A smart Internet of Things water meter control method and system

Abstract

The present invention relates to the field of automatic control technology, and specifically to a smart Internet of Things water meter control method and system, including: obtaining the water consumption of the water meter at the current time node, inputting the water consumption of the current time node into an improved machine learning model, outputting the water consumption status of the current time node, and if the water consumption status is abnormal, controlling the water meter to send out an early warning signal and stop water supply; wherein the improved machine learning model includes iteratively training the machine learning model using a loss function, and calculating the loss function value of the current time node, and the historical loss function value is positively correlated with the abnormal degree of water consumption at the historical time node and negatively correlated with the surge coefficient of water consumption at the historical time node. The present invention solves the potential deficiencies of existing algorithms in detection efficiency and accuracy, which may lead to the problem that the overall performance and reliability of smart water meters are not high.

Description

Technical Field

[0001] The present invention relates to the field of automatic control technology, and more specifically, to a smart Internet of Things water meter control method and system. Background Art

[0002] With the rapid development of smart IoT technology, the functions of traditional water meters are undergoing profound changes, gradually moving towards a more efficient and intelligent direction. At present, many smart water meters use wireless communication technology to upload water consumption data to the cloud platform in a timely manner for centralized management and in-depth analysis. However, existing smart water meter systems still face many challenges in data processing and analysis. Many systems rely on preset algorithms to process the collected water consumption data, which may have certain limitations in practical applications. On the one hand, the preset algorithm lacks flexibility and cannot effectively cope with changing water use scenarios; on the other hand, this method does not pay enough attention to the detection efficiency and accuracy of the algorithm, which may lead to inaccurate results in the prediction and detection of abnormal water use status.

[0003] The patent application document with application publication number CN117336330A discloses a smart water meter system. The patent application document includes a smart water meter and a cloud server. The smart water meter is equipped with a built-in water flow sensor for collecting the user's water consumption data in real time. The cloud server has a data processing module, which analyzes the collected water consumption data by applying a predetermined algorithm to generate personalized water consumption recommendations for the user. Once the user confirms that he accepts these water consumption recommendations, the cloud server will continue to collect real-time water consumption data to evaluate whether the user follows the water consumption recommendations. If the user fails to fully comply with the water consumption recommendations, the system will automatically adjust the valve opening of the smart water meter by issuing a control signal instruction to ensure that the water saving goal is achieved.

[0004] However, the above technical solution only analyzes the collected water usage data based on a predetermined algorithm to achieve control of the smart water meter, but does not fully consider the potential deficiencies of the algorithm in detection efficiency and accuracy, which may lead to low accuracy of the prediction results, thereby limiting the system's ability to promptly identify and respond to abnormal water usage conditions, thereby affecting the overall performance and reliability of the smart water meter. Summary of the invention

[0005] In order to solve the problems affecting the overall performance and reliability of the smart water meter raised in the above background technology, the present invention provides solutions in the following aspects.

[0006] In a first aspect, the present invention provides a method for controlling an intelligent IoT water meter, comprising: obtaining the water consumption of the water meter at a current time node, inputting the water consumption of the current time node into an improved machine learning model, outputting the water consumption status of the current time node, and if the water consumption status is abnormal, controlling the water meter to send out an early warning signal and stop water supply; wherein the improved machine learning model includes a loss function; the loss function for:

[0007] ;

[0008] In the formula, represents an exponential function with the natural constant e as the base, Indicates that the water meter is The historical loss function value of water consumption at each historical time node, Represents the number of historical loss function values ​​of water consumption at multiple historical time nodes, Represents the mean of the historical loss function values ​​of water consumption at multiple historical time nodes; the historical loss function value is positively correlated with the abnormal degree of water consumption at the historical time node and negatively correlated with the surge coefficient of water consumption at the historical time node; the surge coefficient is:

[0009] ;

[0010] In the formula, Indicates that the water meter is The surge coefficient of water consumption at each historical time point, Indicates The water consumption at each historical time point, Represents the average water consumption at multiple historical time nodes.

[0011] The above technical solution predicts the water use status at the current time node by introducing an improved machine learning model, achieving more accurate water use status analysis, effectively reducing the false alarm rate, and improving the response speed and accuracy of the water meter to abnormal water use. When abnormal water use is detected, the system can issue an early warning in time and automatically stop water supply to avoid resource waste and loss caused by water leakage or uncontrolled water use. This intelligent control method not only improves the monitoring accuracy of the water meter, but also enhances the security and automation level of the IoT water meter system.

[0012] Furthermore, the abnormality degree is:

[0013] ;

[0014] In the formula, Indicates that the water meter is The abnormal degree of water consumption at each historical time point, Indicates The surge coefficient of water consumption at each historical time point, Indicates The water consumption at each historical time point, Represents the average of water consumption at multiple historical time nodes, Represents the variance mean of the water consumption surge coefficient at multiple historical time nodes, Represents an exponential function with the natural constant e as the base.

[0015] Furthermore, the abnormality degree is:

[0016] ;

[0017] In the formula, Indicates that the water meter is The abnormal degree of water consumption at each historical time point, Indicates The surge coefficient of water consumption at each historical time point, Indicates The water consumption at each historical time point, Represents the average of water consumption at multiple historical time nodes, Represents the variance mean of the water consumption surge coefficient at multiple historical time nodes, Represents the standard deviation of water consumption at multiple historical time nodes, Represents an exponential function with the natural constant e as the base.

[0018] The above technical solution fully considers the volatility and surge of water consumption by introducing a calculation formula for the degree of abnormality, reflects the abnormal water consumption through the ratio of the surge coefficient to the mean, and uses the variance mean to weaken the noise effect that may exist in abnormality detection. The application of the exponential function further smoothes the detection results, making the judgment of abnormal water consumption more sensitive and robust. Overall, the solution can more accurately evaluate the degree of water use anomaly in historical data, reduce the risk of misjudgment, and improve the system's adaptability and detection accuracy to abnormal situations in different water use scenarios.

[0019] Furthermore, the historical loss function value is:

[0020] ;

[0021] In the formula, Indicates that the water meter is The loss function value of water consumption at each historical time node, Indicates that the water meter is The abnormal degree of water consumption at each historical time point, Indicates The surge coefficient of water consumption at each historical time point, Represents an exponential function with the natural constant e as the base.

[0022] The above technical solution introduces exponential corrections of the degree of abnormality and the surge coefficient in the calculation of the loss function value, so that the loss function can more sensitively reflect the abnormal water consumption. The surge coefficient is attenuated by an exponential function in the calculation, which can effectively balance the excessive impact of sudden surges in water consumption on the loss function and prevent over-response to short-term fluctuations. At the same time, the loss function is combined with the quantification of the degree of abnormality to make the overall evaluation more accurate, especially when dealing with complex water use patterns or scenarios with large fluctuations. The solution significantly improves the accuracy of determining abnormal water use. This improvement method enhances the system's ability to detect abnormal conditions, makes the algorithm more stable and more resistant to interference, and helps to improve the overall performance of the smart water meter system.

[0023] Furthermore, the machine learning model is a long short-term memory neural network model.

[0024] The above technical solution significantly enhances the system's ability to process time series data by adopting a long short-term memory neural network model. Since the long short-term memory neural network model can retain important historical information and filter out irrelevant data, the system can more accurately predict and detect water use anomalies and improve its ability to learn and adapt to water use behavior patterns.

[0025] Furthermore, the water meter is connected to a flow sensor and a wireless communication module for data transmission, the flow sensor is used to obtain the water consumption of the water meter at historical time nodes and current time nodes, and the water consumption is transmitted to the user terminal or management system through the wireless communication module.

[0026] Furthermore, the water usage status includes: a normal state and an abnormal state.

[0027] Furthermore, it also includes iterative training of the machine learning model, specifically:

[0028] Initialize the model parameters, use the chain rule to calculate the gradient of the loss function relative to the model parameters through back propagation; and update the model parameters according to the set learning rate, repeating the above steps until the loss value converges or reaches the preset maximum number of iterations.

[0029] The above technical solution enables the system to adaptively optimize model parameters by introducing iterative training of machine learning models, thereby improving the ability to identify abnormal water use conditions. By calculating the gradient of the loss function through the chain rule and back propagation algorithm, the model can accurately adjust the parameters to minimize the loss function, ensuring that the model performance is gradually improved after each training. By setting the learning rate, the stability and efficiency of the model update are guaranteed to avoid adjusting the parameters too fast or too slow. Through repeated iterative training, the loss value gradually converges, allowing the model to adapt to different water use patterns and environments, with higher generalization ability and anomaly detection accuracy. This solution enhances the intelligence and self-learning ability of the water meter system and optimizes the water use monitoring effect.

[0030] In a second aspect, the present invention provides an intelligent Internet of Things water meter control system, comprising a memory and a processor, wherein the memory stores computer program instructions, and when the computer program instructions are executed by the processor, any one of the above-described intelligent Internet of Things water meter control methods is implemented.

[0031] The beneficial effects of the present invention are:

[0032] The present invention realizes accurate monitoring and anomaly detection of water usage status. The use of improved machine learning models, especially long short-term memory neural networks, enables the system to effectively process time series data and capture the dynamic changes and long-term dependencies of water consumption. By designing loss functions and surge coefficient calculations, the system can keenly identify abnormal water usage, thereby quickly issuing early warnings and automatically stopping water supply, reducing the risk of resource waste. In addition, the iterative training process of the system continuously optimizes model performance and improves the accuracy and reliability of abnormal state identification. Overall, the present invention not only improves the intelligence level of water meters, but also provides a more efficient and safe solution for water management, promoting the development of smart water services. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] By reading the following detailed description with reference to the accompanying drawings, the above and other objects, features and advantages of the exemplary embodiments of the present invention will become readily understood. In the accompanying drawings, several embodiments of the present invention are shown in an exemplary and non-limiting manner, and the same or corresponding reference numerals represent the same or corresponding parts, wherein:

[0034] Figure 1 is a flow chart schematically showing a method for controlling a water meter of an intelligent Internet of Things according to an embodiment of the present invention;

[0035] Figure 2 The figure schematically shows a structural block diagram of an intelligent Internet of Things water meter control system according to an embodiment of the present invention. DETAILED DESCRIPTION

[0036] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present invention.

[0037] The specific embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0038] An embodiment of a smart Internet of Things water meter control method.

[0039] like Figure 1 As shown, a flow chart of a smart IoT water meter control method according to an embodiment of the present invention includes the following steps:

[0040] S1: Get the water consumption of the water meter at the current time node.

[0041] In one embodiment, the water meter is connected to a flow sensor and a wireless communication module for data transmission. The flow sensor is mainly used to obtain the water consumption data of the water meter at each time point in real time to ensure accurate monitoring of water consumption. The flow sensor is installed at the inlet and outlet of the water meter, and can capture the instantaneous flow and cumulative flow data of water flowing through the water meter in real time, and organize these data into data information that meets the standards. In this way, the flow sensor can provide stable and accurate water consumption data under various water pressures, flow rates and environmental conditions.

[0042] When the flow sensor obtains the water consumption data of the current time node, it will transmit the data to the wireless communication module connected to it. As a key device for data transmission, the wireless communication module undertakes the functions of data uploading and remote communication. In order to ensure the stability of data transmission, the wireless communication module can adopt a variety of communication technologies, such as cellular networks, Wi-Fi or low-power wide area networks. This diversified communication method ensures that water meter data can be reliably transmitted in various complex environments. Whether it is a remote management center or a user terminal, water consumption data can be quickly received through the wireless network. During the data transmission process, the wireless communication module will first encode and process the raw data from the flow sensor to ensure that it meets the requirements of the transmission protocol and data format. Subsequently, the encrypted data is transmitted to the user terminal or management system via wireless signals. In order to ensure the security and privacy of data transmission, the system can encrypt the water consumption data through an encryption algorithm.

[0043] S2: Input the water consumption at the current time node into the improved machine learning model, and output the water consumption status at the current time node.

[0044] In one embodiment, taking the improved long short-term memory neural network model as an example, the water consumption at the current time node is input into the improved long short-term memory neural network model to obtain the water consumption status of the water consumption at the current time node;

[0045] The water use status can be divided into a normal state and an abnormal state. In order to improve the system's recognition accuracy of the water use status, the machine learning model can also be iteratively trained. The training process is as follows: first, initialize the parameters of the model; then, use the chain rule to calculate the gradient of the loss function relative to the model parameters through the back propagation algorithm; then, update the model parameters according to the set learning rate. The above process is repeated until the loss value converges to a preset threshold or reaches the set maximum number of iterations. Among them, the number of iterations and the learning rate are set manually.

[0046] S3: The improved machine learning model includes a loss function.

[0047] In one embodiment, the improved machine learning model includes a loss function; taking the long short-term memory neural network model as an example, the loss function of the long short-term memory neural network model is for:

[0048] ;

[0049] in, represents an exponential function with the natural constant e as the base, Indicates that the water meter is The historical loss function value of water consumption at each historical time node, Represents the number of historical loss function values ​​of water consumption at multiple historical time nodes, Represents the mean of the historical loss function values ​​of water consumption at multiple historical time nodes;

[0050] The historical loss function value is positively correlated with the abnormal degree of water consumption at the historical time node, and negatively correlated with the surge coefficient of water consumption at the historical time node;

[0051] The historical loss function value is:

[0052] ;

[0053] In the formula, Indicates that the water meter is The loss function value of water consumption at each historical time node, Indicates that the water meter is The abnormal degree of water consumption at each historical time point, Indicates The surge coefficient of water consumption at each historical time point, Represents an exponential function with the natural constant e as the base.

[0054] The surge factor is:

[0055] ;

[0056] In the formula, Indicates The surge coefficient of water consumption at each historical time point, Indicates The water consumption at each historical time point, Represents the average water consumption at multiple historical time nodes.

[0057] The degree of abnormality is:

[0058] ;

[0059] In the formula, Indicates that the water meter is The abnormal degree of water consumption at each historical time point, Indicates The surge coefficient of water consumption at each historical time point, Indicates The water consumption at each historical time point, Represents the average of water consumption at multiple historical time nodes, Represents the variance mean of the water consumption surge coefficient at multiple historical time nodes, Represents an exponential function with the natural constant e as the base.

[0060] In another embodiment, the abnormality degree is:

[0061] ;

[0062] In the formula, Indicates that the water meter is The abnormal degree of water consumption at each historical time point, Indicates The surge coefficient of water consumption at each historical time point, Indicates The water consumption at each historical time point, Represents the average of water consumption at multiple historical time nodes, Represents the variance mean of the water consumption surge coefficient at multiple historical time nodes, Represents the standard deviation of water consumption at multiple historical time nodes, Represents an exponential function with the natural constant e as the base.

[0063] For example: Assume that the historical time node is 5, the water consumption at the current time node is 120, and the water consumption at the historical time nodes is {90, 100, 105, 95, 100}, then the mean water consumption at the historical time nodes is The water consumption loss function value at each historical time node can be obtained by substituting the above example parameters into the above formula.

[0064] S4: If the water usage status is abnormal, the water meter is controlled to send out an early warning signal and stop the water supply.

[0065] In one embodiment, once the current water usage status is determined to be abnormal, the system will immediately generate an early warning signal. The early warning signal can be issued in a variety of ways, such as the indicator light on the water meter flashing, the buzzer sounding a prompt, or the early warning information is sent to the user terminal or management center through the wireless communication module; the user terminal can be a mobile phone APP, SMS notification or control panel, and the user can receive the abnormal water usage alarm in time through these ways.

[0066] At the same time, in order to avoid wasting water resources or causing greater losses due to abnormal conditions, such as pipe ruptures and water leaks, the system will implement an automatic water supply control strategy. When an abnormal state is triggered, the water supply control device built into the water meter, such as an electric valve, will automatically close and immediately stop the water supply. This water supply stop operation is achieved through a pre-set control program to ensure that the system can respond quickly and take effective measures when an abnormality is detected.

[0067] The solution of the present invention realizes intelligent monitoring and management of water use status by integrating improved machine learning models, anomaly detection mechanisms and flow sensors. The solution uses multi-dimensional data analysis such as loss functions and surge coefficients to accurately identify water use anomalies, significantly improving the detection rate and response speed of abnormal conditions. When the water use status is abnormal, the system can issue an early warning signal in time and automatically stop water supply, effectively avoiding the waste of water resources and potential property losses. In addition, the use of long short-term memory neural network models enhances the system's ability to understand time series data and improves its adaptability to complex water use behaviors, while ensuring the self-optimization of the model through an iterative training mechanism. This series of technical improvements not only improves the intelligence level of water meters, but also provides users with safer, more convenient and efficient water management solutions.

[0068] An embodiment of a smart Internet of Things water meter control system:

[0069] like Figure 2 As shown, a structural block diagram of an intelligent Internet of Things water meter control system according to an embodiment of the present invention includes a processor and a memory.

[0070] The present invention also provides an intelligent Internet of Things water meter control system. Figure 2 As shown, the system includes a processor and a memory, and the memory stores computer program instructions. When the computer program instructions are executed by the processor, an intelligent Internet of Things water meter control method according to the present invention described above is implemented.

[0071] The intelligent Internet of Things water meter control system also includes other components familiar to those skilled in the art, such as a communication interface, and their settings and functions are known in the art, so they will not be described in detail here.

[0072] In the present invention, the aforementioned memory may be any tangible medium containing or storing a program that can be used by or in combination with an instruction execution system, apparatus or device. For example, a computer-readable storage medium may be any appropriate magnetic storage medium or magneto-optical storage medium, such as a resistive random access memory RRAM (Resistive Random Access Memory), a dynamic random access memory DRAM (Dynamic Random Access Memory), a static random access memory SRAM (Static Random-Access Memory), an enhanced dynamic random access memory EDRAM (Enhanced Dynamic Random Access Memory), a high-bandwidth memory HBM (High-Bandwidth Memory), a hybrid memory cube HMC (Hybrid Memory Cube), etc., or any other medium that can be used to store the required information and can be accessed by an application, a module, or both. Any such computer storage medium may be part of a device or accessible or connectable to a device. Any application or module described in the present invention may be implemented using computer-readable / executable instructions that may be stored or otherwise maintained by such a computer-readable medium.

[0073] In the description of this specification, "plurality" or "several" means at least two, such as two, three or more, etc., unless otherwise clearly and specifically defined.

[0074] Although this specification has shown and described a number of embodiments of the present invention, it will be apparent to those skilled in the art that such embodiments are provided by way of example only. Those skilled in the art will conceive of many modifications, changes and alternatives without departing from the ideas and spirit of the present invention. It should be understood that in the practice of the present invention, various alternatives to the embodiments of the present invention described herein may be employed.

Claims

1. A smart Internet of Things water meter control method, characterized in that: include: Obtain the water consumption of the water meter at the current time node, input the water consumption of the current time node into the improved machine learning model, output the water consumption status of the current time node, and if the water consumption status is abnormal, control the water meter to send out an early warning signal and stop water supply; The improved machine learning model includes a loss function; the loss function for: ； In the formula, represents an exponential function with the natural constant e as the base, Indicates that the water meter is The historical loss function value of water consumption at each historical time node, Represents the number of historical loss function values ​​of water consumption at multiple historical time nodes, Represents the mean of the historical loss function values ​​of water consumption at multiple historical time nodes; The historical loss function value is positively correlated with the abnormal degree of water consumption at the historical time node and negatively correlated with the surge coefficient of water consumption at the historical time node; the surge coefficient is: ； In the formula, Indicates that the water meter is The surge coefficient of water consumption at each historical time point, Indicates The water consumption at each historical time point, Represents the average water consumption at multiple historical time nodes.

2. The intelligent Internet of Things water meter control method according to claim 1 is characterized in that: The degree of abnormality is: ； In the formula, Indicates that the water meter is The abnormal degree of water consumption at each historical time point, Indicates The surge coefficient of water consumption at each historical time point, Indicates The water consumption at each historical time point, Represents the average of water consumption at multiple historical time nodes, Represents the variance mean of the water consumption surge coefficient at multiple historical time nodes, Represents an exponential function with the natural constant e as the base.

3. The intelligent Internet of Things water meter control method according to claim 1 is characterized in that: The degree of abnormality is: ； In the formula, Indicates that the water meter is The abnormal degree of water consumption at each historical time point, Indicates The surge coefficient of water consumption at each historical time point, Indicates The water consumption at each historical time point, Represents the average of water consumption at multiple historical time nodes, Represents the variance mean of the water consumption surge coefficient at multiple historical time nodes, Represents the standard deviation of water consumption at multiple historical time nodes, Represents an exponential function with the natural constant e as the base.

4. The intelligent Internet of Things water meter control method according to claim 1 is characterized in that: The historical loss function value is: ； In the formula, Indicates that the water meter is The loss function value of water consumption at each historical time node, Indicates that the water meter is The abnormal degree of water consumption at each historical time point, Indicates The surge coefficient of water consumption at each historical time point, Represents an exponential function with the natural constant e as the base.

5. The intelligent Internet of Things water meter control method according to claim 1 is characterized in that: The machine learning model is a long short-term memory neural network model.

6. The intelligent Internet of Things water meter control method according to claim 1 is characterized in that: The water meter is connected to a flow sensor and a wireless communication module for data transmission. The flow sensor is used to obtain the water consumption of the water meter at historical time nodes and current time nodes. The water consumption is transmitted to the user terminal or management system through the wireless communication module.

7. The intelligent Internet of Things water meter control method according to claim 1 is characterized in that: The water use status includes: a normal state and an abnormal state.

8. The intelligent Internet of Things water meter control method according to claim 1, characterized in that: It also includes iterative training of the machine learning model, specifically: Initialize the model parameters and use the chain rule to calculate the gradient of the loss function with respect to the model parameters through back propagation; The model parameters are updated according to the set learning rate, and the above steps are repeated until the loss value converges or the preset maximum number of iterations is reached.

9. An intelligent Internet of Things water meter control system, characterized in that: The invention comprises a memory and a processor, wherein the memory stores computer program instructions, and when the computer program instructions are executed by the processor, an intelligent Internet of Things water meter control method as described in any one of claims 1 to 8 is implemented.

Images