A remote water meter low-power consumption control method and system based on hierarchical wake-up

Abstract

The present application relates to the technical field of water meter control, and specifically discloses a low-power consumption control method and system for remote water meter based on hierarchical awakening, wherein the present application divides the power supply area of the remote water meter and configures an independent awakening circuit, combines multi-source state sensing to construct an operation state vector, performs stage division and multi-rule fusion based on the operation state vector to determine the business scenario, executes hierarchical awakening and task coordination control according to the business scenario type, constructs a water consumption behavior model through historical operation data to predict the next stage business scenario, corrects the prediction result through an interference identification mechanism, realizes model adaptive optimization based on the comparison between the prediction result and the actual result, and thus forms a closed-loop control mechanism; the present application can reduce power consumption while realizing fine management and service life of the remote water meter.

Description

Technical Field

[0001] This invention relates to the field of water meter control technology, and more specifically, to a low-power control method and system for remote water meters based on hierarchical wake-up. Background Technology

[0002] Existing remote water meter systems typically employ continuous power supply or periodic wake-up modes to achieve remote acquisition and transmission of water usage data. Although some systems implement basic sleep control by introducing low-power microcontrollers, their overall power management remains relatively crude. Most functional modules remain in standby power-consuming mode when not in operation, resulting in high overall power consumption, short battery life, and difficulty in meeting the requirements for long-term maintenance-free operation. Furthermore, they are prone to invalid or excessive wake-ups, further increasing energy consumption. Even when only a partial function needs to be executed, all modules may still be activated, causing unnecessary energy waste.

[0003] In actual operation, the lack of exploration and utilization of water usage patterns makes it impossible to predict future operating conditions. As a result, the system can only passively respond to real-time state changes. Sudden water usage or abnormal events can easily interfere with state recognition, leading to misjudgments in business scenarios and affecting the accuracy of wake-up strategies and system stability.

[0004] Therefore, it is necessary to provide a low-power control method and system for remote water meters based on hierarchical wake-up to solve the above-mentioned technical problems. In order to solve the above problems, a technical solution is provided. Summary of the Invention

[0005] To overcome the aforementioned deficiencies of the prior art, this invention provides a low-power control method and system for remote water meters based on hierarchical wake-up, which addresses the problems of existing remote water meters having crude power management, a single wake-up mechanism, and a lack of behavior prediction and anti-interference capabilities, resulting in high power consumption and insufficient intelligence.

[0006] To achieve the above objectives, the present invention provides the following technical solution: A low-power control method for remote water meters based on hierarchical wake-up includes the following steps: By dividing the power supply of the remote water meter into several power supply areas and equipping each power supply area with a separate wake-up circuit, the operating status data of each power supply area is acquired in real time based on the multi-source sensing unit, and the data is preprocessed to form an operating status vector; the operating status data includes water flow status data, time status data and power consumption status data. Based on the operational status vector within the monitoring period, a phase division is performed, a multi-rule fusion phase discrimination mechanism is constructed, the business scenario type corresponding to each phase is obtained, the wake-up power supply area is controlled according to the business scenario type, and the tasks of the power supply area are dynamically adjusted. By constructing a phased water use behavior model based on historical operational status data, we can analyze and predict the type of business scenario for users in the next phase, and eliminate level identification interference caused by sudden interference scenarios to obtain the actual business scenario type. The accuracy of scenario prediction is confirmed by comparing the actual business scenario type with the predicted business scenario type in the next stage, and the water use behavior model in the stage is optimized in reverse based on the scenario prediction accuracy.

[0007] As a further embodiment of the present invention, the power supply area includes: a core area, a sensor area, a communication area, and a storage area.

[0008] As a further aspect of the present invention, the operational status vector within the monitoring period is used to divide the process into stages, and a multi-rule fusion stage discrimination mechanism is constructed to obtain the business scenario type corresponding to each stage. The specific steps are as follows: A time-series state sequence is constructed based on the operating state vector within the monitoring period. The time-series state sequence includes the water flow state sequence and the power consumption state sequence. By calculating the rate of change of adjacent running state vectors in the time-series state sequence, the time-series state sequence is dynamically divided into stages, and the state change characteristics of each stage are extracted. A multi-rule fusion stage discrimination mechanism is constructed based on state change characteristics to identify business scenarios and obtain the corresponding business scenario type for each stage.

[0009] As a further aspect of the present invention, the time-series state sequence is dynamically divided into stages by calculating the rate of change of adjacent running state vectors in the time-series state sequence, and the state change characteristics of each stage are extracted. The specific steps are as follows: Extract the water flow state sequence and the power consumption state sequence respectively, and calculate the rate of change of the corresponding state sequence, specifically including the rate of change of water flow state and the rate of change of power consumption state. After normalizing the water flow state change rate and power consumption state change rate, a state partitioning model is built to calculate the state partitioning coefficient. Based on the state partitioning coefficient, a dynamic segmentation rule is built to output the state stage set. Calculate the state change characteristics for each state stage. The state change characteristics include state statistical characteristics, state trend characteristics, and stability characteristics.

[0010] As a further aspect of the present invention, the state statistical features include the mean of the water flow state, the variance of the water flow state, the mean of the power consumption state, and the variance of the power consumption state; the state trend features include the water flow state trend features and the power consumption state trend features.

[0011] As a further aspect of the present invention, a dynamic segmentation rule output state stage set is constructed based on the state partitioning coefficient, as detailed below: The state division coefficient is compared with the preset segmentation threshold range. If the state division coefficient is greater than the upper limit of the preset segmentation threshold range, the remote water meter battery is in the sudden change stage; if the state division coefficient is within the preset segmentation threshold range, the remote water meter battery is in the transition buffer stage; if the state division coefficient is less than the upper limit of the preset segmentation threshold range, the remote water meter battery is in the stable stage.

[0012] As a further aspect of the present invention, the multi-rule fusion stage discrimination mechanism includes standby scenario discrimination, stable metering scenario discrimination, dynamic water usage scenario discrimination, and abnormal scenario discrimination.

[0013] As a further aspect of the present invention, the wake-up power supply area is controlled according to the business scenario type, and the tasks of the power supply area are dynamically adjusted. Specifically, the battery wake-up level is determined based on the business scenario type, the wake-up power supply area is determined based on the battery wake-up level, and the corresponding tasks are controlled according to the wake-up power supply area. Among them, standby scenario corresponds to Level 1 wake-up, stable metering scenario corresponds to Level 2 wake-up, dynamic water usage scenario corresponds to Level 3 wake-up, and abnormal scenario corresponds to Level 4 wake-up.

[0014] As a further embodiment of the present invention, the power supply area corresponding to the first-level wake-up is the core area; the power supply area corresponding to the second-level wake-up is the core area and the sensor area; the power supply area corresponding to the third-level wake-up is the core area, the sensor area and the communication area; and the power supply area corresponding to the fourth-level wake-up is the core area, the sensor area, the communication area and the storage area.

[0015] A low-power control system for remote water meters based on hierarchical wake-up includes a power supply area division module, an operation status data acquisition module, a business scenario determination module, a stage scenario prediction module, and an accuracy analysis module. The power supply area division module is used to divide the power supply of the remote water meter into several power supply areas, and equips each power supply area with a separate wake-up circuit. The operation status data acquisition module is used to acquire the operation status data of each power supply area in real time based on the multi-source sensing unit, and form an operation status vector after preprocessing. The business scenario determination module is used to divide the operation status vector within the monitoring period into stages, build a multi-rule fusion stage discrimination mechanism, obtain the business scenario type corresponding to each stage, control the wake-up power supply area according to the business scenario type, and dynamically adjust the tasks of the power supply area to be executed. The phase scenario prediction module is used to build a phase water use behavior model through historical operation status data, analyze and predict the type of business scenario for users in the next phase, and eliminate level identification interference caused by sudden interference scenarios to obtain the actual business scenario type. The accuracy analysis module is used to confirm the scenario prediction accuracy by comparing the actual business scenario type with the predicted business scenario type in the next stage, and to optimize the water use behavior model in the next stage based on the scenario prediction accuracy.

[0016] The technical effects and advantages of this invention, a low-power control method and system for remote water meters based on hierarchical wake-up, are as follows: This invention enables the transformation of remote water meters from traditional fixed wake-up to on-demand hierarchical wake-up, achieving module-level refined management at the power supply level, effectively reducing ineffective power supply and energy waste, thereby significantly reducing overall system power consumption and extending battery life; through stage division and multi-rule fusion discrimination mechanism, the accuracy and stability of business scenario identification are improved, avoiding misjudgment problems caused by single feature judgment; combined with water use behavior model and scenario prediction mechanism, the system has forward-looking decision-making capability, transforming from passive response to active control; through interference identification and elimination mechanism, the robustness of the system under sudden abnormal conditions is enhanced, and through predictive feedback, the model achieves adaptive optimization, maintaining superior performance over a long period of time, significantly improving the intelligence level and operational reliability of remote water meters while reducing power consumption. Attached Figure Description

[0017] Figure 1 A flowchart illustrating a low-power control method for remote water meters based on hierarchical wake-up provided in an embodiment of the present invention; Figure 2 This is a system block diagram of a low-power control system for a remote water meter based on hierarchical wake-up, provided as an embodiment of the present invention. Detailed Implementation

[0018] The technical solutions of this invention will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described technical solutions are only a part of this invention, and not all of it. All other technical solutions obtained by those skilled in the art based on the technical solutions of this invention without inventive effort are within the scope of protection of this invention.

[0019] Example 1: As Figure 1 The diagram shown is a flowchart illustrating a low-power control method for remote water meters based on hierarchical wake-up provided in an embodiment of the present invention. Figure 1The execution entity of the method shown can be a software and / or hardware device. The execution entity of this application can include, but is not limited to, at least one of the following: user equipment, network equipment, etc. User equipment can include, but is not limited to, computers, smartphones, personal digital assistants (PDAs), and the aforementioned electronic devices. Network equipment can include, but is not limited to, a single network server, a server group consisting of multiple network servers, or a cloud based on cloud computing consisting of a large number of computers or network servers. Cloud computing is a type of distributed computing, consisting of a super virtual computer composed of a group of loosely coupled computers. This embodiment does not limit this. Steps S1 to S4 are detailed as follows: Step S1: Divide the power supply of the remote water meter into several power supply areas, equip each power supply area with a separate wake-up circuit, and obtain the operating status data of each power supply area in real time based on the multi-source sensing unit. After preprocessing, an operating status vector is formed. The operating status data includes water flow status data, time status data, and power consumption status data. Step S2: Divide the operation status vector within the monitoring period into stages, construct a multi-rule fusion stage discrimination mechanism, obtain the business scenario type corresponding to each stage, control the wake-up power supply area according to the business scenario type, and dynamically adjust the tasks of the power supply area. Step S3: Construct a phased water use behavior model using historical operational status data, analyze and predict the user's next phase of business scenario type, and eliminate level identification interference caused by sudden interference scenarios to obtain the actual business scenario type. Step S4: Confirm the scenario prediction accuracy by comparing the actual business scenario type with the predicted business scenario type in the next stage, and optimize the water use behavior model in reverse based on the scenario prediction accuracy.

[0020] Preferably, the power supply of the remote water meter is divided into several power supply areas, and a separate wake-up circuit is provided for each power supply area, specifically: The power supply of the remote water meter is divided into several independent power supply zones, each equipped with a separate power switch circuit. The power supply zones include: a core zone (microcontroller core), a sensor zone (water flow sensor and related signal conditioning circuits), a communication zone (wireless communication module and antenna matching circuits), and a storage zone (data storage and related interface circuits). Each power supply zone's power supply status is controlled by a MOSFET switch or a low-power power management IC, achieving precise module-level power management. The wake-up circuit is designed with low power consumption to support several wake-up sources that trigger the power supply area to resume from deep sleep. These wake-up sources include: water flow sensor interrupts, real-time clock alarms, and external communication requests. The wake-up circuit employs low-power comparators and logic gates, achieving a static power consumption as low as 0.1μA, ensuring responsiveness to external events even when the power supply area is in deep sleep.

[0021] In one embodiment of the present invention, taking an integrated remote water meter as an example, its internal power supply system is modularly divided, and the power supply of the whole machine is divided into four independent power supply units: a core area, a sensor area, a communication area, and a storage area. The core area includes a microcontroller and its minimum system circuit, the sensor area includes a water flow sensor and a signal conditioning circuit, the communication area includes a wireless communication module and an antenna matching circuit, and the storage area includes a data memory and an interface circuit. Each power supply area is independently powered by a MOSFET switch or a low-power power management IC, so that each functional module can be powered on or off as needed, thereby realizing module-level fine power management.

[0022] Based on this, a low-power wake-up circuit is designed for each power supply area. This wake-up circuit is normally in an ultra-low-power monitoring state, with its static power consumption controllable to the order of 0.1μA. A multi-source triggering mechanism is implemented through a combination of low-power comparators and logic gate circuits. When the water flow sensor detects a change in flow rate and generates an interrupt signal, the sensor area and core area are powered on to execute data acquisition tasks. When the real-time clock reaches a preset time point, the core area and storage area are powered on to execute timing metering and data recording tasks. When an external communication request arrives, such as a magnetic switch trigger or a near-field interaction signal, the communication area and related modules are powered on to execute data transmission tasks. This invention enables remote water meters to activate only the necessary functional modules in different business scenarios, ensuring real-time response to external events while significantly reducing overall standby power consumption and ineffective energy consumption.

[0023] Preferably, the monitoring period is divided into stages based on the operational status vector, and a multi-rule fusion stage discrimination mechanism is constructed to obtain the business scenario type corresponding to each stage. The specific steps are as follows: A time-series state sequence is constructed based on the operating state vector within the monitoring period. This time-series state sequence includes a water flow state sequence and a power consumption state sequence; the water flow state sequence is... ,in, Here is the water flow state data at time t, where T is the length of the monitoring period. The power consumption state sequence is as follows: ,in, The power consumption status data at time t; By calculating the rate of change of adjacent running state vectors in the time-series state sequence, the time-series state sequence is dynamically divided into stages, and the state change characteristics of each stage are extracted. A multi-rule fusion stage discrimination mechanism is constructed based on state change characteristics to identify business scenarios and obtain the corresponding business scenario type for each stage.

[0024] In one embodiment of the present invention, water flow and power consumption data are continuously collected within a set monitoring period, such as 24 hours, and water flow state sequences are constructed respectively. With power state sequence Each sampling time t corresponds to an operating state vector. Subsequently, the state data from adjacent time points are differentially calculated to obtain the rate of change of water flow and the rate of change of power consumption. Based on a weighted fusion method, the degree of state change is calculated, and the entire time-series data is dynamically divided into stages, dividing the continuous operation process into several state stages with similar change characteristics. For example, during the nighttime period, the water flow data is close to zero for a long period, and the power consumption change is small; this period is classified as a stable stage. However, during the morning peak water consumption period, due to frequent opening and closing of the water flow, the rate of change increases significantly, corresponding to a dynamic change stage.

[0025] For each defined state stage, its statistical features, trend features, and stability features are further extracted. A multi-rule fusion discrimination mechanism is used to identify the business scenario of each stage. When low flow, low power consumption, and high stability are detected, it is determined to be a standby scenario; when the flow is stable and the change trend is gradual, it is determined to be a stable metering scenario; when the flow fluctuates significantly, it is determined to be a dynamic water use scenario; and when both water flow and power consumption show abrupt changes, it is determined to be an abnormal scenario. Through this embodiment of the invention, the continuous and complex water use process can be structured into multiple business scenario stages with clear semantics in a real-world operating environment, providing an accurate basis for subsequent hierarchical wake-up control and low-power optimization.

[0026] It should be further explained that by calculating the rate of change of adjacent running state vectors in the time-series state sequence, the time-series state sequence is dynamically divided into stages, and the state change characteristics of each stage are extracted. The specific steps are as follows: Extract the water flow state sequence and power consumption state sequence separately, and calculate the rate of change of the corresponding state sequences, specifically including the rate of change of the water flow state. and power state change rate ,in, Let be the rate of change of the water flow state at time t. The data represents the water flow state at time t-1. Let be the rate of change of power consumption state at time t. This represents the power consumption status data at time t-1; After normalizing the rate of change of water flow state and the rate of change of power consumption state, a state partitioning model is built to calculate the state partitioning coefficients. Based on the state partitioning coefficients, a dynamic segmentation rule is built to output the set of state stages. , For the i-th state stage, This is the nth state stage; Calculate the state change characteristics for each stage, including state statistical characteristics, state trend characteristics, and stability characteristics; the state statistical characteristic is the mean value of the flow state. Variance of water flow state Average power consumption and power state variance State trend characteristics include water flow state trend characteristics. and power consumption state trend characteristics The formula for calculating stability characteristics is: ; It should be noted that the calculation formula for the state partitioning model is as follows: In the formula: The state partitioning coefficient at time t, These are the weighting coefficients for the water flow state data. These are the weighting coefficients for the power consumption status data.

[0027] Specifically, a set of state stages is output based on dynamic segmentation rules constructed according to state partitioning coefficients, as detailed below: The state division coefficient is compared with the preset segmentation threshold range. If the state division coefficient is greater than the upper limit of the preset segmentation threshold range, the remote water meter battery is in the sudden change stage; if the state division coefficient is within the preset segmentation threshold range, the remote water meter battery is in the transition buffer stage; if the state division coefficient is less than the upper limit of the preset segmentation threshold range, the remote water meter battery is in the stable stage.

[0028] In one embodiment of the present invention, a water flow state sequence and a power consumption state sequence are collected at a fixed sampling period, such as 1 second or 5 seconds. Data from adjacent time points are differentially processed to calculate the water flow state change rate and the power consumption state change rate. Subsequently, the two types of change rates are normalized to eliminate dimensional differences, and a state partitioning coefficient is constructed based on a weighted fusion model. The weighting coefficient can be set according to actual operating conditions, such as water flow sensitivity or power consumption sensitivity. Based on this, the state partitioning coefficients for consecutive time points are compared with a preset segmentation threshold range. When the state partitioning coefficient exceeds the upper threshold, the system is determined to enter a sudden change phase, such as a sudden water usage or communication startup. When the state partitioning coefficient is within the threshold range, it is determined to be a transition buffer phase. When the state partitioning coefficient is below the lower threshold, it is determined to be a stable phase. This dynamically segments the entire time-series data, forming a set of state phases. For each state phase, its state change characteristics are further extracted, including the mean and variance of water flow and power consumption to characterize statistical properties. Simultaneously, trend characteristics are calculated using differential approximation to reflect the direction of change, and stability indicators are used to evaluate the degree of state fluctuation within the phase. The embodiments of the present invention help to transform raw continuous state data into multi-stage feature representations with clear physical meaning, providing a reliable basis for subsequent business scenario discrimination and hierarchical wake-up control. At the same time, in actual operation, it can effectively identify the operating status of water meters under different working conditions such as stable water use, dynamic changes and sudden events.

[0029] This invention, by calculating the rate of change of water flow and power consumption and constructing state classification coefficients, can accurately capture the abrupt changes, transitions, and stable phases during battery operation. This refined analysis is more sensitive than traditional average value or single-point threshold monitoring, and can promptly detect anomalies, such as sudden increases in water flow or abnormal power consumption, preventing potential faults from escalating. Statistical features, trend features, and stability features are extracted within each classified phase, providing a more comprehensive understanding of the battery's operating characteristics. For example, stability indicators can be used to determine whether the battery is in a high-fluctuation state for an extended period, thereby assisting in maintenance decisions.

[0030] By clearly distinguishing between the mutation phase, transition buffer phase, and stable phase, maintenance personnel can schedule inspections and maintenance in a targeted manner, avoiding blind inspections or missing critical anomaly phases, thus improving resource utilization efficiency. The state partitioning model adjusts the importance of water flow and power consumption through weighting coefficients and dynamically sets segmented thresholds, allowing for flexible adjustments based on different types of remote water meters, battery capacities, or operating environments, making the method highly versatile. The time-series information generated by phase partitioning and feature extraction can be directly used in intelligent monitoring systems, supporting anomaly prediction, lifespan estimation, and automated alarms, thereby providing reliable data support for remote operation and maintenance and decision-making.

[0031] Preferably, a multi-rule fusion stage discrimination mechanism is constructed based on state change characteristics to determine business scenarios and obtain the business scenario type corresponding to each stage. The specific steps are as follows: A multi-rule fusion stage discrimination mechanism is constructed based on state change characteristics, including standby scenario discrimination, stable metering scenario discrimination, dynamic water use scenario discrimination, and abnormal scenario discrimination; Specifically, the standby scenario determination is as follows: when the average water flow state... Approaching zero, average power consumption state Below a preset threshold, stability characteristics When the value exceeds the preset threshold, the device is in standby mode. The criteria for determining a stable metering scenario are as follows: when the average value of the water flow state... The variance of the water flow state is greater than zero. Water flow trend characteristics below a preset threshold The value is close to 0, indicating a stable metering scenario. Dynamic water use scenario discrimination specifically refers to: when the water flow state variance Water flow trend characteristics above preset threshold When the water usage exceeds a preset threshold, the system is in a dynamic water usage scenario. The specific method for identifying abnormal scenarios is: the difference in the trend characteristics of water flow at adjacent time points. The difference is greater than the preset difference threshold, and the power consumption state trend characteristic difference is greater than the preset difference threshold. If the difference exceeds the preset threshold, the situation is considered abnormal.

[0032] In this embodiment of the invention, taking the monitoring data of a remote water meter during the daily operation of a residential household as an example, after completing the state stage division, the corresponding state change features are extracted for each stage, including the mean value of water flow state, the variance of water flow state, the mean value of power consumption state, the water flow trend features, the power consumption trend features, and the stability features. Based on the above features, a multi-rule fusion stage discrimination mechanism is constructed to identify the business scenario.

[0033] During periods of no water usage at night, the average water flow is close to zero, power consumption remains at a low level, and stability is high, thus classifying it as a standby scenario. When the user turns on the tap and maintains a stable water flow, such as during continuous washing or device use, the average water flow is greater than zero with minimal fluctuations, and the water flow trend change is close to zero, classifying it as a stable metering scenario. When the user uses water intermittently, such as by frequently turning the tap on and off or using water from multiple points, the variance of the water flow state increases significantly, and the trend change is significant, thus classifying it as a dynamic water usage scenario. In the event of a sudden anomaly, such as a pipe rupture or a sudden surge in flow rate, if the difference in water flow trend characteristics and the difference in power consumption trend characteristics between adjacent moments both exceed preset thresholds, then it is classified as an abnormal scenario.

[0034] This invention, through a multi-rule fusion discrimination mechanism, can accurately classify different water usage states of remote water meters in a real operating environment. This not only improves the accuracy of scene recognition but also effectively avoids misjudgment caused by single feature judgment, providing a reliable basis for subsequent hierarchical wake-up control and low-power management.

[0035] Preferably, the wake-up power supply area is controlled according to the business scenario type, and the tasks in the power supply area are dynamically adjusted. Specifically, the battery wake-up level is determined based on the business scenario type, the wake-up power supply area is determined based on the battery wake-up level, and the corresponding tasks are controlled according to the wake-up power supply area. Among them, standby scenario corresponds to Level 1 wake-up, stable metering scenario corresponds to Level 2 wake-up, dynamic water usage scenario corresponds to Level 3 wake-up, and abnormal scenario corresponds to Level 4 wake-up. Level 1 wake-up corresponds to the core area being the power supply area; Level 2 wake-up corresponds to the core area and the sensor area being the power supply area; Level 3 wake-up corresponds to the core area, the sensor area, and the communication area being the power supply area; and Level 4 wake-up corresponds to the core area, the sensor area, the communication area, and the storage area being the power supply area.

[0036] In one embodiment of the present invention, during the 24-hour operation of a remote water meter installed in a residential household, after determining the business scenario, the system first determines the corresponding battery wake-up level based on the current business scenario type, and accordingly performs graded wake-up control on each power supply area, while driving the corresponding functional tasks to execute. When a standby scenario is identified at night or during a long period of no water usage, a level one wake-up is triggered, only powering the core area to keep the microcontroller in a minimum power consumption state, mainly performing clock maintenance, wake-up monitoring, and basic status recording tasks, thereby minimizing power consumption; when the user begins to use water stably, such as keeping the faucet running for a long time or using the water heater, a stable metering scenario is identified, corresponding to a level two wake-up, activating the sensor area on top of the core area, powering on the water flow sensor and signal conditioning circuit, and performing continuous flow acquisition and metering calculation tasks, while maintaining a low communication frequency to reduce energy consumption; when frequent changes in the user's water usage behavior are detected, such as in the kitchen and... Alternating water usage in the bathroom or multiple faucet cycles are considered dynamic water usage scenarios, triggering a Level 3 wake-up. Building upon the previous steps, this further activates the communication area, putting the wireless communication module into operation. Based on a preset strategy, it selectively uploads data or interacts with the backend system to ensure the real-time nature of critical data. Conversely, in the event of sudden anomalies, such as pipe leaks, abnormally high flow rates, or externally triggered communication requests, this is considered an abnormal scenario, triggering a Level 4 wake-up. This fully activates the core area, sensor area, communication area, and storage area, not only performing high-frequency data acquisition and real-time communication but also storing and logging abnormal data locally for subsequent analysis and tracing.

[0037] During the wake-up process at each level, task execution can be dynamically adjusted based on real-time status. For example, the sampling frequency can be reduced when the water flow stabilizes, and data transmission can be delayed when communication demand decreases, thereby further reducing energy consumption while ensuring functional responsiveness. This embodiment of the invention, through the aforementioned business scenario-driven hierarchical wake-up and task collaborative control mechanism, achieves on-demand power supply and optimized task scheduling for remote water meters under different operating conditions, significantly improving system energy efficiency and operational reliability.

[0038] Preferably, a phased water use behavior model is constructed using historical operational status data to analyze and predict the user's next phase of business scenario type, and interference from sudden disturbance scenarios that cause level identification interference is eliminated to obtain the actual business scenario type. The specific steps are as follows: By constructing a phased water use behavior model based on historical operational status data, extracting the time distribution characteristics of each historical phase, and combining the current time distribution characteristics, the business scenario type for the next phase can be predicted. The predicted business scenario type is verified based on the real-time business scenario type, and sudden interference scenarios are identified and the predicted business scenario type of sudden interference scenarios is eliminated; sudden interference scenarios include scenarios such as water pipe rupture.

[0039] In this embodiment of the invention, historical operational status data is first divided and labeled into stages to form a historical dataset containing each stage and its corresponding business scenario type. Furthermore, the frequency and duration of different business scenarios within each time period are statistically analyzed to extract temporal distribution features. For example, statistics show that users are mostly in stable metering or dynamic water usage scenarios in the morning and evening, while they are mostly in standby scenarios late at night. Based on these temporal distribution features, combined with the current time interval, the most likely business scenario type for the next stage can be predicted, such as predicting dynamic water usage or stable metering scenarios during the morning peak period.

[0040] Subsequently, the current business scenario type is acquired in real time and compared with the prediction results. When a significant deviation is detected, the real-time state change characteristics are further analyzed to identify whether a sudden interference scenario exists. For example, if the original prediction is a stable metering scenario, but a sudden change in water flow trend and a synchronous abnormal increase in power consumption are detected in real time, it can be judged as a sudden abnormal scenario, such as a water pipe rupture or an abnormally large flow leakage. For such sudden interference scenarios, the influence of the original prediction results on the control strategy will be eliminated, and the real-time judgment result will be directly used as the actual business scenario type, triggering the corresponding high-level wake-up and emergency handling mechanisms. Through the processing flow of this embodiment of the invention, not only can the historical water usage patterns of users be used to achieve forward-looking prediction of future scenarios, but it can also effectively avoid the interference of sudden anomalies on scenario identification, thereby improving the accuracy of business scenario determination and the stability of system operation.

[0041] Preferably, the scenario prediction accuracy is confirmed by comparing the actual business scenario type with the predicted business scenario type in the next stage. The specific steps are as follows: The actual business scenario types for the next phase are compared and confirmed to be consistent with the predicted business scenario types after eliminating unexpected interference scenarios. The consistency comparison results are statistically analyzed, and the scenario prediction accuracy is calculated. ,in, For scene prediction accuracy, To compare the number of times the results are consistent, To predict the frequency of different business scenario types.

[0042] In one embodiment of the present invention, the remote water meter, during a continuous operation period such as a week or a month, outputs the corresponding predicted business scenario type based on a historical behavior model at each stage. Simultaneously, it combines real-time operating status and an interference identification mechanism to obtain the actual business scenario type after eliminating sudden anomalies. Within a statistical period, for example by day or by a fixed number of stages, the predicted result and the actual judgment result for each stage are compared one by one. When they match, it is recorded as a valid match, i.e., counted in the consistency count; otherwise, it is recorded as inconsistent but not counted in the consistency count. The total number of predictions is accumulated. For example, if 100 stage scenario predictions are performed in a certain day, and 85 of the prediction results match the actual business scenario type, then according to the formula... The scenario prediction accuracy within the statistical period was found to be 85%. This invention provides a quantitative assessment of the predictive performance of the water use behavior model in actual operation and offers a basis for subsequent model parameter optimization. For example, when the accuracy falls below a preset threshold, a model update or weight adjustment mechanism can be triggered to further improve prediction accuracy and the overall intelligence level of the system.

[0043] Preferably, a low-power control method for remote water meters based on hierarchical wake-up further includes: Battery health and remaining capacity are assessed by periodically monitoring battery voltage and internal resistance. The monitoring employs a low-power ADC and simplified measurement circuitry to ensure that the monitoring process itself does not significantly increase system power consumption. Based on battery status data, operating parameters are dynamically adjusted, such as reducing communication frequency, lowering processor frequency, or increasing sleep time when the battery is low, to extend battery life. Battery status information is also prioritized for transmission, ensuring the management platform can promptly understand the device's power status. Operating parameters are dynamically adjusted based on environmental conditions, such as temperature. In low-temperature environments, battery performance degrades, necessitating reduced functional requirements or extended communication intervals; in suitable temperatures, normal operating parameters are restored. This environmental adaptability ensures the system maintains optimal energy efficiency and reliability under various conditions. It supports receiving remote configuration commands via wireless communication and dynamically adjusting power consumption control parameters. The management platform can remotely adjust the wake-up frequency, sampling strategy, and communication parameters according to actual application needs and battery status. This remote configuration capability enables the system to maintain optimal energy efficiency in different application scenarios, while also facilitating parameter optimization for batch devices.

[0044] Example 2: In one embodiment of the present invention, a low-power control method for remote water meters based on hierarchical wake-up is deployed in an urban residential community. The remote water meter is installed at the user's inlet pipe, powered by a built-in battery and operating continuously. To achieve synergistic optimization of low power consumption and high reliability, the power supply of the entire device is first modularized into four independent power supply units: a core area, a sensor area, a communication area, and a storage area. The core area includes a microcontroller and basic control circuits to maintain the minimum operating capacity of the system; the sensor area includes a water flow sensor and signal conditioning circuits for collecting water usage data; the communication area includes a wireless communication module and its radio frequency matching circuit for remote data transmission; and the storage area includes non-volatile memory and interface circuits for data caching and log recording. Each power supply area is controlled by an independent power switch circuit and equipped with a low-power wake-up circuit, enabling each module to be powered on as needed under different business requirements. Meanwhile, the system collects real-time operational status data, including water flow status data, time status data, and power consumption status data, through a multi-source sensing unit. The collected data is then preprocessed, including normalization and time-series alignment, to construct a unified operational status vector, providing a foundation for subsequent analysis.

[0045] By continuously acquiring the operational state vector sequence within a set monitoring period, such as 24 hours, and dynamically dividing the process into stages based on state changes at adjacent times, the system calculates the rate of change of water flow and the rate of change of power consumption. A weighted fusion method is then used to construct state division coefficients, segmenting the continuous data to divide the entire operation into multiple state stages with similar characteristics. For example, during the nighttime period when there is no water usage, both water flow and power consumption are relatively stable with minimal changes, and this period is classified as a stable stage. During the morning peak water usage period, the flow rate changes significantly due to frequent user usage, thus classifying this period as a dynamic stage. For each defined state stage, its state change characteristics are further extracted, including the mean, variance, trend characteristics, and stability indicators of water flow and power consumption. Based on these characteristics, a multi-rule fusion stage discrimination mechanism is constructed to identify the business scenario for each stage. In practical applications, when the average water flow is close to zero, the power consumption is low, and the stability is high, this stage is determined to be a standby scenario; when the average water flow is greater than zero and the change is stable, it is determined to be a stable metering scenario; when the water flow fluctuates greatly and the trend changes significantly, it is determined to be a dynamic water use scenario; and when both water flow and power consumption show abrupt changes, it is determined to be an abnormal scenario.

[0046] After identifying the business scenario, a tiered wake-up control strategy is executed based on different scenario types. For example, in standby scenarios, only the core area is powered to maintain an ultra-low power consumption state; in stable metering scenarios, the sensor area is activated to achieve continuous data acquisition; in dynamic water usage scenarios, the communication area is further activated to support necessary data uploads; and in abnormal scenarios, all power supply areas, including the storage area, are fully activated to achieve high-frequency data acquisition, real-time communication, and abnormal data recording. Simultaneously, during each wake-up stage, the task execution strategy is dynamically adjusted based on the real-time status, such as reducing the sampling frequency when the water flow is stable and delaying data transmission when communication demand is low, thereby further reducing energy consumption.

[0047] To further enhance the system's intelligence and proactive control capabilities, a phased water usage behavior model is constructed based on historical operational data. Specifically, by statistically analyzing historical phase data, the distribution characteristics of different business scenarios across various time periods are extracted, such as morning and evening peak water usage and nighttime low usage, and a temporal distribution probability model is established. During actual operation, the type of business scenario that may occur in the next phase is predicted based on the current time characteristics. For example, in the morning, it is predicted that users will enter a dynamic water usage or stable metering scenario. However, in actual operation, sudden interference scenarios may occur, such as water pipe ruptures or abnormally large flow surges. Such situations can cause deviations between the prediction results and the actual situation. Through a real-time status verification mechanism, the prediction results are verified, and sudden interference scenarios are identified by combining the abrupt changes in water flow and power consumption characteristics. Once an anomaly is detected, the influence of the original prediction results is eliminated, and the real-time judgment result is directly used as the actual business scenario type, triggering the corresponding high-level wake-up and emergency handling mechanisms, thereby ensuring rapid response capabilities in abnormal situations.

[0048] During continuous system operation, to evaluate the predictive performance of the behavioral model and achieve adaptive optimization, the actual business scenario type in the next stage is compared with the predicted business scenario type, and the consistency between the two is statistically analyzed. For example, within a statistical period, several scenario predictions are performed, and the number of times the predicted result matches the actual result is recorded as the number of valid matches. The scenario prediction accuracy is calculated by combining the total number of predictions. If the prediction accuracy is high, it indicates that the current behavioral model can reflect the user's water usage patterns well; if the prediction accuracy is low, the model parameters will be adjusted according to the error situation, such as optimizing the time distribution weights or updating the state transition probabilities, thereby continuously improving the prediction accuracy. Through the above feedback mechanism, the behavioral model achieves dynamic adaptive optimization, enabling it to continuously evolve with changes in users' water usage habits.

[0049] The embodiments of this invention, through the synergistic effect of a series of technical means such as power supply partitioning and multi-source sensing, dynamic stage division and multi-rule scenario discrimination, hierarchical wake-up and task collaborative control, water use behavior prediction and interference elimination, and prediction feedback optimization, can effectively achieve the unity of low-power operation and intelligent control in actual remote water meter applications. This not only significantly reduces energy consumption and extends battery life, but also improves the accuracy and stability of business scenario identification, and has good engineering application value and promotion prospects.

[0050] Example 3: A low-power control system for remote water meters based on hierarchical wake-up, including a power supply area division module, an operation status data acquisition module, a business scenario determination module, a stage scenario prediction module, and an accuracy analysis module; the power supply area division module is connected to the operation status data acquisition module, the operation status data acquisition module is connected to the business scenario determination module, the business scenario determination module is connected to the stage scenario prediction module, and the stage scenario prediction module is connected to the accuracy analysis module. The power supply area division module is used to divide the power supply of the remote water meter into several power supply areas, and equips each power supply area with a separate wake-up circuit. The operation status data acquisition module is used to acquire the operation status data of each power supply area in real time based on the multi-source sensing unit, and form an operation status vector after preprocessing. The business scenario determination module is used to divide the operation status vector within the monitoring period into stages, build a multi-rule fusion stage discrimination mechanism, obtain the business scenario type corresponding to each stage, control the wake-up power supply area according to the business scenario type, and dynamically adjust the tasks of the power supply area to be executed. The phase scenario prediction module is used to build a phase water use behavior model through historical operation status data, analyze and predict the type of business scenario for users in the next phase, and eliminate level identification interference caused by sudden interference scenarios to obtain the actual business scenario type. The accuracy analysis module is used to confirm the scenario prediction accuracy by comparing the actual business scenario type with the predicted business scenario type in the next stage, and to optimize the water use behavior model in the next stage based on the scenario prediction accuracy.

[0051] like Figure 2 The diagram shown is a system block diagram of a low-power control system for a remote water meter based on hierarchical wake-up according to an embodiment of the present invention, which can be used to execute... Figure 1 The steps in the method embodiments shown are implemented in a similar manner and have similar technical effects, and will not be repeated here.

[0052] Through the above embodiments, this invention divides the power supply area and combines multi-source sensing units to acquire water flow status, time status, and power consumption status data to construct a unified operating state vector. Then, based on the operating state vector, dynamic stages are divided within the monitoring period, and a multi-rule fusion mechanism is used to determine the business scenario for each stage, achieving accurate identification of different operating conditions such as standby, stable metering, dynamic water use, and anomalies. A graded wake-up control strategy is executed according to the business scenario type, providing on-demand power to different power supply areas and dynamically adjusting the task execution of each module. Historical operating data is used to construct a stage-based water use behavior model, which, combined with time distribution characteristics, predicts the business scenario type for the next stage. Real-time verification and interference identification mechanisms are used to eliminate the impact of sudden anomalies, obtaining a more accurate actual business scenario. The prediction accuracy is calculated by comparing the predicted results with the actual results, and the behavior model is optimized in reverse based on this accuracy, forming a continuously adaptive closed-loop control system.

[0053] This invention enables remote water meters to transition from traditional fixed wake-up to on-demand, tiered wake-up. It achieves module-level refined management at the power supply level, effectively reducing ineffective power supply and energy waste, thereby significantly reducing overall system power consumption and extending battery life. Through stage-based segmentation and multi-rule fusion discrimination mechanisms, it improves the accuracy and stability of business scenario identification, avoiding misjudgments caused by single-feature judgments. Combining water usage behavior models and scenario prediction mechanisms, it enables the system to have forward-looking decision-making capabilities, shifting from passive response to proactive control. Through interference identification and elimination mechanisms, it enhances the system's robustness in sudden anomalies, and through predictive feedback, it achieves adaptive model optimization, maintaining superior performance over the long term. This significantly improves the intelligence level and operational reliability of remote water meters while reducing power consumption.

[0054] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any changes or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application.

[0055] Finally: The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A low-power control method for remote water meters based on hierarchical wake-up, characterized in that, Includes the following steps: By dividing the power supply of the remote water meter into several power supply areas and equipping each power supply area with a separate wake-up circuit, the operating status data of each power supply area is acquired in real time based on the multi-source sensing unit, and the data is preprocessed to form an operating status vector; the operating status data includes water flow status data, time status data and power consumption status data. Based on the operational status vector within the monitoring period, a phase division is performed, a multi-rule fusion phase discrimination mechanism is constructed, the business scenario type corresponding to each phase is obtained, the wake-up power supply area is controlled according to the business scenario type, and the tasks of the power supply area are dynamically adjusted. By constructing a phased water use behavior model based on historical operational status data, we can analyze and predict the type of business scenario for users in the next phase, and eliminate level identification interference caused by sudden interference scenarios to obtain the actual business scenario type. The accuracy of scenario prediction is confirmed by comparing the actual business scenario type with the predicted business scenario type in the next stage, and the water use behavior model in the stage is optimized in reverse based on the scenario prediction accuracy.

2. The low-power control method for remote water meters based on hierarchical wake-up as described in claim 1, characterized in that, The power supply area includes: the core area, the sensor area, the communication area, and the storage area.

3. The low-power control method for remote water meters based on hierarchical wake-up as described in claim 1, characterized in that, Based on the operational status vector within the monitoring period, a phase division is performed, and a multi-rule fusion phase discrimination mechanism is constructed to obtain the business scenario type corresponding to each phase. The specific steps are as follows: A time-series state sequence is constructed based on the operating state vector within the monitoring period. The time-series state sequence includes the water flow state sequence and the power consumption state sequence. By calculating the rate of change of adjacent running state vectors in the time-series state sequence, the time-series state sequence is dynamically divided into stages, and the state change characteristics of each stage are extracted. A multi-rule fusion stage discrimination mechanism is constructed based on state change characteristics to identify business scenarios and obtain the corresponding business scenario type for each stage.

4. The low-power control method for remote water meters based on hierarchical wake-up as described in claim 3, characterized in that, By calculating the rate of change of adjacent running state vectors in the time-series state sequence, the time-series state sequence is dynamically divided into stages, and the state change characteristics of each stage are extracted. The specific steps are as follows: Extract the water flow state sequence and the power consumption state sequence respectively, and calculate the rate of change of the corresponding state sequence, specifically including the rate of change of water flow state and the rate of change of power consumption state. After normalizing the water flow state change rate and power consumption state change rate, a state partitioning model is built to calculate the state partitioning coefficient. Based on the state partitioning coefficient, a dynamic segmentation rule is built to output the state stage set. Calculate the state change characteristics for each state stage. The state change characteristics include state statistical characteristics, state trend characteristics, and stability characteristics.

5. The low-power control method for remote water meters based on hierarchical wake-up as described in claim 4, characterized in that, State statistical characteristics include the mean of water flow state, the variance of water flow state, the mean of power consumption state, and the variance of power consumption state; state trend characteristics include the trend characteristics of water flow state and the trend characteristics of power consumption state.

6. The low-power control method for remote water meters based on hierarchical wake-up as described in claim 4, characterized in that, Based on the state partitioning coefficients, a dynamic segmentation rule is constructed to output a set of state stages, as follows: The state division coefficient is compared with the preset segmentation threshold range. If the state division coefficient is greater than the upper limit of the preset segmentation threshold range, the remote water meter battery is in the sudden change stage; if the state division coefficient is within the preset segmentation threshold range, the remote water meter battery is in the transition buffer stage; if the state division coefficient is less than the upper limit of the preset segmentation threshold range, the remote water meter battery is in the stable stage.

7. The low-power control method for remote water meters based on hierarchical wake-up as described in claim 1, characterized in that, The multi-rule fusion stage discrimination mechanism includes standby scenario discrimination, stable metering scenario discrimination, dynamic water usage scenario discrimination, and abnormal scenario discrimination.

8. The low-power control method for remote water meters based on hierarchical wake-up as described in claim 1, characterized in that, The wake-up power supply area is controlled according to the business scenario type, and the tasks in the power supply area are dynamically adjusted. Specifically, the battery wake-up level is determined based on the business scenario type, the wake-up power supply area is determined based on the battery wake-up level, and the corresponding tasks are controlled based on the wake-up power supply area. Among them, standby scenario corresponds to Level 1 wake-up, stable metering scenario corresponds to Level 2 wake-up, dynamic water usage scenario corresponds to Level 3 wake-up, and abnormal scenario corresponds to Level 4 wake-up.

9. A low-power control method for remote water meters based on hierarchical wake-up as described in claim 8, characterized in that, Level 1 wake-up corresponds to the core area being the power supply area; Level 2 wake-up corresponds to the core area and the sensor area being the power supply area; Level 3 wake-up corresponds to the core area, the sensor area, and the communication area being the power supply area; and Level 4 wake-up corresponds to the core area, the sensor area, the communication area, and the storage area being the power supply area.

10. A low-power control system for remote water meters based on hierarchical wake-up, applied to the low-power control method for remote water meters based on hierarchical wake-up as described in any one of claims 1-9, characterized in that, It includes a power supply area division module, an operation status data acquisition module, a business scenario determination module, a stage scenario prediction module, and an accuracy analysis module; The power supply area division module is used to divide the power supply of the remote water meter into several power supply areas, and equips each power supply area with a separate wake-up circuit. The operation status data acquisition module is used to acquire the operation status data of each power supply area in real time based on the multi-source sensing unit, and form an operation status vector after preprocessing. The business scenario determination module is used to divide the operation status vector within the monitoring period into stages, build a multi-rule fusion stage discrimination mechanism, obtain the business scenario type corresponding to each stage, control the wake-up power supply area according to the business scenario type, and dynamically adjust the tasks of the power supply area to be executed. The phase scenario prediction module is used to build a phase water use behavior model through historical operation status data, analyze and predict the type of business scenario for users in the next phase, and eliminate level identification interference caused by sudden interference scenarios to obtain the actual business scenario type. The accuracy analysis module is used to confirm the scenario prediction accuracy by comparing the actual business scenario type with the predicted business scenario type in the next stage, and to optimize the water use behavior model in the next stage based on the scenario prediction accuracy.

Images