Meter reading information prediction system
By adopting a targeted structural design artificial intelligence mechanism in the intelligent water meter prediction system, the meter reading range of intelligent water meter is predicted based on the automatic encoder neural network model, the problem of inability to effectively predict the meter reading range of intelligent water meter in the existing technology is solved, and the prediction effect of high accuracy and reliability is achieved.
Patent Information
- Application Number
- CN202510132127.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-02-06
- Publication Date
- 2025-05-13
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
The existing technology lacks mature technical solutions and cannot effectively predict the range of meter reading values of smart water meters in the set time interval, resulting in the inability to judge the validity of meter reading values.
Using an artificial intelligence mechanism with targeted structural design, the meter reading range of intelligent water meters is predicted based on the automatic encoder neural network model. The model obtains the automatic encoder neural network after completing the learning through multiple learning operations, and intelligently analyzes it based on the time length of the set time interval, multiple related contents of the intelligent water meter, and historical meter reading values.
Accurate prediction of the range of meter reading values of smart water meter is realized, providing key reference information for judging the validity of meter reading values, and improving the stability and reliability of predictions.
Smart Images
Figure CN119990450A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of artificial intelligence, and in particular to a meter reading information prediction system. Background Art
[0002] Artificial Intelligence (AI) is a new technology science that studies and develops theories, methods, technologies and application systems for simulating, extending and expanding human intelligence. It is a branch of computer science. Artificial Intelligence aims to understand the essence of intelligence and produce a new intelligent machine that can respond in a similar way to human intelligence. Research in this field includes robotics, speech recognition, image recognition, natural language processing and expert systems. Artificial Intelligence can simulate the information process of human consciousness and thinking. It is not human intelligence, but it can think like a human and may even exceed human intelligence.
[0003] As a newly developed technology, artificial intelligence still has many gaps that need to be filled. For example, how to apply artificial intelligence to various sub-sectors that do not yet have mature technical solutions. For example, if the artificial intelligence mechanism can be used to intelligently predict the meter reading value range of the current smart water meter in the set time period of the day, then it can be determined whether the meter reading value of the current smart water meter in the set time period of the day is a valid value. Obviously, there is a lack of corresponding mature technical solutions in the existing technology. Summary of the invention
[0004] In order to solve the technical problems in the prior art, the present invention provides a meter reading information prediction system, which intelligently predicts the meter reading value range of the current smart water meter in the set time interval of the day by adopting an artificial intelligence mechanism with targeted structural design. The intelligent prediction is based on an autoencoder neural network model, wherein multiple learning operations are performed on the autoencoder neural network in succession to obtain the autoencoder neural network after completing multiple learning operations and output it as the autoencoder neural network model. The number of learning operations completed by the autoencoder neural network model is proportional to the time length of the set time interval, thereby realizing the targeted design of the structure of the autoencoder neural network model, and providing key reference information for judging whether the meter reading value of the current smart water meter in the set time interval of the day is a valid value.
[0005] According to the present invention, a meter reading information prediction system is provided, the system comprising: The first grabbing component is used to obtain the caliber, type code value, overload flow, working voltage and pressure loss level of the current smart water meter, and output the caliber, type code value, overload flow, working voltage and pressure loss level of the current smart water meter as multiple related contents of the current smart water meter; The second grabbing component is used to obtain the meter reading values corresponding to the set time intervals of each day in the history of the current smart water meter, and the set time interval of each day has the same position on the time axis of that day; A sequence conversion component is used to successively perform multiple learning operations on the autoencoder neural network to obtain the autoencoder neural network after completing the multiple learning operations and output it as an autoencoder neural network model, wherein the number of learning operations completed by the autoencoder neural network model is proportional to the time length of the set time interval; An information reading mechanism is connected to the first grabbing component, the second grabbing component and the order conversion component respectively, and is used to use the automatic encoder neural network model to intelligently analyze two prediction values corresponding to the maximum value and the minimum value of the meter reading value of the current smart water meter in the set time interval of the day according to the time length of the set time interval, multiple related contents of the current smart water meter and the meter reading values corresponding to the set time intervals of each day in the history of the current smart water meter; A data transmission mechanism, connected to the information reading mechanism, is used to provide two prediction values corresponding to the maximum and minimum values of the meter reading values of the current smart water meter in the set time interval of the day to a remote big data server through a frequency division transmission channel; The second grabbing component is used to obtain the meter reading values corresponding to the set time intervals of each day in the history of the current smart water meter, and the position of each set time interval on the time axis of the day is the same, including: the number of days of each selected historical day is positively correlated with the caliber of the current smart water meter; Among them, the second capture component is used to obtain the meter reading values corresponding to the set time intervals of each historical day of the current smart water meter. The set time interval of each day has the same position on the time axis of that day and also includes: using a parameter mapping function to represent the parameter mapping relationship between the selected historical days and the caliber of the current smart water meter.
[0006] It can be seen that the present invention has at least the following three significant substantive features: First, an artificial intelligence mechanism is used to intelligently predict the meter reading value range of the current smart water meter in the set time interval of the day. The intelligent prediction is based on an autoencoder neural network model, wherein multiple learning operations are performed on the autoencoder neural network in succession to obtain an autoencoder neural network after completing multiple learning operations and output it as an autoencoder neural network model. The number of learning operations completed by the autoencoder neural network model is proportional to the length of the set time interval, thereby realizing a targeted design of the structure of the autoencoder neural network model; Second: multiple basic data are introduced to serve the intelligent prediction of the meter reading value range of the current smart water meter in the set time interval of the day. The multiple basic data include the caliber, type code value, overload flow, working voltage and pressure loss level of the current smart water meter, and also include the meter reading values corresponding to the set time intervals of each day in the history of the current smart water meter. The full and comprehensive setting of the above multiple basic data ensures the stability and reliability of the intelligent prediction results; Third: Specifically, when obtaining the meter reading values corresponding to the set time intervals of each historical day of the current smart water meter, the set time interval of each day has the same position on the time axis of that day, and the number of days of the selected historical days is positively correlated with the caliber of the current smart water meter.
[0007] The meter reading information prediction system of the present invention is logically reliable and has certain pertinence. By adopting an artificial intelligence mechanism with a pertinent structural design to intelligently predict the meter reading value range of the current smart water meter in the set time interval of the day, it provides key reference information for judging whether the meter reading value of the current smart water meter in the set time interval of the day is a valid value. BRIEF DESCRIPTION OF THE DRAWINGS
[0008] The embodiments of the present invention will be described below with reference to the accompanying drawings, wherein: Figure 1 FIG. 1 is an internal structure diagram of a meter reading information prediction system according to a first embodiment of the present invention.
[0009] Figure 2 FIG. 1 is an internal structure diagram of a meter reading information prediction system according to a second embodiment of the present invention.
[0010] Figure 3 FIG. 4 is an internal structure diagram of a meter reading information prediction system according to a third embodiment of the present invention. DETAILED DESCRIPTION
[0011] The following is a detailed description of an embodiment of the meter reading information prediction system of the present invention with reference to the accompanying drawings.
[0012] Figure 1 The internal structure diagram of the meter reading information prediction system according to the first embodiment of the present invention is shown, and the system includes: The first grabbing component is used to obtain the caliber, type code value, overload flow, working voltage and pressure loss level of the current smart water meter, and output the caliber, type code value, overload flow, working voltage and pressure loss level of the current smart water meter as multiple related contents of the current smart water meter; For example, the first capture component is used to obtain the caliber, type code value, overload flow, working voltage and pressure loss level of the current smart water meter, and output the caliber, type code value, overload flow, working voltage and pressure loss level of the current smart water meter as multiple associated contents of the current smart water meter, including: the first capture component includes multiple data acquisition components, which are used to respectively acquire the caliber, type code value, overload flow, working voltage and pressure loss level of the current smart water meter; The second grabbing component is used to obtain the meter reading values corresponding to the set time intervals of each day in the history of the current smart water meter, and the set time interval of each day has the same position on the time axis of that day; A sequence conversion component is used to successively perform multiple learning operations on the autoencoder neural network to obtain the autoencoder neural network after completing the multiple learning operations and output it as an autoencoder neural network model, wherein the number of learning operations completed by the autoencoder neural network model is proportional to the time length of the set time interval; An information reading mechanism is connected to the first grabbing component, the second grabbing component and the order conversion component respectively, and is used to use the automatic encoder neural network model to intelligently analyze two prediction values corresponding to the maximum value and the minimum value of the meter reading value of the current smart water meter in the set time interval of the day according to the time length of the set time interval, multiple related contents of the current smart water meter and the meter reading values corresponding to the set time intervals of each day in the history of the current smart water meter; A data transmission mechanism, connected to the information reading mechanism, is used to provide two prediction values corresponding to the maximum and minimum values of the meter reading values of the current smart water meter in the set time interval of the day to a remote big data server through a frequency division transmission channel; The second grabbing component is used to obtain the meter reading values corresponding to the set time intervals of each day in the history of the current smart water meter, and the position of each set time interval on the time axis of the day is the same, including: the number of days of each selected historical day is positively correlated with the caliber of the current smart water meter; Among them, the second grabbing component is used to obtain the meter reading values corresponding to the set time intervals of each historical day of the current smart water meter, and the position of the set time interval of each day on the time axis of the day is the same, and also includes: using a parameter mapping function to represent the parameter mapping relationship of the number of days of each selected historical day and the caliber of the current smart water meter; And wherein, the automatic encoder neural network model is used to intelligently analyze two prediction values corresponding to the maximum value and the minimum value of the meter reading value of the current smart water meter in the set time interval of the day according to the time length of the set time interval, multiple related contents of the current smart water meter, and each meter reading value corresponding to the set time interval of each day in the history of the current smart water meter, including: the time length of the set time interval, multiple related contents of the current smart water meter, and each meter reading value corresponding to the set time interval of each day in the history of the current smart water meter are converted into binary values respectively and then synchronously input into the automatic encoder neural network model; And wherein, the time length of the set time interval, multiple related contents of the current smart water meter, and the meter reading values corresponding to the set time intervals of each historical day of the current smart water meter are converted into binary values and then synchronously input into the automatic encoder neural network model, including: using a numerical simulation mode to realize the simulation and testing of the automatic encoder neural network model.
[0013] Figure 2 FIG. 1 is an internal structure diagram of a meter reading information prediction system according to a second embodiment of the present invention.
[0014] exist Figure 2 In, with Figure 1 Differently, the meter reading information prediction system shown in the second embodiment of the present invention may further include: A connection service device, for connecting the information reading mechanism, the first grabbing component, the second grabbing component and the order conversion component, respectively, to receive data from the information reading mechanism, the first grabbing component, the second grabbing component and the order conversion component, respectively, and to send parallel data from the connection processing component to the information reading mechanism, the first grabbing component, the second grabbing component and the order conversion component, respectively; Among them, the connection service device is used for connecting the information reading mechanism, the first grabbing component, the second grabbing component and the order conversion component respectively, so as to receive data from the information reading mechanism, the first grabbing component, the second grabbing component and the order conversion component respectively, and send parallel data from the connection processing component to the information reading mechanism, the first grabbing component, the second grabbing component and the order conversion component respectively, including: the connection processing component is a serial communication connection interface.
[0015] Figure 3 FIG. 4 is an internal structure diagram of a meter reading information prediction system according to a third embodiment of the present invention.
[0016] exist Figure 3 In, with Figure 1Differently, the meter reading information prediction system shown in the third embodiment of the present invention may further include: The instruction transceiver device is respectively connected to the information reading mechanism, the first grabbing component, the second grabbing component and the order conversion component, and is used to receive user control instructions for the information reading mechanism, the first grabbing component, the second grabbing component and the order conversion component.
[0017] Next, the specific structure of the meter reading information prediction system of the present invention will be further described.
[0018] In the meter reading information prediction system according to any embodiment of the present invention: The information reading mechanism, the first grabbing component, the second grabbing component and the sequence conversion component are connected to the same oscillation actuator to obtain the timing data provided by the oscillation actuator, and the information reading mechanism, the first grabbing component, the second grabbing component and the sequence conversion component are respectively connected to the same content storage chip, and the content storage chip is one of FLASH flash memory, SDRAM memory chip and DDR memory chip.
[0019] In the meter reading information prediction system according to any embodiment of the present invention: The information reading mechanism, the first grabbing component, the second grabbing component and the order conversion component are respectively connected to the same IIC control bus, and are used to receive various control commands sent by the IIC control bus, and the various control commands are used to respectively configure the operation data of the information reading mechanism, the first grabbing component, the second grabbing component and the order conversion component; The information reading mechanism, the first grabbing component, the second grabbing component and the sequence conversion component are respectively connected to the IIC control bus and are used to receive various control instructions sent through the IIC control bus.
[0020] In the meter reading information prediction system according to any embodiment of the present invention: The information reading mechanism, the first grabbing component, the second grabbing component and the order conversion component are respectively connected to the same MCU controller, and are used for switching the information reading mechanism, the first grabbing component, the second grabbing component and the order conversion component between the sleep mode and the working mode respectively under the control of the same MCU controller; Wherein, the information reading mechanism, the first grabbing component, the second grabbing component and the order conversion component are respectively connected to the same MCU controller, and are used for switching the information reading mechanism, the first grabbing component, the second grabbing component and the order conversion component between the sleep mode and the working mode respectively under the control of the same MCU controller, including: the same MCU controller is designed based on the ARM13 core; And wherein, the information reading mechanism, the first grabbing component, the second grabbing component and the order conversion component are respectively connected to the same MCU controller, and are used for, under the control of the same MCU controller, the information reading mechanism, the first grabbing component, the second grabbing component and the order conversion component to switch between sleep mode and working mode respectively. It also includes: the information reading mechanism, the first grabbing component, the second grabbing component and the order conversion component use the same quartz oscillator device to provide the information reading mechanism, the first grabbing component, the second grabbing component and the order conversion component with the clock waveforms they need respectively.
[0021] In addition, in the meter reading information prediction system, the automatic encoder neural network model is used to intelligently analyze the meter reading values corresponding to the maximum and minimum values of the meter reading values of the current smart water meter in the set time interval of the day according to the time length of the set time interval, multiple related contents of the current smart water meter, and the meter reading values corresponding to the set time intervals of each day in the history of the current smart water meter. The two prediction values also include: the two prediction values corresponding to the maximum and minimum values of the meter reading values of the current smart water meter in the set time interval of the day output by the automatic encoder neural network model are in the form of binary values.
[0022] Although the present invention has been described with reference to the structures disclosed herein, it is not limited to the details shown, and this application is intended to cover such modifications or changes as fall within the scope of the following claims.
Claims
1. A meter reading information prediction system, characterized in that: The system comprises: The first grabbing component is used to obtain the caliber, type code value, overload flow, working voltage and pressure loss level of the current smart water meter, and output the caliber, type code value, overload flow, working voltage and pressure loss level of the current smart water meter as multiple related contents of the current smart water meter; The second grabbing component is used to obtain the meter reading values corresponding to the set time intervals of each day in the history of the current smart water meter, and the set time interval of each day has the same position on the time axis of that day; A sequence conversion component is used to successively perform multiple learning operations on the autoencoder neural network to obtain the autoencoder neural network after completing the multiple learning operations and output it as an autoencoder neural network model, wherein the number of learning operations completed by the autoencoder neural network model is proportional to the time length of the set time interval; An information reading mechanism is connected to the first grabbing component, the second grabbing component and the order conversion component respectively, and is used to use the automatic encoder neural network model to intelligently analyze two prediction values corresponding to the maximum value and the minimum value of the meter reading value of the current smart water meter in the set time interval of the day according to the time length of the set time interval, multiple related contents of the current smart water meter and the meter reading values corresponding to the set time intervals of each day in the history of the current smart water meter; A data transmission mechanism, connected to the information reading mechanism, is used to provide two prediction values corresponding to the maximum and minimum values of the meter reading values of the current smart water meter in the set time interval of the day to a remote big data server through a frequency division transmission channel; The second grabbing component is used to obtain the meter reading values corresponding to the set time intervals of each day in the history of the current smart water meter, and the position of each set time interval on the time axis of the day is the same, including: the number of days of each selected historical day is positively correlated with the caliber of the current smart water meter; Among them, the second capture component is used to obtain the meter reading values corresponding to the set time intervals of each historical day of the current smart water meter. The set time interval of each day has the same position on the time axis of that day and also includes: using a parameter mapping function to represent the parameter mapping relationship between the selected historical days and the caliber of the current smart water meter.
2. The meter reading information prediction system according to claim 1, characterized in that: The automatic encoder neural network model is used to intelligently analyze two prediction values corresponding to the maximum value and the minimum value of the meter reading value of the current smart water meter in the set time interval of the day according to the time length of the set time interval, multiple related contents of the current smart water meter, and each meter reading value corresponding to the set time interval of each day in the history of the current smart water meter, including: performing binary value conversion on the time length of the set time interval, multiple related contents of the current smart water meter, and each meter reading value corresponding to the set time interval of each day in the history of the current smart water meter, and then synchronously inputting them into the automatic encoder neural network model; Among them, the time length of the set time interval, multiple related contents of the current smart water meter and the meter reading values corresponding to the set time intervals of each historical day of the current smart water meter are converted into binary values and then synchronously input into the automatic encoder neural network model, including: using a numerical simulation mode to realize the simulation and testing of the automatic encoder neural network model.
3. The meter reading information prediction system according to claim 2, characterized in that: The system further comprises: A connection service device, for connecting the information reading mechanism, the first grabbing component, the second grabbing component and the order conversion component, respectively, to receive data from the information reading mechanism, the first grabbing component, the second grabbing component and the order conversion component, respectively, and to send parallel data from the connection processing component to the information reading mechanism, the first grabbing component, the second grabbing component and the order conversion component, respectively; Among them, the connection service device is used for connecting the information reading mechanism, the first grabbing component, the second grabbing component and the order conversion component respectively, so as to receive data from the information reading mechanism, the first grabbing component, the second grabbing component and the order conversion component respectively, and send parallel data from the connection processing component to the information reading mechanism, the first grabbing component, the second grabbing component and the order conversion component respectively, including: the connection processing component is a serial communication connection interface.
4. The meter reading information prediction system according to claim 3, characterized in that: The system further comprises: The instruction transceiver device is respectively connected to the information reading mechanism, the first grabbing component, the second grabbing component and the order conversion component, and is used to receive user control instructions for the information reading mechanism, the first grabbing component, the second grabbing component and the order conversion component.
5. The meter reading information prediction system according to any one of claims 2 to 4, characterized in that: The information reading mechanism, the first grabbing component, the second grabbing component and the sequence conversion component are connected to the same oscillation actuator to obtain the timing data provided by the oscillation actuator, and the information reading mechanism, the first grabbing component, the second grabbing component and the sequence conversion component are respectively connected to the same content storage chip, and the content storage chip is one of FLASH flash memory, SDRAM memory chip and DDR memory chip.
6. The meter reading information prediction system according to any one of claims 2 to 4, characterized in that: The information reading mechanism, the first grabbing component, the second grabbing component and the order conversion component are respectively connected to the same IIC control bus, and are used to receive various control commands sent by the IIC control bus, and the various control commands are used to respectively configure the various operating data of the information reading mechanism, the first grabbing component, the second grabbing component and the order conversion component.
7. The meter reading information prediction system according to claim 6, characterized in that: The information reading mechanism, the first grabbing component, the second grabbing component and the order conversion component are respectively connected to the IIC control bus and are used to receive various control instructions sent through the IIC control bus.
8. The meter reading information prediction system according to any one of claims 2 to 4, characterized in that: The information reading mechanism, the first grabbing component, the second grabbing component and the order conversion component are respectively connected to the same MCU controller, and are used to switch the information reading mechanism, the first grabbing component, the second grabbing component and the order conversion component between sleep mode and working mode under the control of the same MCU controller.
9. The meter reading information prediction system according to claim 8, characterized in that: The information reading mechanism, the first grabbing component, the second grabbing component and the order conversion component are respectively connected to the same MCU controller, and are used for switching the information reading mechanism, the first grabbing component, the second grabbing component and the order conversion component between the sleep mode and the working mode under the control of the same MCU controller, including: the same MCU controller is designed based on the ARM13 core; Among them, the information reading mechanism, the first grabbing component, the second grabbing component and the order conversion component are respectively connected to the same MCU controller, and are used for switching between sleep mode and working mode respectively under the control of the same MCU controller. It also includes: the information reading mechanism, the first grabbing component, the second grabbing component and the order conversion component use the same quartz oscillator device to provide the information reading mechanism, the first grabbing component, the second grabbing component and the order conversion component with the clock waveforms they need respectively.