Online calibration method for household water meter

Abstract

The invention provides an online calibration method for a household water meter, which comprises the following steps of: taking a closestool as a standard device for calibrating the household meter, collecting the maximum water consumption of the closestool as standard data for online calibration, and calibrating actually measured data of the maximum water consumption of the closestool by using the standard data. Wherein the process of identifying the maximum water consumption of the closestool comprises the following steps: firstly identifying the working mode of the water tank, then further identifying the water consumption mode of the closestool from the working mode of the water tank, and further extracting the maximum water consumption of the closestool. According to the method, the closestool used by each family is adopted as a calibration standard device of the household water meter, the characteristics of the closestool water consumption mode are recognized through long-term tracking analysis of the user water consumption mode, accurate extraction of the closestool water consumption is achieved on the basis, and a new development direction is provided for water meter calibration.

Description

Technical Field

[0001] The invention belongs to the technical field of data processing, and in particular relates to a calibration method for a water meter. Background Art

[0002] According to the standard, household water meters must be periodically disassembled and verified or calibrated on a standard calibration device. The existing household water meter verification or calibration method mainly includes removing it from the pipeline and taking it to the measurement standard device for verification or calibration. The existing method has the following disadvantages:

[0003] Huge workload: There are a huge number of meters in households. Disassembly and verification are a huge workload with low social benefits. If the meters are not disassembled and verified, it will take 6 years to know the metering status, which is not conducive to the management of production and sales. The mandatory replacement every six years causes a lot of waste.

[0004] Low detection efficiency: Traditional detection methods are time-consuming and require a large amount of water for detection, which leads to low detection efficiency and cannot meet the needs of modern society for efficient detection.

[0005] High testing cost: Since traditional verification methods require a large amount of water, the cost of testing is increased, which undoubtedly increases the economic burden for large-scale water meter verification work.

[0006] ·Backward technology: Some traditional verification methods still use manual data recognition or simple automated equipment, lacking modern data collection and processing technology, which affects the accuracy and comprehensiveness of data collection and the overall efficiency and accuracy of verification.

[0007] Technical content

[0008] In view of this, the present invention proposes an online calibration method for household water meters, which uses the toilet installed in every household as a "quasi-metering standard bucket" to perform "quasi-calibration" detection on the household meter.

[0009] The specific technical means are as follows:

[0010] An online calibration method for a household water meter uses a toilet as a standard device for calibrating the household meter, collects the maximum water consumption of the toilet as standard data for online calibration, and uses the standard data to calibrate the actual measured data of the maximum water consumption of the toilet.

[0011] A standard instrument is an instrument used to test a certain object.

[0012] Furthermore, the identification process of the toilet water usage pattern is as follows:

[0013] Step 1, identifying the working mode of the water tank. Specifically, the instantaneous flow time series data is converted into an image, the working mode of the water tank is identified by a deep learning method, and the identified instantaneous flow time series is marked. The working mode of the water tank is that the water consumption data presents a periodic rectangular wave that rises linearly from zero to a certain value within a period of time, and then drops linearly to zero after a period of time.

[0014] Step 2: extract water use data characteristics of a water use process in the water tank working mode, wherein the water use data characteristics include the rate of change of instantaneous water flow over time, water use time, water use duration, and water consumption data;

[0015] Step 3: further identify the toilet water consumption mode from the water tank working mode. Specifically, cluster analysis is performed on the water tank water consumption data characteristics. The toilet water consumption mode is the one that meets all the association rules of the toilet mode. The association rules of the toilet mode include: the water consumption data of the toilet does not exceed the maximum water consumption of the toilet standard; the water consumption time of the toilet is less than the time threshold t s Seconds, 50 ≤ t s ≤70; the interval between two rectangular waves presented by toilet water is greater than the interval threshold t inter Hours, t inter ≥3; The toilet water usage time includes 6-9 am, 11-13 pm, 16-19 pm, and 21-23 pm; The change rate of the toilet water instantaneous flow rate over time is less than the change threshold t c , t c ≤0.1.

[0016] Furthermore, the method for obtaining standard data for online calibration is that, at the initial stage of installation of the household meter, the household meter records the user's water consumption for a period of time, identifies the user's toilet water consumption pattern, and uses the maximum toilet water consumption obtained under the pattern as the standard data for online calibration.

[0017] Furthermore, the method for obtaining the measured data of the maximum water consumption of the toilet is that during the household meter monitoring period, the household meter records the user's water consumption, identifies the user's toilet water use pattern therefrom, and uses the maximum water consumption of the toilet obtained under the pattern as the measured data of the maximum water consumption of the toilet.

[0018] Furthermore, the household meter periodically sends the household water consumption to the big data platform, and the big data platform recognizes the toilet water consumption pattern and obtains the maximum water consumption data value of the toilet under this pattern.

[0019] Furthermore, the method for obtaining the standard data for online calibration is: obtaining it through actual testing or data reporting.

[0020] Furthermore, standard data for online calibration are obtained at the initial stage of installation of the household meter.

[0021] Furthermore, the water consumption of the target user is monitored over a period of time, and when there is no water leakage, online calibration is performed.

[0022] Beneficial Effects

[0023] 1. The water capacity of a long-term qualified toilet is certain, generally designed to be 6 liters, or 0.006m3. The present invention adopts the toilet used by every household as a calibration standard for household water meters, which has a basis for wide use.

[0024] 2. Since the user has various water-using appliances at home, it is necessary to identify the toilet water use mode through a water use mode recognition method, including the water use time, instantaneous water flow, duration, and cumulative water volume during use, excluding other water use situations at the same time, and calculating the toilet water volume separately. The present invention identifies the characteristics of the toilet water use mode through long-term tracking and analysis of the user's water use mode, and on this basis realizes the accurate extraction of the toilet water volume.

[0025] 3. The household meter used in the present invention has the measurement accuracy and resolution to complete the standard water volume, thereby ensuring the accuracy of the standard capacity. Specific technical indicators include the measurement sampling frequency (seconds or 0.5 seconds), and the measurement scale or resolution needs to reach 0.00001m3.

[0026] 4. With the support of the upper system, the present invention identifies the maximum water consumption of the toilet in a data-driven manner and performs online calibration of the online water meter, which significantly improves the efficiency and accuracy of the calibration.

[0027] 5. The present invention proposes a new development direction for water meter calibration. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] Figure 1 , flow chart of the method of the present invention. DETAILED DESCRIPTION

[0029] Assume that a household user has a household water meter (DN15 or DN20) installed and a toilet that works properly. The toilet is used as a standard for calibrating the household meter. After the float valve is set, the capacity of the qualified and watertight toilet container is determined, and it meets the requirements of being a standard. In the early stage of using a new water meter, the water meter is still within the mandatory verification period. Use it to monitor the long-term stability of the toilet capacity and the water capacity value. If this value is stable for a long time, it meets the requirements of a standard.

[0030] The online calibration method for domestic water meters consists of two phases:

[0031] In the initial stage of household meter installation, the household meter records the user's water consumption data for a period of time. The household meter sends the water consumption data to the big data platform, which identifies the toilet water usage pattern and uses the maximum toilet water consumption extracted under this pattern as the standard data for online calibration. The household meter has second or millisecond measurement capabilities.

[0032] During the household meter monitoring period, the household meter sends the user's water consumption to the big data platform at a fixed time every day. The purpose of sending it at a fixed time rather than in real time is to reduce the energy consumption of the household meter. The toilet water use pattern is identified on the big data platform, and the maximum toilet water consumption extracted under this pattern is used as the actual measured data of the maximum toilet water consumption.

[0033] For ordinary residents, there must be a period of no water use every day. If there is no zero flow period during 24 hours (or 12 or 1 hour, which can be set), it can be determined that the user has a water leak, and the meter can issue an alarm to remind maintenance personnel to repair or replace it. If there is no water leak, the household water meter is calibrated online, that is, the error is calculated between the actual water consumption measurement value of the household water meter and the standard data of the online calibration established at the beginning of installation. If the error exceeds the limit, it is regarded as the household water meter online calibration error exceeds the tolerance, and the household meter has potential risks.

[0034] The error range can be determined by setting a standard value (threshold) or using a dynamic standard value of toilet error big data or further using dynamic toilet water consumption standard data. If the error exceeds the limit of the error range, it is considered that the online calibration error of the household water meter exceeds the tolerance. Using dynamic toilet water consumption standard data to calculate the error and perform online calibration of the toilet can reduce the error of online calibration using big data.

[0035] As time goes by, the working state of the float valve that controls the toilet will also change, which will cause changes in the maximum water consumption of the toilet. This change may cause an alarm, which can be checked manually. Compared with traditional calibration methods, it still greatly reduces the calibration cost.

[0036] The identification of the toilet water use pattern includes the following process:

[0037] Step 1: Based on the convolutional neural network, the working mode of the water tank of each device measured by the water meter is identified. Specifically, the time series data of the instantaneous flow rate is converted into an image, and the deep learning method, especially the convolutional neural network, is used to effectively identify the working mode of the water tank and mark the water meter's water consumption time series.

[0038] The user's water usage pattern includes toilet water usage pattern, washing machine water usage pattern, dishwasher water usage pattern, water heater water usage pattern, and daily life water usage pattern. Daily life water usage pattern includes washing hands, washing vegetables and cooking, washing dishes, etc. Among them, the water consumption data of toilets, washing machines, dishwashers, and water heaters show a periodic rectangular wave that rises straight from zero to a certain value within a period of time, and then quickly drops to zero after a period of time. Therefore, it is necessary to first identify the water usage pattern of toilets, washing machines, dishwashers, and water heaters, that is, to identify the working mode of the water tank.

[0039] Taking advantage of the meter's second-level metering capability, when no water is used, the instantaneous flow is 0, and when water is used, the flow is greater than 0. The following is an expanded description of various water use modes:

[0040] For example: when the water heater switch is turned on, the flow rate must be greater than a certain flow rate before the fire can be ignited for heating (different seasons, different temperature adjustments, and estimated set flow rates are different, which may also be related to the capacity of the water heater). Usually, the shower time is about twenty minutes, and the flow rate may be adjusted in the middle. The curves of the instantaneous flow rate of water use and the water consumption per second have their own patterns. Specifically, when the water is turned on, there is an instantaneous ignition flow rate greater than a certain value, and then water is continuously supplied for showering for more than 1 minute. After that, the water temperature is adjusted, and the flow rate can vary in size. Next, the appropriate water temperature is reached and used continuously for more than 5 minutes, until the faucet is turned off to end the shower. The entire duration is generally more than 10 minutes, and there may be slight differences among different people in different regions and families.

[0041] Toilet water use mode: It is turned on by triggering the toilet drain switch. After the toilet is drained, the toilet float switch is turned on to draw water from the water supply pipe. The amount of water drawn is related to the amount of water adjusted by the toilet float, and the water drawing time is continuous and stable. After the water is full, the float switch is closed and the toilet stops drawing water immediately. The way the toilet uses water is a quasi-rectangular wave. The amount of water used by the toilet is a fixed value.

[0042] Although the toilet may drain water in small or large amounts or not completely, after recognizing this toilet mode, there is always a chance to obtain the toilet to drain all the water in the water tank container and obtain the maximum water consumption. The maximum water consumption is regarded as a fixed amount of flushing water for the toilet.

[0043] Different models of toilets (6L and 4L) have different water consumption values. The water consumption of the same model is related to the installation and adjustment method of the toilet float valve. It can be considered that the maximum water consumption value of each toilet is different. The maximum water extraction time of a toilet is basically within the range of 50 seconds to 70 seconds.

[0044] The water usage mode of the washing machine, because there is only one faucet, the washing machine has a fixed flow rate when working, but the water usage time is different each time. According to the different working modes set by the washing machine, each mode consists of different time periods to form the complete working process of the washing machine, and the water usage is intermittent. Assuming that the maximum instantaneous flow rate of the washing machine is equivalent to the maximum instantaneous flow rate of the water heater, the working time of the washing machine is at least 15 minutes and can be as long as several hours (with drying type). The maximum instantaneous flow rate of the water pulses in the middle is the same, but the time is different.

[0045] For example, the time for washing hands is generally within the range of a few seconds to 1 to 2 minutes, and it is usually not at the maximum instantaneous flow rate of the faucet. After people develop the habit of saving water, they will actively reduce the water flow rate to obtain a good water experience (when the water flow rate is large, water may splash everywhere, affecting the user experience), and generally, it can be washed clean once, and there will be no need to wash it twice.

[0046] For example: when washing vegetables and cooking (usually occurs within the time range of cooking), the water can be used for a relatively long time, and the water use is also continuous and intermittent, with long and short water use times. Sometimes the flow rate is large, such as rinsing vegetables, and sometimes the flow rate is small for rinsing vegetables. It usually lasts for 30 minutes. This situation occurs regularly, usually during meal times (cooking time for breakfast, lunch and dinner), and it rarely occurs that the maximum instantaneous flow rate lasts for a long time.

[0047] For example, dishwashing usually occurs in a period of time after the cooking time range, either manually or by machine. In most cases, the instantaneous flow rate of dishwashing will not reach the maximum, and there is usually continuous water, occasionally the water is turned off, and the dishwasher has to finish the dishwashing work continuously (usually this is the case, occasionally some people will soak the dishes before washing, this mode can be ignored). Usually a family of three will finish washing dishes in 3 to 5 minutes. Occasionally when there are many people, it will take longer, but usually it can be completed within 10 to 20 minutes.

[0048] The difference between washing machine water use and toilet water use is: to complete a complete washing process, there will be multiple water tank water use modes in the middle, and the amount of water used each time may be different, and the total completion time is 15 minutes or more.

[0049] Step 2: extract water use data characteristics of a water use process in the water tank working mode, wherein the water use data characteristics include the rate of change of instantaneous water flow over time, water use time, water use duration, and water consumption data;

[0050] Step 3: further identify the toilet water consumption mode from the water tank working mode. Specifically, cluster analysis is performed on the water tank water consumption data characteristics. The toilet water consumption mode is the one that meets all the association rules of the toilet mode. The association rules of the toilet mode include: the water consumption data of the toilet does not exceed the maximum water consumption of the toilet standard, which is 6L for the traditional toilet and 4L for the water-saving toilet; the water consumption time of the toilet is less than the time threshold t s Seconds, 50 ≤ t s ≤70; the interval between two rectangular waves presented by toilet water is greater than the interval threshold t inter Hours, t inter ≥3, the toilet water use presents a rectangular wave, which usually appears once. When it appears next time, there is a long interval in between. The number of rectangular waves that appear rarely appears in a relatively fixed time range. The toilet water use time includes 6-9 am, 11-13 pm, 16-19 pm, and 21-23 pm. The time points of toilet water use are usually in line with residents' living habits, such as in the morning, before going to bed, before meals, etc. The statistical probability of its occurrence is high, and the periodicity (daily / weekly) presents a stable time range; the rate of change of the instantaneous flow rate of toilet water over time is less than the change threshold t c , t c ≤0.1.

[0051] In addition, standard data for online calibration can also be obtained by filling in data information or by actually testing the toilet.

[0052] Note: The real-time data monitored by the instrument may be the water usage of the water heater, toilet, washing machine, hand washing, vegetable washing, cooking, dish washing, etc., or two or more actions together, which becomes mixed mode water usage. However, there is always a chance that the toilet works alone every day (especially in the middle of the night or in the morning).

[0053] Other conditions: When a household user installs two toilets and uses one total meter for measurement, the total meter can also be calibrated online according to the previous assumptions. The difference lies in the reprocessing of the two errors, which can be judged separately or as one error.

Claims

1. An online calibration method for a household water meter, characterized in that: The toilet is used as a standard device for calibrating household meters, the maximum water consumption of the toilet is collected as standard data for online calibration, and the standard data is used to calibrate the measured data of the maximum water consumption of the toilet.

2. The method for online calibration of a household water meter according to claim 1, characterized in that: The identification process of toilet water use pattern is as follows: Step 1, identifying the working mode of the water tank, wherein the working mode of the water tank is that the water consumption data presents a periodic rectangular wave-like shape, which rises linearly from zero to a certain value within a period of time, and then drops linearly to zero after a period of time; Step 2, extracting water use data characteristics of a water use process in a water tank working mode; Step 3: Further identify the toilet water usage mode from the water tank working mode.

3. The method for online calibration of a household water meter according to claim 1, characterized in that: Identifying the working mode of the water tank specifically includes converting the instantaneous flow time series data into an image, using a deep learning method to identify the working mode of the water tank, and marking the identified instantaneous flow time series.

4. The method for online calibration of a household water meter according to claim 1, characterized in that: The water use data characteristics include the rate of change of instantaneous water flow over time, water use time, water use duration, and water consumption data.

5. The method for online calibration of a household water meter according to claim 1, characterized in that: Identifying the toilet water usage pattern specifically includes clustering analysis of the water tank water usage data characteristics. The toilet water usage pattern is the one that meets all the association rules of the toilet pattern. The association rules of the toilet pattern include: the water usage data of the toilet does not exceed the maximum water usage of the toilet standard; the water usage time of the toilet is less than the time threshold t s Seconds, 50 ≤ t s ≤70; the interval between two rectangular waves presented by toilet water is greater than the interval threshold t inter Hours, t inter ≥3; The toilet water usage time includes 6-9 am, 11-13 pm, 16-19 pm, and 21-23 pm; The change rate of the toilet water instantaneous flow rate over time is less than the change threshold t c , t c ≤0.

1.

6. The method for online calibration of a household water meter according to any one of claims 1 to 5, characterized in that: The household meter periodically sends the user's water consumption to the big data platform, which identifies the toilet water consumption pattern and obtains the maximum water consumption data value of the toilet under this pattern.

7. The method for online calibration of a household water meter according to claim 6, characterized in that: The method for obtaining standard data for online calibration is that, at the initial stage of installation of the household meter, the household meter records the user's water consumption for a period of time, from which the user's toilet water consumption pattern is identified, and the maximum toilet water consumption obtained under this pattern is used as the standard data for online calibration.

8. The method for online calibration of a household water meter according to claim 6, characterized in that: The method for obtaining the measured data of the maximum water consumption of a toilet is that during the household meter monitoring period, the household meter records the user's water consumption, identifies the user's toilet water use pattern therefrom, and uses the maximum water consumption of the toilet obtained under the pattern as the measured data of the maximum water consumption of the toilet.

9. The method for online calibration of a household water meter according to claim 1, characterized in that: The method for obtaining standard data for online calibration is: obtaining it through actual testing or data reporting at the initial stage of household meter installation.

Images