phase measured value; During the titration reaction stage, the color phase measured value of the solution after each drop of titration reagent is collected, and when the number of drops reaches the target sample amount, the preset ARIMA model is trained using the obtained color phase measured value; During the titration reaction stage, the color phase measured value of the solution after each drop of titration reagent is collected, and when the number of drops reaches the target sample amount, the preset ARIMA model is trained using the obtained color phase measured value; During the titration reaction stage, the color phase measured value of the solution after each drop of titration reagent is collected, and when the number of drops reaches the target sample amount, the preset ARIMA model is trained using the obtained color phase measured value; During the titration reaction stage, the color phase measured value of the solution after each drop of titration reagent is collected, and when the number of drops reaches the target sample amount, the preset ARIMA model is trained using the obtained color phase measured value; During the titration reaction stage, the color phase measured value of the solution after each drop of titration reagent is collected, and when the number of drops reaches the target sample amount, the preset ARIMA model is trained using the obtained color phase measured value; During the titration reaction stage, the color phase measured value of the solution after each drop of titration reagent is collected, and when the number of drops reaches the target sample amount, the preset ARIMA model is trained using the obtained color phase measured value; During the titration reaction stage, the color phase measured value of the solution after each drop of titration reagent is collected, and when the number of drops reaches the target sample amount, the preset ARIMA model is trained using the obtained color phase measured value; During the titration reaction stage, the color phase measured value of the solution after each drop of titration reagent is collected, and comparing the hue prediction value with the corresponding hue measured value, judging whether the comparison difference is greater than a set first threshold value; when the comparison difference is greater than the set first threshold value, determining that the current titration point is a color mutation inflection point; when the comparison difference is not greater than the set first threshold value, supplementing the solution hue value of the current titration point to the training set of the ARIMA model, and iteratively training the ARIMA model through the updated training set, so that the iteratively trained ARIMA model predicts the hue prediction value after the next addition of the titration reagent.

5. The method of claim 1, wherein, when the comparison difference is greater than the set first threshold value, determining that the current titration point is a color mutation inflection point, and collecting at least three consecutive hue measured values after the color mutation inflection point as a dynamic training set to train the preset linear regression model, including: in the case that the comparison difference between the hue prediction value and the hue measured value is greater than the set first threshold value, determining the current titration point as a color mutation inflection point, and recording the number of cyclic titrations to reach the color mutation inflection point; judging whether the difference amplitude of the at least three consecutive hue measured values after the color mutation inflection point exceeds a set amplitude threshold value; when the difference amplitude of the at least three consecutive hue measured values after the color mutation inflection point does not exceed the set amplitude threshold value, outputting a control instruction for ending titration; when the difference amplitude of the at least three consecutive hue measured values after the color mutation inflection point exceeds the set amplitude threshold value, collecting the hue measured values obtained after the color mutation inflection point as a dynamic training set to train the preset linear regression model.

6. The method of claim 1, wherein, based on the trained linear regression model, iteratively performing the hue value prediction and the titration reagent addition operation three times, until the average difference between the consecutive three times of hue prediction values and the hue measured values is less than a set second threshold value, determining that the first addition point in the consecutive three addition points is a color stable inflection point, including: predicting the hue prediction value of the solution to be titrated after the three times of titration operation through the trained linear regression model, and synchronously acquiring the hue measured value of the solution to be titrated; judging whether the average difference between the consecutive three times of hue prediction values and the hue measured values is less than the set second threshold value; when the average difference between the consecutive three times of hue prediction values and the hue measured values is less than the set second threshold value, determining that the first addition point in the consecutive three addition points is a color stable inflection point, and recording the number of cyclic titrations to reach the color stable inflection point and outputting a control instruction for ending titration; when the average difference between the consecutive three times of hue prediction values and the hue measured values is not less than the set second threshold value, replacing the earliest three times of consecutive hue measured values in the dynamic training set with the consecutive three times of hue measured values, and iteratively training the linear regression model through the updated training set, so that the iteratively trained linear regression model predicts the hue prediction value after the next three times of addition of the titration reagent.

7. A color titration recognition system based on a combined prediction model, characterized in that, including: an ARIMA model training module, configured to acquire the hue measured value of the solution after each addition of the titration reagent in the titration reaction stage, and train a preset ARIMA model by using the acquired hue measured value when the number of additions reaches a target sample size; The first titration color judgment module is configured to compare the measured hue after each drop of titration reagent in the cyclic titration process with the predicted hue value predicted by the trained ARIMA model, and determine whether the comparison difference is greater than a first threshold value; The color mutation inflection point determination module is configured to determine the current titration point as a color mutation inflection point when the comparison difference is greater than the first threshold value, and collect at least three consecutive measured hue values after the color mutation inflection point as a dynamic training set to train the preset linear regression model; The second titration color judgment module is configured to iteratively perform the hue value prediction and the titration reagent drop operation three times based on the trained linear regression model, and determine whether the average difference between the consecutive three times of hue prediction value and the measured hue value is less than a second threshold value; The color smooth inflection point determination module is configured to determine the first drop point in the three consecutive drop points as a color smooth inflection point when the average difference between the consecutive three times of hue prediction value and the measured hue value is less than the second threshold value.

8. A titration device, characterized by It comprises: A quantitative liquid adding device; An industrial camera; A motion assembly connected with the industrial camera, used to drive the industrial camera to a specified position to enable the industrial camera to collect image frames of the solution to be titrated; A controller connected with the quantitative liquid adding device, the industrial camera and the motion assembly, respectively, used to execute the color titration recognition method based on the combined prediction model according to any one of claims 1-6 based on the collected image frames.

9. The apparatus of claim 8, wherein, The quantitative liquid adding device comprises a peristaltic pump, a plunger pump and a titration solution warehouse; The peristaltic pump is connected with the titration solution warehouse pipeline, used to add titration reagent to the titration solution warehouse; The plunger pump is connected with the titration solution warehouse pipeline and electrically connected with the color titration recognition device, used to control the drop of a corresponding amount of titration reagent into the solution to be titrated according to the control instruction output by the color titration recognition device.

10. The apparatus of claim 8, wherein, It also comprises: A background light source and a light shield; The background light source is connected with the controller, used to dynamically adjust the background brightness of the color recognition area according to the original hue value of the solution to be titrated output by the controller.

Images