A method and system for evaluating water quality of drinking water

Abstract

The present application relates to the field of digital data processing technology, and specifically to a method and system for evaluating the water quality of drinking water, which includes: obtaining the detection values ​​of various parameters at the water source of the site and various indicators at the water plant at each collection time; constructing a feature set, trend feature vector and feature description set of each indicator of the water plant every day; determining the contribution coefficient of each feature in the feature description set of each indicator; determining the weight coefficient of each indicator of the water plant; calculating the excess risk of each indicator every day; determining the water quality health index every day, and evaluating the water quality of drinking water. The present application can measure the degree of impact of water quality on the health of the population and improve the accuracy of water quality evaluation.

Description

Technical Field

[0001] The present application relates to the technical field of digital data processing, and in particular to a water quality assessment method and system for drinking water. Background Art

[0002] At present, the demand for drinking water is constantly increasing. The quality of drinking water is related to human health. In order to ensure healthy drinking water, people have high requirements for the quality of drinking water. Although there are indicators such as the Water Quality Index (WQI) to evaluate and analyze water quality, these indicators can only reflect the water quality of a single water body in a certain period of time, without considering the trend of water quality changes at different times and the comprehensive impact of the water source water body on the water quality of the water plant. In addition, it cannot reflect the impact of the quality of drinking water on human health. At present, the construction of a health risk assessment system for the impact of drinking water quality on the population at home and abroad has not yet matured, and there is a lack of a relatively complete set of drinking water-health risk assessment and early warning prediction technical indicators that can be promoted in many places. Summary of the invention

[0003] In order to solve the above technical problems, a method and system for evaluating the water quality of drinking water are provided to solve the existing problems.

[0004] The solution to the technical problem of this application is to provide a method and system for evaluating the quality of drinking water, including the following steps:

[0005] In a first aspect, an embodiment of the present application provides a method for assessing the quality of drinking water, the method comprising the following steps:

[0006] Obtain the detection values ​​of various parameters at the water source of the site and various indicators at the water plant at each collection time, and construct the feature set, trend feature vector and feature description set of each indicator of the water plant every day;

[0007] For the turbidity in each indicator of the water plant, the characteristic description set of turbidity of all days is clustered to obtain multiple clusters; the discreteness of each feature in the characteristic description set of turbidity within the same cluster, and the discreteness of each feature in the characteristic description set of turbidity between different clusters are analyzed, and the intra-class variance and inter-class variance of each feature in the characteristic description set of turbidity are calculated;

[0008] Determining a contribution coefficient of each feature in a feature description set of turbidity based on the intra-class variance and the inter-class variance;

[0009] Determine the turbidity weight coefficient according to the difference of elements in the characteristic vector of each parameter in the characteristic set of turbidity of the water plant every day and the contribution coefficient of the corresponding characteristic vector;

[0010] For other indicators in the water plant indicators, the turbidity weight coefficient determination method is adopted to obtain the weight coefficient of each indicator of the water plant;

[0011] Based on the weight coefficient of each indicator of the water plant and the detection value of each indicator, the excess risk of each indicator is calculated every day; the water quality health index is determined every day, and the water quality of drinking water is evaluated.

[0012] Preferably, the method for constructing the feature set, trend feature vector and feature description set of each indicator of the water plant every day is:

[0013] Collect the detection values ​​of parameters related to each indicator at the water source of each station every day, and construct the characteristic vectors of each parameter in the order of collection time;

[0014] The characteristic vectors of all relevant parameters of each indicator are combined into a characteristic set of each indicator of the water plant every day;

[0015] Filling processing is performed on each indicator of the water plant on two consecutive days, and a vector with the same dimension as the characteristic vector is filled, which is recorded as the trend characteristic vector of each indicator of the water plant on each day;

[0016] The feature vector of each parameter in the feature set and the trend feature vector are combined to form a feature description set of each indicator of the water plant every day.

[0017] Preferably, the calculation method of the intra-class variance of each feature in the feature description set of turbidity is:

[0018] The variance of each feature of the feature description set of turbidity in each cluster is recorded as the first variance;

[0019] The mean of the first variance of each feature of all clusters is used as the intra-class variance of each feature in the feature description set of turbidity.

[0020] Preferably, the method for calculating the inter-class variance of each feature in the turbidity feature description set is:

[0021] The mean of each feature in the feature description set of turbidity in each cluster is recorded as the first mean of each feature;

[0022] The variance of the first mean of each feature of all clusters is used as the inter-class variance of each feature in the feature description set of turbidity.

[0023] Preferably, the contribution coefficient of each feature in the feature description set for determining turbidity is the ratio between the between-class variance and the within-class variance.

[0024] Preferably, the determining of the turbidity weight coefficient comprises:

[0025] Each feature in the feature description set of turbidity is a feature vector and a trend feature vector of each parameter in the feature set, and a contribution coefficient of the feature vector of each parameter in the feature set of turbidity and a contribution coefficient of the trend feature vector of turbidity are determined;

[0026] The ratio between the contribution coefficient of the characteristic vector of each parameter in the turbidity characteristic set and the contribution coefficient of the trend characteristic vector of the turbidity is taken as the characteristic contribution rate of each parameter;

[0027] The mean of the differences of all adjacent elements in the feature vector of each parameter in the feature set of the water plant turbidity for each day is taken as the average relative change of each parameter in the feature set of the water plant turbidity for each day;

[0028] The product of the characteristic contribution rate and the average relative change is used as the water source detection influence coefficient of each parameter in the characteristic set of turbidity of the water plant every day;

[0029] The mean of the water source detection influence coefficients of all parameters in the feature set of the water plant's turbidity every day is used as the turbidity weight of the water plant every day; the mean of the turbidity weights of the water plant on all days is used as the turbidity weight coefficient.

[0030] Preferably, the calculation formula for the excess risk of each indicator every day is: ,in, For the water plant Day indicator excess risk, For water plant indicators The risk factor, For the water plant Day indicator The detection value of .

[0031] Preferably, the calculation method of the risk coefficient of each indicator in the water plant is:

[0032] The product of the test value of each indicator of the water plant every day and the weight coefficient of the corresponding indicator is used as the quality influence coefficient of each indicator of the water plant every day;

[0033] The water quality assessment index of the water plant every day is in a positive synergistic relationship with the quality influence coefficients of all indicators;

[0034] The product of the water quality assessment index and the detection value of each indicator of the water plant every day is used as the risk coefficient of each indicator of the water plant.

[0035] Preferably, the water quality health index is determined every day, and the calculation formula is: ,in For the Water quality health index of Tianshui Plant, is the number of indicators of the water plant, represents the maximum value function, The number of days for collection.

[0036] In a second aspect, an embodiment of the present application also provides a water quality assessment system for drinking water, comprising a memory, a processor, and a computer program stored in the memory and running on the processor, wherein when the processor executes the computer program, the steps of any one of the above-mentioned methods for assessing the water quality of drinking water are implemented.

[0037] This application has at least the following beneficial effects:

[0038] Based on the parameter data at the water source of the site, this application constructs a feature set, trend feature vector and feature description set of each indicator of the water plant every day, and performs clustering, calculates the contribution coefficient of each feature in the feature description set of each indicator, and determines the feature contribution rate of each parameter. The beneficial effect is that water quality with higher similarity is classified as one category, and the contribution rate of each parameter at the water source of the site to the clustering is measured; the weight coefficient of each indicator of the water plant is calculated according to the feature contribution rate, and the water quality assessment index is determined. The beneficial effect is that the influence of the detection value of each parameter at the water source of the site on the water quality of the water plant is taken into account, so as to better assess the water quality of the water plant; finally, the excess risk of each indicator is calculated, and a drinking water-health risk assessment model is constructed to calculate the daily water quality health index. The beneficial effect is that the influence of the content of each indicator on the water quality is taken into account, and the health degree of water quality is quantified, and the health degree of water quality is measured every day, so that the daily water quality assessment is more accurate. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] The following is a detailed description of a water quality assessment method for drinking water of the present application in conjunction with the accompanying drawings.

[0040] Figure 1 A flow chart of the steps of a method for assessing the quality of drinking water provided in this application;

[0041] Figure 2 A flowchart of the steps of the method for obtaining the contribution coefficient of each feature in the turbidity feature description set provided in the present application;

[0042] Figure 3 This is a flowchart of the steps of the method for obtaining the turbidity weight coefficient provided in this application. DETAILED DESCRIPTION

[0043] In order to make the purpose, technical solution and advantages of this application more clear, the following is a further detailed description of a water quality assessment method and system for drinking water proposed in this application in conjunction with the accompanying drawings and implementation examples. It should be understood that the specific embodiments described herein are only used to explain this application and are not used to limit this application.

[0044] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs.

[0045] See also Figure 1 , which shows a flow chart of a method for evaluating the quality of drinking water provided by an embodiment of the present application, the method comprising the following steps:

[0046] Step 1: Obtain the detection values ​​of various parameters at the water source of the site and various indicators at the water plant at each collection time.

[0047] Through the established water source monitoring stations and water plants, water quality data at the water sources and water plants of the target area over the years are collected at different collection time intervals. 12 parameters are obtained at the water sources, namely: water temperature, pH value, dissolved oxygen, conductivity, turbidity, permanganate index, ammonia nitrogen, total phosphorus, comprehensive toxicity, lead, copper, and cadmium; 4 conventional indicators are obtained at the water plants, namely: turbidity, total bacteria count, residual chlorine, and chemical oxygen demand. The test values ​​of the 12 parameters at the water sources and the 4 indicators at the water plants at each collection time are obtained.

[0048] Preferably, as an embodiment of the present application, water quality data at the site water source and water plant in the target area from 2015 to 2019 are collected; the collection time interval at the site water source is 4 hours, and the collection time interval at the water plant is 24 hours.

[0049] It should be noted that when collecting water quality data at the water plant, the water collected is the water quality after being treated by the water plant. The treated water from the water plant will be transmitted through the pipeline to the end of the water network. The end of the water network is also the tap water used in daily life, while the water source at the site collects natural surface water from the water source monitoring site.

[0050] At this point, the detection values ​​of various parameters at the water source of the site and various indicators at the water plant at each collection time are obtained.

[0051] Step 2: Construct the feature set, trend feature vector and feature description set of each indicator of the water plant every day.

[0052] Since the water network ends connect thousands of households, it is difficult to detect water quality, and the time interval for water quality detection is long and uneven; while the site water source is natural surface water, and most of the water that people come into contact with on a daily basis is treated by the water plant. The monitoring cycle is relatively appropriate and uniform, and it is also close to people's health. Therefore, compared with using the site water source or the end of the water network, it is more reasonable to judge the water quality through the four conventional indicators of the water plant. However, since the detection cycle of the water plant is longer than the site water source, but the water in the water plant is constantly obtained from the site water source, processed by the water plant and transported to the end of the water network, the time difference also makes it certain that there are certain defects in using only the four conventional indicators of the water plant to judge the water quality; at the same time, when conventionally using the water plant detection indicators to judge the water quality, the weights of each indicator are often the same, which ignores the impact of the changes of each indicator over a period of time on the water quality. In order to avoid this situation, improve the accuracy of using the four conventional indicators of the water plant to judge the water quality, consider combining the indicators of the site water source, and further optimize the method of water quality assessment. It is generally believed that turbidity reflects the content of suspended solids in water, and is also related to pollutants such as water temperature, permanganate index, and heavy metals (lead, copper, cadmium). The reason why turbidity is related to human health is essentially due to the content of these substances. These substances will be eliminated in the water treatment process of the water plant, but they will not be completely eliminated, and the water treatment process of the water plant cannot adjust the specific details of the water treatment according to the real-time changes in the content of these substances. Therefore, the time-series change trend of these substances detected at the water source of the site will also affect the turbidity of the water plant; secondly, residual chlorine refers to the water The effective chlorine compounds that still exist in the water after disinfection are affected by factors such as water temperature, pH value, dissolved oxygen, permanganate index, ammonia nitrogen, total phosphorus, and comprehensive toxicity; chemical oxygen demand refers to the amount of oxygen required to oxidize all oxidizable organic matter and some inorganic matter in the water sample under certain conditions using a strong oxidant, and is related to physical properties or chemical substances such as water temperature, pH value, permanganate index, ammonia nitrogen, and total phosphorus in the water; and the total number of bacteria is related to water temperature, pH value, dissolved oxygen, conductivity, permanganate index, ammonia nitrogen, total phosphorus, and heavy metals in the water.

[0053] Based on the above analysis, the feature set, trend feature vector and feature description set of each indicator of the water plant are constructed every day. The specific method is as follows:

[0054] Collect the detection values ​​of parameters related to each indicator at the water source of each station every day, and construct the characteristic vectors of each parameter in the order of collection time;

[0055] The characteristic vectors of all relevant parameters of each indicator are combined into a characteristic set of each indicator of the water plant every day;

[0056] Filling processing is performed on each indicator of the water plant on two consecutive days, and a vector with the same dimension as the characteristic vector is filled, which is recorded as the trend characteristic vector of each indicator of the water plant on each day;

[0057] The feature vector of each parameter in the feature set and the trend feature vector are combined to form a feature description set of each indicator of the water plant every day.

[0058] It should be noted that the binary filling refers to the assumption that the data and data Missing in between values, then calculate the data and data The mean of The missing value at the position where express Round up, and so on, until all missing values ​​at all positions are filled.

[0059] In this embodiment, the turbidity of the water plant is related to the detection values ​​of parameters such as the water temperature, permanganate index, and heavy metals (lead, copper, and cadmium) at the water source of the station. Therefore, based on the above method, a feature set of the turbidity of the water plant every day, a trend feature vector of the turbidity, and a feature description set of the turbidity are constructed;

[0060] Secondly, the residual chlorine in the water plant is related to the water temperature, pH value, dissolved oxygen, permanganate index, ammonia nitrogen, total phosphorus, comprehensive toxicity and other parameters at the water source of the station. The feature set of residual chlorine in the water plant every day, the trend feature vector of residual chlorine and the feature description set of residual chlorine are constructed;

[0061] The chemical oxygen demand of the water plant is related to the water temperature, pH value, permanganate index, ammonia nitrogen, total phosphorus and other parameters at the water source of the site. The feature set of the chemical oxygen demand of the water plant every day, the trend feature vector of the chemical oxygen demand and the feature description set of the chemical oxygen demand are constructed;

[0062] The total bacterial count in a water plant is related to the water temperature, pH value, dissolved oxygen, conductivity, permanganate index, ammonia nitrogen, total phosphorus, heavy metals and other parameters at the water source of the site. A feature set of the total bacterial count in the water plant every day, a trend feature vector of the total bacterial count and a feature description set of the total bacterial count are constructed.

[0063] At this point, the feature set, trend feature vector and feature description set of each indicator of the water plant are obtained every day.

[0064] Step 3, analyze the discreteness of each feature in the feature description set of turbidity within the same cluster, and the discreteness of each feature in the feature description set of turbidity between different clusters, calculate the intra-class variance and inter-class variance of each feature in the feature description set of turbidity; determine the contribution coefficient of each feature in the feature description set of turbidity.

[0065] Taking turbidity among the indicators of the water plant as an example, further analysis is carried out based on the constructed feature set of daily turbidity of the water plant, the trend feature vector of turbidity and the feature description set of turbidity.

[0066] By analyzing the difference in the contribution rate of each parameter at the water source to the turbidity of water quality relative to the contribution rate of the turbidity trend change at the water plant, the contribution of the detection amount of each parameter at the water source to the turbidity at the water plant is analyzed, and the contribution coefficient of each feature is calculated. The specific method is:

[0067] Clustering the characteristic description set of turbidity on all days to obtain multiple clusters;

[0068] Preferably, as an embodiment of the present application, a K-means clustering algorithm is used for clustering to obtain 5 clusters, wherein the K-means clustering algorithm is a well-known technology and the specific process is not repeated any more.

[0069] It is understandable that this embodiment only provides one clustering algorithm, namely, the K-means clustering algorithm. As other implementation methods, implementers can use other clustering algorithms in the prior art for clustering, such as hierarchical clustering algorithms, etc. This embodiment does not impose any special restrictions on this.

[0070] The variance of each feature of the feature description set of turbidity in each cluster is recorded as the first variance; the mean of the first variance of each feature of all clusters is taken as the intra-class variance of each feature in the feature description set of turbidity.

[0071] The mean of each feature in the feature description set of turbidity within each cluster is recorded as the first mean of each feature; the variance of the first mean of each feature of all clusters is taken as the between-class variance of each feature in the feature description set of turbidity.

[0072] In this embodiment, the intra-class variance and inter-class variance of each feature in the feature description set of turbidity are specifically calculated as follows:

[0073]

[0074]

[0075] in, is the first in the turbidity feature description set The between-class variance of features, For the The first characterization set of the turbidity within the cluster The mean of the features, is the first characterization set of turbidity in all clusters The mean of the features; The first character in the feature description set representing turbidity The intra-class variance of the features, Indicates The first characterization set of the turbidity within the cluster The variance of the features, K represents the number of clusters.

[0076] It should be noted that the conventional variance and mean are calculated for a set of discrete data, while in this embodiment, the feature vector of each feature represents a set of vectors, and each cluster contains feature vectors of multiple features, but the dimensions of the feature vectors are consistent. By calculating the variance and mean of all feature vectors corresponding to each feature of each cluster, all variances of each feature of each cluster are accumulated as the variance and mean of each feature of each cluster; for example, The clusters are clustered into the feature description set of two days of turbidity, where the feature description set of one day of turbidity is , the characteristic description set of turbidity on another day is ,in, and Indicates The first Features, and Indicates The first Features, The first characterization set of the turbidity within the cluster The mean and variance of the features are: vector The mean and variance of , ,vector The mean and variance of , ,Will and of and as the first The first characterization set of the turbidity within the cluster The mean of the features will be and of and as the first The first characterization set of the turbidity within the cluster The variance of a feature.

[0077] The ratio of the inter-class variance to the intra-class variance of each feature in the turbidity feature description set is used as the contribution coefficient of each feature in the turbidity feature description set;

[0078] Among them, each feature in the turbidity feature description set includes the feature vector of each parameter of the turbidity feature set and the trend feature vector of the turbidity of the water plant. Therefore, the contribution coefficient of the feature vector of each parameter in the turbidity feature set and the contribution coefficient of the trend feature vector of the turbidity of the water plant can be obtained;

[0079] In this embodiment, the turbidity feature description set includes the feature vectors of each parameter of the turbidity feature set and the trend feature vector of the turbidity, wherein the turbidity feature set has 6 parameter feature vectors, and therefore, the turbidity feature description set includes a total of 7 feature vectors. Thus, the contribution coefficient of the feature vector of each parameter in the turbidity feature set and the contribution coefficient of the trend feature vector of the turbidity can be obtained.

[0080] Further, the step flow chart of the method for obtaining the contribution coefficient of each feature in the turbidity feature description set provided in the present application is as follows: Figure 2 shown.

[0081] It should be noted that if the contribution rate of the feature is large during clustering, the change of the feature in the same class is not obvious, the stability of the feature in the same class is high, that is, the intra-class variance of the feature is small; and the feature has obvious differences in different classes, that is, the inter-class variance of the feature is large, so the contribution coefficient is relatively large. Conversely, if the contribution rate of the feature is small during classification, the contribution coefficient of the feature is relatively small.

[0082] At this point, the contribution coefficient of each feature in the feature description set of turbidity is obtained.

[0083] Step 4, determining the turbidity weight coefficient according to the difference of elements in the characteristic vector of each parameter in the characteristic set of turbidity of the water plant every day and the contribution coefficient of the corresponding characteristic vector; for other indicators in the various indicators of the water plant, the method for determining the turbidity weight coefficient is used to obtain the weight coefficient of each indicator of the water plant.

[0084] Based on the above analysis, the content of indicators in the water quality at the water source of the site will affect the water quality at the water plant. Furthermore, according to the differences in various parameters at the water source of the site, the degree of turbidity affecting the water quality at the water plant is obtained, and the turbidity weight coefficient is calculated. The specific method is:

[0085] The ratio between the contribution coefficient of the characteristic vector of each parameter of the turbidity characteristic set and the contribution coefficient of the trend characteristic vector of the turbidity is taken as the characteristic contribution rate of each parameter;

[0086] The mean of the differences of all adjacent elements in the feature vector of each parameter in the feature set of the water plant turbidity for each day is taken as the average relative change of each parameter in the feature set of the water plant turbidity for each day;

[0087] The product of the characteristic contribution rate and the average relative change is used as the water source detection influence coefficient of each parameter in the characteristic set of turbidity of the water plant every day;

[0088] The mean of the water source detection influence coefficients of all parameters in the feature set of the turbidity of the water plant every day is used as the turbidity weight of the water plant every day; the mean of the turbidity weights of all days of the water plant is used as the turbidity weight coefficient;

[0089] Further, the step flow chart of the method for obtaining the turbidity weight coefficient provided in the present application is as follows: Figure 3 shown.

[0090] It should be noted that if the contribution of the characteristic vector of a certain parameter is greater than the contribution of the trend characteristic vector of turbidity, the greater the characteristic contribution rate. If the average relative change is greater, it means that the differences in various parameters at the water source of the site are large, then the water source detection influence coefficient is greater, indicating that the water quality is more turbid, that is, the turbidity weight is greater.

[0091] For other indicators in the water plant, such as residual chlorine, chemical oxygen demand and total bacteria count, among which residual chlorine is related to water temperature, pH value, dissolved oxygen, permanganate index, ammonia nitrogen, total phosphorus, comprehensive toxicity and other parameters at the water source of the site;

[0092] Secondly, the chemical oxygen demand is related to the water temperature, pH value, permanganate index, ammonia nitrogen, total phosphorus and other parameters at the water source of the site;

[0093] Finally, the total bacterial count is related to parameters such as water temperature, pH, dissolved oxygen, conductivity, permanganate index, ammonia nitrogen, total phosphorus, and heavy metals at the water source of the site;

[0094] Based on the above analysis, the method for determining the turbidity weight coefficient is used to obtain the residual chlorine weight, chemical oxygen demand weight and total bacteria weight of the water plant every day;

[0095] At this point, the weight coefficients of each indicator of the water plant are obtained.

[0096] Step 5, based on the weight coefficient of each indicator and the detection value of each indicator of the water plant, calculate the excess risk of each indicator every day, determine the daily water quality health index, and evaluate the water quality of drinking water.

[0097] Based on the above analysis, according to the weight coefficients of different indicators in the water plant and the detection values ​​of each indicator, the risk of exceeding the standard for each indicator is analyzed to further evaluate the water quality of the water plant and calculate the excess risk of each indicator every day. The specific method is as follows:

[0098] The product of the test value of each indicator of the water plant every day and the weight coefficient of the corresponding indicator is used as the quality influence coefficient of each indicator of the water plant every day;

[0099] The water quality assessment index of the water plant every day is in a positive synergistic relationship with the quality influence coefficients of all indicators;

[0100] Preferably, as an embodiment of the present application, the water quality assessment index of the water plant every day is the sum of the quality influence coefficients of all indicators.

[0101] The product of the water quality assessment index and the detection value of each indicator of the water plant every day is used as the risk coefficient of each indicator of the water plant;

[0102] Furthermore, the excess risk of each indicator of the water plant is calculated every day. The calculation method is:

[0103]

[0104] in, For the water plant Day indicator excess risk, For water plant indicators The risk factor, For the water plant Day indicator The detection value of .

[0105] It should be noted that the higher the detection value of each indicator, the higher the excess risk of the indicator, the lower the water quality, and the greater the impact on people's health and life.

[0106] The daily Water Quality Health Index (WQHI) is constructed based on the excess risk of each indicator every day, specifically:

[0107]

[0108] in, For the Water quality health index for the day, For the Day indicator excess risk, is the number of indicators of the water plant, represents the maximum value function, The number of days for collection.

[0109] Preferably, in one embodiment of the present application, the number of each indicator of the water plant is 4, which are turbidity, total bacteria count, residual chlorine and chemical oxygen demand.

[0110] Based on the water quality health index, an evaluation level is constructed. A value of 0-3 indicates low risk, 4-6 indicates medium risk, 7-10 indicates high risk, and greater than 10 indicates extremely high risk.

[0111] Based on the same inventive concept as the above method, an embodiment of the present application also provides a water quality assessment system for drinking water, comprising a memory, a processor, and a computer program stored in the memory and running on the processor, wherein the processor implements the steps of any one of the above-mentioned methods for assessing the water quality of drinking water when executing the computer program.

[0112] It should be understood that although Figure 1 The steps in the flowchart are shown in sequence as indicated by the arrows, but these steps are not necessarily executed in the order indicated by the arrows. Unless otherwise specified in this document, there is no strict order restriction for the execution of these steps, and these steps can be executed in other orders. Moreover, Figure 1 At least part of the steps may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily executed at the same time, but can be executed at different times. The execution order of these sub-steps or stages is not necessarily sequential, but can be executed in turn or alternately with other steps or at least part of the sub-steps or stages of other steps.

[0113] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0114] The above-mentioned embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the present application. It should be pointed out that, for those of ordinary skill in the art, several modifications and improvements can be made without departing from the concept of the present application. Therefore, any simple modification, equivalent change and modification made to the above embodiments according to the technical essence of the present application without departing from the content of the technical solution of the present application, shall fall within the protection scope of the technical solution of the present application.

Claims

1. A method for evaluating the quality of drinking water, characterized in that: The method comprises the following steps: Obtain the detection values ​​of various parameters at the water source of the site and various indicators at the water plant at each collection time, and construct the feature set, trend feature vector and feature description set of each indicator of the water plant every day; For the turbidity in each indicator of the water plant, the characteristic description set of turbidity of all days is clustered to obtain multiple clusters; the discreteness of each feature in the characteristic description set of turbidity within the same cluster, and the discreteness of each feature in the characteristic description set of turbidity between different clusters are analyzed, and the intra-class variance and inter-class variance of each feature in the characteristic description set of turbidity are calculated; Determining a contribution coefficient of each feature in a feature description set of turbidity based on the intra-class variance and the inter-class variance; Determine the turbidity weight coefficient according to the difference of elements in the characteristic vector of each parameter in the characteristic set of turbidity of the water plant every day and the contribution coefficient of the corresponding characteristic vector; For other indicators in the water plant indicators, the turbidity weight coefficient determination method is adopted to obtain the weight coefficient of each indicator of the water plant; Based on the weight coefficient of each indicator of the water plant and the detection value of each indicator, the excess risk of each indicator is calculated every day; the water quality health index is determined every day to evaluate the water quality of drinking water; The method for constructing the feature set, trend feature vector and feature description set of each indicator of the water plant every day is as follows: Collect the detection values ​​of parameters related to each indicator at the water source of each station every day, and construct the characteristic vectors of each parameter in the order of collection time; The characteristic vectors of all relevant parameters of each indicator are combined into a characteristic set of each indicator of the water plant every day; Filling processing is performed on each indicator of the water plant on two consecutive days, and a vector with the same dimension as the characteristic vector is filled, which is recorded as the trend characteristic vector of each indicator of the water plant on each day; The feature vectors of each parameter in the feature set and the trend feature vector form a feature description set of each indicator of the water plant every day; The step of determining the turbidity weight coefficient comprises: Each feature in the feature description set of turbidity is a feature vector and a trend feature vector of each parameter in the feature set, and a contribution coefficient of the feature vector of each parameter in the feature set of turbidity and a contribution coefficient of the trend feature vector of turbidity are determined; The ratio between the contribution coefficient of the characteristic vector of each parameter in the turbidity characteristic set and the contribution coefficient of the trend characteristic vector of the turbidity is taken as the characteristic contribution rate of each parameter; The mean of the differences of all adjacent elements in the feature vector of each parameter in the feature set of the water plant turbidity for each day is taken as the average relative change of each parameter in the feature set of the water plant turbidity for each day; The product of the characteristic contribution rate and the average relative change is used as the water source detection influence coefficient of each parameter in the characteristic set of turbidity of the water plant every day; The mean of the water source detection influence coefficients of all parameters in the feature set of the water plant's turbidity every day is used as the turbidity weight of the water plant every day; the mean of the turbidity weights of the water plant on all days is used as the turbidity weight coefficient.

2. A method for evaluating the quality of drinking water according to claim 1, characterized in that: The calculation method of the intra-class variance of each feature in the feature description set of turbidity is: The variance of each feature of the feature description set of turbidity in each cluster is recorded as the first variance; The mean of the first variance of each feature of all clusters is used as the intra-class variance of each feature in the feature description set of turbidity.

3. A method for evaluating the quality of drinking water according to claim 1, characterized in that: The calculation method of the inter-class variance of each feature in the feature description set of turbidity is: The mean of each feature in the feature description set of turbidity in each cluster is recorded as the first mean of each feature; The variance of the first mean of each feature of all clusters is used as the inter-class variance of each feature in the feature description set of turbidity.

4. A method for evaluating the quality of drinking water according to claim 1, characterized in that: The contribution coefficient of each feature in the feature description set for determining turbidity is the ratio between the between-class variance and the within-class variance.

5. A method for evaluating the quality of drinking water according to claim 1, characterized in that: The calculation formula for the excess risk of each indicator every day is: ,in, For the water plant Day indicator excess risk, For water plant indicators The risk factor, For the water plant Day indicator The detection value of .

6. A method for evaluating the quality of drinking water according to claim 5, characterized in that: The calculation method of the risk coefficient of each indicator in the water plant is: The product of the test value of each indicator of the water plant every day and the weight coefficient of the corresponding indicator is used as the quality influence coefficient of each indicator of the water plant every day; The water quality assessment index of the water plant every day is in a positive synergistic relationship with the quality influence coefficients of all indicators; The product of the water quality assessment index and the detection value of each indicator of the water plant every day is used as the risk coefficient of each indicator of the water plant.

7. A method for evaluating the quality of drinking water according to claim 5, characterized in that: The water quality health index for each day is determined by the following calculation formula: ,in For the Water quality health index of Tianshui Plant, is the number of indicators of the water plant, represents the maximum value function, The number of days for collection.

8. A water quality assessment system for drinking water, comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that: When the processor executes the computer program, the steps of a method for assessing water quality of drinking water as described in any one of claims 1 to 7 are implemented.

Images