Water meter reading information verification system

Abstract

The invention relates to a water meter reading information verification system, which comprises an object verification mechanism used for intelligently identifying an upper limit prediction value and a lower limit prediction value of a reclaimed water meter reading total number corresponding to a current time segment of a target block based on an AI identification model according to multiple items of basic information screened in a targeted manner; and the state alarm mechanism is used for sending a meter reading information alarm signal when the actual total water meter reading number of the target block in the current time segment is not between the upper limit prediction value and the lower limit prediction value. The water meter reading information verification system is stable in operation and intelligent in design. The AI identification model with a customized structure can be adopted to accurately determine the upper limit prediction value and the lower limit prediction value of the reclaimed water meter reading total number corresponding to the current time segment of the target block, so that a reference range is provided for reliability identification of the meter reading data of the target block in the future time segment.

Description

Technical Field

[0001] The invention relates to the field of meter reading, and in particular to a water meter reading information verification system. Background Art

[0002] Greywater meters and tap water meters have the same functions, but their design, materials, accuracy, installation location and reading methods are different to suit the special needs of greywater systems. Greywater treatment costs are lower, and governments may provide preferential policies to encourage the use of greywater. The pricing structure of water charges may vary depending on the type of water and the amount of water used. Please check local regulations and charging standards for details. Greywater meters and tap water meters are both instruments used to measure and record water usage. They are usually mechanical or electronic, and can rotate or electronically count according to the amount of water flowing through.

[0003] However, there is a large deviation in the total amount of grey water usage obtained from meter readings within the set time period of each block. The reason is that there are more users of grey water in larger blocks, there is a possibility of leakage during water use, and there are errors in remote meter reading of grey water meters. How to determine the total amount of grey water meter readings for each block within the set time period obtained by remote meter reading is an important reference data for allocating limited grey water resources between blocks and calculating grey water fees. Summary of the invention

[0004] In order to solve technical problems in related fields, the present invention provides a water meter reading information verification system, which can accurately determine the upper and lower predicted values ​​of the total number of grey water meter readings corresponding to the target block in the current time segment. Specifically, the preset time length, various grey water configuration information of the target block, and the total number of grey water meter readings corresponding to the target block in multiple past time segments are synchronously input into an AI identification model, and the AI ​​identification model is executed to obtain the upper and lower predicted values ​​of the total number of grey water meter readings corresponding to the target block in the current time segment output by the AI ​​identification model, thereby completing the intelligent prediction of the value range of the meter reading data of the target block in the future time segments, and providing a reference range for the reliability identification of the meter reading data of the target block in the future time segments.

[0005] According to the present invention, a water meter reading information verification system is provided, the system comprising: A continuous learning component, for training the deep residual network for each time to obtain the deep residual network after each training and output it as an AI identification model, wherein the number of times the deep residual network is trained is positively correlated with the number of reclaimed water users in the target block; The first capture component is used to obtain the total number of water meter readings corresponding to the target block in multiple past time segments, where the multiple past time segments are before the current time segment and the duration of each time segment is equal and is a preset time length; The second capture component is used to obtain the block area, the number of greywater users, the number of residential buildings and the length of greywater pipelines of the target block, so as to output various greywater configuration information of the target block; The object verification mechanism is connected to the continuous learning component, the first capture component and the second capture component respectively, and is used to synchronously input the preset time length, various types of greywater configuration information of the target block and the total number of greywater meter readings corresponding to the target block in multiple past time segments into the AI ​​identification model, and execute the AI ​​identification model to obtain the upper limit prediction value and the lower limit prediction value of the total number of greywater meter readings corresponding to the target block in the current time segment output by the AI ​​identification model; A state alarm mechanism, connected to the object verification mechanism, for sending a meter reading information alarm signal when the total number of actual water meter readings in the target block in the current time segment is not between the upper limit prediction value and the lower limit prediction value, and for sending a meter reading information valid signal when the total number of actual water meter readings in the target block in the current time segment is between the upper limit prediction value and the lower limit prediction value; Among them, when the total number of actual grey water meter readings in the target block in the current time segment is not between the upper limit prediction value and the lower limit prediction value, a meter reading information alarm signal is issued, and when the total number of actual grey water meter readings in the target block in the current time segment is between the upper limit prediction value and the lower limit prediction value, a meter reading information valid signal is issued, including: the total number of actual grey water meter readings in the target block in the current time segment is the sum of the meter reading data of each grey water meter corresponding to each grey water user in the target block within the current time segment.

[0006] Therefore, the present invention has the following outstanding technical effects: First, the deep residual network is trained each time to obtain the deep residual network after each training and output as the AI ​​identification model. The number of times the deep residual network is trained is positively correlated with the number of reclaimed water users in the target block, so as to customize AI identification models with different structures for different blocks. Second: obtain the total number of water meter readings corresponding to the target block in multiple past time segments, where the multiple past time segments are before the current time segment and the duration of each time segment is equal, all of which are preset time lengths. Also obtain the block area, number of water users, number of residential buildings and length of water pipelines of the target block as the output of various water configuration information of the target block, thereby providing comprehensive basic information for subsequent intelligent prediction operations; Third: synchronously input the preset time length, various reclaimed water configuration information of the target block, and the total number of reclaimed water meter readings corresponding to the target block in multiple past time segments into the AI ​​identification model, and execute the AI ​​identification model to obtain the upper limit prediction value and the lower limit prediction value of the total number of reclaimed water meter readings corresponding to the target block in the current time segment output by the AI ​​identification model, thereby completing the intelligent prediction of the value range of meter reading data in the future time segments of the target block; Fourthly: when the total number of actual recycled water meter readings in the target block in the current time segment is not between the upper limit prediction value and the lower limit prediction value of the intelligent prediction, a meter reading information alarm signal is issued; and when the total number of actual recycled water meter readings in the target block in the current time segment is between the upper limit prediction value and the lower limit prediction value, a meter reading information valid signal is issued, thereby further ensuring the reliability and validity of the meter reading information in each time segment.

[0007] The water meter reading information verification system of the present invention is stable in operation and intelligent in design. Since the customized AI identification model can be used to accurately determine the upper and lower predicted values ​​of the total number of water meter readings corresponding to the target block in the current time segment, a reference range is provided for the reliability identification of the meter reading data of the target block in the future time segment. BRIEF DESCRIPTION OF THE DRAWINGS

[0008] The embodiments of the present invention will be described below with reference to the accompanying drawings, wherein:

[0009] Figure 1 The figure is a structural block diagram of a water meter reading information verification system according to the first embodiment of the present invention.

[0010] Figure 2 The figure is a structural block diagram of a water meter reading information verification system according to the second embodiment of the present invention.

[0011] Figure 3 The structure block diagram of the water meter reading information verification system according to the third embodiment of the present invention is shown. DETAILED DESCRIPTION

[0012] The embodiment of the water meter reading information verification system of the present invention will be described in detail below with reference to the accompanying drawings.

[0013] Figure 1 The structure block diagram of the water meter reading information verification system according to the first embodiment of the present invention is shown, and the system includes: A continuous learning component, for training the deep residual network for each time to obtain the deep residual network after each training and output it as an AI identification model, wherein the number of times the deep residual network is trained is positively correlated with the number of reclaimed water users in the target block; Specifically, a continuous learning component is used to perform training on a deep residual network for each time to obtain a deep residual network after each training and output it as an AI identification model, and the number of times the deep residual network performs each training is positively correlated with the number of reclaimed water users in the target block, including: using a programmable logic device to implement the continuous learning component, used to perform training on a deep residual network for each time to obtain a deep residual network after each training and output it as an AI identification model, and the number of times the deep residual network performs each training is positively correlated with the number of reclaimed water users in the target block; The first capture component is used to obtain the total number of water meter readings corresponding to the target block in multiple past time segments, where the multiple past time segments are before the current time segment and the duration of each time segment is equal and is a preset time length; The second capture component is used to obtain the block area, the number of greywater users, the number of residential buildings and the length of greywater pipelines of the target block, so as to output various greywater configuration information of the target block; The object verification mechanism is connected to the continuous learning component, the first capture component and the second capture component respectively, and is used to synchronously input the preset time length, various types of greywater configuration information of the target block and the total number of greywater meter readings corresponding to the target block in multiple past time segments into the AI ​​identification model, and execute the AI ​​identification model to obtain the upper limit prediction value and the lower limit prediction value of the total number of greywater meter readings corresponding to the target block in the current time segment output by the AI ​​identification model; A state alarm mechanism, connected to the object verification mechanism, for sending a meter reading information alarm signal when the total number of actual water meter readings in the target block in the current time segment is not between the upper limit prediction value and the lower limit prediction value, and for sending a meter reading information valid signal when the total number of actual water meter readings in the target block in the current time segment is between the upper limit prediction value and the lower limit prediction value; Wherein, when the total number of actual water meter readings in the target block in the current time segment is not between the upper limit prediction value and the lower limit prediction value, a meter reading information alarm signal is issued, and when the total number of actual water meter readings in the target block in the current time segment is between the upper limit prediction value and the lower limit prediction value, a meter reading information valid signal is issued, including: the total number of actual water meter readings in the target block in the current time segment is the sum of meter reading data of each water meter corresponding to each water user in the target block in the current time segment; The first capture component is used to obtain the total number of water meter readings corresponding to the target block in the past multiple time segments, wherein the past multiple time segments are before the current time segment and the duration of each time segment is equal, and the preset time length includes: the number of the past multiple time segments is proportional to the number of water users in the target block; Among them, the first capture component is used to obtain the total number of each grey water meter reading corresponding to the target block in multiple past time segments, wherein the multiple past time segments are before the current time segment and the duration of each time segment is equal, and all are preset time lengths. It also includes: for each time segment, the total number of grey water meter readings corresponding to it is the sum of the grey water consumption data of each grey water user in the corresponding target block; And wherein, a continuous learning component is used to perform training on the deep residual network each time to obtain the deep residual network after each training and output it as an AI identification model, and the number of times the deep residual network performs each training is positively correlated with the number of grey water users in the target block, including: using an information conversion function to represent the information conversion relationship between the number of times the deep residual network performs each training and the number of grey water users in the target block.

[0014] Figure 2 The figure is a structural block diagram of a water meter reading information verification system according to the second embodiment of the present invention.

[0015] Compared to Figure 1 , Figure 2 The water meter reading information verification system in the invention may also include: a directional acquisition component, disposed near the object verification mechanism, the continuous learning component, the first capture component, and the second capture component and connected to the object verification mechanism, the continuous learning component, the first capture component, and the second capture component respectively; Among them, the directional acquisition component is arranged near the object verification mechanism, the continuous learning component, the first capture component and the second capture component and is respectively connected to the object verification mechanism, the continuous learning component, the first capture component and the second capture component, including: the directional acquisition component is used to realize on-site measurement of the current of the object verification mechanism, the continuous learning component, the first capture component and the second capture component respectively.

[0016] Figure 3 The structure block diagram of the water meter reading information verification system according to the third embodiment of the present invention is shown.

[0017] Compared to Figure 1 , Figure 3 The water meter reading information verification system in the invention may also include: a parameter parsing component, which is disposed near the object verification mechanism, the continuous learning component, the first capture component, and the second capture component and is connected to the object verification mechanism, the continuous learning component, the first capture component, and the second capture component respectively; Among them, the parameter parsing component is arranged near the object verification mechanism, the continuous learning component, the first capture component and the second capture component and is respectively connected to the object verification mechanism, the continuous learning component, the first capture component and the second capture component, including: the parameter parsing component is used to respectively realize the on-site measurement of the current heat dissipation of the object verification mechanism, the continuous learning component, the first capture component and the second capture component.

[0018] Next, the specific structure of the water meter reading information verification system of the present invention will be further described.

[0019] In the water meter reading information verification system according to various embodiments of the present invention: Using an FPGA device to perform image data processing on the output data of the object verification mechanism, the continuous learning component, the first capture component, and the second capture component to obtain output processing data corresponding to the object verification mechanism, the continuous learning component, the first capture component, and the second capture component respectively; Wherein, using an FPGA device to perform image data processing on the output data of the object verification mechanism, the continuous learning component, the first capture component and the second capture component to obtain the output processing data corresponding to the object verification mechanism, the continuous learning component, the first capture component and the second capture component respectively includes: performing guided filtering processing on the output data of the object verification mechanism, the continuous learning component, the first capture component and the second capture component to obtain the output processing data corresponding to the object verification mechanism, the continuous learning component, the first capture component and the second capture component respectively; Wherein, using an FPGA device to perform image data processing on the output data of the object verification mechanism, the continuous learning component, the first capture component and the second capture component to obtain the output processing data corresponding to the object verification mechanism, the continuous learning component, the first capture component and the second capture component respectively includes: performing box filtering processing on the output data of the object verification mechanism, the continuous learning component, the first capture component and the second capture component to obtain the output processing data corresponding to the object verification mechanism, the continuous learning component, the first capture component and the second capture component respectively; Wherein, using an FPGA device to perform image data processing on the output data of the object verification mechanism, the continuous learning component, the first capture component and the second capture component to obtain the output processing data corresponding to the object verification mechanism, the continuous learning component, the first capture component and the second capture component respectively comprises: performing wavelet filtering processing on the output data of the object verification mechanism, the continuous learning component, the first capture component and the second capture component to obtain the output processing data corresponding to the object verification mechanism, the continuous learning component, the first capture component and the second capture component respectively; And wherein, using an FPGA device to perform image data processing on the output data of the object verification mechanism, the continuous learning component, the first capture component and the second capture component to obtain the output processing data corresponding to the object verification mechanism, the continuous learning component, the first capture component and the second capture component respectively includes: performing bilinear interpolation processing on the output data of the object verification mechanism, the continuous learning component, the first capture component and the second capture component to obtain the output processing data corresponding to the object verification mechanism, the continuous learning component, the first capture component and the second capture component respectively.

[0020] In addition, in the water meter reading information verification system, an information conversion function is used to represent the information conversion relationship in which the number of times the deep residual network performs each training is positively correlated with the number of grey water users in the target block, including: in the information conversion function, the number of grey water users in the target block is the input information of the information conversion function, and the number of times the deep residual network performs each training corresponding to the number of grey water users in the target block is the output information of the information conversion function.

[0021] While the invention has been particularly shown and described with reference to exemplary embodiments thereof, it will be understood by those skilled in the art that various changes in form and details may be made therein without departing from the spirit and scope of the invention as defined by the claims.

Claims

1. A water meter reading information verification system, characterized in that: The system comprises: A continuous learning component, for training the deep residual network for each time to obtain the deep residual network after each training and output it as an AI identification model, wherein the number of times the deep residual network is trained is positively correlated with the number of reclaimed water users in the target block; The first capture component is used to obtain the total number of water meter readings corresponding to the target block in multiple past time segments, where the multiple past time segments are before the current time segment and the duration of each time segment is equal and is a preset time length; The second capture component is used to obtain the block area, the number of greywater users, the number of residential buildings and the length of greywater pipelines of the target block, so as to output various greywater configuration information of the target block; The object verification mechanism is connected to the continuous learning component, the first capture component and the second capture component respectively, and is used to synchronously input the preset time length, various types of greywater configuration information of the target block and the total number of greywater meter readings corresponding to the target block in multiple past time segments into the AI ​​identification model, and execute the AI ​​identification model to obtain the upper limit prediction value and the lower limit prediction value of the total number of greywater meter readings corresponding to the target block in the current time segment output by the AI ​​identification model; A status alarm mechanism is connected to the object verification mechanism, and is used to issue a meter reading information alarm signal when the total number of actual grey water meter readings in the target block in the current time segment is not between the upper limit prediction value and the lower limit prediction value, and is also used to issue a meter reading information valid signal when the total number of actual grey water meter readings in the target block in the current time segment is between the upper limit prediction value and the lower limit prediction value, including: the total number of actual grey water meter readings in the target block in the current time segment is the sum of the meter reading data of each grey water meter corresponding to each grey water user in the target block within the current time segment.

2. The water meter reading information verification system according to claim 1, characterized in that: The first capture component is used to obtain the total number of each water meter reading corresponding to the target block in the past multiple time segments, wherein the past multiple time segments are before the current time segment and the duration of each time segment is equal, and all are preset time lengths, including: the number of the past multiple time segments is proportional to the number of water users in the target block; Among them, the first capture component is used to obtain the total number of each grey water meter reading corresponding to the target block in multiple past time segments, wherein the multiple past time segments are before the current time segment and the duration of each time segment is equal, and all are preset time lengths. It also includes: for each time segment, the total number of grey water meter readings corresponding to it is the sum of the grey water consumption data of each grey water user in the corresponding target block; Among them, the continuous learning component is used to train the deep residual network multiple times to obtain the deep residual network after each training and output it as the AI ​​identification model. The number of times the deep residual network performs each training is positively correlated with the number of grey water users in the target block, including: using an information conversion function to represent the information conversion relationship between the number of times the deep residual network performs each training and the number of grey water users in the target block.

3. The water meter reading information verification system according to claim 2, characterized in that: The system further comprises: a directional acquisition component, disposed near the object verification mechanism, the continuous learning component, the first capture component, and the second capture component and connected to the object verification mechanism, the continuous learning component, the first capture component, and the second capture component respectively; Among them, the directional acquisition component is arranged near the object verification mechanism, the continuous learning component, the first capture component and the second capture component and is respectively connected to the object verification mechanism, the continuous learning component, the first capture component and the second capture component, including: the directional acquisition component is used to realize on-site measurement of the current of the object verification mechanism, the continuous learning component, the first capture component and the second capture component respectively.

4. The water meter reading information verification system according to claim 2, characterized in that: The system further comprises: a parameter parsing component, which is disposed near the object verification mechanism, the continuous learning component, the first capture component, and the second capture component and is connected to the object verification mechanism, the continuous learning component, the first capture component, and the second capture component respectively; Among them, the parameter parsing component is arranged near the object verification mechanism, the continuous learning component, the first capture component and the second capture component and is respectively connected to the object verification mechanism, the continuous learning component, the first capture component and the second capture component, including: the parameter parsing component is used to respectively realize the on-site measurement of the current heat dissipation of the object verification mechanism, the continuous learning component, the first capture component and the second capture component.

5. The water meter reading information verification system according to any one of claims 2 to 4, characterized in that: An FPGA device is used to perform image data processing on the output data of the object verification mechanism, the continuous learning component, the first capture component and the second capture component to obtain output processing data corresponding to the object verification mechanism, the continuous learning component, the first capture component and the second capture component respectively.

6. The water meter reading information verification system according to claim 5, characterized in that: Using an FPGA device to perform image data processing on the output data of the object verification mechanism, the continuous learning component, the first capture component and the second capture component to obtain the output processing data corresponding to the object verification mechanism, the continuous learning component, the first capture component and the second capture component respectively includes: performing guided filtering processing on the output data of the object verification mechanism, the continuous learning component, the first capture component and the second capture component to obtain the output processing data corresponding to the object verification mechanism, the continuous learning component, the first capture component and the second capture component respectively.

7. The water meter reading information verification system according to claim 5, characterized in that: Using an FPGA device to perform image data processing on the output data of the object verification mechanism, the continuous learning component, the first capture component and the second capture component to obtain the output processing data corresponding to the object verification mechanism, the continuous learning component, the first capture component and the second capture component respectively includes: performing box filtering processing on the output data of the object verification mechanism, the continuous learning component, the first capture component and the second capture component to obtain the output processing data corresponding to the object verification mechanism, the continuous learning component, the first capture component and the second capture component respectively.

8. The water meter reading information verification system according to claim 5, characterized in that: Using an FPGA device to perform image data processing on the output data of the object verification mechanism, the continuous learning component, the first capture component and the second capture component to obtain the output processing data corresponding to the object verification mechanism, the continuous learning component, the first capture component and the second capture component respectively includes: performing wavelet filtering processing on the output data of the object verification mechanism, the continuous learning component, the first capture component and the second capture component to obtain the output processing data corresponding to the object verification mechanism, the continuous learning component, the first capture component and the second capture component respectively.

9. The water meter reading information verification system according to claim 5, characterized in that: Using an FPGA device to perform image data processing on the output data of the object verification mechanism, the continuous learning component, the first capture component and the second capture component to obtain the output processing data corresponding to the object verification mechanism, the continuous learning component, the first capture component and the second capture component respectively includes: performing bilinear interpolation processing on the output data of the object verification mechanism, the continuous learning component, the first capture component and the second capture component to obtain the output processing data corresponding to the object verification mechanism, the continuous learning component, the first capture component and the second capture component respectively.

Images