Method and system for automatic water meter reading

Abstract

The present invention relates to the field of automatic meter reading technology, and solves the technical problems of low efficiency and poor real-time performance of meter reading in the prior art, lack of monitoring of abnormal water use data and intuitive display of water use characteristics, and particularly relates to a method for automatic meter reading of water meter data, the method comprising the following steps: S1, collecting raw water meter data, performing preliminary abnormal value removal on the raw data, and obtaining non-abnormal data; S2, performing deep preprocessing on the non-abnormal data to obtain preprocessed data; S3, dividing the preprocessed data into multiple water volume data Wa according to time periods (t) , calculate water volume data Wa (t) The present invention can reduce the dimensionality curse problem by performing dimensionality reduction processing on water use data. While improving the performance of the method, it can also enable users to obtain a three-dimensional visual image of the water use situation, allowing users to obtain water use information more intuitively and quickly.

Description

Technical Field

[0001] The present invention relates to the technical field of automatic meter reading, and in particular to a method and system for automatic water meter data reading. Background Art

[0002] As a key tool for water resource management, the accuracy and real-time nature of water meter data are crucial for energy allocation, billing, and resource optimization. Traditional water meter data collection methods rely primarily on regular manual on-site meter readings. This process involves on-site staff recording the readings of each water meter and then manually entering them into the management system. Although this method is technically simple to implement, manual meter reading consumes a large amount of manpower. As the number of water meters increases, the required manpower and time costs increase linearly, and real-time performance is poor. Existing meter reading methods use digital equipment to obtain water volume, but the data obtained from sensors is prone to errors, and it is impossible to monitor abnormal water consumption behaviors such as water theft and leakage. When users obtain water consumption data, they cannot intuitively obtain water consumption, resulting in a poor user experience. Summary of the Invention

[0003] In response to the shortcomings of the existing technology, the present invention provides a method and system for automatic meter reading of water meter data, which solves the technical problems of low meter reading efficiency and poor real-time performance in the existing technology, lack of monitoring of abnormal water usage data and intuitive display of water usage characteristics, and achieves the purpose of improving meter reading efficiency and real-time performance, regularly monitoring water usage data and improving the intuitiveness of water usage data.

[0004] To solve the above technical problems, the present invention provides the following technical solution: a method for automatically reading water meter data, the method comprising the following steps:

[0005] S1. Collect the original data of the water meter at any stage of operation, and perform preliminary outlier removal on the original data to obtain data without abnormalities;

[0006] S2. Perform deep preprocessing on the data without abnormalities to obtain preprocessed data;

[0007] S3, divide the pre-processed data into multiple water volume data Wa according to time period (t) , calculate water volume data Wa (t) characteristic variables;

[0008] S4. Classify users according to characteristic variables to obtain user classification sets, and obtain water volume prediction data Yc through the time convolution network method w ;

[0009] S5. Based on the water volume forecast data Yc w Calculate the dimensionality reduction value Jw z , and based on the dimensionality reduction value Jwz Generate a three-dimensional visual graph;

[0010] S6. Send the visualized three-dimensional image to the user data center.

[0011] Preferably, in step S1, the specific implementation steps are as follows:

[0012] S11. Arrange the original data in ascending order to generate a processing sequence C a , according to the processing sequence C a Calculate the first extreme value A1 and the second extreme value A2 using the following formula:

[0013] A1=(C max +C min )×25%

[0014] A2=(C max +C min )×75%

[0015] Among them, C max and C min Represents the processing sequence C a The maximum and minimum values ​​in ;

[0016] S12, calculating the measured value B based on the first extreme value A1 and the second extreme value A2 c , the calculation formula is:

[0017] B c =A2-A1

[0018] Among them, B c represents the cth measurement value;

[0019] S13, according to the measured value B c Calculation is used to represent the processing sequence C a The maximum value J in the normal data range jd and the minimum value J jx , the calculation formula is:

[0020] J jd =A2+1.5×B c

[0021] J jx =A1-1.5×B c

[0022] Among them, J jd represents the d-th maximum value, J jx represents the xth minimum value;

[0023] S14, according to the maximum value J jd and the minimum value J jxFor processing sequence C a The outliers are removed;

[0024] If C a ≤J jx or C a ≥J jd , then process sequence C a The outliers are removed and the remaining K bits are obtained. cy ;

[0025] If J jx <C a <J jd , then process sequence C a Normal value and retained;

[0026] S15, the K containing the vacant position cy Processing sequence C a The data sequence is restored and data without abnormalities is obtained.

[0027] Preferably, in step S2, the specific implementation steps are as follows:

[0028] S21, according to the vacant position K cy Determine adjacent vacancies K cy+1 and K cy-1 , calculate the K bits used to fill the empty space in the data without abnormality cy Fill value Tc h , the calculation formula is:

[0029]

[0030] Among them, Tc h Indicates the hth filling value;

[0031] S22, fill value Tc h Fill in the remaining K spaces without abnormal data cy On the above, we get no missing data Ws f ;

[0032] S23. Calculate the detection value fw for no missing data z , the calculation formula is:

[0033]

[0034] Among them, erf -1 represents the inverse function of the error function, and p represents the cumulative probability value;

[0035] S24, obtain the number of data without missing data by counting method, and calculate the number of data without missing data according to the detection value fw z Calculate the percentile value Fw d , the calculation formula is:

[0036]

[0037] Among them, Fw d represents the dth percentile value, fw z represents the zth detection value;

[0038] S25, no missing data Ws f Sort by small to large to get the sorted data set P;

[0039] S26, according to the percentile value Fw d The test coefficient H is obtained by table lookup method and the number of data N, and the test value Jh is calculated according to the test coefficient H b , the calculation formula is:

[0040]

[0041]

[0042] Among them, Jh b represents the bth test value, Indicates no missing data Ws f Preferably, in step S26, the specific implementation steps are as follows:

[0043] S261, according to the number of data N and the test value Jh b The critical value p is obtained by table lookup method;

[0044] S262. Determine whether there is missing data Ws based on the critical value p f Is it a normal distribution?

[0045] If p<0.05, it means that the data does not obey the normal distribution, and the process goes to step S263;

[0046] If p≥0.05, it means that the data obeys the normal distribution, and the process goes to step S265;

[0047] S263. Calculate the non-positive test value Jy under non-normal distribution z , the calculation formula is:

[0048]

[0049] Among them, Ws f-1 and Ws1 represent the f-1th and first non-missing data, respectively;

[0050] S264, according to the non-positive test value Jy z For non-missing data Ws f Perform outlier detection;

[0051] If Jyz <2, it is a normal value and preprocessed data is generated;

[0052] If Jy z ≥2, it is an abnormal value, calculate the first replacement value Th e And replace the abnormal value; S265, calculate the normal test value Zj under normal distribution c , the calculation formula is:

[0053]

[0054] Where s represents no missing data Ws f The sample standard deviation of

[0055] S266, according to the normal inspection value Zj c For non-missing data Ws f Perform outlier detection;

[0056] If Zj c <2, it is a normal value and preprocessed data is generated;

[0057] If Zj c ≥2, it is an abnormal value, calculate the second replacement value Te h And replace the outliers.

[0058] Preferably, the first replacement value Th e and the second replacement value Te h The calculation formula is:

[0059]

[0060] Among them, Ws f+1 and Ws f-1 Respectively represent the non-missing data Ws f There are no missing data for the f+1th and f-1th on both sides.

[0061] Preferably, in step S3, the specific implementation steps are as follows:

[0062] S31. Define the water volume data of the t-th time period as Wa (t) , calculate water volume data Wa (t) The periodic model Zq m , the calculation formula is:

[0063] Zq m =D·sin(ρt+δ)+G

[0064] Where D represents the model coefficient, ρ represents the periodic coefficient, δ represents the initial phase, and G represents the model offset value;

[0065] S32. Calculate weekly water consumption Zz y , the calculation formula is:

[0066]

[0067] Among them, n represents the water volume data Wa in one week (t) The number of Wa (t) Represents the water volume data for the tth period;

[0068] S33. Calculate the fluctuation period T of water consumption based on the periodic coefficient ρ b , the calculation formula is:

[0069]

[0070] Among them, T b represents the bth fluctuation period;

[0071] S34. Calculate the growth rate θ of water volume data z , the calculation formula is:

[0072]

[0073] Among them, Zq m+1 Represents the periodic model value of the m+1th water volume data.

[0074] Preferably, in step S4, the specific implementation steps are as follows:

[0075] S41, the characteristic variables include weekly water consumption Zz y , Fluctuation period T b and the growth rate θ z , the weekly water consumption Zz in the same period y , Fluctuation period T b and the growth rate θ z Integrate into water use vector Yw s , the expression is:

[0076] Yw s =[Zz y , T b ,θ z ]

[0077] Among them, Yw s represents the sth water consumption vector;

[0078] S42, according to the water use vector Yw s Calculation standard value B z , the calculation formula is:

[0079]

[0080] in, Represents the water use vector Yw s The average of m represents the water use vector Yw s the number of

[0081] S43, a standard value B z As the initial center Zx c , according to the standard value B z and the initial center Zx c Calculate the numerical spacing Rj s , the calculation formula is:

[0082] Rj s =||B z -Zx c || 2

[0083] Among them, Rj s Indicates the sth numerical interval;

[0084] S43, change the initial center Zx c Repeat step S43 to obtain multiple numerical intervals Rj s , calculate the optimal spacing value Rzj k , and the optimal spacing value Rzj k The corresponding initial center Zx c As a classification center fl x , the optimal spacing value Rzj k The expression is:

[0085]

[0086] Where L represents the numerical spacing Rj s the number of

[0087] S44, according to the classification center fl x Calculate the group mean value WC s , the calculation formula is:

[0088]

[0089] Among them, B represents the standard value B z and classification center fl x Number of groups;

[0090] S45, according to the group average value WC s Calculate the optimal classification value Zy v , the calculation formula is:

[0091]

[0092] Among them, Zy vrepresents the vth optimal classification value, Indicates the number of Q categories divided into the s-1th population average;

[0093] S46, repeat step S45 to obtain multiple optimal classification values ​​Zy v , the optimal classification value Zy v The number of categories Q corresponding to the minimum value in is taken as the optimal classification scheme, and the user classification set is obtained;

[0094] S47, construct a water consumption prediction model by using the time convolution network method, input the user classification data into the water consumption prediction model, and obtain the water consumption prediction data Yc w .

[0095] Preferably, in step S5, the specific implementation steps are as follows:

[0096] S51, the water volume forecast data Yc w Transformed into an m×n analysis matrix Fx through data conversion method j , the expression is:

[0097]

[0098] Among them, R 11 Represents the analysis matrix Fx j The data value of the first row and first column in the table, n represents the water volume forecast data Yc w The number of characteristic variables, m represents the water volume prediction data Yc in each characteristic variable w the number of

[0099] S52. Calculate and analyze the matrix Fx j The variance matrix Xf j , the calculation formula is:

[0100]

[0101] Among them, Xf j represents the j-th variance matrix;

[0102] S53, through the characteristic decomposition method of the variance matrix Xf j Perform eigendecomposition to obtain the characteristic matrix Tz j and the characteristic matrix Tz j The corresponding diagonal value λ i ;

[0103] S54, according to the characteristic matrix Tz j Calculate the dimensionality reduction value Jw z , the calculation formula is:

[0104]

[0105] Among them, F j Represents the analysis matrix Fx j The mean value of

[0106] S55, according to the contribution value λ i and the characteristic matrix Tz j Calculate the weight value Qz d , the calculation formula is:

[0107]

[0108] Among them, Qz d Represents the dth weight value;

[0109] S56, according to the weight value Qz d The dimension reduction value Jw z Sort from large to small, and map the sorting results into three-dimensional space to obtain a visual three-dimensional graph.

[0110] The technical solution also provides a system for the above-mentioned automatic meter reading method, the system comprising:

[0111] The preliminary processing module collects the raw data of the water meter at any stage of operation and performs preliminary outlier removal on the raw data to obtain data without abnormalities;

[0112] The preprocessing module performs deep preprocessing on the data without abnormalities to obtain preprocessed data;

[0113] Feature module, which divides the pre-processed data into multiple water volume data Wa according to time periods (t) , calculate water volume data Wa (t) characteristic variables;

[0114] The classification prediction module classifies users according to the characteristic variables to obtain the user classification set, and obtains the water volume prediction data Yc through the time convolution network method w ;

[0115] Dimensionality reduction module, based on water volume prediction data Yc w Calculate the dimensionality reduction value Jw z , and based on the dimensionality reduction value Jw z Generate a three-dimensional visual graph;

[0116] The visualization module sends the visualized three-dimensional graph to the user data center.

[0117] By means of the above technical solution, the present invention provides a method for automatic water meter reading, which has at least the following beneficial effects:

[0118] 1. The present invention completes the preparation of user water usage data through preliminary processing and preprocessing. In order to prevent large errors in user water usage data, outlier detection in preprocessing can preliminarily discover abnormal water usage values, and then mark the abnormal user water usage data.

[0119] 2. The present invention can complete the regular monitoring of water usage data by regularly acquiring water usage data, and obtain and calculate multiple water usage characteristics of water usage data, which can more accurately analyze the user's water usage habits, facilitate comprehensive prediction of the user's water usage situation, and provide the user's future water usage situation, so that the user can not only quickly and efficiently obtain the water usage situation, but also know the future water consumption, thereby improving the user's usage experience.

[0120] 3. The present invention can reduce the data dimension and alleviate the dimensionality disaster problem by performing dimensionality reduction processing on water usage data. While improving the performance of the method, it can also make the reduced-dimensional data more vivid and comprehensive. The water usage situation obtained by the user is a three-dimensional visual image, allowing the user to obtain water usage information more intuitively and quickly. BRIEF DESCRIPTION OF THE DRAWINGS

[0121] The drawings described herein are used to provide a further understanding of the present application and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0122] Figure 1 This is a flow chart of a method for automatic water meter reading according to the present invention;

[0123] Figure 2 The present invention is a structural block diagram of a water meter data automatic meter reading system. DETAILED DESCRIPTION

[0124] To make the above-mentioned objectives, features, and advantages of the present invention more clearly understood, the present invention is further described below in detail with reference to the accompanying drawings and specific embodiments. This will enable a full understanding of how this application uses technical means to solve technical problems and achieve technical effects, and to implement the invention accordingly.

[0125] Due to the low efficiency and poor real-time performance of existing meter reading technologies, there is a lack of monitoring of abnormal water use data and the lack of technical issues in visually displaying water use characteristics. Please refer to Figure 1 This embodiment provides a method for automatically reading water meter data, which can improve meter reading efficiency and real-time performance, regularly monitor water consumption data, and improve the intuitiveness of water consumption data. The method includes the following steps:

[0126] S1. Collect the raw data of the water meter at any stage of operation and perform preliminary outlier removal on the raw data to obtain data without abnormalities. After obtaining the raw data of the water meter, due to the influence of the device's sensors or the environment, the raw data may contain abnormal values, which need to be processed. The specific implementation steps are as follows:

[0127] S11. Arrange the original data in ascending order to generate a processing sequence C a , according to the processing sequence C a Calculate the first extreme value A1 and the second extreme value A2 using the following formula:

[0128] A1=(C max +C min )×25%

[0129] A2=(C max +C min )×75%

[0130] Among them, C max and C min Represents the processing sequence C a The maximum and minimum values ​​in ;

[0131] S12, calculating the measured value B based on the first extreme value A1 and the second extreme value A2 c , the calculation formula is:

[0132] B c =A2-A1

[0133] Among them, B c represents the cth measurement value;

[0134] S13, according to the measured value B c Calculation is used to represent the processing sequence C a The maximum value J in the normal data range jd and the minimum value J jx , the calculation formula is:

[0135] J jd =A2+1.5×B c

[0136] J jx =A1-1.5×B c

[0137] Among them, J jd represents the d-th maximum value, J jx represents the xth minimum value;

[0138] S14, according to the maximum value J jd and the minimum value J jx For processing sequence Ca The outliers are removed;

[0139] If C a ≤J jx or C a ≥J jd , then process sequence C a The outliers are removed and the remaining K bits are obtained. cy ;

[0140] If J jx <C a <J jd , then process sequence C a Normal value and retained;

[0141] S15, the K containing the vacant position cy Processing sequence C a The data sequence is restored and no abnormal data is obtained. Through preliminary processing and preprocessing, the preparation of user water use data is completed. In order to prevent large errors in user water use data, outlier detection in preprocessing can initially detect water use anomalies and then mark the abnormal water use data.

[0142] S2. Perform deep preprocessing on the data without abnormalities to obtain preprocessed data. After removing the abnormal values, there may still be data that does not conform to the normal distribution due to the influence of the acquired circuit or environment, which affects the calculation of the subsequent steps. In order to facilitate the accurate calculation of the subsequent steps, the specific implementation steps are as follows:

[0143] S21, according to the vacant position K cy Determine adjacent vacancies K cy+1 and K cy-1 , calculate the K bits used to fill the empty space in the data without abnormality cy Fill value Tc h , the calculation formula is:

[0144]

[0145] Among them, Tc h Indicates the hth filling value;

[0146] S22, fill value Tc h Fill in the remaining K spaces without abnormal data cy On the above, we get no missing data Ws f ;

[0147] S23. Calculate the detection value fw for no missing data z , the calculation formula is:

[0148]

[0149] Among them, erf -1 represents the inverse function of the error function, and p represents the cumulative probability value;

[0150] S24, obtain the number of data without missing data by counting method, and calculate the number of data without missing data according to the detection value fw z Calculate the percentile value Fw d , the calculation formula is:

[0151]

[0152] Among them, Fw d represents the dth percentile value, fw z represents the zth detection value;

[0153] S25, no missing data Ws f Sort by small to large to get the sorted data set P;

[0154] S26, according to the percentile value Fw d The test coefficient H is obtained by table lookup method and the number of data N, and the test value Jh is calculated according to the test coefficient H b , the calculation formula is:

[0155]

[0156] Among them, Jh b represents the bth test value, Indicates no missing data Ws f The specific implementation steps are as follows: The table lookup method is a common method for obtaining the test coefficient H, which can be used to test whether the data conforms to the normal distribution or the skewed distribution in subsequent steps. It will not be described in detail here.

[0157] S261, according to the number of data N and the test value Jh b The critical value p is obtained by table lookup method;

[0158] S262. Determine whether there is missing data Ws based on the critical value p f Is it a normal distribution?

[0159] If p<0.05, it means that the data does not obey the normal distribution, and the process goes to step S263;

[0160] If p≥0.05, it means that the data obeys the normal distribution, and the process goes to step S265;

[0161] S263. Calculate the non-positive test value Jy under non-normal distribution z , the calculation formula is:

[0162]

[0163] Among them, Ws f-1 and Ws1 represent the f-1th and 1st non-missing data respectively; the non-normal distribution can be understood as a skewed distribution.

[0164] S264, according to the non-positive test value Jy z For non-missing data Ws f Perform outlier detection;

[0165] If Jy z <2, it is a normal value and preprocessed data is generated;

[0166] If Jy z ≥2, it is marked as an outlier and the first replacement value Th is calculated. e And replace the abnormal value, the first replacement value Th e and the second replacement value Te h The calculation formula is:

[0167]

[0168] Among them, Ws f+1 and Ws f-1 Respectively represent the non-missing data Ws f There are no missing data for the f+1th and f-1th on both sides.

[0169] S265. Calculate the normal test value Zj under normal distribution c , the calculation formula is:

[0170]

[0171] Where s represents no missing data Ws f The sample standard deviation of

[0172] S266, according to the normal inspection value Zj c For non-missing data Ws f Perform outlier detection;

[0173] If Zj c <2, it is a normal value and preprocessed data is generated;

[0174] If Zj c ≥2, it is marked as an outlier and the second replacement value Te is calculated. h And replace the abnormal values, through preliminary processing and preprocessing, complete the preparation of user water use data. In order to prevent large errors in user water use data, through the abnormal value detection in preprocessing, the abnormal values ​​of water use can be initially discovered, and then the abnormal water use data of users can be marked.

[0175] S3, divide the pre-processed data into multiple water volume data Wa according to time period (t) , calculate water volume data Wa (t) In order to monitor the water volume regularly and prevent water leakage when users use water, the specific implementation steps are as follows:

[0176] S31. Define the water volume data of the t-th time period as Wa (t) , calculate water volume data Wa (t) The periodic model Zq m , the calculation formula is:

[0177] Zq m =D·sin(ρt+δ)+G

[0178] Where D represents the model coefficient, ρ represents the periodic coefficient, δ represents the initial phase, and G represents the model offset value;

[0179] S32. Calculate weekly water consumption Zz y , the calculation formula is:

[0180]

[0181] Among them, n represents the water volume data Wa in one week (t) The number of Wa (t) Represents the water volume data for the tth period;

[0182] S33. Calculate the fluctuation period T of water consumption based on the periodic coefficient ρ b , the calculation formula is:

[0183]

[0184] Among them, T b represents the bth fluctuation period;

[0185] S34. Calculate the growth rate θ of water volume data z , the calculation formula is:

[0186]

[0187] Among them, Zq m+1 It represents the periodic model value of the m+1th water volume data. By acquiring the water usage data at regular intervals, it is possible to complete the regular monitoring of the water usage data and obtain multiple water usage characteristics for calculating the water usage data. This can more accurately analyze the user's water usage habits, facilitate a comprehensive prediction of the user's water usage situation, and provide the user's future water usage situation, so that the user can not only obtain the water usage situation quickly and efficiently, but also know the future water consumption, thereby improving the user's usage experience.

[0188] S4. Classify users according to characteristic variables to obtain user classification sets, and obtain water volume prediction data Yc through the time convolution network method w To provide users with a more comprehensive understanding of their water usage, give them a general understanding of their future water usage, and enhance user experience, the specific implementation steps are as follows:

[0189] S41, characteristic variables include weekly water consumption Zz y , Fluctuation period T b and the growth rate θ z , the weekly water consumption Zz in the same period y , Fluctuation period T b and the growth rate θ z Integrate into water use vector Yw s , the expression is:

[0190] Yw s =[Zz y , T b ,θ z ]

[0191] Among them, Yw s represents the sth water consumption vector; weekly water consumption Zz y , Fluctuation period T b and the growth rate θ z These are just three types of analysis variables for water consumption. In practice, water consumption will be analyzed through more types of variables. The present invention uses these three variables as an example for explanation.

[0192] S42, according to the water use vector Yw s Calculation standard value B z , the calculation formula is:

[0193]

[0194] in, Represents the water use vector Yw s The average of m represents the water use vector Yw s the number of

[0195] S43, a standard value B z As the initial center Zx c , according to the standard value B z and the initial center Zx c Calculate the numerical spacing Rj s , the calculation formula is:

[0196] Rj s =||B z -Zx c || 2

[0197] Among them, Rj s Indicates the sth numerical interval;

[0198] S43, change the initial center Zx c Repeat step S43 to obtain multiple numerical intervals Rj s , calculate the optimal spacing value Rzj k , and the optimal spacing value Rzj k The corresponding initial center Zx c As a classification center fl x , the optimal spacing value Rzj k The expression is:

[0199]

[0200] Where L represents the numerical spacing Rj s the number of

[0201] S44, according to the classification center fl x Calculate the group mean value WC s , the calculation formula is:

[0202]

[0203] Among them, B represents the standard value B z and classification center fl x Number of groups;

[0204] S45, according to the group average value WC s Calculate the optimal classification value Zy v , the calculation formula is:

[0205]

[0206] Among them, Zy v represents the vth optimal classification value, Indicates the number of Q categories divided into the s-1th population average;

[0207] S46, repeat step S45 to obtain multiple optimal classification values ​​Zy v , the optimal classification value Zy v The number of categories Q corresponding to the minimum value in is taken as the optimal classification scheme, and the user classification set is obtained;

[0208] S47, construct a water consumption prediction model by using the time convolution network method, input the user classification data into the water consumption prediction model, and obtain the water consumption prediction data Yc wThe temporal convolutional network method is a model for predicting water consumption by combining the time dimension with the convolutional neural network. It is a commonly used method for establishing models and will not be described in detail here. By acquiring water consumption data at regular intervals, it is possible to complete regular monitoring of water consumption data and obtain multiple water consumption features for calculating water consumption data. It can more accurately analyze the user's water consumption habits, facilitate a comprehensive prediction of the user's water consumption, and provide the user's future water consumption, so that the user can not only quickly and efficiently obtain the water consumption situation, but also know the future water consumption, thereby improving the user's experience.

[0209] S5. Based on the water volume forecast data Yc w Calculate the dimensionality reduction value Jw z , and based on the dimensionality reduction value Jw z Generate a visual three-dimensional graph; obtain the water volume prediction data Yc through the water use prediction model w Contains data of multiple dimensions, which will increase the algorithm operation steps in the analysis process. In order to solve this problem, it is necessary to calculate the water volume prediction data Yc w To perform dimensionality reduction, the specific implementation steps are as follows:

[0210] S51, the water volume forecast data Yc w Transformed into an m×n analysis matrix Fx through data conversion method j , the expression is:

[0211]

[0212] Among them, R 11 Represents the analysis matrix Fx j The data value of the first row and first column in the table, n represents the water volume forecast data Yc w The number of characteristic variables, m represents the water volume prediction data Yc in each characteristic variable w the number of

[0213] S52. Calculate and analyze the matrix Fx j The variance matrix Xf j , the calculation formula is:

[0214]

[0215] Among them, Xf j represents the j-th variance matrix;

[0216] S53, through the characteristic decomposition method of the variance matrix Xf j Perform eigendecomposition to obtain the characteristic matrix Tz j and the characteristic matrix Tz j The corresponding diagonal value λ i ;

[0217] S54, according to the characteristic matrix Tz j Calculate the dimensionality reduction value Jw z , the calculation formula is:

[0218] Jw z =(Fx j -F j )×Tz j

[0219] Among them, F j Represents the analysis matrix Fx j The mean value of

[0220] S55, according to the contribution value λ i and the characteristic matrix Tz j Calculate the weight value Qz d , the calculation formula is:

[0221]

[0222] Among them, Qz d Represents the dth weight value;

[0223] S56, according to the weight value Qz d The dimension reduction value Jw z The data is sorted from large to small, and the sorting results are mapped into three-dimensional space to obtain a three-dimensional visualization. By reducing the dimensionality of water use data, the data dimension can be reduced and the dimensional disaster problem can be reduced. While improving the performance of the method, the reduced-dimensional data can also be visualized more vividly and comprehensively, so that the water use situation obtained by the user is a three-dimensional visualization image, allowing the user to obtain water use information more intuitively and quickly.

[0224] S6. Send the visualized three-dimensional image to the user data center, which will display the visualized three-dimensional image to the user through a display screen. During the data transmission and storage process, in order to ensure the privacy of user data, the present invention also encrypts the data through a hash algorithm. Since the hash algorithm is a commonly used encryption method, it will not be described here.

[0225] See also Figure 2 , which is a structural block diagram of the water meter data automatic meter reading system provided by this embodiment. The automatic meter reading system includes a preliminary processing module, a preprocessing module, a feature module, a classification prediction module, a dimensionality reduction module, and a visualization module.

[0226] The preliminary processing module is used to collect the original data of the water meter at any stage of operation and perform preliminary outlier removal on the original data to obtain normal data; the preprocessing module is used to perform in-depth preprocessing on the normal data to obtain preprocessed data; the feature module is used to divide the preprocessed data into multiple water volume data according to time periods.(t) , calculate water volume data Wa (t) The classification prediction module is used to classify users according to the characteristic variables to obtain user classification sets, and obtain water volume prediction data Yc through the time convolution network method. w ; Dimensionality reduction module, used to predict data Yc based on water volume w Calculate the dimensionality reduction value Jw z , and based on the dimensionality reduction value Jw z Generate a three-dimensional visualization graph; a visualization module is used to send the three-dimensional visualization graph to a user data center.

[0227] Those skilled in the art will appreciate that all or part of the steps in the above-mentioned embodiment methods can be accomplished by instructing the relevant hardware through a program. Therefore, the present application may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Furthermore, the present application may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0228] The above embodiments provide a detailed introduction to the present invention. Specific examples are used herein to illustrate the principles and implementation methods of the present invention. The description of the above embodiments is only used to help understand the method of the present invention and its core ideas. At the same time, for those skilled in the art, according to the ideas of the present invention, there may be changes in the specific implementation methods and application scopes. In summary, the contents of this specification should not be understood as limiting the present invention.

Claims

1. A method for automatic water meter reading, characterized in that: The method comprises the following steps: S1. Collect the original data of the water meter at any stage of operation, and perform preliminary outlier removal on the original data to obtain data without abnormalities; S2. Perform deep preprocessing on the data without abnormalities to obtain preprocessed data. The specific implementation steps are as follows: S21, according to the vacant position K cy Determine adjacent vacancies K cy+1 and K cy-1 , calculate the K bits used to fill the empty space in the data without abnormality cy Fill value Tc h , the calculation formula is: Among them, Tc h Indicates the hth filling value; S22, fill value Tc h Fill in the remaining K spaces without abnormal data cy On the above, we get the missing data Ws f ; S23. Calculate the detection value fw for no missing data z , the calculation formula is: Among them, erf -1 represents the inverse function of the error function, and p represents the cumulative probability value; S24, obtain the number of data without missing data by counting method, and calculate the number of data without missing data according to the detection value fw z Calculate the percentile value Fw d , the calculation formula is: Among them, Fw d represents the dth percentile value, fw z represents the zth detection value; S25, no missing data Ws f Sort by small to large to get the sorted data set P; S26, according to the percentile value Fw d The test coefficient H is obtained by table lookup method and the number of data N, and the test value Jh is calculated according to the test coefficient H b , the calculation formula is: Among them, Jh b represents the bth test value, Indicates no missing data Ws f the average of S3, divide the pre-processed data into multiple water volume data Wa according to time period (t) , calculate water volume data Wa (t) characteristic variables; S4. Classify users according to characteristic variables to obtain user classification sets, and obtain water volume prediction data Yc through the time convolution network method w ; S5. Based on the water volume forecast data Yc w Calculate the dimensionality reduction value Jw z , and based on the dimensionality reduction value Jw z Generate a three-dimensional visual graph; S6. Send the visualized three-dimensional image to the user data center.

2. The automatic meter reading method according to claim 1, characterized in that: In step S1, the specific implementation steps are as follows: S11. Arrange the original data in ascending order to generate a processing sequence C a , according to the processing sequence C a Calculate the first extreme value A1 and the second extreme value A2 using the following formula: A1=(C max +C min )×25% A2=(C max +C min )×75% Among them, C max and C min Represents the processing sequence C a The maximum and minimum values ​​in ; S12, calculating the measured value B based on the first extreme value A1 and the second extreme value A2 c , the calculation formula is: <h2 style=";text-align:left;direction:ltr">B<h2 style=";text-align:left;direction:ltr"> c <h2 style=";text-align:left;direction:ltr"> =A2-A1 Among them, B c represents the cth measurement value; S13, according to the measured value B c Calculation is used to represent the processing sequence C a The maximum value J in the normal data range jd and the minimum value J jx , the calculation formula is: <h2 style=";text-align:left;direction:ltr">J<h2 style=";text-align:left;direction:ltr"> jd <h2 style=";text-align:left;direction:ltr"> =A2+1.5×B<h2 style=";text-align:left;direction:ltr"> c J jx =A1-1.5×B c Among them, J jd represents the d-th maximum value, J jx represents the xth minimum value; S14, according to the maximum value J jd and the minimum value J jx For processing sequence C a The outliers are removed; If C a ≤J jx or C a ≥J jd , then process sequence C a The outliers are removed and the remaining K bits are obtained. cy ; If J jx <C a <J jd , then process sequence C a Normal value and retained; S15, the K containing the vacant position cy Processing sequence C a The data sequence is restored and data without abnormalities is obtained.

3. The automatic meter reading method according to claim 1, characterized in that: In step S26, the specific implementation steps are as follows: S261, according to the number of data N and the test value Jh b The critical value p is obtained by table lookup method; S262. Determine whether there is missing data Ws based on the critical value p f Is it a normal distribution? If p<0.05, it means that the data does not obey the normal distribution, and the process goes to step S263; If p≥0.05, it means that the data obeys the normal distribution, and the process goes to step S265; S263. Calculate the non-positive test value Jy under non-normal distribution z , the calculation formula is: Among them, Ws f-1 and Ws1 represent the f-1th and first non-missing data, respectively; S264, according to the non-positive test value Jy z For non-missing data Ws f Perform outlier detection; If Jy z <2, it is a normal value and preprocessed data is generated; If Jy z ≥2, it is marked as an outlier and the first replacement value Th is calculated. e And replace the outliers; S265. Calculate the normal test value Zj under normal distribution c , the calculation formula is: Where s represents no missing data Ws f The sample standard deviation of S266, according to the normal inspection value Zj c For non-missing data Ws f Perform outlier detection; If Zj c <2, it is a normal value and preprocessed data is generated; If Zj c ≥2, it is marked as an outlier and the second replacement value Te is calculated. h And replace the outliers.

4. The automatic meter reading method according to claim 3, characterized in that: The first replacement value Th e and the second replacement value Te h The calculation formula is: Among them, Ws f+1 and Ws f-1 Respectively represent the non-missing data Ws f There are no missing data for the f+1th and f-1th on both sides.

5. The automatic meter reading method according to claim 1, characterized in that: In step S3, the specific implementation steps are as follows: S31. Define the water volume data of the t-th time period as Wa (t) , calculate water volume data Wa (t) The periodic model Zq m , the calculation formula is: <h2 style=";text-align:left;direction:ltr">Zq<h2 style=";text-align:left;direction:ltr"> m <h2 style=";text-align:left;direction:ltr"> =D·sin(ρt+δ)+G Where D represents the model coefficient, ρ represents the periodic coefficient, δ represents the initial phase, and G represents the model offset value; S32. Calculate weekly water consumption Zz y , the calculation formula is: Among them, n represents the water volume data Wa in one week (t) The number of Wa (t) Represents the water volume data for the tth period; S33. Calculate the fluctuation period T of water consumption based on the periodic coefficient ρ b , the calculation formula is: Among them, T b represents the bth fluctuation period; S34. Calculate the growth rate θ of water volume data z , the calculation formula is: Among them, Zq m+1 Represents the periodic model value of the m+1th water volume data.

6. The automatic meter reading method according to claim 1, characterized in that: In step S4, the specific implementation steps are as follows: S41, the characteristic variables include weekly water consumption Zz y , Fluctuation period T b and the growth rate θ z , the weekly water consumption Zz in the same period y , Fluctuation period T b and the growth rate θ z Integrate into water use vector Yw s , the expression is: <h2 style=";text-align:left;direction:ltr">Yw<h2 style=";text-align:left;direction:ltr"> s <h2 style=";text-align:left;direction:ltr"> =[Zz<h2 style=";text-align:left;direction:ltr"> y <h2 style=";text-align:left;direction:ltr"> ,T<h2 style=";text-align:left;direction:ltr"> b <h2 style=";text-align:left;direction:ltr"> ,θ<h2 style=";text-align:left;direction:ltr"> z <h2 style=";text-align:left;direction:ltr"> ] Among them, Yw s represents the sth water consumption vector; S42, according to the water use vector Yw s Calculation standard value B z , the calculation formula is: in, Represents the water use vector Yw s The average of m represents the water use vector Yw s the number of S43, a standard value B z As the initial center Zx c , according to the standard value B z and the initial center Zx c Calculate the numerical spacing Rj s , the calculation formula is: Rj s =||B z -Zx c || 2 Among them, Rj s Indicates the sth numerical interval; S43, change the initial center Zx c Repeat step S43 to obtain multiple numerical intervals Rj s , calculate the optimal spacing value Rzj k , and the optimal spacing value Rzj k The corresponding initial center Zx c As a classification center fl x , the optimal spacing value Rzj k The expression is: Where L represents the numerical spacing Rj s the number of S44, according to the classification center fl x Calculate the group mean value WC s , the calculation formula is: Among them, B represents the standard value B z and classification center fl x Number of groups; S45, according to the group average value WC s Calculate the optimal classification value Zy v , the calculation formula is: Among them, Zy v represents the vth optimal classification value, Indicates the number of Q categories divided into the s-1th population average; S46, repeat step S45 to obtain multiple optimal classification values ​​Zy v , the optimal classification value Zy v The number of categories Q corresponding to the minimum value in is taken as the optimal classification scheme, and the user classification set is obtained; S47, construct a water consumption prediction model by using the time convolution network method, input the user classification data into the water consumption prediction model, and obtain the water consumption prediction data Yc w .

7. The automatic meter reading method according to claim 1, characterized in that: In step S5, the specific implementation steps are as follows: S51, the water volume forecast data Yc w Transformed into an m×n analysis matrix Fx through data conversion method j , the expression is: Among them, R 11 Represents the analysis matrix Fx j The data value of the first row and first column in the table, n represents the water volume forecast data Yc w The number of characteristic variables, m represents the water volume prediction data Yc in each characteristic variable w the number of S52. Calculate and analyze the matrix Fx j The variance matrix Xf j , the calculation formula is: Among them, Xf j represents the j-th variance matrix; S53, through the characteristic decomposition method of the variance matrix Xf j Perform eigendecomposition to obtain the characteristic matrix Tz j and the characteristic matrix Tz j The corresponding diagonal value λ i ; S54, according to the characteristic matrix Tz j Calculate the dimensionality reduction value Jw z , the calculation formula is: Jw z =(Fx j -F j )×Tz j Among them, F j Represents the analysis matrix Fx j The mean value of S55, according to the contribution value λ i and the characteristic matrix Tz j Calculate the weight value Qz d , the calculation formula is: Among them, Qz d Represents the dth weight value; S56, according to the weight value Qz d The dimension reduction value Jw z Sort from large to small, and map the sorting results into three-dimensional space to obtain a visual three-dimensional graph.

8. A system applied to the automatic meter reading method according to any one of claims 1 to 7, characterized in that: The system includes: The preliminary processing module is used to collect the original data of the water meter at any stage of operation and perform preliminary outlier removal on the original data to obtain data without abnormalities; The preprocessing module is used to perform deep preprocessing on the data without abnormalities to obtain preprocessed data; Feature module, used to divide the pre-processed data into multiple water volume data Wa according to time periods (t) , calculate water volume data Wa (t) characteristic variables; The classification prediction module is used to classify users according to feature variables to obtain user classification sets, and obtain water volume prediction data Yc through the time convolution network method. w ; Dimensionality reduction module, used to predict data Yc based on water volume w Calculate the dimensionality reduction value Jw z , and based on the dimensionality reduction value Jw z Generate a three-dimensional visual graph; The visualization module is used to send the visualized three-dimensional graph to the user data center.

Images