An AI intelligent permanganate index detection method, device, equipment and medium

Abstract

The present application relates to the field of water quality monitoring technology, and in particular to an AI intelligent permanganate index detection method, device, equipment and medium, the method comprising: after controlling the target instrument to use an excess of potassium permanganate solution for back titration, obtaining the titration signal data obtained by the back titration collected by the target instrument, wherein the back titration adopts a timed and quantitative equal titration mode; drawing an image according to the titration signal data to obtain a signal image with time as the horizontal coordinate and signal value as the vertical coordinate; inputting the signal image into the permanganate index endpoint recognition model to obtain the first titration endpoint of the test sample; the permanganate index endpoint recognition model is obtained by training the network model based on the training samples, and the training samples include: multiple back titration signal training images and corresponding titration endpoint labels; according to the first titration endpoint, calculating the permanganate index value of the test sample. The present application can improve the accuracy of determining the titration endpoint.

Description

Technical Field

[0001] The present application relates to the field of water quality monitoring technology, and in particular to an AI intelligent permanganate index detection method, device, equipment and medium. Background Art

[0002] Permanganate index is one of the important indicators commonly used in water quality monitoring, which is used to evaluate the content of organic and inorganic reducing substances in water. In the determination process of permanganate index, endpoint identification is a key step, which is directly related to the accuracy and reliability of the determination results.

[0003] End point identification methods mostly rely on physical signal changes during chemical titration, such as potential, light intensity, etc., to determine whether the titration reaction has reached the predetermined endpoint. Taking ORP potential mutation as an example, during the titration process, by monitoring the potential change of the solution, when the potential value undergoes a significant mutation, it is considered that the titration reaction has reached the endpoint. The specific operation is: first add an excess of sodium oxalate solution to the water sample to be tested for reduction reaction, and then add potassium permanganate solution drop by drop for back titration. During the titration process, the instrument monitors the potential value in real time. When the potential value reaches the preset endpoint threshold, the titration is stopped immediately and the titration time is recorded. The permanganate index can be converted based on the titration time and the amount of potassium permanganate solution consumed. The light intensity mutation rule determines the endpoint by monitoring the change in solution color (i.e., the change in light intensity) during the titration process.

[0004] Although this identification method ensures the uniqueness of the endpoint, it is very easy to cause misjudgment once the process is disturbed. In addition, since the mode of stopping the titration immediately when the conditions are met is used, if abnormal data appears during the titration process (such as a sudden jump in the potential value, abnormal fluctuation in light intensity, etc.), the instrument may not be able to identify and analyze these abnormal data in time, affecting the accuracy of the measurement results.

[0005] Therefore, how to accurately locate the titration endpoint has become a technical problem that needs to be solved urgently by those skilled in the art. Summary of the invention

[0006] The purpose of this application is to provide an AI intelligent permanganate index detection method, device, equipment and medium, which can improve the accuracy of titration endpoint determination.

[0007] In the first aspect, an AI intelligent permanganate index detection method is provided, comprising:

[0008] After the target instrument is controlled to perform back titration using an excess of potassium permanganate solution, titration signal data obtained by the back titration collected by the target instrument is obtained, wherein a timed and quantitative equal titration mode is used during the back titration;

[0009] Draw an image according to the titration signal data to obtain a signal image with time as the horizontal axis and signal value as the vertical axis;

[0010] Inputting the signal image into a permanganate index endpoint recognition model to obtain a first titration endpoint of the test sample; the permanganate index endpoint recognition model is obtained by training a network model based on training samples, and the training samples include: a plurality of back titration signal training images and corresponding titration endpoint labels;

[0011] According to the first titration endpoint, the permanganate index value of the test sample is calculated.

[0012] In a preferred example, the present application can be further configured as follows: the training process of the permanganate index endpoint recognition model includes:

[0013] Obtain training samples of various working condition types;

[0014] Input the back titration signal training image into the network model to perform endpoint recognition and obtain the training titration endpoint;

[0015] Determine the loss function value of the network model based on the training titration endpoint and the titration endpoint label;

[0016] The parameters of the network model are corrected according to the loss function value, and the network model is trained based on the training samples, and the network parameters of the network model are updated until the loss function value of the network model is no greater than the preset loss threshold, thereby obtaining a permanganate index endpoint recognition model.

[0017] In a preferred example, the present application can be further configured as follows: training the network model based on the training samples, updating the network parameters of the network model until the loss function value of the network model is not greater than the preset loss threshold, and after obtaining the permanganate index endpoint recognition model, it also includes:

[0018] Acquire a test sample; the test sample includes: a titration signal test image and a corresponding titration endpoint label;

[0019] Using the titration signal test image, multiple measurements are performed through the permanganate index endpoint recognition model to obtain multiple test titration endpoints;

[0020] Calculate accuracy and precision based on multiple test titration endpoints and titration endpoint labels;

[0021] When the accuracy is less than the preset accuracy threshold and the precision is less than the preset precision threshold, the permanganate index endpoint recognition model is used as the final permanganate index endpoint recognition model;

[0022] Otherwise, the permanganate index endpoint recognition model is trained again until a permanganate index endpoint recognition model is obtained with an accuracy less than a preset accuracy threshold and a precision less than a preset precision threshold.

[0023] In a preferred example, the present application may be further configured as follows: after inputting the signal image into the permanganate index endpoint recognition model to obtain the first titration endpoint of the test sample, the present application may further include:

[0024] generating a target curve according to the titration signal data;

[0025] Performing peak search analysis on the target curve to obtain characteristic information, wherein the characteristic information includes the existence of a minimum peak and the peak position corresponding to the minimum peak; determining a second titration endpoint of the test sample according to the characteristic information;

[0026] Using the second titration endpoint to verify the first titration endpoint;

[0027] When the verification is passed, calculating the permanganate index value of the test sample according to the first titration endpoint;

[0028] When the verification fails, a first prompt message is generated.

[0029] In a preferred example, the present application can be further configured as follows: performing peak search analysis on the target curve to obtain characteristic information, wherein the characteristic information includes the existence of a very small peak and the peak position corresponding to the very small peak; determining the second titration endpoint of the test sample according to the characteristic information, including:

[0030] Performing smoothing processing on the titration signal data to obtain a smooth curve; and performing peak search analysis on the smooth curve to obtain first characteristic information, wherein the first characteristic information includes the existence of a first minimum peak and a peak position corresponding to the first minimum peak;

[0031] Performing differential processing on the titration signal data to obtain a differential curve; and performing peak search analysis on the differential curve to obtain second characteristic information, wherein the second characteristic information includes the existence of a second minimum peak and a peak position corresponding to the second minimum peak;

[0032] A second titration endpoint of the potassium permanganate solution is determined according to at least one of the first characteristic information and the second characteristic information.

[0033] In a preferred example, the present application may be further configured as follows: after inputting the signal image into the permanganate index endpoint recognition model to obtain the first titration endpoint of the test sample, the present application may further include:

[0034] After the target instrument is controlled to perform back titration using an excess of potassium permanganate solution, multiple dimensional parameters obtained by back titration collected by the target instrument are obtained;

[0035] Predicting the titration endpoint according to multiple dimensional parameters to obtain the third titration endpoint;

[0036] Verifying the first titration endpoint using the third titration endpoint;

[0037] When the verification is passed, calculating the permanganate index value of the test sample according to the first titration endpoint;

[0038] When the verification fails, a second prompt message is generated.

[0039] In a preferred example, the present application may be further configured as follows: after the target instrument is controlled to use an excess of potassium permanganate solution for back titration, before obtaining the multi-dimensional parameters obtained by the back titration collected by the target instrument, the method may further include:

[0040] Acquire multiple initial dimension parameters during the back titration process of test samples of various working conditions in history;

[0041] According to whether the parameter change of the initial dimension parameter is greater than the preset change threshold at the titration endpoint;

[0042] All target dimensional parameters are used as the multiple dimensional parameters, wherein the target dimensional parameters are initial dimensional parameters that are greater than a preset change threshold.

[0043] In the second aspect, an AI intelligent permanganate index detection device is provided, comprising:

[0044] An acquisition module is used to acquire titration signal data obtained by back titration collected by the target instrument after the target instrument is controlled to perform back titration with an excess of potassium permanganate solution, wherein a timed and quantitative equal titration mode is adopted during the back titration;

[0045] An image drawing module is used to draw an image according to the titration signal data to obtain a signal image with time as the horizontal axis and signal value as the vertical axis;

[0046] A titration endpoint determination module is used to input the signal image into a permanganate index endpoint recognition model to obtain a first titration endpoint of the test sample; the permanganate index endpoint recognition model is obtained by training a network model based on training samples, and the training samples include: a plurality of back titration signal training images and corresponding titration endpoint labels;

[0047] A calculation module is used to calculate the permanganate index value of the test sample according to the first titration endpoint.

[0048] In a third aspect, an electronic device is provided, including:

[0049] one or more processors;

[0050] Memory;

[0051] One or more applications, wherein one or more applications are stored in a memory and configured to be executed by one or more processors, and one or more programs are configured to: perform operations corresponding to the method shown in any possible implementation of the first aspect.

[0052] In a fourth aspect, a computer-readable storage medium is provided, wherein the storage medium stores at least one instruction, at least one program, a code set or an instruction set, and the at least one instruction, at least one program, a code set or an instruction set is loaded by a processor and executes the steps of the method shown in any possible implementation of the first aspect.

[0053] In a fifth aspect, a computer program product is provided, comprising a computer program, wherein when the computer program is executed by a processor, operations corresponding to the method shown in any possible implementation manner in the first aspect are implemented.

[0054] In summary, the method provided by this application includes the following beneficial technical effects:

[0055] In the back titration test process, the method adopts a timed and quantitative titration mode to replace the existing endpoint self-recognition mode, and draws a signal image based on the collected data after the titration is completed; a neural network is used to identify the signal image to facilitate the rapid and accurate determination of the titration endpoint, and the volume of potassium permanganate consumed in the back titration process is converted based on the titration endpoint, thereby accurately calculating the permanganate index value of the measured sample.

[0056] In addition, the present application also provides an AI intelligent permanganate index detection device, equipment and medium, all of which have the above-mentioned beneficial technical effects. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] In order to more clearly illustrate the embodiments of the present application or the technical solutions of the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0058] Figure 1 This is a flow chart of an AI intelligent permanganate index detection method provided in an embodiment of the present application;

[0059] Figure 2A schematic diagram of a process flow of another AI intelligent permanganate index detection method provided in an embodiment of the present application;

[0060] Figure 3 A schematic diagram of a titration endpoint provided in an embodiment of the present application;

[0061] Figure 4 It is a structural schematic diagram of an AI intelligent permanganate index detection device provided in an embodiment of the present application;

[0062] Figure 5 It is a structural schematic diagram of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0063] This specific embodiment is merely an explanation of the present application and is not a limitation of the present application. After reading this specification, a person skilled in the art may make non-creative modifications to the present embodiment as needed, but such modifications are protected by the patent law as long as they are within the scope of the present application.

[0064] It should be noted that in the optional embodiments of the present application, the object information and other related data involved, when the embodiments in the present application are applied to specific products or technologies, need to obtain the permission or consent of the object, and the collection, use and processing of the relevant data need to comply with the relevant laws, regulations and standards of the relevant countries and regions. In other words, if the embodiments of the present application involve data related to the object, it needs to be obtained with the authorization and consent of the object, the authorization and consent of the relevant departments, and in compliance with the relevant laws, regulations and standards of the country and region. If personal information is involved in the embodiments, the acquisition of all personal information needs to obtain the consent of the individual. If sensitive information is involved, the separate consent of the information subject needs to be obtained. The embodiments also need to be implemented with the authorization and consent of the object.

[0065] In order to make the purpose, technical solution and advantages of the embodiments of the present application clearer, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0066] In addition, the term "and / or" in this article is only a description of the association relationship of associated objects, indicating that there can be three relationships. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. In addition, the character " / " in this article, unless otherwise specified, generally means that the associated objects before and after are in an "or" relationship.

[0067] The endpoint recognition methods of existing permanganate index instruments mostly use ORP potential mutation, light intensity mutation and other methods, that is, by presetting the endpoint threshold range or using preset coefficients or related judgment modes to achieve the determination of the endpoint and the output of the test results. Although this recognition method ensures the uniqueness of the endpoint, it is very easy to cause misjudgment once the process is disturbed, and because the mode of stopping the titration immediately when the conditions are met is used, there will be abnormal data that cannot be recognized and analyzed during the titration process. Often, the final result is distorted and cannot be traced due to some abnormal conditions, making it difficult to locate the abnormal situation, thereby increasing the labor cost of troubleshooting.

[0068] The present invention adopts a timed and quantitative titration mode plus an endpoint determination mode determined by a post-algorithm, completely records the signal changes during the titration process, and uses an image recognition or mathematical fitting mode to identify and calculate the endpoint of the acquired titration curve, which can trace the signal changes during the entire titration process, reduce data anomalies caused by abnormal interference, improve maintenance efficiency and reduce maintenance difficulty.

[0069] Among them, the amount and order of the quantitative water sample, potassium permanganate, acid solution and sodium oxalate reagent can refer to the relevant technology.

[0070] In the embodiment of the present application, the testing process of the permanganate index includes:

[0071] The sample to be tested is mixed with sulfuric acid solution and potassium permanganate solution, and the temperature is raised to 94.6-94.9°C according to tideal=94.831-77.737*exp(-t / 113.082), and an oxidation reaction is carried out during the heating process;

[0072] The volume ratio of the test sample, sulfuric acid solution and potassium permanganate solution is 20:1:2; the concentration of potassium permanganate solution is 0.31-0.33 g / L, the concentration of sulfuric acid solution is 485-495 g / L, and the concentration of sodium oxalate solution is 0.83-0.85 g / L.

[0073] In the process of adding excess sodium oxalate solution for reduction reaction and then adding potassium permanganate solution for back titration, a timed and quantitative equal titration mode is used to replace the real-time titration mode when direct titration until the potential value suddenly changes, that is, an excess of potassium permanganate solution is used for back titration during the titration process and the signal value of the entire titration process is recorded in detail to obtain a complete titration process curve; this method can be used to comprehensively monitor the titration process, reduce the probability of misjudgment, improve the accuracy of endpoint recognition, and reduce the difficulty of troubleshooting.

[0074] According to the obtained titration process curve, the post-judgment mode is used to apply image recognition and curve mathematical simulation algorithms to obtain the titration endpoint information to calculate the permanganate index value;

[0075] ;

[0076] Where, Y is the permanganate index, mg / L;

[0077] X is the titration time of the sample, s;

[0078] C2 is the standard sample concentration of 80-100% of the range, mg / L; the range is 0-5 mg / L;

[0079] C1 is the standard sample concentration of the range of 0-20%, mg / L;

[0080] t2 is the titration time of the standard sample concentration of 80-100% of the range, s;

[0081] t1 is the standard sample concentration of 0-20% of the range, mg / L.

[0082] Specifically, the present application embodiment provides an AI intelligent permanganate index detection method, such as Figure 2 As shown, the method provided in the embodiment of the present application can be executed by an electronic device, and the electronic device is a server, which can be an independent physical server, or a server cluster or distributed system composed of multiple physical servers, or a cloud server providing cloud computing services. The terminal device can be a smart phone, a tablet computer, a laptop computer, a desktop computer, a target instrument, i.e., an electrode permanganate index online analyzer, etc., but is not limited thereto. The terminal device and the electronic device can be directly or indirectly connected via wired or wireless communication, which is not limited in the embodiment of the present application. The method includes:

[0083] S101, after controlling the target instrument to perform back titration with an excess of potassium permanganate solution, obtaining titration signal data obtained by the back titration collected by the target instrument,

[0084] Among them, the back titration adopts the equal titration mode of timed quantitative titration;

[0085] In the embodiment of the present application, the signal acquisition board of the target instrument collects and records the signal during the titration process, so that the electronic device obtains the titration signal data of the back titration of the excess potassium permanganate solution in the equal titration mode. The titration signal data includes: one of the ORP signal, the light intensity signal, and the double platinum electrode signal.

[0086] The signal value of the entire titration process is collected and recorded, and 3-10 data are collected per second. According to the specific range difference, the collection time can be adaptively adjusted, and the collection time range can be 300-800 seconds.

[0087] S102, drawing an image according to the titration signal data to obtain a signal image with time as the horizontal axis and signal value as the vertical axis;

[0088] After the titration signal data is obtained, the recorded titration signal data is analyzed and plotted, and the obtained titration signal value is plotted with time as the horizontal axis and the signal value as the vertical axis.

[0089] Furthermore, the titration signal data may be filtered by median filtering, Gaussian filtering or the like to remove noise interference, and then an image may be drawn using the denoised titration signal data to obtain a signal image with time as the horizontal axis and signal value as the vertical axis.

[0090] S103, inputting the signal image into a permanganate index endpoint recognition model to obtain a first titration endpoint of the test sample;

[0091] The permanganate index endpoint recognition model is obtained by training the network model based on training samples, and the training samples include: multiple back titration signal training images and corresponding titration endpoint labels;

[0092] In the embodiment of the present application, the titration endpoint is judged and identified by the permanganate index endpoint recognition model to accurately identify the titration endpoint. The characteristic points of the titration curve are extracted using the trained model and converted into the required titration endpoint information, and then the corresponding permanganate index value (CODmn) is calculated based on the titration endpoint value.

[0093] Machine learning or deep learning algorithms are used to train and learn the data in the titration process to establish a model for endpoint recognition. The model can automatically identify the titration endpoint. This method is highly adaptive and can automatically adjust and optimize according to the characteristics of different samples. At the same time, the algorithm can handle complex data relationships and improve the accuracy of judgment.

[0094] S104. Calculate the permanganate index value of the test sample according to the first titration endpoint.

[0095] It can be seen that in the embodiment of the present application, the permanganate index test process is used to perform sample pretreatment, sampling, reagent addition, and oxidation-reduction process. In the final back titration test process, a timed and quantitative titration mode is used to replace the existing endpoint self-recognition mode. After the titration is completed, a signal image is drawn based on the collected data; a neural network is used to identify the signal image to quickly and accurately determine the titration endpoint, and the volume of potassium permanganate consumed in the back titration process is converted based on the titration endpoint, thereby accurately calculating the permanganate index value of the measured sample.

[0096] In a possible implementation of the embodiment of the present application, the training process of the permanganate index endpoint recognition model includes:

[0097] Obtain training samples of various working condition types;

[0098] Input the back titration signal training image into the network model to perform endpoint recognition and obtain the training titration endpoint;

[0099] Determine the loss function value of the network model based on the training titration endpoint and the titration endpoint label;

[0100] The parameters of the network model are corrected according to the loss function value, and the network model is trained based on the training samples, and the network parameters of the network model are updated until the loss function value of the network model is no greater than the preset loss threshold, thereby obtaining a permanganate index endpoint recognition model.

[0101] In the embodiment of the present application, a large amount of titration curve data under various working conditions is obtained, the coordinates of the feature points in the image are manually annotated, and these coordinates are used as labels for training data. The various working conditions include but are not limited to: lake water, river water and seawater samples in different basins and regions.

[0102] The network model is trained using the labeled data and the parameters of the network model are continuously adjusted to improve the recognition accuracy of the model, wherein the network model is a neural network model, and the specific structure is not limited in the embodiment of the present application, and the user can set it according to actual needs. Among them, the preset loss threshold can be set according to experience, and is not limited in the present embodiment.

[0103] A possible implementation of the embodiment of the present application includes training the network model based on the training sample, updating the network parameters of the network model until the loss function value of the network model is not greater than the preset loss threshold, and obtaining the permanganate index endpoint recognition model, further comprising:

[0104] Acquire a test sample; the test sample includes: a titration signal test image and a corresponding titration endpoint label;

[0105] Using the titration signal test image, multiple measurements are performed through the permanganate index endpoint recognition model to obtain multiple test titration endpoints;

[0106] Calculate accuracy and precision based on multiple test titration endpoints and titration endpoint labels;

[0107] When the accuracy is less than the preset accuracy threshold and the precision is less than the preset precision threshold, the permanganate index endpoint recognition model is used as the final permanganate index endpoint recognition model;

[0108] Otherwise, the permanganate index endpoint recognition model is trained again until a permanganate index endpoint recognition model is obtained with an accuracy less than a preset accuracy threshold and a precision less than a preset precision threshold.

[0109] In the embodiment of the present application, the formula for calculating accuracy is:

[0110] ;

[0111] in: is the indication error of the standard solution, %; used to characterize accuracy. The smaller the value, the better the accuracy.

[0112] is the average value of n measurements, mg / L;

[0113] C is the concentration of the standard solution, mg / L.

[0114] In the embodiment of the present application, the formula for calculating precision is:

[0115] ;

[0116] in: is repeatability, %, used to characterize precision. The smaller the value, the better the precision.

[0117] is the i-th measurement value, mg / L

[0118] is the average value of n measurements, mg / L;

[0119] n is the number of measurements.

[0120] Among them, the preset accuracy threshold and the preset precision threshold can be set by the user according to actual needs.

[0121] It can be seen that in the embodiment of the present application, after the permanganate index endpoint recognition model is obtained, the test sample can be used for testing to determine the accuracy and precision of the permanganate index endpoint recognition model. When the corresponding threshold is met, it can be used as the final model. Otherwise, iterative training can be performed to improve the accuracy and precision of the permanganate index endpoint recognition model until a model that meets the requirements is obtained.

[0122] A possible implementation method of the embodiment of the present application is to combine Figure 2 , Figure 2 A schematic flow chart of another AI intelligent permanganate index detection method provided in an embodiment of the present application, wherein the signal image is input into the permanganate index endpoint recognition model, and after obtaining the first titration endpoint of the test sample, it also includes:

[0123] Generate a target curve based on the titration signal data;

[0124] Perform peak search analysis on the target curve to obtain characteristic information, which includes the existence of a very small peak and the peak position corresponding to the very small peak; determine the second titration endpoint of the test sample based on the characteristic information;

[0125] Using the second titration endpoint to verify the first titration endpoint;

[0126] When the verification is passed, the permanganate index value of the test sample is calculated according to the first titration endpoint;

[0127] When the verification fails, a first prompt message is generated.

[0128] In the embodiment of the present application, after analyzing the collected and recorded titration process data, the titration curve is analyzed and processed. The titration curve is analyzed by mathematical methods (such as derivative analysis, curve fitting, etc.), and the inflection point or mutation point of the curve change is found as the titration end point. For example, taking the double platinum electrode signal as an example, see Figure 3 , Figure 3 A schematic diagram of a titration endpoint provided in an embodiment of the present application, wherein the position in the circle is the second titration endpoint. This method can make full use of all the data in the titration process to accurately judge the endpoint, and the method is relatively simple and easy to implement. This method works better when the titration curve has obvious characteristics.

[0129] It can be seen that in the embodiment of the present application, the curve analysis method can also be used to perform a second verification on the first titration endpoint. When the difference between the second titration endpoint and the first titration endpoint is within the preset difference range, it is determined that the verification is passed, otherwise the verification fails, so as to improve the accuracy of the titration endpoint determination.

[0130] A possible implementation of the embodiment of the present application is to perform peak analysis on the target curve to obtain characteristic information, where the characteristic information includes the existence of a minimum peak and the peak position corresponding to the minimum peak; and determine the second titration endpoint of the test sample according to the characteristic information, including:

[0131] Performing smoothing processing on the titration signal data to obtain a smooth curve; and performing peak search analysis on the smooth curve to obtain first characteristic information, wherein the first characteristic information includes the existence of a first minimum peak and a peak position corresponding to the first minimum peak;

[0132] Performing differential processing on the titration signal data to obtain a differential curve; and performing peak search analysis on the differential curve to obtain second characteristic information, the second characteristic information including the existence of the second minimum peak and the peak position corresponding to the second minimum peak;

[0133] Determine a second titration endpoint of the potassium permanganate solution based on at least one of the first characteristic information and the second characteristic information.

[0134] The order of smooth curve peak search and differential dotted line peak search is no longer limited in the present embodiment of the application; in one possible case, both are performed simultaneously; in another possible case, the peak is first searched according to the smooth curve. If there is a first minimum peak and the number of the first minimum peak is 1, the titration end point of the potassium permanganate solution is determined to be the peak position corresponding to the first minimum peak, and the differential curve peak search process is not required at this time; otherwise, the differential curve peak search is performed, and the final titration end point is determined based on the information of the two minimum peaks. Either method is acceptable, and the user can set it according to actual needs.

[0135] The second case is further explained by taking it as an example. Specifically, smoothing is performed on the titration signal data to obtain a smooth curve; and peak-finding analysis is performed on the smooth curve to obtain first characteristic information, the first characteristic information including the existence of the first minimum peak and the peak position corresponding to the first minimum peak; if the first minimum peak exists and the number of the first minimum peak is 1, then the first titration end point of the potassium permanganate solution is determined to be the peak position corresponding to the first minimum peak; otherwise, the titration signal data is differentiated to obtain a differential curve; and peak-finding analysis is performed on the differential curve to obtain second characteristic information, the second characteristic information including the existence of the second minimum peak and the peak position corresponding to the second minimum peak; based on the first characteristic information and the second characteristic information, the second titration end point of the potassium permanganate solution is determined.

[0136] Wherein, determining the second titration endpoint of the potassium permanganate solution according to the first characteristic information and the second characteristic information includes: if there is no first minimum peak and no second minimum peak, determining the second titration endpoint of the potassium permanganate solution as the data collection endpoint; if there is no first minimum peak and there is a second minimum peak, determining the second titration endpoint of the potassium permanganate solution as the peak position corresponding to the second minimum peak; if there is a first minimum peak and the number of first minimum peaks is greater than 1, determining the target first minimum peak from at least two first minimum peaks according to the peak position corresponding to the second minimum peak, and determining the peak position corresponding to the target first minimum peak as the second titration endpoint of the potassium permanganate solution.

[0137] Among them, according to the peak position corresponding to the second minimum peak, the target first minimum peak is determined from at least two first minimum peaks, including: when the number of second minimum peaks is 1, the first minimum peak closest to the peak position corresponding to the second minimum peak is selected from at least two first minimum peaks as the target first minimum peak; when the number of second minimum peaks is greater than 1, the first minimum peak closest to the peak position corresponding to the target second minimum peak is selected from at least two first minimum peaks as the target first minimum peak, and the target second minimum peak is the second minimum peak with the smallest peak value.

[0138] A possible implementation of the embodiment of the present application includes, after inputting the signal image into the permanganate index endpoint recognition model to obtain the first titration endpoint of the test sample, further comprising:

[0139] After the target instrument is controlled to perform back titration using an excess of potassium permanganate solution, multiple dimensional parameters obtained by back titration collected by the target instrument are obtained;

[0140] Predicting the titration endpoint according to multiple dimensional parameters to obtain the third titration endpoint;

[0141] Using the third titration endpoint to verify the first titration endpoint;

[0142] When the verification is passed, the permanganate index value of the test sample is calculated according to the first titration endpoint;

[0143] When the verification fails, a second prompt message is generated.

[0144] Among them, the multiple dimensional parameters are the parameters that undergo a significant change at the titration endpoint, and the titration endpoint can be determined by the change in the parameters, and the first titration endpoint can be verified.

[0145] Exemplarily, when the titration signal data is a dual platinum electrode signal, the multiple dimensional parameters may include: ORP potential, light intensity, and temperature. When the titration signal data is ORP potential, the multiple dimensional parameters may include: dual platinum electrode signal, light intensity, and temperature. When the titration signal data is light intensity, the multiple dimensional parameters may include: dual platinum electrode signal, ORP potential, and temperature.

[0146] When the multiple parameters meet the preset conditions, it is determined that the titration reaches the endpoint to obtain a third titration endpoint.

[0147] It can be seen that in the embodiment of the present application, multiple dimensional parameters are used to assist in verifying the first titration endpoint, which reduces the misjudgment caused by interference with a single parameter and improves the accuracy and reliability of determining the titration endpoint.

[0148] Of course, it is understandable that the accuracy of the first titration end point determined by the model can be verified by one or two verification models of the second titration end point determined by curve analysis and the third titration end point determined by multi-dimensional parameters. The final titration end point can be determined by one successful verification, or the titration end point can be determined only when both verifications are successful. The embodiment of the present application is no longer limited, and the user can set it according to actual conditions.

[0149] A possible implementation of the embodiment of the present application, 7. The method according to claim 6, characterized in that, after controlling the target instrument to use an excess of potassium permanganate solution for back titration, before obtaining the multi-dimensional parameters obtained by the back titration collected by the target instrument, further comprises:

[0150] Acquire multiple initial dimension parameters during the back titration process of test samples of various working conditions in history;

[0151] According to whether the parameter change of the initial dimension parameter is greater than the preset change threshold at the titration endpoint;

[0152] All target dimension parameters are taken as multiple dimension parameters, wherein the target dimension parameters are initial dimension parameters that are greater than a preset change threshold.

[0153] Among them, the preset change threshold can be set according to actual needs, as long as the dimensional parameter can be determined from multiple initial dimensional parameters.

[0154] Based on any of the above embodiments, by comparing the accuracy and precision of the method of determining the first titration endpoint by the model involved in the embodiments of the present application and the method of determining the second titration endpoint by curve analysis with the prior art, see Table 1, it can be seen that the method of determining the first titration endpoint by the model involved in the embodiments of the present application and the method of determining the second titration endpoint by curve analysis have excellent accuracy and precision.

[0155] Table 1

[0156]

[0157] In summary, this application provides a more robust and comprehensive solution for the improvement of the endpoint recognition method of the permanganate index instrument. Although the traditional endpoint recognition methods such as ORP potential mutation or light intensity mutation are simple and direct, they are prone to misjudgment in complex environments or with interference factors, and it is difficult to trace the abnormal details during the titration process. The core of the present invention is to introduce the method of "timing and quantitative titration mode plus post-algorithm judgment", which has the following significant advantages:

[0158] (1) Enhanced anti-interference ability: By performing titration at a fixed time and in a quantitative manner, the system reduces the possibility of misjudgment of immediate reactions due to external interference. The post-judgment mode allows the system to conduct a comprehensive analysis based on the data of the entire titration process after the titration is completed, thereby more accurately identifying the endpoint and effectively reducing the misjudgment rate.

[0159] (2) Improve data reliability: Completely record the signal changes during the titration process, which not only provides rich data support for the identification of the endpoint, but also enables any abnormal data to be recorded and traced. This helps to quickly locate the problem when the data is abnormal and reduce decision-making errors caused by insufficient data information.

[0160] (3) Improve maintenance efficiency and reduce difficulty: Since the entire titration process can be traced, when the system fails or the data is abnormal, maintenance personnel can quickly locate the cause of the problem and take targeted solutions. Compared with the traditional "trial and error" maintenance, this data-driven maintenance method not only improves efficiency, but also reduces maintenance difficulty and cost.

[0161] (4) Flexibility and adaptability: Post-judgment modes such as image recognition or mathematical fitting enable the system to make adaptive adjustments based on the characteristics of different samples. This flexibility ensures the stability and accuracy of the system under different environments and conditions.

[0162] In summary, the endpoint identification method of the present invention has significant advantages and potential in the application of permanganate index instruments. It not only improves the accuracy and reliability of monitoring results, but also reduces maintenance costs and difficulties, providing strong support for technological progress in the field of water quality monitoring.

[0163] The following is an introduction to a device provided in an embodiment of the present application. The device described below and the method described above can be referred to each other. The device 200 of this embodiment is set in an electronic device. Figure 4 , Figure 4 It is a structural block diagram of a device of one embodiment of the present application, comprising:

[0164] An acquisition module 210 is used to acquire titration signal data obtained by the target instrument after the target instrument is controlled to perform back titration using an excess of potassium permanganate solution, wherein a timed and quantitative equal titration mode is adopted during the back titration;

[0165] An image drawing module 220 is used to draw an image according to the titration signal data to obtain a signal image with time as the horizontal axis and signal value as the vertical axis;

[0166] The titration endpoint determination module 230 is used to input the signal image into the permanganate index endpoint recognition model to obtain the first titration endpoint of the test sample; the permanganate index endpoint recognition model is obtained by training the network model based on the training samples, and the training samples include: multiple back titration signal training images and corresponding titration endpoint labels;

[0167] The calculation module 240 is used to calculate the permanganate index value of the test sample according to the first titration endpoint.

[0168] In one achievable manner, the present invention further includes:

[0169] Model training module, used to obtain training samples of various working conditions;

[0170] Input the back titration signal training image into the network model to perform endpoint recognition and obtain the training titration endpoint;

[0171] Determine the loss function value of the network model based on the training titration endpoint and the titration endpoint label;

[0172] The parameters of the network model are corrected according to the loss function value, and the network model is trained based on the training samples, and the network parameters of the network model are updated until the loss function value of the network model is no greater than the preset loss threshold, thereby obtaining a permanganate index endpoint recognition model.

[0173] In one achievable manner, the present invention further includes:

[0174] The test module is used to obtain a test sample; the test sample includes: a titration signal test image and a corresponding titration endpoint label;

[0175] Using the titration signal test image, multiple measurements are performed through the permanganate index endpoint recognition model to obtain multiple test titration endpoints;

[0176] Calculate accuracy and precision based on multiple test titration endpoints and titration endpoint labels;

[0177] When the accuracy is less than the preset accuracy threshold and the precision is less than the preset precision threshold, the permanganate index endpoint recognition model is used as the final permanganate index endpoint recognition model;

[0178] Otherwise, the permanganate index endpoint recognition model is trained again until a permanganate index endpoint recognition model is obtained with an accuracy less than a preset accuracy threshold and a precision less than a preset precision threshold.

[0179] In one achievable manner, the titration endpoint determination module is further used for:

[0180] Generate a target curve based on the titration signal data;

[0181] Perform peak search analysis on the target curve to obtain characteristic information, which includes the existence of a very small peak and the peak position corresponding to the very small peak; determine the second titration endpoint of the test sample based on the characteristic information;

[0182] Using the second titration endpoint to verify the first titration endpoint;

[0183] The device also includes:

[0184] The first execution module is used to calculate the permanganate index value of the test sample according to the first titration endpoint when the verification is passed; and generate the first prompt information when the verification fails.

[0185] In an achievable manner, the titration endpoint determination module is further used to: perform smoothing processing according to the titration signal data to obtain a smooth curve; and perform peak search analysis on the smooth curve to obtain first characteristic information, wherein the first characteristic information includes the existence of the first minimum peak and the peak position corresponding to the first minimum peak;

[0186] Performing differential processing on the titration signal data to obtain a differential curve; and performing peak search analysis on the differential curve to obtain second characteristic information, the second characteristic information including the existence of the second minimum peak and the peak position corresponding to the second minimum peak;

[0187] Determine a second titration endpoint of the potassium permanganate solution based on at least one of the first characteristic information and the second characteristic information.

[0188] In an achievable manner, the titration endpoint determination module is further used to: obtain multiple dimensional parameters obtained by back titration collected by the target instrument after the target instrument is controlled to use an excess of potassium permanganate solution for back titration;

[0189] Predicting the titration endpoint according to multiple dimensional parameters to obtain the third titration endpoint;

[0190] Using the third titration endpoint to verify the first titration endpoint;

[0191] The device also includes:

[0192] A second execution module is used to calculate the permanganate index value of the test sample according to the first titration endpoint when the verification is passed;

[0193] When the verification fails, a second prompt message is generated.

[0194] In one achievable manner, the device further includes:

[0195] A multiple initial dimension parameter acquisition module is used to acquire multiple initial dimension parameters in the back titration process of test samples of various working conditions in history;

[0196] A judgment module is used to determine whether the parameter change of the initial dimension parameter is greater than a preset change threshold at the titration endpoint;

[0197] The multiple dimensional parameter module is used to treat all target dimensional parameters as multiple dimensional parameters, wherein the target dimensional parameters are initial dimensional parameters greater than a preset change threshold.

[0198] An electronic device is provided in an embodiment of the present application, such as Figure 5 As shown, Figure 5The electronic device 300 shown includes: a processor 301 and a memory 303. The processor 301 and the memory 303 are connected, such as through a bus 302. Optionally, the electronic device 300 may also include a transceiver 304. It should be noted that in actual applications, the transceiver 304 is not limited to one, and the structure of the electronic device 300 does not constitute a limitation on the embodiments of the present application.

[0199] The processor 301 may be a CPU (Central Processing Unit), a general-purpose processor, a DSP (Digital Signal Processor), an ASIC (Application Specific Integrated Circuit), an FPGA (Field Programmable Gate Array) or other programmable logic devices, transistor logic devices, hardware components or any combination thereof. It may implement or execute various exemplary logic blocks, modules and circuits described in conjunction with the disclosure of this application. The processor 301 may also be a combination that implements computing functions, such as a combination of one or more microprocessors, a combination of a DSP and a microprocessor, etc.

[0200] The bus 302 may include a path to transmit information between the above components. The bus 302 may be a PCI (Peripheral Component Interconnect) bus or an EISA (Extended Industry Standard Architecture) bus. The bus 302 may be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 5 Only one thick line is used in the diagram, but this does not mean that there is only one bus or only one type of bus.

[0201] The memory 303 may be a ROM (Read Only Memory) or other types of static storage devices that can store static information and instructions, a RAM (Random Access Memory) or other types of dynamic storage devices that can store information and instructions, or an EEPROM (Electrically Erasable Programmable Read Only Memory), a CD-ROM (Compact Disc Read Only Memory) or other optical disk storage, optical disk storage (including compressed optical disk, laser disk, optical disk, digital versatile disk, Blu-ray disk, etc.), a magnetic disk storage medium or other magnetic storage device, or any other medium that can be used to carry or store the desired program code in the form of instructions or data structures and can be accessed by a computer, but is not limited thereto.

[0202] The memory 303 is used to store the application code for executing the solution of the present application, and the execution is controlled by the processor 301. The processor 301 is used to execute the application code stored in the memory 303 to implement the contents shown in the above method embodiment.

[0203] Figure 5 The electronic device shown is merely an example and should not bring any limitation to the functions and scope of use of the embodiments of the present application.

[0204] An embodiment of the present application provides a computer-readable storage medium, on which a computer program is stored. When the computer-readable storage medium is run on a computer, the computer can execute the corresponding content in the aforementioned method embodiment.

[0205] An embodiment of the present application provides a computer program product, including a computer program, which implements the corresponding contents of the aforementioned method embodiment when the computer program is executed by a processor.

[0206] It should be understood that, although the steps in the flowchart of the accompanying drawings are displayed in sequence as indicated by the arrows, these steps are not necessarily executed in sequence in the order indicated by the arrows. Unless otherwise specified herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least a part of the steps in the flowchart of the accompanying drawings may include multiple sub-steps or multiple stages, and these sub-steps or stages are not necessarily executed at the same time, but can be executed at different times, and their execution order is not necessarily sequential, but can be executed in turn or alternately with other steps or at least a part of the sub-steps or stages of other steps.

[0207] The above are only some implementation methods of the present application. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present application. These improvements and modifications should also be regarded as the scope of protection of the present application.

Claims

1. An AI intelligent permanganate index detection method, characterized in that: include: After the target instrument is controlled to perform back titration using an excess of potassium permanganate solution, titration signal data obtained by the back titration collected by the target instrument is obtained, wherein a timed and quantitative equal titration mode is used during the back titration; Draw an image according to the titration signal data to obtain a signal image with time as the horizontal axis and signal value as the vertical axis; Inputting the signal image into a permanganate index endpoint recognition model to obtain a first titration endpoint of the test sample; the permanganate index endpoint recognition model is obtained by training a network model based on training samples, and the training samples include: a plurality of back titration signal training images and corresponding titration endpoint labels; According to the first titration endpoint, the permanganate index value of the test sample is calculated.

2. The AI ​​intelligent permanganate index detection method according to claim 1, characterized in that: The training process of the permanganate index endpoint recognition model includes: Obtain training samples of various working condition types; Input the back titration signal training image into the network model to perform endpoint recognition and obtain the training titration endpoint; Determine the loss function value of the network model based on the training titration endpoint and the titration endpoint label; The parameters of the network model are corrected according to the loss function value, and the network model is trained based on the training samples, and the network parameters of the network model are updated until the loss function value of the network model is no greater than the preset loss threshold, thereby obtaining a permanganate index endpoint recognition model.

3. The AI ​​intelligent permanganate index detection method according to claim 2, characterized in that: The network model is trained based on the training samples, and the network parameters of the network model are updated until the loss function value of the network model is not greater than the preset loss threshold, and the permanganate index endpoint recognition model is obtained, and the method further includes: Acquire a test sample; the test sample includes: a titration signal test image and a corresponding titration endpoint label; Using the titration signal test image, multiple measurements are performed through the permanganate index endpoint recognition model to obtain multiple test titration endpoints; Calculate accuracy and precision based on multiple test titration endpoints and titration endpoint labels; When the accuracy is less than the preset accuracy threshold and the precision is less than the preset precision threshold, the permanganate index endpoint recognition model is used as the final permanganate index endpoint recognition model; Otherwise, the permanganate index endpoint recognition model is trained again until a permanganate index endpoint recognition model is obtained with an accuracy less than a preset accuracy threshold and a precision less than a preset precision threshold.

4. The AI ​​intelligent permanganate index detection method according to claim 1, characterized in that: After inputting the signal image into the permanganate index endpoint recognition model to obtain the first titration endpoint of the test sample, the method further includes: generating a target curve according to the titration signal data; Performing peak search analysis on the target curve to obtain characteristic information, wherein the characteristic information includes the existence of a minimum peak and the peak position corresponding to the minimum peak; determining a second titration endpoint of the test sample according to the characteristic information; Verifying the first titration endpoint using the second titration endpoint; When the verification is passed, calculating the permanganate index value of the test sample according to the first titration endpoint; When the verification fails, a first prompt message is generated.

5. The AI ​​intelligent permanganate index detection method according to claim 4, characterized in that: Performing peak search analysis on the target curve to obtain characteristic information, wherein the characteristic information includes the existence of a minimum peak and the peak position corresponding to the minimum peak; determining a second titration endpoint of the test sample according to the characteristic information, comprising: Performing smoothing processing on the titration signal data to obtain a smooth curve; and performing peak search analysis on the smooth curve to obtain first characteristic information, wherein the first characteristic information includes the existence of a first minimum peak and a peak position corresponding to the first minimum peak; Performing differential processing on the titration signal data to obtain a differential curve; and performing peak search analysis on the differential curve to obtain second characteristic information, wherein the second characteristic information includes the existence of a second minimum peak and a peak position corresponding to the second minimum peak; A second titration endpoint of the potassium permanganate solution is determined according to at least one of the first characteristic information and the second characteristic information.

6. The AI ​​intelligent permanganate index detection method according to claim 1, characterized in that: After inputting the signal image into the permanganate index endpoint recognition model to obtain the first titration endpoint of the test sample, the method further includes: After the target instrument is controlled to perform back titration using an excess of potassium permanganate solution, multiple dimensional parameters obtained by back titration collected by the target instrument are obtained; Predicting the titration endpoint according to multiple dimensional parameters to obtain the third titration endpoint; Verifying the first titration endpoint using the third titration endpoint; When the verification is passed, calculating the permanganate index value of the test sample according to the first titration endpoint; When the verification fails, a second prompt message is generated.

7. The AI ​​intelligent permanganate index detection method according to claim 6, characterized in that: After the target instrument is controlled to use an excess of potassium permanganate solution for back titration, before obtaining the multi-dimensional parameters obtained by the back titration collected by the target instrument, it also includes: Acquire multiple initial dimension parameters during the back titration process of test samples of various working conditions in history; According to whether the parameter change of the initial dimension parameter is greater than the preset change threshold at the titration endpoint; All target dimensional parameters are used as the multiple dimensional parameters, wherein the target dimensional parameters are initial dimensional parameters that are greater than a preset change threshold.

8. An AI intelligent permanganate index detection device, characterized in that: include: An acquisition module is used to acquire titration signal data obtained by back titration collected by the target instrument after the target instrument is controlled to perform back titration with an excess of potassium permanganate solution, wherein a timed and quantitative equal titration mode is adopted during the back titration; An image drawing module is used to draw an image according to the titration signal data to obtain a signal image with time as the horizontal axis and signal value as the vertical axis; A titration endpoint determination module is used to input the signal image into a permanganate index endpoint recognition model to obtain a first titration endpoint of the test sample; the permanganate index endpoint recognition model is obtained by training a network model based on training samples, and the training samples include: a plurality of back titration signal training images and corresponding titration endpoint labels; A calculation module is used to calculate the permanganate index value of the test sample according to the first titration endpoint.

9. An electronic device, characterized in that: include: one or more processors; Memory; One or more applications, wherein the one or more applications are stored in the memory and configured to be executed by the one or more processors, and the one or more applications are configured to: execute the steps of the AI ​​intelligent permanganate index detection method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: The storage medium stores at least one instruction, at least one program, code set or instruction set, and the at least one instruction, at least one program, code set or instruction set is loaded by the processor and executes the steps of the AI ​​intelligent permanganate index detection method according to any one of claims 1 to 7.

Images