A tap water quality detection and analysis instrument system

Abstract

The present invention discloses a tap water quality detection analyzer system, which belongs to the field of water quality detection technology, and solves the problem that when the flowing tap water is detected by the analyzer, a relatively large amount of data will be brought, which will cause the analyzer system to operate slowly, and the flowing tap water will have a significant impact on the turbidity, microbial content, and dissolved oxygen content of the tap water due to its flow rate and water temperature. The continuously flowing tap water is detected; a random number algorithm is used to randomly extract multiple data points, and the similarity of the multiple data points is compared to screen out a few data points and retain the majority data points; the flow rate V, water temperature T, and turbidity N of the tap water are extracted from the majority data points. The present invention can randomly extract the data after the detection, reduce the load in the operation process of the system, and at the same time adjust the influence of the flowing tap water and water temperature on the turbidity, microbial content, and dissolved oxygen content of the tap water, so that the final output detection result will not be affected.

Description

Technical Field

[0001] The invention relates to the technical field of water quality detection, and in particular to a tap water quality detection and analysis instrument system. Background Art

[0002] Tap water quality testing is a key public health safeguard to ensure that water supply is safe and meets health standards. By regularly testing water sources, factors that may affect water quality, such as microbial contamination, chemical pollutants, suspended solids and dissolved substances, can be discovered and treated in a timely manner. Testing indicators usually include turbidity, microbial content, dissolved oxygen content, pH value, heavy metal content, etc. Modern water quality testing technology covers everything from traditional laboratory testing to real-time online monitoring systems. These technologies use a variety of sensors and analytical instruments to evaluate water quality parameters. They are highly automated and can provide continuous and real-time data to help monitoring personnel respond quickly to changes in water quality. In addition, water quality testing also involves the collection, analysis and reporting of data, which are essential for formulating water quality management strategies and improving water treatment processes. With the advancement of science and technology, water quality testing is moving towards a more accurate and efficient direction to protect public health and environmental safety.

[0003] As an indispensable resource in daily life, tap water has passed strict testing and treatment by water plants before flowing through thousands of households to ensure that it basically meets health and safety standards. However, in the field of industrial production, especially in the textile industry, which has special requirements for water quality, it is far from enough to rely solely on water plant treatment. The requirements for water quality in the textile printing and dyeing process are extremely strict. Not only must the turbidity be controlled at an extremely low level to ensure the purity and uniformity of the color of the fabric, but the microbial content must also be maintained within a safe range to prevent the fabric from being contaminated during the production process. In addition, the level of dissolved oxygen content also directly affects the stability of the printing and dyeing effect and the durability of the fabric. Therefore, in order to ensure product quality, many textile manufacturers will purchase additional professional tap water quality analysis instruments to regularly conduct detailed tests on the water quality entering the production process. These tests usually include the collection of water samples, and the determination of multiple key indicators such as turbidity, microbial content, and dissolved oxygen content through analytical instruments to ensure that each batch of tap water meets the strict standards required for production. Current testing methods often involve manual sampling, placing tap water into special containers, and then using analyzers for in-depth analysis. Although this process is cumbersome, it is crucial to ensuring product quality.

[0004] However, for tap water, there is a significant difference between the tap water that just flows out of the tap and the tap water that continues to flow out. Therefore, the continuously flowing tap water is tested. However, there are some problems that will affect the final output of the water quality test results. On the one hand, the flowing tap water will bring a large amount of data when it is tested by the analyzer, which will cause the analyzer system to operate slowly. On the other hand, the flowing tap water will have a significant impact on the turbidity, microbial content and dissolved oxygen content of the tap water due to its flow rate and water temperature, which will cause the final output result to be inaccurate.

[0005] Therefore, a tap water quality detection and analysis instrument system is proposed to solve or alleviate the above problems. Summary of the invention

[0006] The purpose of the present invention is to solve the shortcomings of the prior art and to propose a tap water quality detection and analysis instrument system.

[0007] In order to achieve the above object, the present invention adopts the following technical solutions:

[0008] A tap water quality detection and analysis instrument system comprises the following steps:

[0009] Testing of continuously running tap water;

[0010] Use random number algorithm to randomly extract multiple data points, compare the similarity of multiple data points, filter out a few data points, and retain the majority data points;

[0011] Extract the water quality characteristics of flow rate V, water temperature T, turbidity N, microbial content C and dissolved oxygen content DO of tap water from public data points;

[0012] The water quality characteristics are normalized and the flow rate V and water temperature T in the water quality characteristics are selected as input characteristics, and the turbidity N, microbial content C and dissolved oxygen content DO are selected as output characteristics;

[0013] The water quality characteristics are used as a training set and imported into the neural network model for training, and finally an adjustment model is output. The adjustment model includes , where n is the number of neurons in the hidden layer of the neural network model, is the weight from the i-th neuron in the previous layer to the k-th neuron in the L-th layer, is the bias of the kth neuron in the Lth layer, is the activation function used in layer L;

[0014] Finally, the input characteristics of water temperature T being normal temperature and flow rate V being 0 are input into the adjustment model and the adjusted turbidity N, microbial content C, and dissolved oxygen content DO are output respectively.

[0015] Preferably, the method further comprises the following steps:

[0016] After changing the continuous flow rate of tap water, a random number algorithm is used to randomly extract multiple supplementary data points, and the similarity of multiple supplementary data points is compared to screen out a few supplementary data points and retain the majority of supplementary data points;

[0017] The flow rate V, water temperature T, turbidity N, microbial content C and dissolved oxygen content DO of tap water were extracted from the public supplementary data points to supplement the water quality characteristics;

[0018] Normalize the supplementary water quality characteristics and select the flow rate V and water temperature T in the supplementary water quality characteristics as input features, and select turbidity N, microbial content C and dissolved oxygen content DO;

[0019] The supplementary water quality characteristics are imported as training sets into the neural network model for training.

[0020] Preferably, the output adjustment model comprises the following steps:

[0021] The flow rate V and water temperature T are used as input features through weights Weighted sum;

[0022] Plus the bias term of the corresponding neuron ;

[0023] The result of weighted summation is passed through The activation function performs nonlinear transformation to obtain the final activation output .

[0024] Preferably, the neural network model comprises an input layer, a hidden layer, and an output layer, the input layer has 2 nodes, each hidden layer has n nodes, and the output layer has 3 nodes.

[0025] Preferably, the adjusted turbidity , is the weight matrix corresponding to turbidity N, which is related to the input feature Do matrix multiplication.

[0026] Preferably, the adjusted microbial content , is the weight matrix corresponding to the microbial content C, which is related to the input feature Do matrix multiplication.

[0027] Preferably, the adjusted dissolved oxygen content , is the weight matrix corresponding to the microbial content C, which is related to the input feature Do matrix multiplication.

[0028] Preferably, the step of randomly extracting multiple data points using a random number algorithm comprises the following steps:

[0029] The basic data points R obtained from the continuous flowing tap water detection are used as a library;

[0030] For i from 1 to n it is defined as , Represents the i-th position of the basic data point R, stores the i-th data point, For data stream;

[0031] Generate a random number j ranging from 1 to i: jU(1,i), where U(1,i) represents a random number uniformly distributed in the range 1 to i;

[0032] If j ≤ n, replace the jth element in the base data point R with the current data point: , For data stream;

[0033] The elements finally retained in the basic data point R are the multiple randomly extracted data points.

[0034] Preferably, the similarity comparison of the plurality of data points, screening out a minority of data points, and retaining the majority of data points comprises the following steps:

[0035] Convert data points to data arrays;

[0036] Use the Frobenius norm to calculate the similarity of adjacent data arrays and associate the similarity with the data points;

[0037] Use similarity comparison thresholds to classify data points;

[0038] A similarity graph is constructed using the similarity between each pair of data points. It is an undirected graph where the nodes represent data points and the weight of the edge is the similarity.

[0039] Use graph clustering algorithm to perform cluster analysis on similarity graph and classify similar data points into the same category;

[0040] Count the number of data points in all categories, and filter out the category with the largest number of data points as the popular data point.

[0041] The present invention has the following beneficial effects:

[0042] After completing the water quality test, the present invention can select data through a random sampling method, effectively reducing the system's computing pressure when processing a large amount of data. In addition, the present invention can also adjust and calibrate the flow rate of tap water and the influence of its water temperature on water quality indicators, including turbidity, microbial content and dissolved oxygen content. Through this adjustment, the accuracy and reliability of the test results are ensured and are not affected by fluctuations in data volume or environmental changes. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the embodiments. It should be understood that the following drawings only show certain embodiments of the present invention and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without creative work.

[0044] Figure 1 It is a structural block diagram of the present invention. DETAILED DESCRIPTION

[0045] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Generally, the components of the embodiments of the present invention described and shown in the drawings here can be arranged and designed in various different configurations.

[0046] Therefore, the following detailed description of the embodiments of the present invention provided in the accompanying drawings is not intended to limit the scope of the invention claimed for protection, but merely represents selected embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0047] It should be noted that similar reference numerals and letters denote similar items in the following drawings, and therefore, once an item is defined in one drawing, further definition and explanation thereof is not required in subsequent drawings.

[0048] In the description of the present invention, it should be understood that the terms "center", "up", "down", "left", "right", "vertical", "horizontal", "inside", "outside", etc. indicate orientations or positional relationships based on the orientations or positional relationships shown in the accompanying drawings, or are the orientations or positional relationships in which the inventive product is conventionally placed when in use, or are the orientations or positional relationships conventionally understood by those skilled in the art. They are only for the convenience of describing the present invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be understood as a limitation on the present invention.

[0049] Furthermore, the terms “first”, “second”, “third”, etc. are merely used for distinguishing descriptions and are not to be understood as indicating or implying relative importance.

[0050] In the description of the present invention, it is also necessary to explain that, unless otherwise clearly specified and limited, the terms "set", "install", "connect", and "connect" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be indirectly connected through an intermediate medium, or it can be the internal communication of two elements. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0051] A tap water quality detection and analysis instrument system, such as Figure 1 As shown, the following steps are included:

[0052] Testing of continuously running tap water;

[0053] Use random number algorithm to randomly extract multiple data points, compare the similarity of multiple data points, filter out a few data points, and retain the majority data points;

[0054] Extract the water quality characteristics of flow rate V, water temperature T, turbidity N, microbial content C and dissolved oxygen content DO of tap water from public data points;

[0055] The water quality characteristics are normalized and the flow rate V and water temperature T in the water quality characteristics are selected as input characteristics, and the turbidity N, microbial content C and dissolved oxygen content DO are selected as output characteristics;

[0056] The water quality characteristics are used as a training set and imported into the neural network model for training, and finally an adjustment model is output. The adjustment model includes , where n is the number of neurons in the hidden layer of the neural network model, is the weight from the i-th neuron in the previous layer to the k-th neuron in the L-th layer, is the bias of the kth neuron in the Lth layer, is the activation function used in layer L;

[0057] Finally, the input characteristics of water temperature T being normal temperature and flow rate V being 0 are input into the adjustment model and the adjusted turbidity N, microbial content C, and dissolved oxygen content DO are output respectively.

[0058] By using a random number algorithm to extract data points from continuously flowing tap water and screening out representative data points through similarity comparison, the problem of slow system operation caused by the huge amount of data in the process of testing flowing tap water by the water quality analyzer is effectively solved. The system further normalizes the water quality characteristics in the extracted data points, trains them through a neural network model, and outputs an adjustment model to ensure the accuracy and reliability of the test results. This method not only reduces the system's computing burden, but also improves the efficiency and quality of data processing through intelligent algorithms, making the test results more accurate and meeting the strict requirements of different industrial fields for water quality, especially the textile printing and dyeing industry, which has extremely high requirements for water quality.

[0059] Preferably, the method further comprises the following steps:

[0060] After changing the continuous flow rate of tap water, a random number algorithm is used to randomly extract multiple supplementary data points, and the similarity of multiple supplementary data points is compared to screen out a few supplementary data points and retain the majority of supplementary data points;

[0061] The flow rate V, water temperature T, turbidity N, microbial content C and dissolved oxygen content DO of tap water were extracted from the public supplementary data points to supplement the water quality characteristics;

[0062] Normalize the supplementary water quality characteristics and select the flow rate V and water temperature T in the supplementary water quality characteristics as input features, and select turbidity N, microbial content C and dissolved oxygen content DO;

[0063] The supplementary water quality characteristics are imported as training sets into the neural network model for training.

[0064] By changing the flow rate of tap water and re-extracting supplementary data points, the system can adapt to different water flow conditions, ensuring accurate and reliable water quality data in all situations. This approach makes the system more adaptable and flexible, able to cope with the challenges brought by changes in tap water flow rate, thereby providing more comprehensive and accurate water quality analysis results. In addition, by incorporating supplementary data points into the training set, the neural network model can learn more abundant supplementary water quality characteristics, further improving the model's predictive ability and the accuracy of water quality detection.

[0065] Preferably, the output adjustment model comprises the following steps:

[0066] The flow rate V and water temperature T are used as input features through weights Weighted sum;

[0067] Plus the bias term of the corresponding neuron ;

[0068] The result of weighted summation is passed through The activation function performs nonlinear transformation to obtain the final activation output .

[0069] The output adjustment model uses flow rate and water temperature as input features, and generates adjusted water quality indicators through weighted summation, bias term addition, and nonlinear transformation of activation functions. The advantage of this process is that it can relatively accurately simulate the impact of water flow characteristics on water quality indicators and make effective adjustments through the model. This not only improves the scientificity and systematicness of water quality test results, but also enhances the predictability and controllability of test results. Even when the flow rate and water temperature of tap water change, it can ensure that the output turbidity, microbial content, and dissolved oxygen content test results remain stable and accurate, meeting high-standard water quality testing requirements.

[0070] Preferably, the neural network model comprises an input layer, a hidden layer, and an output layer, the input layer has 2 nodes, each hidden layer has n nodes, and the output layer has 3 nodes.

[0071] Preferably, the adjusted turbidity , is the weight matrix corresponding to turbidity N, which is related to the input feature Do matrix multiplication;

[0072] The adjusted microbial content , is the weight matrix corresponding to the microbial content C, which is related to the input feature Do matrix multiplication;

[0073] The adjusted dissolved oxygen content , is the weight matrix corresponding to the microbial content C, which is related to the input feature Do matrix multiplication.

[0074] Through the matrix multiplication operation of the weight matrix and the input features, the turbidity, microbial content and dissolved oxygen content in the tap water quality are adjusted. The advantage of this method is that it can accurately quantify the complex relationship between each water quality index and the flow rate and water temperature, thereby providing a water quality prediction value. This method not only improves the accuracy of water quality detection, but also enhances the adaptability of the model to water quality changes under different environmental conditions. In addition, through the unified weight matrix processing, the consistency of the adjustment of each water quality index is guaranteed, and the efficiency and effect of the detection process are further optimized.

[0075] Preferably, the step of randomly extracting multiple data points using a random number algorithm comprises the following steps:

[0076] The basic data points R obtained from the continuous flowing tap water detection are used as a library;

[0077] For i from 1 to n it is defined as , Represents the i-th position of the basic data point R, stores the i-th data point, For data stream;

[0078] Generate a random number j ranging from 1 to i: jU(1,i), where U(1,i) represents a random number uniformly distributed in the range 1 to i;

[0079] If j ≤ n, replace the jth element in the base data point R with the current data point: , For data stream;

[0080] The elements finally retained in the basic data point R are the multiple randomly extracted data points.

[0081] The method of extracting data points using a random number algorithm effectively ensures the representativeness and randomness of data points by randomly selecting and replacing elements from the basic data point library. Without increasing the system computing burden as a major premise, key data points can be quickly screened out from a large amount of data, thereby providing a high-quality data set for subsequent data analysis and model training. In addition, by setting the range and replacement conditions for random number generation, the present invention can remove outliers and edge data points while ensuring the diversity of data points, thereby ensuring the quality and reliability of data points. This efficient data extraction and screening mechanism provides a solid data foundation for the accurate detection and analysis of tap water quality, and improves the efficiency and effectiveness of detection work.

[0082] Preferably, the similarity comparison of the plurality of data points, screening out a minority of data points, and retaining the majority of data points comprises the following steps:

[0083] Convert data points to data arrays;

[0084] Use the Frobenius norm to calculate the similarity of adjacent data arrays and associate the similarity with the data points;

[0085] Use similarity comparison thresholds to classify data points;

[0086] A similarity graph is constructed using the similarity between each pair of data points. It is an undirected graph where the nodes represent data points and the weight of the edge is the similarity.

[0087] Use graph clustering algorithm to perform cluster analysis on similarity graph and classify similar data points into the same category;

[0088] Count the number of data points in all categories, and filter out the category with the largest number of data points as the popular data point.

[0089] The method for performing similarity comparison and clustering on multiple data points realizes the quantification of the similarity between data points by converting the data points into data arrays and calculating the Frobenius norm of adjacent data arrays. In this way, the most representative popular data points can be identified and retained, while those abnormal points that are greatly different from the mainstream data points are excluded. By setting a similarity threshold and constructing a similarity graph, the present invention can effectively cluster similar data points, thereby ensuring the diversity of data points while enhancing the internal consistency of the data set.

[0090] This similarity-based clustering analysis not only improves the scientificity and rationality of data point selection, but also provides a more accurate and reliable data basis for subsequent data analysis and model training, ensuring the accuracy and stability of tap water quality test results.

[0091] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A tap water quality detection and analysis instrument system, comprising the following steps: Testing of continuously running tap water; Use random number algorithm to randomly extract multiple data points, compare the similarity of multiple data points, filter out a few data points, and retain the majority data points; Extract the water quality characteristics of flow rate V, water temperature T, turbidity N, microbial content C and dissolved oxygen content DO of tap water from public data points; The water quality characteristics are normalized and the flow rate V and water temperature T in the water quality characteristics are selected as input characteristics, and the turbidity N, microbial content C and dissolved oxygen content DO are selected as output characteristics; The water quality characteristics are used as a training set and imported into the neural network model for training, and finally an adjustment model is output. The adjustment model includes ,in, n is the number of neurons in the hidden layer of the neural network model, is the weight from the i-th neuron in the previous layer to the k-th neuron in the L-th layer, is the bias of the kth neuron in the Lth layer, is the activation function used in layer L; Finally, the input characteristics of water temperature T being normal temperature and flow rate V being 0 are input into the adjustment model and the adjusted turbidity N, microbial content C, and dissolved oxygen content DO are output respectively; The adjusted turbidity , is the weight matrix corresponding to turbidity N, which is related to the input feature Do matrix multiplication; the adjusted microbial content , is the weight matrix corresponding to the microbial content C, which is related to the input feature Do matrix multiplication; the adjusted dissolved oxygen content , Corresponding to the dissolved oxygen content The weight matrix, which is related to the input features Do matrix multiplication, where is the neuronal bias corresponding to turbidity, is the neuronal bias corresponding to microbial content, for the neuronal bias corresponding to dissolved oxygen content; The method of randomly extracting multiple data points using a random number algorithm comprises the following steps: The basic data points R obtained from the continuous flowing tap water detection are used as a library; For i from 1 to n it is defined as , Represents the i-th position of the basic data point R, stores the i-th data point, For data stream; Generate a random number j ranging from 1 to i: jU(1,i), where U(1,i) represents a random number uniformly distributed in the range 1 to i; If j ≤ n, replace the jth element in the base data point R with the current data point: , For data stream; The elements finally retained in the basic data point R are the multiple randomly extracted data points; The similarity comparison of the plurality of data points, screening out a minority of data points, and retaining the majority of data points comprises the following steps: Convert data points to data arrays; Use the Frobenius norm to calculate the similarity of adjacent data arrays and associate the similarity with the data points; Use similarity comparison thresholds to classify data points; A similarity graph is constructed using the similarity between each pair of data points. It is an undirected graph where the nodes represent data points and the weight of the edge is the similarity. Use graph clustering algorithm to perform cluster analysis on similarity graph and classify similar data points into the same category; Count the number of data points in all categories, and filter out the category with the largest number of data points as the popular data point.

2. A tap water quality detection and analysis instrument system according to claim 1, characterized in that: It also includes the following steps, After changing the continuous flow rate of tap water, a random number algorithm is used to randomly extract multiple supplementary data points, and the similarity of multiple supplementary data points is compared to screen out a few supplementary data points and retain the majority of supplementary data points; The flow rate V, water temperature T, turbidity N, microbial content C and dissolved oxygen content DO of tap water were extracted from the public supplementary data points to supplement the water quality characteristics; Normalize the supplementary water quality characteristics and select the flow rate V and water temperature T in the supplementary water quality characteristics as input features, and select turbidity N, microbial content C and dissolved oxygen content DO; The supplementary water quality characteristics are imported as training sets into the neural network model for training.

3. A tap water quality detection and analysis instrument system according to claim 1, characterized in that: The output adjustment model comprises the following steps: The flow rate V and water temperature T are used as input features through weights Weighted sum; Plus the bias term of the corresponding neuron ; The result of weighted summation is passed through The activation function performs nonlinear transformation to obtain the final activation output .

4. A tap water quality detection and analysis instrument system according to claim 1, characterized in that: The neural network model includes an input layer, a hidden layer, and an output layer. The input layer has 2 nodes, each hidden layer has n nodes, and the output layer has 3 nodes.

Images