Water supply management system based on hydraulic simulation data analysis

Abstract

The invention relates to the technical field of water supply management, in particular to a water supply management system based on hydraulic simulation data analysis, and the system comprises a water flow detection module which is used for collecting three index data, including water flow data, water flow pressure data and water quality parameter data, of each monitoring point of a water supply network in real time through a sensor; a data preprocessing module; the data elimination module is used for eliminating burr data in various index data and eliminating pre-counted fault state data through a burr identification algorithm; according to the invention, real-time analysis and monitoring are carried out on the operation data of the pipe network, rules and data characteristics among various data are quickly found, the requirements of real-time processing and quick analysis are met, abnormal data can be judged more quickly, a management strategy can be formulated in time, and the scientificity and accuracy of decision making are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of water supply management, and in particular to a water supply management system based on hydraulic simulation data analysis. Background Art

[0002] In traditional pipe network management, production departments can only focus on the pressure, flow, and water quality at specific pipe network monitoring points within the water supply area. This lacks a holistic perspective and prevents a comprehensive understanding of the pipe network's operational status and anomalies. Equipment anomalies can also impact production department analysis of the pipe network. This situation limits the effectiveness of pipe network monitoring, troubleshooting, and preventive measures, while also impacting the accuracy of subsequent analysis. Traditional management systems are inefficient in processing massive amounts of data and struggle to quickly identify patterns and characteristics. This makes it impossible to meet the demands of real-time processing and rapid analysis, and it cannot predict trends in indicator data based on abnormal data changes. Summary of the Invention

[0003] In view of the above problems, the present invention proposes a water supply management system based on hydraulic simulation data analysis, comprising: a water flow detection module: used to collect three indicator data of each monitoring point of the water supply network in real time through sensors, including water flow data, water flow pressure data and water quality parameter data;

[0004] Data preprocessing module; used to remove burr data from various indicator data and remove pre-statisticed fault status data through burr recognition algorithm;

[0005] The data analysis module is used to analyze the pre-processed indicator data using the 3σ principle, obtain abnormal indicator data, and obtain the correlation between other indicator data related to the abnormal indicator data based on the simulation data, and formulate management strategies based on the correlation between the abnormal indicator data and other indicator data related to the abnormal indicator data.

[0006] Preferably, the burr identification algorithm is used to remove burr data from various indicator data, including: extracting n single indicator data within a certain time period of the monitoring point, and the data point values are sequentially , calculate the average fluctuation avg within the time period:

[0007]

[0008] Set the glitch judgment conditions, including setting three average fluctuation values arg1, arg2, and arg3 through empirical analysis, and satisfying the following conditions:

[0009]

[0010]

[0011] and simultaneously satisfy one of the following two sub-conditions:

[0012]

[0013]

[0014] Eliminate the data of the identified burr points.

[0015] Preferably, analyzing the preprocessed data of each index using the 3σ principle includes: calculating the mean μ, standard deviation σ and outliers of the data of each index; among which, the calculation of the mean μ includes:

[0016]

[0017] wherein, represents the value of a single data point, and n represents the sample size within a set time period;

[0018] Calculate the standard deviation σ reflecting the degree of dispersion of the index data:

[0019]

[0020] The determination of outliers includes: if < μ - 3σ or > μ + 3σ, then is an outlier.

[0021] Preferably, the correlation degrees between the data of each index include: the linear correlation degree r between the water flow rate of a single data point and the water flow pressure, and the linear correlation degree R between the water flow rate of a single data point and the water quality parameters; and the calculation methods of each correlation degree are respectively:

[0022] a. Calculate the linear correlation degree r between the water flow rate of a single data point and the water flow pressure based on the Pearson correlation coefficient:

[0023]

[0024] wherein, r > 0 is a positive correlation, r < 0 is a negative correlation, |r| ≥ 0.7 is a strong correlation, 0.3 ≤ |r| < 0.7 is a medium correlation, and |r| < 0.3 is a weak correlation; represents the measured water flow rate data of a single data point, represents the measured water flow pressure data of a single data point, represents the standard deviation of the water flow rate within a predetermined time, represents the standard deviation of the water flow pressure within a predetermined time, represents the mean value of the water flow rate within a predetermined time, represents the mean value of the water flow pressure within a predetermined time; m represents the sample size within a predetermined time.

[0025] b. Calculate the linear correlation R between water flow and water quality parameters at a single data point based on the Pearson correlation coefficient:

[0026]

[0027] Among them, r>0 is positive correlation, r<0 is negative correlation, |r|≥0.7 is strong correlation, 0.3≤|r|<0.7 is moderate correlation, and |r|<0.3 is weak correlation. Represented as a single data point of measured water flow pressure data, Indicates the standard deviation of water flow pressure within a predetermined time. Indicates the average water flow pressure within a predetermined time.

[0028] Preferably, formulating a management strategy based on the correlation between abnormal indicator data and other indicator data related to the abnormal indicator data includes: if the analyzed water flow data is abnormal, then calculating the linear correlation r between the water flow rate and the water flow pressure of a single data point or the linear correlation R between the water flow rate and the water quality parameter of a single data point, and formulating a management strategy based on the two calculated correlation values; if the analyzed water flow pressure data is abnormal, then calculating the linear correlation r between the water flow rate and the water flow pressure of a single data point, and formulating a management strategy based on the calculated correlation value; if the water quality parameter data is abnormal, then calculating the linear correlation R between the water flow rate and the water quality parameter of a single data point, and formulating a management strategy based on the calculated correlation value.

[0029] The beneficial effects of this application are: this application conducts real-time analysis and monitoring of pipeline network operation data, quickly discovers the patterns and data characteristics between various data, meets the needs of real-time processing and rapid analysis, can judge abnormal data more quickly, and formulate management strategies in a timely manner, thereby improving the scientific nature and accuracy of decision-making. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] Figure 1 Shown is a schematic diagram of this application management system. DETAILED DESCRIPTION

[0031] In order to enable those skilled in the art to better understand the technical solution of this application,

[0032] The present invention is further described in detail with reference to the drawings and preferred embodiments.

[0033] Water flow detection module: used to collect three indicator data of each monitoring point of the water supply network in real time through sensors, including water flow data, water flow pressure data and water quality parameter data;

[0034] Data preprocessing module; used to remove burr data from various indicator data and remove pre-statisticed fault status data through burr recognition algorithm;

[0035] A data analysis module is used to analyze the preprocessed index data by using the 3σ principle, obtain abnormal index data, calculate the correlation degree between the abnormal index data and other item index data based on the simulation data, and formulate a management strategy based on the abnormal index data and the correlation degree between the abnormal index data and other item index data.

[0036] Through the burr recognition algorithm, the burr data in each index data is removed, including: extracting n single index data within a certain time period of a monitoring point, and the data point values are in turn , calculate the average fluctuation avg within the time period:

[0037]

[0038] Set the burr judgment conditions, including setting three average fluctuation values arg1, arg2, and arg3 through empirical analysis, and satisfying the following formula conditions:

[0039]

[0040]

[0041] And at the same time satisfy one of the following two sub-conditions:

[0042]

[0043]

[0044] The burr removal method of this application is specifically applied as follows. Set the average fluctuation of the past day as avg, there are n time collection points in the past day, and set three average fluctuation values arg1, arg2, and arg3 for each index data:

[0045] dataType: Flow rate arg1 = 7.5, arg2 = 2.5, arg3 = 7.5;

[0046] dataType: Pressure arg1 = 4.0, arg2 = 2.0, arg3 = 7.5;

[0047] dataType: Turbidity (water quality) arg1 = 5.0, arg2 = 2.0, arg3 = 7.5;

[0048] Successively substitute the index data information of three consecutive collection points into the burr judgment conditions. If the burr judgment conditions are met, the data of the middle collection point will be removed. When rotating, the removed point will be replaced by the middle value of the previous point and the next point.

[0049] Analyzing the preprocessed index data using the 3σ principle includes: calculating the mean μ, standard deviation σ, and outliers of each index data; among which, the calculation of the mean μ includes:

[0050]

[0051] Among them, represents the value of a single data point, and n represents the sample size within a set time period;

[0052] Calculating the standard deviation σ that reflects the dispersion degree of the index data:

[0053]

[0054] The determination of outliers includes: if < μ−3σ or >μ + 3σ, then is an outlier.

[0055] Preferably, the correlation degree between each index data includes: the linear correlation degree r between the water flow rate of a single data point and the water flow pressure, and the linear correlation degree R between the water flow rate of a single data point and the water quality parameters; and the calculation methods of each correlation degree are respectively:

[0056] a. Calculating the linear correlation degree r between the water flow rate of a single data point and the water flow pressure based on the Pearson correlation coefficient:

[0057]

[0058] Among them, r > 0 is a positive correlation, r < 0 is a negative correlation, |r|≥0.7 is a strong correlation, 0.3≤|r|<0.7 is a medium correlation, and |r|<0.3 is a weak correlation; represents the measured water flow rate data of a single data point, represents the measured water flow pressure data of a single data point, represents the standard deviation of the water flow rate within a predetermined time, represents the standard deviation of the water flow pressure within a predetermined time, represents the average value of the water flow rate within a predetermined time, represents the average value of the water flow pressure within a predetermined time; m represents the sample size within a predetermined time.

[0059] b. Calculating the linear correlation degree R between the water flow rate of a single data point and the water quality parameters based on the Pearson correlation coefficient:

[0060]

[0061] Among them, r > 0 is a positive correlation, r < 0 is a negative correlation, |r|≥0.7 is a strong correlation, 0.3≤|r|<0.7 is a medium correlation, |r|<0.3 is a weak correlation, Represents the water flow pressure data of a single measured data point, Represents the standard deviation of the water flow pressure within a predetermined time, Represents the average value of the water flow pressure within a predetermined time.

[0062] The data of water quality parameters includes residual chlorine and turbidity. The water flow pressure and water quality parameters show a weakly insignificant correlation. Therefore, the correlation between the water flow pressure and water quality parameters will no longer be analyzed.

[0063] Formulating management strategies based on the correlation between abnormal index data and other item index data related to the abnormal index data includes: If it is analyzed that the water flow rate data is abnormal, then calculate the linear correlation coefficient r between the water flow rate of a single data point and the water flow pressure or the linear correlation coefficient R between the water flow rate of a single data point and the water quality parameters, and formulate management strategies based on the two calculated correlation coefficient values. For example, when the water flow rate data suddenly increases abnormally and the linear correlation coefficient r between the water flow rate of a single data point and the water flow pressure strengthens, the abnormal reason may be pipeline breakage and leakage or unconventional operations of upstream and downstream gates. At the same time, if the linear correlation coefficient R between the water flow rate of a single data point and the water quality parameters strengthens, it is judged that there may be reverse flushing of the water flow or too high flow velocity, causing impurities in the pipeline to be washed down from the pipe wall, increasing the turbidity and affecting the water quality.

[0064] In the case of stable flow and stable turbidity, a slight increase or decrease in the flow rate will not affect the turbidity. However, when the flow rate surges or drops sharply, it will cause a sharp increase in the turbidity, forming a positive or negative correlation and strengthening the correlation. Therefore, the change in the flow velocity will cause the turbidity to increase.

[0065] If it is analyzed that the water flow pressure data is abnormal, calculate the linear correlation coefficient r between the water flow rate of a single data point and the water flow pressure, and formulate management strategies based on the calculated correlation coefficient value; if the water quality parameter data is abnormal, calculate the linear correlation coefficient R between the water flow rate of a single data point and the water quality parameters, and formulate management strategies based on the calculated correlation coefficient value. For example, when the turbidity rises, if the linear correlation coefficient R between the water flow rate of a single data point and the water quality parameters strengthens, it is determined that there may be water flow fluctuations caused by water plant regulation, pump station regulation, unconventional operations of gates, or pipeline breakage and leakage, resulting in an increase in turbidity.

[0066] Formulating management strategies based on the correlation between abnormal data and other item index data related to the abnormal index data helps to accurately infer the type of abnormality and timely formulate targeted strategies.

[0067] The above are only the preferred embodiments of the present invention. It should be noted that for those of ordinary skill in the art, without departing from the principle of the present invention, several improvements and refinements can be made, and these improvements and refinements should also be regarded as the protection scope of the present invention.

Claims

1. A water supply management system based on the analysis of hydraulic simulation data, characterized in that Including: A water flow detection module: used to collect in real time through sensors three - index data of each monitoring point in the water supply network, including water flow data, water flow pressure data, and water quality parameter data; A data pre - processing module; Used to eliminate the spike data in each index data and eliminate the pre - statistically fault - state data through a spike recognition algorithm; A data analysis module, used to analyze each index data after pre - processing using the 3σ principle, obtain abnormal index data, and calculate the correlation degree between the abnormal index data and other index data related to the abnormal index data based on simulated data. A management strategy is formulated based on the abnormal index data and the correlation degree between the abnormal index data and other index data related to it. For example, if the water flow data is analyzed as abnormal, calculate the linear correlation degree r between the water flow of a single data point and the water flow pressure or the linear correlation degree R between the water flow of a single data point and the water quality parameter, and formulate a management strategy based on the two calculated correlation degree values; If the water flow pressure data is analyzed as abnormal, calculate the linear correlation degree r between the water flow of a single data point and the water flow pressure, and formulate a management strategy based on the calculated correlation degree value; If the water quality parameter data is analyzed as abnormal, calculate the linear correlation degree R between the water flow of a single data point and the water quality parameter, and formulate a management strategy based on the calculated correlation degree value.

2. The water supply management system based on hydraulic simulation data analysis according to claim 1, characterized in that Through the burr recognition algorithm, the burr data in each index data is eliminated, including: extracting n single index data within a certain time period for a monitoring point, and the data point values are successively , calculating the average fluctuation avg within the time period: ； Set the conditions for judging spikes, including setting three average fluctuation values arg1, arg2, and arg3 through empirical analysis, and satisfying the following formula conditions: ； ； And simultaneously satisfy one of the following two sub - conditions: ； ； Eliminate the data of the identified spike points.

3. The water supply management system based on hydraulic simulation data analysis according to claim 1, characterized in that Analyzing each index data after pre - processing using the 3σ principle includes: obtaining the mean μ, standard deviation σ, and abnormal values of each index data; among them, the calculation of the mean μ includes: ； Among them, represents the value of a single data point, and n represents the sample size within a set time period; Obtaining the standard deviation σ reflecting the dispersion degree of the index data: ； The determination of outliers includes: if <μ - 3σ or >μ + 3σ, then it is an outlier.

4. The water supply management system based on hydraulic simulation data analysis according to claim 1, characterized in that The correlation degrees between each index data include: the linear correlation degree r between the water flow of a single data point and the water flow pressure and the linear correlation degree R between the water flow of a single data point and the water quality parameter; and the calculation methods of each correlation degree are respectively: a. Calculate the linear correlation degree r between the water flow of a single data point and the water flow pressure based on the Pearson correlation coefficient: ； Among them, r > 0 represents positive correlation, r < 0 represents negative correlation, |r| ≥ 0.7 represents strong correlation, 0.3 ≤ |r| < 0.7 represents medium correlation, and |r| < 0.3 represents weak correlation; represents the water flow rate data of a single measured data point, represents the water flow pressure data of a single measured data point, represents the standard deviation of water flow rate within a predetermined time, represents the standard deviation of water flow pressure within a predetermined time, represents the average value of water flow rate within a predetermined time, represents the average value of water flow pressure within a predetermined time; m represents the sample size within a predetermined time; b. Calculate the linear correlation degree R between the water flow of a single data point and the water quality parameter based on the Pearson correlation coefficient: ； Among them, r > 0 indicates positive correlation, r < 0 indicates negative correlation, |r| ≥ 0.7 indicates strong correlation, 0.3 ≤ |r| < 0.7 indicates medium correlation, and |r| < 0.3 indicates weak correlation. represents the measured water flow pressure data of a single data point represents the standard deviation of the water flow pressure within a predetermined time represents the mean value of the water flow pressure within a predetermined time.

Images