Intelligent water meter data analysis method and system

Abstract

The invention relates to the field of intelligent water meter data analysis, in particular to an intelligent water meter data analysis method and system. The intelligent water meter data analysis system comprises a data acquisition module, a fault risk assessment module, an intelligent water meter classification module and a fault detection module. According to the method, the total water consumption, the water quality data, the water consumption behavior data and the historical fault frequency of the intelligent water meter are comprehensively analyzed, and the fault risk of the intelligent water meter of each user is evaluated in a customized mode; the personalized evaluation mechanism not only improves the accuracy of fault detection, but also can provide more accurate fault early warning for the water use habits and environmental conditions of different users, thereby effectively avoiding the problems of false alarm and missing alarm possibly caused by an excessively single detection method.

Description

Technical Field

[0001] The present invention relates to the field of intelligent water meter data analysis, and in particular to an intelligent water meter data analysis method and system. Background Art

[0002] In today's rapidly developing urbanization, smart water meters are key components of modern water supply systems. Their stability and reliability are crucial to ensuring residents' water safety and the efficient operation of water supply systems. However, smart water meters may fail during use due to a variety of reasons. If these failures are not discovered and handled in a timely manner, they will not only affect users' normal water use, but may also lead to waste of water resources and reduced efficiency of water supply systems. Traditional water meter fault detection methods mainly rely on regular manual inspections or simple automatic monitoring systems. These methods have problems such as low detection efficiency and lack of personalized evaluation.

[0003] In view of these problems, the present invention proposes a smart water meter data analysis method, which can perform personalized fault risk assessment according to the water usage characteristics of each user, thereby achieving more accurate and efficient fault detection and early warning; the core of the present invention is to evaluate the fault risk of each smart water meter by collecting and analyzing multi-dimensional data of smart water meters, and give corresponding evaluation values; according to the comparison between the evaluation values ​​and the preset risk thresholds, the smart water meters are divided into two categories of high fault risk and low fault risk, and differentiated fault detection strategies are implemented; thereby, personalized fault risk assessment can be performed according to the water usage characteristics of each user, thereby improving the pertinence and accuracy of fault detection; at the same time, detection resources are reasonably allocated, detection efficiency is improved, and unnecessary detection costs are reduced. Summary of the invention

[0004] The present invention evaluates the failure risk of the smart water meter for each user in a customized manner by comprehensively analyzing the total water consumption, water quality data, water use behavior data and historical failure times of the smart water meter. This personalized evaluation mechanism not only improves the accuracy of fault detection, but also can provide more accurate fault warnings based on the water use habits and environmental conditions of different users, thereby effectively avoiding the false alarm and missed alarm problems that may be caused by an overly single detection method.

[0005] A smart water meter data analysis method, comprising: At the beginning of any preset time period, for N smart water meters corresponding to N users in the management area, obtain the total water consumption data, water quality data, water use behavior data and historical fault times of any user's smart water meter in the previous preset time period; wherein the water quality data includes water hardness data and water turbidity data, and the water use behavior data includes the total number of water valve starts and stops and the daily water consumption data; For any smart water meter corresponding to a user, the impact degree is evaluated based on the total water consumption data, water quality data, water use behavior data and the number of historical faults, and the water consumption impact score, water quality impact score, water use behavior impact score and historical fault impact score are obtained; the fault risk assessment value is obtained based on the water consumption impact score, water quality impact score, water use behavior impact score and historical fault impact score; Based on the fault risk assessment value of each smart water meter, a smart water meter with a fault risk assessment value greater than or equal to a preset risk threshold is marked as a high fault risk smart water meter; a smart water meter with a fault risk assessment value lower than the risk threshold is marked as a low fault risk smart water meter; A primary fault detection strategy is implemented for all smart water meters with high fault risk, and a secondary fault detection strategy is implemented for all smart water meters with low fault risk.

[0006] Preferably, the impact degree is evaluated based on the total water consumption data, water quality data, water use behavior data and the number of historical faults, and the water consumption impact score, water quality impact score, water use behavior impact score and historical fault impact score are obtained. The specific operations are as follows: Obtain water consumption impact score. The specific operation is: obtain the total water consumption data of all smart water meters corresponding to users in the previous preset time period , i=1, 2, ..., N: Calculate the total water consumption data of all users The average ; For any user, if the user's total water consumption data Less than average , then the water consumption impact score of the user's corresponding smart water meter is 0; if the user's total water consumption data Greater than or equal to the average , then use the formula Calculate and obtain the water consumption impact score of the smart water meter corresponding to the user; To obtain the water quality impact score, the specific operations are as follows: set the monitoring time point, obtain the water hardness data and water turbidity data of the environment where each smart water meter is located at any monitoring time point; set the minimum water hardness standard value , Maximum water hardness standard value , minimum water turbidity standard value and the maximum water turbidity standard value ; For any smart water meter, based on all the water hardness data and water turbidity data obtained in the last preset time period, the average value of all water hardness data and water turbidity data is calculated to obtain the overall water hardness. and overall water turbidity ; If the overall water hardness Less than the minimum water hardness standard value , then the water hardness impact score of the smart water meter is 0; if the overall water hardness Greater than or equal to the minimum water hardness standard value , then use the formula Calculate the water hardness impact score of the smart water meter; if the overall water turbidity Less than the minimum water turbidity standard value , then the water turbidity impact score of the smart water meter is 0; if the overall water turbidity Greater than or equal to the minimum water turbidity standard value , then use the formula Calculate the water turbidity impact score of the smart water meter; use the formula Calculate and obtain the water quality impact score of the smart water meter; Get the water use behavior impact score. The specific operation is: Get the total number of water valve starts and stops for all users in the previous preset time period , and calculate the total number of starts and stops of water valves for all users The average ; For any user, if the total number of times the user's water valve is started and stopped Less than average , then the water valve start-stop impact score of the user's corresponding smart water meter is 0; if the total number of times the user's water valve is started and stopped is Greater than or equal to the average , then use the formula Calculate and obtain the water valve start and stop impact score of the smart water meter corresponding to the user; Set the maximum water consumption fluctuation index and minimum water use fluctuation index ; For any user, obtain the user's daily water consumption data within the last preset time period , j = 1, 2, ..., n; n represents the number of days in a preset time period; calculate the water consumption data of the user in the previous preset time period The average ; Using the formula Calculate and obtain the user's water consumption fluctuation index; if the user's water consumption fluctuation index is less than the minimum water consumption fluctuation index , then the user's corresponding smart water meter water consumption fluctuation impact score is 0; if the user's water consumption fluctuation index is greater than or equal to the minimum water consumption fluctuation index , then use the formula Calculate and obtain the water consumption fluctuation impact score of the smart water meter corresponding to the user; Using the formula Calculate and obtain the water use behavior impact score of the user's corresponding smart water meter; Get the historical fault impact score. The specific operation is: Get the historical fault count of all users using the smart water meter in the last c preset time periods , for any smart water meter, using the formula Calculate and obtain the historical fault impact score of the smart water meter.

[0007] Preferably, the fault risk assessment value is obtained according to the water consumption impact score, water quality impact score, water use behavior impact score and historical fault impact score. The specific operation is as follows: For any smart water meter, the water consumption impact score of the smart water meter is obtained. , Water Quality Impact Score , Water use behavior impact score and historical failure impact score , using the formula Calculate and obtain the fault risk assessment value of the smart water meter; , , and The water consumption impact score is , Water Quality Impact Score , Water use behavior impact score and historical failure impact score The weight of .

[0008] Preferably, a first-level fault detection strategy is implemented for all high-failure-risk smart water meters, and the specific operations are as follows: For any high-fault-risk smart water meter, the water flow, water inlet pressure, water outlet pressure and pressure difference are obtained at the first fault monitoring time point; the time interval between every two adjacent first fault monitoring time points is t; if the water flow is greater than or equal to the preset flow threshold, it means that the current first fault monitoring time point is in a water-using state, then it is determined whether the water inlet pressure, water outlet pressure and pressure difference are respectively within the normal range of the corresponding water-using state, if so, no operation is performed; if not, a fault warning is issued; if the water flow is less than the preset flow threshold, it means that the current first fault monitoring time point is in a non-water-using state, then it is determined whether the water inlet pressure, water outlet pressure and pressure difference are respectively within the normal range of the corresponding non-water-using state, if so, no operation is performed; if not, a fault warning is issued; During the current preset time period, the smart water meter self-test data is obtained at a frequency of time interval T, and it is determined whether there is an abnormality. If so, an early warning is issued; if not, no operation is performed.

[0009] Preferably, a secondary fault detection strategy is implemented for all low-failure-risk smart water meters, and the specific operations are as follows: For any low-fault-risk smart water meter, the water flow, water inlet pressure, water outlet pressure and pressure difference are obtained at the second fault monitoring time point; the time interval between every two adjacent second fault monitoring time points is 4t; if the water flow is greater than or equal to the preset flow threshold, it means that the current second fault monitoring time point is in a water-using state, then it is determined whether the water inlet pressure, water outlet pressure and pressure difference are respectively within the normal range of the corresponding water-using state, if so, no operation is performed; if not, a fault warning is issued; if the water flow is less than the preset flow threshold, it means that the current second fault monitoring time point is in a non-water-using state, then it is determined whether the water inlet pressure, water outlet pressure and pressure difference are respectively within the normal range of the corresponding non-water-using state, if so, no operation is performed; if not, a fault warning is issued; During the current preset time period, the smart water meter self-test data is obtained at a frequency of 4T, and it is determined whether there is an abnormality. If so, an early warning is issued; if not, no operation is performed.

[0010] Preferably, the preset risk threshold is obtained through a genetic algorithm.

[0011] Preferably, the preset risk threshold is obtained by a genetic algorithm, and the specific operation is as follows: Step 1: Set the population size, randomly generate several candidate risk thresholds in the range of 0 to 1, form the initial population with the randomly generated candidate risk thresholds, and set the maximum number of iterations; Step 2: For any candidate risk threshold, use the formula The fitness value K of the candidate risk threshold is calculated, where D represents the false negative rate generated when the candidate risk threshold is applied for risk classification, and the false negative rate is the proportion of those that are assessed as low risk but actually have faults; this formula indicates that the optimization goal of the candidate risk threshold is to minimize the false negative rate; the higher the fitness value obtained by calculation, the better the performance of the candidate risk threshold; Step 3: Perform selection, crossover and mutation operations on the population to generate a new generation of population; Step 4: Repeat step 3 to iteratively update the population until the maximum number of iterations is reached. The population with the best fitness in the last generation is the preset risk threshold.

[0012] Preferably, a smart water meter data analysis system comprises: The data acquisition module is used to obtain the total water consumption data, water quality data, water use behavior data and historical fault count of any user's smart water meter in the previous preset time period for N smart water meters corresponding to N users in the management area at the beginning of any preset time period; The fault risk assessment module includes a water consumption impact scoring unit, a water quality impact scoring unit, a water use behavior impact scoring unit and a historical fault impact scoring unit, which are used to evaluate the impact degree based on the acquired total water consumption data, water quality data, water use behavior data and the number of historical faults, and obtain the water consumption impact score, water quality impact score, water use behavior impact score and historical fault impact score; and obtain the fault risk assessment value based on the water consumption impact score, water quality impact score, water use behavior impact score and historical fault impact score; The smart water meter classification module is used to mark the smart water meters with a fault risk assessment value greater than or equal to a preset risk threshold as high fault risk smart water meters according to the fault risk assessment value of each smart water meter; and mark the smart water meters with a fault risk assessment value lower than the risk threshold as low fault risk smart water meters; The fault detection module is used to implement the primary fault detection strategy for all smart water meters with high fault risks and the secondary fault detection strategy for all smart water meters with low fault risks.

[0013] The present invention has the following advantages: 1. The present invention evaluates the failure risk of smart water meters for each user in a customized manner by comprehensively analyzing the total water consumption, water quality data, water use behavior data, and historical failure times of smart water meters. This personalized evaluation mechanism not only improves the accuracy of fault detection, but also provides more accurate fault warnings based on the water use habits and environmental conditions of different users, thereby effectively avoiding the false alarm and missed alarm problems that may be caused by overly single detection methods.

[0014] 2. The present invention conducts risk assessment on smart water meters. This method can intelligently distinguish between smart water meters with high fault risks and smart water meters with low fault risks, and execute different levels of fault detection strategies accordingly. This differentiated detection method not only improves the efficiency of fault detection, but also optimizes the use of detection resources, ensuring that under limited resource conditions, those water meters that are more likely to fail can be tested more frequently, thereby reducing maintenance costs while ensuring the stable operation of the water supply system. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 This is a schematic diagram of the structure of the smart water meter data analysis system used in an embodiment of the present invention. DETAILED DESCRIPTION

[0016] In order to enable persons skilled in the art to better understand the technical solutions in the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention.

[0017] Embodiment 1, a smart water meter data analysis method, comprising:

[0018] At the beginning of any preset time period, for N smart water meters corresponding to N users in the management area, the total water consumption data, water quality data, water use behavior data and historical fault times of any user's smart water meter in the previous preset time period are obtained; the preset time period refers to a fixed time period set by the system for data collection and analysis, which can be set to one month or any other suitable time length; among them, water quality data includes water hardness data and water turbidity data, water hardness data refers to the concentration of minerals (such as calcium and magnesium ions) contained in water, which is usually used to measure the hardness of water; water hardness has an impact on human health and the performance of household water equipment; water turbidity data refers to the concentration of suspended particles in water, reflecting the clarity of water; water with high turbidity may contain more pollutants; water use behavior data includes the total number of water valve starts and stops and daily water consumption data; by collecting the above data, the smart water meter data analysis system can conduct a comprehensive analysis of the water use situation of each user, thereby evaluating the failure risk of each smart water meter and taking corresponding detection strategies;

[0019] For any smart water meter corresponding to a user, the impact degree is evaluated based on the total water consumption data, water quality data, water use behavior data and the number of historical failures, and the water consumption impact score, water quality impact score, water use behavior impact score and historical failure impact score are obtained; the water consumption impact score can help identify users who may face a higher risk of failure due to extreme water use patterns (such as overuse or underuse); the water quality impact score helps identify users who may face a higher risk of failure due to water quality problems; the water use behavior impact score helps identify users who may face a higher risk of failure due to irregular or abnormal water use behavior; the historical failure impact score helps identify users who may face a higher risk of failure due to frequent failures in the past; based on the water consumption impact score, water quality impact score, water use behavior impact score and historical failure impact score, the failure risk assessment value is obtained;

[0020] The reason why total water consumption data, water quality data, water use behavior data and historical failure times are used for evaluation here is that excessive water consumption may cause the internal sensors of smart water meters to work at the limit state for a long time, thereby increasing the risk of sensor damage; minerals contained in high-hardness water may be deposited on the mechanical parts of the water meter, causing wear or jamming of the parts, affecting the normal operation of the water meter, and at the same time, suspended particles in the water may enter the electronic components of the water meter, causing short circuits or damage to the electronic components; suspended matter contained in high-turbidity water may block the channels inside the water meter, affecting the passage of water, and thus affecting the metering accuracy of the water meter; in terms of water use behavior, frequent water valve starts and stops may increase the wear of the mechanical parts of the water meter; on the other hand, water meters with a large number of historical failures directly indicate that the user's usage method or the water meter may have some inherent problems or weaknesses, which may increase the risk of future failures;

[0021] Based on the fault risk assessment value of each smart water meter, a smart water meter whose fault risk assessment value is greater than or equal to a preset risk threshold is marked as a high fault risk smart water meter; a smart water meter whose fault risk assessment value is lower than the risk threshold is marked as a low fault risk smart water meter; the preset risk threshold is a pre-set value used to compare the fault risk assessment values ​​of smart water meters to distinguish high-risk and low-risk smart water meters;

[0022] A first-level fault detection strategy is implemented for all smart water meters with high failure risks. Through more intensive monitoring, potential problems of high-risk water meters can be discovered and handled in a timely manner, reducing the possibility of failures. A second-level fault detection strategy is implemented for all smart water meters with low failure risks, which can reduce the frequency and intensity of detection, thereby rationally allocating resources and reducing maintenance costs.

[0023] The impact degree is evaluated based on the total water consumption data, water quality data, water use behavior data and the number of historical faults, and the water consumption impact score, water quality impact score, water use behavior impact score and historical fault impact score are obtained. The specific operations are as follows: Obtain water consumption impact score. The specific operation is: obtain the total water consumption data of all smart water meters corresponding to users in the previous preset time period , i=1, 2, ..., N: Calculate the total water consumption data of all users The average ; For any user, if the user's total water consumption data Less than average , then the water consumption impact score of the user's corresponding smart water meter is 0; if the user's total water consumption data Greater than or equal to the average , then use the formula Calculate and obtain the water consumption impact score of the smart water meter corresponding to the user; To obtain the water quality impact score, the specific operations are as follows: set the monitoring time point, obtain the water hardness data and water turbidity data of the environment where each smart water meter is located at any monitoring time point; set the minimum water hardness standard value , Maximum water hardness standard value , minimum water turbidity standard value and the maximum water turbidity standard value ; For any smart water meter, based on all the water hardness data and water turbidity data obtained in the last preset time period, the average value of all water hardness data and water turbidity data is calculated to obtain the overall water hardness. and overall water turbidity ; If the overall water hardness Less than the minimum water hardness standard value , then the water hardness impact score of the smart water meter is 0; if the overall water hardness Greater than or equal to the minimum water hardness standard value , then use the formula Calculate the water hardness impact score of the smart water meter; if the overall water turbidity Less than the minimum water turbidity standard value , then the water turbidity impact score of the smart water meter is 0; if the overall water turbidity Greater than or equal to the minimum water turbidity standard value , then use the formula Calculate the water turbidity impact score of the smart water meter; use the formula Calculate and obtain the water quality impact score of the smart water meter; Get the water use behavior impact score. The specific operation is: Get the total number of water valve starts and stops for all users in the previous preset time period , and calculate the total number of starts and stops of water valves for all users The average ; For any user, if the total number of times the user's water valve is started and stopped Less than average , then the water valve start-stop impact score of the user's corresponding smart water meter is 0; if the total number of times the user's water valve is started and stopped is Greater than or equal to the average , then use the formula Calculate and obtain the water valve start and stop impact score of the smart water meter corresponding to the user; Set the maximum water consumption fluctuation index and minimum water use fluctuation index ; For any user, obtain the user's daily water consumption data within the last preset time period , j = 1, 2, ..., n; n represents the number of days in a preset time period; calculate the water consumption data of the user in the previous preset time period The average ; Using the formula Calculate and obtain the user's water consumption fluctuation index; if the user's water consumption fluctuation index is less than the minimum water consumption fluctuation index , then the user's corresponding smart water meter water consumption fluctuation impact score is 0; if the user's water consumption fluctuation index is greater than or equal to the minimum water consumption fluctuation index , then use the formula Calculate and obtain the water consumption fluctuation impact score of the smart water meter corresponding to the user; Using the formula Calculate and obtain the water use behavior impact score of the user's corresponding smart water meter; Get the historical fault impact score. The specific operation is: Get the historical fault count of all users using the smart water meter in the last c preset time periods , for any smart water meter, using the formula Calculate and obtain the historical fault impact score of the smart water meter.

[0024] Obtain the fault risk assessment value based on the water consumption impact score, water quality impact score, water use behavior impact score, and historical fault impact score. The specific operations are as follows: For any smart water meter, the water consumption impact score of the smart water meter is obtained. , Water Quality Impact Score , Water use behavior impact score and historical failure impact score , using the formula Calculate and obtain the fault risk assessment value of the smart water meter; , , and The water consumption impact score is , Water Quality Impact Score , Water use behavior impact score and historical failure impact score The weight here can be set according to the actual situation. For example, although water consumption will affect the smart water meter, considering the performance of smart water meters in the existing technology, the water consumption impact score can be Set a smaller weight; water quality problems (such as high hardness or high turbidity) may damage the mechanical parts and sensors of the water meter, affecting its accuracy and life, and should be assigned a moderate weight; changes in water use behavior (such as frequent switching of water valves, large fluctuations in water use) may indicate changes in user water use habits or potential problems with the water meter; a medium weight can be assigned because water use behavior, although important, may not directly affect the performance of the water meter as much as water quality; historical fault records provide direct evidence of the past performance of the water meter, and frequent failures may indicate user usage problems or design defects, and should be assigned a higher weight; The first-level fault detection strategy is implemented for all high-failure-risk smart water meters. The specific operations are as follows: For any high-fault-risk smart water meter, the water flow, water inlet pressure, water outlet pressure and pressure difference are obtained at the first fault monitoring time point; the time interval between every two adjacent first fault monitoring time points is t; if the water flow is greater than or equal to the preset flow threshold, it means that the current first fault monitoring time point is in a water-using state, then it is determined whether the water inlet pressure, water outlet pressure and pressure difference are respectively within the normal range of the corresponding water-using state, if so, no operation is performed; if not, a fault warning is issued; if the water flow is less than the preset flow threshold, it means that the current first fault monitoring time point is in a non-water-using state, then it is determined whether the water inlet pressure, water outlet pressure and pressure difference are respectively within the normal range of the corresponding non-water-using state, if so, no operation is performed; if not, a fault warning is issued; During the current preset time period, the smart water meter self-test data is obtained at a frequency of time interval T, and it is determined whether there is an abnormality. If so, an early warning is issued; if not, no operation is performed.

[0025] The secondary fault detection strategy is implemented for all low-failure-risk smart water meters. The specific operations are as follows: For any low-fault-risk smart water meter, the water flow, water inlet pressure, water outlet pressure and pressure difference are obtained at the second fault monitoring time point; the time interval between every two adjacent second fault monitoring time points is 4t; if the water flow is greater than or equal to the preset flow threshold, it means that the current second fault monitoring time point is in a water-using state, then it is determined whether the water inlet pressure, water outlet pressure and pressure difference are respectively within the normal range of the corresponding water-using state, if so, no operation is performed; if not, a fault warning is issued; if the water flow is less than the preset flow threshold, it means that the current second fault monitoring time point is in a non-water-using state, then it is determined whether the water inlet pressure, water outlet pressure and pressure difference are respectively within the normal range of the corresponding non-water-using state, if so, no operation is performed; if not, a fault warning is issued; During the current preset time period, the smart water meter self-test data is obtained at a frequency of 4T, and it is determined whether there is an abnormality. If so, an early warning is issued; if not, no operation is performed.

[0026] The preset risk threshold is obtained through genetic algorithm.

[0027] The preset risk threshold is obtained through genetic algorithm. The specific operations are as follows: Step 1: Set the population size, randomly generate several candidate risk thresholds in the range of 0 to 1, form the initial population with the randomly generated candidate risk thresholds, and set the maximum number of iterations; Step 2: For any candidate risk threshold, use the formula The fitness value K of the candidate risk threshold is calculated, where D represents the false negative rate generated when the candidate risk threshold is applied for risk classification, and the false negative rate is the proportion of those that are assessed as low risk but actually have faults; this formula indicates that the optimization goal of the candidate risk threshold is to minimize the false negative rate; the higher the fitness value obtained by calculation, the better the performance of the candidate risk threshold; Step 3: Perform selection, crossover and mutation operations on the population to generate a new generation of population; Step 4: Repeat step 3 to iteratively update the population until the maximum number of iterations is reached. The population with the best fitness in the last generation is the preset risk threshold.

[0028] Embodiment 2, a smart water meter data analysis system, such as Figure 1 As shown, including: The data acquisition module is used to obtain the total water consumption data, water quality data, water use behavior data and historical fault count of any user's smart water meter in the previous preset time period for N smart water meters corresponding to N users in the management area at the beginning of any preset time period; The fault risk assessment module includes a water consumption impact scoring unit, a water quality impact scoring unit, a water use behavior impact scoring unit and a historical fault impact scoring unit, which are used to evaluate the impact degree based on the acquired total water consumption data, water quality data, water use behavior data and the number of historical faults, and obtain the water consumption impact score, water quality impact score, water use behavior impact score and historical fault impact score; and obtain the fault risk assessment value based on the water consumption impact score, water quality impact score, water use behavior impact score and historical fault impact score; The smart water meter classification module is used to mark the smart water meters with a fault risk assessment value greater than or equal to a preset risk threshold as high fault risk smart water meters according to the fault risk assessment value of each smart water meter; and mark the smart water meters with a fault risk assessment value lower than the risk threshold as low fault risk smart water meters; The fault detection module is used to implement the primary fault detection strategy for all smart water meters with high fault risks and the secondary fault detection strategy for all smart water meters with low fault risks.

[0029] It should be understood that those skilled in the art can make improvements or changes based on the above description, and all such improvements and changes should fall within the scope of protection of the appended claims of the present invention. Parts not described in detail in this specification belong to the prior art known to those skilled in the art.

Claims

1. A smart water meter data analysis method, characterized in that: include: At the beginning of any preset time period, for N smart water meters corresponding to N users in the management area, obtain the total water consumption data, water quality data, water use behavior data and historical fault times of any user's smart water meter in the previous preset time period; wherein the water quality data includes water hardness data and water turbidity data, and the water use behavior data includes the total number of water valve starts and stops and the daily water consumption data; For any smart water meter corresponding to a user, the impact degree is evaluated based on the total water consumption data, water quality data, water use behavior data and the number of historical faults, and the water consumption impact score, water quality impact score, water use behavior impact score and historical fault impact score are obtained; the fault risk assessment value is obtained based on the water consumption impact score, water quality impact score, water use behavior impact score and historical fault impact score; Based on the fault risk assessment value of each smart water meter, a smart water meter with a fault risk assessment value greater than or equal to a preset risk threshold is marked as a high fault risk smart water meter; a smart water meter with a fault risk assessment value lower than the risk threshold is marked as a low fault risk smart water meter; A primary fault detection strategy is implemented for all smart water meters with high fault risk, and a secondary fault detection strategy is implemented for all smart water meters with low fault risk.

2. The intelligent water meter data analysis method according to claim 1, characterized in that: The impact degree is evaluated based on the total water consumption data, water quality data, water use behavior data and the number of historical faults, and the water consumption impact score, water quality impact score, water use behavior impact score and historical fault impact score are obtained. The specific operations are as follows: Obtain water consumption impact score. The specific operation is: obtain the total water consumption data of all smart water meters corresponding to users in the previous preset time period , i=1, 2, ..., N: Calculate the total water consumption data of all users The average ; For any user, if the user's total water consumption data Less than average , then the water consumption impact score of the user's corresponding smart water meter is 0; if the user's total water consumption data Greater than or equal to the average , then use the formula Calculate and obtain the water consumption impact score of the smart water meter corresponding to the user; To obtain the water quality impact score, the specific operations are as follows: set the monitoring time point, obtain the water hardness data and water turbidity data of the environment where each smart water meter is located at any monitoring time point; set the minimum water hardness standard value , Maximum water hardness standard value , minimum water turbidity standard value and the maximum water turbidity standard value ; For any smart water meter, based on all the water hardness data and water turbidity data obtained in the last preset time period, the average values ​​of all water hardness data and water turbidity data are calculated to obtain the overall water hardness. and overall water turbidity ; If the overall water hardness Less than the minimum water hardness standard value , then the water hardness impact score of the smart water meter is 0; if the overall water hardness Greater than or equal to the minimum water hardness standard value , then use the formula Calculate the water hardness impact score of the smart water meter; if the overall water turbidity Less than the minimum water turbidity standard value , then the water turbidity impact score of the smart water meter is 0; if the overall water turbidity Greater than or equal to the minimum water turbidity standard value , then use the formula Calculate the water turbidity impact score of the smart water meter; use the formula Calculate and obtain the water quality impact score of the smart water meter; Get the water use behavior impact score. The specific operation is: Get the total number of water valve starts and stops for all users in the previous preset time period , and calculate the total number of times the water valves of all users are started and stopped The average ; For any user, if the total number of times the user's water valve is started and stopped Less than average , then the water valve start-stop impact score of the user's corresponding smart water meter is 0; if the total number of times the user's water valve is started and stopped is Greater than or equal to the average , then use the formula Calculate and obtain the water valve start and stop impact score of the smart water meter corresponding to the user; Set the maximum water consumption fluctuation index and minimum water use fluctuation index ; For any user, obtain the user's daily water consumption data within the last preset time period , j = 1, 2, ..., n; n represents the number of days in a preset time period; calculate the water consumption data of the user in the previous preset time period The average ; Using the formula Calculate the user's water consumption fluctuation index; if the user's water consumption fluctuation index is less than the minimum water consumption fluctuation index , then the user's corresponding smart water meter water consumption fluctuation impact score is 0; if the user's water consumption fluctuation index is greater than or equal to the minimum water consumption fluctuation index , then use the formula Calculate and obtain the water consumption fluctuation impact score of the smart water meter corresponding to the user; Using the formula Calculate and obtain the water use behavior impact score of the user's corresponding smart water meter; Get the historical fault impact score. The specific operation is: Get the historical fault count of all users using the smart water meter in the last c preset time periods , for any smart water meter, using the formula Calculate and obtain the historical fault impact score of the smart water meter.

3. A smart water meter data analysis method according to claim 2, characterized in that: Obtain the fault risk assessment value based on the water consumption impact score, water quality impact score, water use behavior impact score, and historical fault impact score. The specific operations are as follows: For any smart water meter, the water consumption impact score of the smart water meter is obtained. , Water Quality Impact Score , Water use behavior impact score and historical failure impact score , using the formula Calculate and obtain the fault risk assessment value of the smart water meter; , , and The water consumption impact score is , Water Quality Impact Score , Water use behavior impact score and historical failure impact score The weight of .

4. The intelligent water meter data analysis method according to claim 3, characterized in that: The first-level fault detection strategy is implemented for all high-failure-risk smart water meters. The specific operations are as follows: For any high-fault-risk smart water meter, the water flow, water inlet pressure, water outlet pressure and pressure difference are obtained at the first fault monitoring time point; the time interval between every two adjacent first fault monitoring time points is t; If the water flow rate is greater than or equal to the preset flow rate threshold, it means that the current first fault monitoring time point is in the water use state, then it is determined whether the water inlet pressure, water outlet pressure and pressure difference are respectively within the corresponding normal range of the water use state. If so, no operation is performed; if not, a fault warning is issued; If the water flow rate is less than the preset flow rate threshold, it means that the current first fault monitoring time point is in a non-water state, then it is determined whether the water inlet pressure, water outlet pressure and pressure difference are respectively within the corresponding normal range of the non-water state. If so, no operation is performed; if not, a fault warning is issued; In the current preset time period, the self-test data of the smart water meter is obtained at a frequency of time interval T, and it is determined whether there is an abnormality. If so, an early warning is issued; If not, no action is taken.

5. The method for analyzing data of a smart water meter according to claim 4, characterized in that: The secondary fault detection strategy is implemented for all low-failure-risk smart water meters. The specific operations are as follows: For any low-fault-risk smart water meter, the water flow, water inlet pressure, water outlet pressure and pressure difference are obtained at the second fault monitoring time point; the time interval between every two adjacent second fault monitoring time points is 4t; if the water flow is greater than or equal to the preset flow threshold, it means that the current second fault monitoring time point is in the water use state, then it is determined whether the water inlet pressure, water outlet pressure and pressure difference are respectively within the corresponding normal range of the water use state. If so, no operation is performed; if not, a fault warning is issued; If the water flow rate is less than the preset flow rate threshold, it means that the current second fault monitoring time point is in a non-water state, then it is determined whether the water inlet pressure, water outlet pressure and pressure difference are respectively within the corresponding normal range of the non-water state. If so, no operation is performed; if not, a fault warning is issued; During the current preset time period, the smart water meter self-test data is obtained at a frequency of 4T, and it is determined whether there is an abnormality. If so, an early warning is issued; if not, no operation is performed.

6. A smart water meter data analysis method according to claim 5, characterized in that: The preset risk threshold is obtained through genetic algorithm.

7. A smart water meter data analysis method according to claim 6, characterized in that: The preset risk threshold is obtained through genetic algorithm. The specific operations are as follows: Step 1: Set the population size, randomly generate several candidate risk thresholds in the range of 0 to 1, form the initial population with the randomly generated candidate risk thresholds, and set the maximum number of iterations; Step 2: For any candidate risk threshold, use the formula The fitness value K of the candidate risk threshold is calculated, where D represents the false negative rate generated when the candidate risk threshold is applied for risk classification, and the false negative rate is the proportion of those that are assessed as low risk but actually have faults; this formula indicates that the optimization goal of the candidate risk threshold is to minimize the false negative rate; the higher the fitness value obtained by calculation, the better the performance of the candidate risk threshold; Step 3: Perform selection, crossover and mutation operations on the population to generate a new generation of population; Step 4: Repeat step 3 to iteratively update the population until the maximum number of iterations is reached. The population with the best fitness in the last generation is the preset risk threshold.

8. A smart water meter data analysis system, characterized in that: The system is applied to the smart water meter data analysis method according to any one of claims 1 to 7, comprising: The data acquisition module is used to obtain the total water consumption data, water quality data, water use behavior data and historical fault count of any user's smart water meter in the previous preset time period for N smart water meters corresponding to N users in the management area at the beginning of any preset time period; The fault risk assessment module includes a water consumption impact scoring unit, a water quality impact scoring unit, a water use behavior impact scoring unit and a historical fault impact scoring unit, which are used to evaluate the impact degree based on the acquired total water consumption data, water quality data, water use behavior data and the number of historical faults, and obtain the water consumption impact score, water quality impact score, water use behavior impact score and historical fault impact score; and obtain the fault risk assessment value based on the water consumption impact score, water quality impact score, water use behavior impact score and historical fault impact score; The smart water meter classification module is used to mark the smart water meters with a fault risk assessment value greater than or equal to a preset risk threshold as high fault risk smart water meters according to the fault risk assessment value of each smart water meter; and mark the smart water meters with a fault risk assessment value lower than the risk threshold as low fault risk smart water meters; The fault detection module is used to implement the primary fault detection strategy for all smart water meters with high fault risks and the secondary fault detection strategy for all smart water meters with low fault risks.

Images