Waterway detection and control method, circuit and application

Abstract

The invention provides a waterway detection and control method, circuit and application. The method comprises the following steps: arranging various types of sensors on a water tank, realizing multi-dimensional acquisition of the water level in the water tank through the sensors, and fusing and calibrating the acquired water level data to obtain preprocessed water level data; acquiring the preprocessed water level data, inputting the preprocessed water level data into a dynamic waterway control adjustment strategy, and dynamically adjusting a waterway according to a water level change trend and a waterway state; meanwhile, the water level data is compared with a preset multi-level water level threshold value, and when the water level data is lower than a safety threshold value in the multi-level water level threshold value, an alarm mechanism is automatically triggered; acquiring water quality data from the water level data, comparing the water quality data with a preset safety range, and automatically cleaning water when the water quality data exceeds the safety range; and adjusting the dynamic waterway control adjustment strategy. The system comprises a data acquisition module, a dynamic adjustment module and an iterative learning module. The water level and the water quality are monitored and controlled in all directions.

Description

Technical Field

[0001] The present invention relates to the technical field of waterway intelligent control, and in particular to a waterway detection and control method, circuit and application. Background Art

[0002] A water dispenser is an intelligent terminal device that provides drinking water. The water tank is a container for storing water in the water dispenser, and the waterway is a pipe system connecting the water tank and the water dispenser. The water tank and the pipe together constitute the water supply system of the water dispenser. The reliability and safety of the waterway are crucial to the function of the water dispenser and directly affect the user experience. The control of the waterway is crucial to the normal operation and user experience of the water dispenser. By controlling the flow of the waterway, the water dispenser can provide a stable water flow in different usage scenarios. Water dispensers usually need to provide cold water, room temperature water and hot water. The control of the waterway can ensure that water of different temperatures can be accurately transported from the water tank to the water dispenser, and reach the temperature required by the user through the heating or cooling system. The control of the waterway can also help monitor and maintain water quality. For example, by setting a filter device in the waterway, impurities and harmful substances in the water can be removed to ensure that the water users drink is clean. Reasonable waterway control can reduce energy waste. For example, through an intelligent control system, the water dispenser can automatically shut down or reduce energy consumption when no one is using it, thereby achieving the purpose of energy saving and environmental protection. The control of the water circuit can also help monitor the operating status of the system; when the water circuit is blocked, leaking or in other abnormal conditions, the control system can issue an early warning in time to remind the user to perform maintenance and avoid larger failures. In short, the control of the water circuit is not only related to the normal operation of the water dispenser, but also directly affects the user's drinking experience and health and safety. Through scientific and reasonable design and control, it can be ensured that the water dispenser can work efficiently and stably in various environments. However, the existing technology has a relatively single means of water circuit detection and does not achieve all-round detection of the water circuit, resulting in the accuracy of the output water circuit detection results needs to be further improved.

[0003] Prior art 1, application number: CN201711042578.3 discloses a water circuit control system and method for a drinking water device on beauty equipment, including a water tank, a heating device, a temperature display device, an alarm device, a heating device, and a microprocessor. The microprocessor is respectively connected to the heating device, the temperature display device, the alarm device, and the temperature sensor. The heating device and the heating device are respectively connected to the water tank. Although the water level meter is omitted, the cost of the drinking water device on the beauty equipment is reduced, and it can more accurately detect whether the water circuit control system of the drinking water device on the beauty equipment is normal. However, the water circuit control is only realized to be normal, and the function is relatively simple, and its function needs to be further enriched and improved.

[0004] Prior art 2, application number: CN201210135031.9 discloses an intelligent water machine detection device, including a device housing, a circuit control system and a water circuit execution system controlled by an electrical signal of the circuit control system are arranged in the device housing, the circuit control system includes a power supply line for providing power transmission for the product to be detected and a control circuit connected in parallel with the power supply line, and the control circuit is installed with an intelligent controller, which can manually set the detection parameters inside the intelligent controller according to the detection requirements, and the intelligent controller harness electrical signal controls the on and off of the electric actuator in the water circuit execution system, thereby controlling the water flow between the water circuit execution system and the product to be detected to meet the test requirements. Although the detection accuracy is high, the application range is wide, the operation is safe and convenient, and it can effectively improve the detection rate of the product and the factory qualified rate of the product. However, the lack of alarm prompts and cleaning prompts has led to the need for further improvement in the intelligence of the detection.

[0005] Prior art three, application number: CN201410584895.8 discloses an intelligent Internet of Things system adapted to a water purification device, including: a water quality pH value detection module, a water quality TDS and hardness detection module, a flow statistics module, a water channel control module, a water leakage detection module, a core operation processing module, a data storage module, a screen control module, a safety alarm module, a sensor error calibration module, a mobile terminal communication configuration module, a cloud service communication module and an intelligent consumable life calculation module. Although it can be adapted to most of the water purifiers on the market, and can complete the functions of intelligent water quality detection, intelligent calculation of consumable life, kitchen water leakage detection and protection, water flow consumption and regularity statistics, kitchen combustible gas detection and alarm, and intelligent calibration of detection sensors without relying on the water purifier itself. However, it is not integrated with the water dispenser, which increases the use cost of the water dispenser and is inconvenient to operate.

[0006] At present, the existing technologies 1, 2 and 3 have the problems of low intelligent degree of waterway detection, simple functions and failure to fully and effectively reflect the waterway detection results. Therefore, the present invention provides a waterway detection and control method, circuit and application. Summary of the invention

[0007] In order to solve the above technical problems, the present invention provides a waterway detection and control method, comprising the following steps: Various types of sensors are placed on the water tank to collect the water level in multiple dimensions. The collected water level data are fused and calibrated to obtain pre-processed water level data. Obtain pre-processed water level data and input it into the dynamic waterway control adjustment strategy to dynamically adjust the waterway according to the water level change trend and waterway status; The data after dynamic adjustment of the waterway is input into the machine learning model to learn the data and dynamic waterway control adjustment strategy. Through continuous iterative learning, the dynamic waterway control adjustment strategy is adjusted.

[0008] Optionally, the process of deploying multiple types of sensors on the water tank includes the following steps: Obtain the geometric characteristics of the water tank, including the height, diameter, wall thickness and inlet and outlet positions of the water tank. According to the height and diameter of the water tank, preliminarily determine the number and distribution of monitoring points, divide the water tank into three areas: upper, middle and lower, and set several monitoring points in each area; Analyze the inlet and outlet locations of the water tank to determine the water flow path and retention area; set up multiple monitoring points based on the geometric characteristics of the water tank and the water flow path, and each monitoring point should contain at least one type of sensor; The pre-processed water level data is input into the dynamic waterway control adjustment strategy, and the position and parameters of the sensor are dynamically adjusted according to the water level change trend and waterway status; Water level change trend prediction formula: In the formula, represents the predicted water level change, represents the observed water level change, represents the last predicted water level change, represents the smoothing coefficient; Sensor position adjustment: In the formula, represents the adjusted sensor position, Indicates the original sensor position, Indicates the position adjustment amount.

[0009] Optionally, the water level data is compared with a preset multi-level water level threshold. When the water level data is lower than a safety threshold in the multi-level water level threshold, an alarm mechanism is automatically triggered. Water quality data is obtained from the water level data and compared with a preset safety range. When the safety range is exceeded, the water quality is automatically cleaned.

[0010] Optionally, the process of dynamically adjusting the waterway includes the following steps: Perform time series analysis on water level data to extract water level change trends and periodic fluctuation target features; use the target feature time series prediction model to predict future water level change trends and generate prediction curves; According to the design capacity of the water tank and the current water level, the waterway capacity is dynamically evaluated through fuzzy reasoning; based on the stability evaluation method, the waterway is dynamically evaluated through probabilistic reasoning; combined with the data from the water quality sensor, it is evaluated whether the water quality meets the safety standards and whether water cleaning is required; According to the water level change trend and waterway status, multi-level water level thresholds are set, including safety thresholds and warning thresholds; adaptive threshold setting is used in combination with historical water level data and real-time water level data to dynamically adjust the thresholds; based on the comparison results between the current water level and the threshold, the corresponding dynamic waterway control adjustment strategy is generated; at the same time, when the water level is lower than the safety threshold, the alarm mechanism is automatically triggered; according to the generated dynamic waterway control adjustment strategy, the waterway is adjusted in real time; and the implementation effect of the adjustment strategy is evaluated by real-time monitoring of water level and water quality data.

[0011] Optionally, the process of dynamically assessing the waterway includes the following steps: Obtain the design capacity data of the water tank as the basic parameter for evaluation; collect the current water level data in real time as the dynamic input for evaluation; perform fuzzy processing on the current water level data and convert it into fuzzy language variables; perform fuzzy processing on the design capacity of the water tank and convert it into fuzzy language variables; Based on expert knowledge and historical data, a fuzzy rule base is constructed; the fuzzified water level and capacity are input into the fuzzy reasoning engine, and the fuzzy evaluation results of the waterway capacity are obtained through reasoning through the fuzzy rule base; the fuzzy evaluation results are defuzzified and converted into specific values ​​as the dynamic evaluation results of the waterway capacity; Obtain historical and real-time water level data as the basic input for evaluation; combine the data from water quality sensors to obtain water quality status as an auxiliary input for evaluation; define the nodes of the Bayesian network as variables for evaluation; Construct a conditional probability table to describe the conditional probability relationship between each node; input water level and water quality data into the Bayesian network inference engine, and obtain the stability assessment results of the waterway through probabilistic reasoning; output the stability assessment results of the waterway, and generate corresponding decision support information based on the assessment results.

[0012] Optionally, the process of defining the nodes of a Bayesian network includes the following steps: The node definition of the Bayesian network is: water level is a primary variable, affecting the waterway capacity; water quality is a secondary variable, indirectly affecting the waterway capacity through waterway stability; waterway stability is an intermediate variable, connecting water quality and waterway capacity; waterway capacity is a tertiary variable, and the final output is the result.

[0013] Optionally, in a directed acyclic graph, nodes represent different entities or states, edges represent dependencies or causal relationships between nodes, and each edge has a direction, indicating a unidirectional relationship from one node to another; this edge indicates that node H has a direct impact on node C, and the path indicates that node Q indirectly affects node C through node S.

[0014] Optionally, a conditional probability table is established: waterway stability is based on a conditional probability table of water quality, describing the probability distribution of waterway stability under different water quality conditions; waterway capacity is based on a conditional probability table of water level H and waterway stability, describing the probability distribution of waterway capacity under different water levels and waterway stability conditions.

[0015] The present invention provides a waterway detection and control circuit, comprising: The data acquisition module is responsible for placing various types of sensors on the water tank, realizing multi-dimensional collection of the water level in the water tank through the sensors, fusing and calibrating the collected water level data, and obtaining the pre-processed water level data; The dynamic adjustment module is responsible for obtaining the pre-processed water level data and inputting it into the dynamic waterway control adjustment strategy. It dynamically adjusts the waterway according to the water level change trend and waterway status. At the same time, it compares the water level data with the preset multi-level water level threshold. When the water level data is lower than the safety threshold in the multi-level water level threshold, the alarm mechanism is automatically triggered. The water quality data is obtained from the water level data and compared with the preset safety range. When it exceeds the safety range, the water quality is automatically cleaned. The iterative learning module is responsible for inputting the data after dynamic adjustment of the waterway into the machine learning model, learning the data and dynamic waterway control adjustment strategy, and adjusting the dynamic waterway control adjustment strategy through continuous iterative learning.

[0016] The present invention provides an application of a water channel detection and control method, in which an ultrasonic sensor and a capacitive sensor are arranged in a water tank of a water dispenser; the ultrasonic sensor is used to measure the water level height, and the capacitive sensor is used to detect the liquid level change; the data of the two sensors are fused and calibrated to obtain water level data; the pre-processed water level data is input into a dynamic water channel control adjustment strategy; and the water channel is dynamically adjusted according to the water level change trend and the water channel status.

[0017] The multi-dimensional water level data acquisition and preprocessing of the present invention uses various types of sensors (such as ultrasonic sensors, pressure sensors, capacitive sensors and water quality sensors) to collect water level data from different dimensions and improve the accuracy and reliability of the data; by fusing and calibrating the collected data, the sensor error is eliminated to obtain more accurate preprocessed water level data. The significance achieved: through multi-sensor fusion and data calibration, the water level change can be monitored more accurately, providing a reliable data basis for subsequent dynamic control; the addition of water quality sensors enables drinking water equipment (water dispensers, etc.) to monitor water quality in real time, timely discover and deal with water quality problems, and ensure the stability and safety of water quality. Dynamic waterway control and alarm mechanism, according to the water level change trend and waterway status, the drinking water equipment can dynamically adjust the waterway to ensure the stability of the water level; by comparing the water level data with the preset multi-level water level threshold, the drinking water equipment can automatically trigger the alarm mechanism when the water level is lower than the safety threshold, and promptly remind the operator; when the water quality data exceeds the safety range, the drinking water equipment can automatically clean the water quality to ensure that the water quality is always in a safe state. Significance achieved: Dynamic adjustment of water routes and automatic alarm mechanisms enable drinking water equipment to respond to changes in water level and water quality in real time, ensuring stable operation of the water routes; automatic alarm and water quality cleaning functions effectively reduce human operational errors and improve the safety and reliability of drinking water equipment. The machine learning model optimizes the control strategy, inputs the data after dynamic adjustment of the water route into the machine learning model, and optimizes the dynamic water route control adjustment strategy through continuous iterative learning; the machine learning model can adaptively adjust the control strategy based on historical data and real-time data to improve the intelligence level of drinking water equipment. Significance achieved: Through the application of machine learning models, drinking water equipment can continuously optimize control strategies, realize intelligent management, reduce human intervention, and improve management efficiency; machine learning models enable drinking water equipment to continuously improve, adapt to different water route environments and needs, and improve the adaptability and flexibility of drinking water equipment.

[0018] Other features and advantages of the present invention will be described in the following description, and partly become apparent from the description, or understood by practicing the present invention. The purpose and other advantages of the present invention can be realized and obtained by the structures particularly pointed out in the written description and the accompanying drawings.

[0019] The technical solution of the present invention is further described in detail below through the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings: Figure 1 This is a flow chart of the waterway detection and control method in Example 1 of the present invention; Figure 2 A process diagram of arranging various types of sensors on a water tank in Example 2 of the present invention; Figure 3 This is a process diagram of dynamically adjusting the waterway in Example 3 of the present invention; Figure 4 This is a process diagram of dynamically evaluating a waterway in Embodiment 4 of the present invention; Figure 5 A process diagram of defining nodes of a Bayesian network in Embodiment 5 of the present invention; Figure 6 is a process diagram for obtaining a stability evaluation result of a waterway in Example 6 of the present invention; Figure 7 is a process diagram of outputting the evaluation result of waterway stability in Example 7 of the present invention; Figure 8 A process diagram of adjusting the dynamic waterway control adjustment strategy in Example 8 of the present invention; Fig. 9 This is a block diagram of the water channel detection and control circuit in Example 9 of the present invention. DETAILED DESCRIPTION

[0021] The preferred embodiments of the present invention are described below in conjunction with the accompanying drawings. It should be understood that the preferred embodiments described herein are only used to illustrate and explain the present invention, and are not used to limit the present invention.

[0022] The terms used in the embodiments of the present application are only for the purpose of describing specific embodiments, and are not intended to limit the embodiments of the present application. The singular forms of "a", "said" and "the" used in the embodiments of the present application are also intended to include plural forms, unless the context clearly indicates other meanings. It should also be understood that the term "and / or" used in this article refers to and includes any or all possible combinations of one or more associated listed items.

[0023] When the following description relates to the accompanying drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present application. On the contrary, they are only examples of devices and methods consistent with some aspects of the present application. In the description of the present application, it should be understood that the terms "first", "second", "third", etc. are only used to distinguish similar objects, and do not have to be used to describe a specific order or sequence, nor can they be understood as indicating or implying relative importance. For those of ordinary skill in the art, the specific meanings of the above terms in the present application can be understood according to the specific circumstances.

[0024] Example 1: Figure 1 As shown, an embodiment of the present invention provides a waterway detection and control method, comprising the following steps: S100: Various types of sensors are arranged on the water tank, and multi-dimensional water level data in the water tank is collected through the sensors. The collected water level data is fused and calibrated to obtain pre-processed water level data. Among them, various types of sensors include ultrasonic sensors, pressure sensors, capacitive sensors and water quality sensors; ultrasonic sensors are suitable for water level measurement, pressure sensors are suitable for depth measurement, capacitive sensors are suitable for liquid level detection, and water quality sensors are used to monitor water quality parameters such as pH value and dissolved oxygen; the selection of each sensor should be based on its functional characteristics and applicable scenarios; S200: Obtain the pre-processed water level data, input it into the dynamic waterway control adjustment strategy, and dynamically adjust the waterway according to the water level change trend and the waterway status; at the same time, compare the water level data with the preset multi-level water level threshold, and automatically trigger the alarm mechanism when the water level data is lower than the safety threshold in the multi-level water level threshold; obtain water quality data from the water level data, compare it with the preset safety range, and automatically clean the water quality when it exceeds the safety range; S300: Input the data after the dynamic adjustment of the waterway into the machine learning model, learn the data and the dynamic waterway control adjustment strategy, and adjust the dynamic waterway control adjustment strategy through continuous iterative learning.

[0025] The working principle and beneficial effects of the above technical solution are as follows: in this embodiment, first, various types of sensors are arranged on the water tank, and multi-dimensional collection of the water level in the water tank is realized through the sensors, and the collected water level data is fused and calibrated to obtain pre-processed water level data; wherein, the various types of sensors include ultrasonic sensors, pressure sensors, capacitive sensors and water quality sensors, etc.; secondly, the pre-processed water level data is obtained and input into the dynamic waterway control adjustment strategy, and the waterway is dynamically adjusted according to the water level change trend and the waterway status; at the same time, the water level data is compared with the preset multi-level water level threshold, and when the water level data is lower than the safety threshold in the multi-level water level threshold, the alarm mechanism is automatically triggered; water quality data is obtained from the water level data, and compared with the preset safety range, and when it exceeds the safety range, the water quality is automatically cleaned; finally, the data after the dynamic adjustment of the waterway is input into the machine learning model, the data and the dynamic waterway control adjustment strategy are learned, and the dynamic waterway control adjustment strategy is adjusted through continuous iterative learning. In the above scheme, step S100 multi-dimensional water level data collection and preprocessing uses various types of sensors (such as ultrasonic sensors, pressure sensors, capacitive sensors and water quality sensors) to collect water level data from different dimensions and improve the accuracy and reliability of the data; by fusing and calibrating the collected data, the sensor error is eliminated to obtain more accurate preprocessed water level data. The significance achieved: through multi-sensor fusion and data calibration, the water level change can be monitored more accurately, providing a reliable data basis for subsequent dynamic control; the addition of water quality sensors enables drinking water equipment (water dispensers, etc.) to monitor water quality in real time, timely discover and deal with water quality problems, and ensure the stability and safety of water quality. Step S200 dynamic waterway control and alarm mechanism, according to the water level change trend and waterway status, the drinking water equipment can dynamically adjust the waterway to ensure the stability of the water level; by comparing the water level data with the preset multi-level water level threshold, the drinking water equipment can automatically trigger the alarm mechanism when the water level is lower than the safety threshold, and promptly remind the operator; when the water quality data exceeds the safety range, the drinking water equipment can automatically clean the water quality to ensure that the water quality is always in a safe state. Significance achieved: Dynamic adjustment of waterways and automatic alarm mechanisms enable drinking water equipment to respond to changes in water level and water quality in real time, ensuring stable operation of the waterways; automatic alarm and water quality cleaning functions effectively reduce human operation errors and improve the safety and reliability of drinking water equipment. Step S300 The machine learning model optimizes the control strategy, inputs the data after dynamic adjustment of the waterway into the machine learning model, and optimizes the dynamic waterway control adjustment strategy through continuous iterative learning; the machine learning model can adaptively adjust the control strategy based on historical data and real-time data to improve the intelligence level of drinking water equipment.Significance achieved: Through the application of machine learning models, drinking water equipment can continuously optimize control strategies, realize intelligent management, reduce human intervention, and improve management efficiency; machine learning models enable drinking water equipment to be continuously improved, adapt to different waterway environments and needs, and improve the adaptability and flexibility of drinking water equipment.

[0026] In summary, this embodiment realizes all-round monitoring and control of water level and water quality through multi-sensor fusion, dynamic control, automatic alarm and machine learning optimization. It can improve monitoring accuracy, respond to water level changes in real time and ensure water quality safety, and continuously improve the performance of drinking water equipment through intelligent management. It not only improves the safety and reliability of waterway drinking water equipment, but also provides strong support for intelligent water resources management.

[0027] Example 2: Figure 2 As shown, based on Example 1, the process of arranging multiple types of sensors on the water tank provided by the embodiment of the present invention includes the following steps: S101: obtaining geometric characteristics of the water tank, including the height, diameter, wall thickness, and inlet and outlet positions of the water tank, and preliminarily determining the number and distribution of monitoring points based on the height and diameter of the water tank, dividing the water tank into three areas: upper, middle, and lower, and setting a number of monitoring points in each area; S102: Analyze the inlet and outlet positions of the water tank to determine the path and retention area of ​​the water flow; set multiple monitoring points according to the geometric characteristics of the water tank and the water flow path, and each monitoring point should include at least one type of sensor; A monitoring point is set at the bottom of the water tank to monitor the water level and water quality; the pressure sensor is used to measure the water pressure at the bottom of the water tank, which indirectly reflects the water level; the water quality sensor is used to monitor the water quality parameters at the bottom, such as pH value, dissolved oxygen and conductivity; A monitoring point is set in the middle of the water tank to monitor the dynamic changes of the water level; the ultrasonic sensor is used to measure the water level height, which has high accuracy and response speed; the temperature sensor is used to monitor the water temperature and help analyze the water flow path and water quality changes; A monitoring point is set at the top of the water tank, mainly used to monitor the upper limit of the water level and water quality; the capacitive sensor is used to measure the water level height, which is suitable for high-precision liquid level detection; the gas sensor is used to monitor the gas composition at the top of the water tank, such as oxygen and carbon dioxide, to help analyze water quality changes; S103: Input the pre-processed water level data into the dynamic waterway control adjustment strategy, and dynamically adjust the position and parameters of the sensor according to the water level change trend and the waterway status.

[0028] The working principle and beneficial effects of the above technical solution are as follows: the present embodiment first obtains the geometric characteristics of the water tank, including the height, diameter, wall thickness and inlet and outlet positions of the water tank, and preliminarily determines the number and distribution of monitoring points according to the height and diameter of the water tank, and divides the water tank into three areas: upper, middle and lower, and sets a number of monitoring points in each area; secondly, analyzes the inlet and outlet positions of the water tank to determine the path and retention area of ​​the water flow; according to the geometric characteristics of the water tank and the water flow path, sets a number of monitoring points, and each monitoring point should contain at least one type of sensor; finally, inputs the pre-processed water level data into the dynamic waterway control adjustment strategy, and dynamically adjusts the position and parameters of the sensor according to the water level change trend and the waterway state. Step S101 of the above solution obtains the geometric characteristics of the water tank, and by obtaining the geometric characteristics of the water tank, such as height, diameter, wall thickness and inlet and outlet positions, the number and distribution of monitoring points can be accurately determined; the water tank is divided into three areas: upper, middle and lower, and a number of monitoring points are set in each area to ensure that all key parts of the water tank are fully covered. Significance achieved: Ensure that all areas of the water tank can be effectively monitored, improve the comprehensiveness and accuracy of monitoring; optimize the layout of sensors through reasonable zoning settings, improve monitoring efficiency and data reliability. Step S102 analyzes the inlet and outlet positions of the water tank, and determines the path and retention area of ​​the water flow by analyzing the inlet and outlet positions, so as to reasonably set the monitoring points; configure various types of sensors at each monitoring point, such as pressure sensors, water quality sensors, ultrasonic sensors, temperature sensors, capacitive sensors and gas sensors, to ensure the collection of multi-dimensional data. Significance achieved: Through multi-sensor configuration, dynamic monitoring of water level, water quality, temperature and gas composition in the water tank is realized, and the real-time and accuracy of monitoring are improved; multi-sensor configuration can provide rich data to help comprehensively analyze the operating status of the water tank and water quality changes. Step S103 dynamic waterway control adjustment strategy, pre-process the collected water level data to improve the accuracy and reliability of the data; dynamically adjust the position and parameters of the sensor according to the water level change trend and waterway status to ensure the flexibility and adaptability of the monitoring system. Significance achieved: Through dynamic adjustment strategies, it is possible to respond to changes in the water tank in real time and improve the response speed and accuracy of the monitoring system; dynamic adjustment strategies help optimize the operating status of the water tank and improve the operating efficiency and safety of the water tank.

[0029] In this embodiment, S101 obtains the geometric characteristics of the water tank: The calculation formula for the number of monitoring points is: In the formula, Indicates the number of monitoring points, Indicates the water tank height, Indicates the diameter of the water tank, Indicates the distance between monitoring points; Calculation formula for the number of monitoring points in each zone: In the formula, Indicates the number of monitoring points in each area; S102 indicates the inlet and outlet positions of the analytical water tank: Water flow path analysis formula: In the formula, represents the length of the water flow path, , , represents the coordinates of the water inlet, , , Indicates the coordinates of the water outlet; The calculation formula of retention area is: In the formula represents the volume of the retention area, Indicates the height of the detention area; S102 indicates sensor settings: Pressure sensor water pressure calculation formula: In the formula, Indicates water pressure, represents the density of water, represents the acceleration due to gravity, Indicates water level height; Ultrasonic sensor water level calculation formula: In the formula, Indicates the water level measured by the ultrasonic sensor. represents the speed of sound, represents the ultrasonic wave propagation time; Capacitive sensor water level calculation formula: In the formula, Indicates the capacitance value, is the dielectric constant of vacuum, represents the relative dielectric constant, represents the electrode area, Indicates the electrode spacing; Gas sensor concentration calculation formula: In the formula, Indicates the gas concentration, represents the number of moles of gas, Indicates the volume of the water tank; S103 Dynamic Waterway Control Adjustment Strategy Water level change trend prediction formula: In the formula, represents the predicted water level change, represents the observed water level change, represents the last predicted water level change, represents the smoothing coefficient; Sensor position adjustment: In the formula, represents the adjusted sensor position, Indicates the original sensor position, Indicates the position adjustment amount.

[0030] In summary, this embodiment achieves comprehensive, dynamic and multi-dimensional monitoring of the water tank through accurate geometric characteristic acquisition, reasonable water flow path analysis and dynamic waterway control adjustment strategy. It not only improves the accuracy and real-time performance of monitoring, but also optimizes the operating state of the water tank, ensuring the safe and efficient operation of the water tank.

[0031] Example 3: Figure 3 As shown, based on Example 1, the process of dynamically adjusting the waterway provided in the embodiment of the present invention includes the following steps: S201: Perform time series analysis on water level data to extract target features such as water level change trend and periodic fluctuation; use the target feature time series prediction model to predict the future water level change trend and generate a prediction curve; S202: Based on the design capacity of the water tank and the current water level, dynamically evaluate the waterway capacity through fuzzy reasoning; dynamically evaluate the waterway through probabilistic reasoning based on the stability evaluation method; and evaluate whether the water quality meets the safety standards and whether water cleaning is required in combination with the data from the water quality sensor; S203: According to the water level change trend and waterway status, set multi-level water level thresholds, including safety thresholds and warning thresholds; use adaptive threshold setting combined with historical water level data and real-time water level data to dynamically adjust the threshold; generate the corresponding dynamic waterway control adjustment strategy based on the comparison result between the current water level and the threshold; at the same time, when the water level is lower than the safety threshold, automatically trigger the alarm mechanism; according to the generated dynamic waterway control adjustment strategy, make real-time adjustments to the waterway; and evaluate the implementation effect of the adjustment strategy by real-time monitoring of water level and water quality data.

[0032] The working principle and beneficial effects of the above technical solution are as follows: In this embodiment, first, a time series analysis is performed on the water level data to extract target features such as the change trend and periodic fluctuation of the water level; the target feature time series prediction model is used to predict the future water level change trend and generate a prediction curve; secondly, according to the design capacity of the water tank and the current water level, the waterway capacity is dynamically evaluated through fuzzy reasoning; based on the stability evaluation method, the waterway is dynamically evaluated through probabilistic reasoning; combined with the data of the water quality sensor, it is evaluated whether the water quality meets the safety standards and whether water cleaning is required; finally, according to the water level change trend and the waterway status, a multi-level water level threshold is set, including a safety threshold and a warning threshold, etc.; an adaptive threshold setting is used to combine historical water level data and real-time water level data to dynamically adjust the threshold; according to the comparison result of the current water level and the threshold, a corresponding dynamic waterway control adjustment strategy is generated; at the same time, when the water level is lower than the safety threshold, an alarm mechanism is automatically triggered; according to the generated dynamic waterway control adjustment strategy, the waterway is adjusted in real time; and the execution effect of the adjustment strategy is evaluated by real-time monitoring of water level and water quality data. Step S201 of the above scheme is data analysis and prediction. Through time series analysis, target features such as water level change trends and periodic fluctuations are extracted to provide basic data for prediction and adjustment. Time series prediction models such as ARIMA and LSTM are used to predict future water level change trends, generate prediction curves, and improve the accuracy and reliability of predictions. Significance achieved: Through accurate predictions, water level change trends can be warned in advance, providing sufficient time and data support for adjustment strategies; based on the prediction results, water resource allocation can be optimized to avoid waste and shortage of resources and improve the operation efficiency of waterways. Step S202 is waterway status assessment. Through fuzzy reasoning, waterway capacity is dynamically evaluated to improve the flexibility and accuracy of the assessment and adapt to complex waterway conditions; based on the stability assessment method of the Bayesian network, the stability of the waterway is dynamically evaluated through probabilistic reasoning to improve the scientificity and reliability of the assessment; combined with the data of the water quality sensor, the water quality status is evaluated in real time to ensure that the water quality meets safety standards and avoid the impact of water quality problems on the waterway. Significance achieved: By dynamically evaluating the capacity and stability of waterways, the safe operation of waterways can be ensured to avoid risks such as overflow and drying up; water quality can be monitored in real time, water can be cleaned in a timely manner to ensure water safety and improve the overall health of waterways.Step S203 generates and executes dynamic adjustment strategies, adopts adaptive threshold setting method, combines historical water level data and real-time water level data, dynamically adjusts thresholds, and improves the adaptive ability of the system; generates corresponding dynamic waterway control adjustment strategies according to the comparison results between the current water level and the threshold, such as increasing or decreasing the water inflow, starting the drainage system, etc., to improve the accuracy and efficiency of the adjustment; when the water level is lower than the safety threshold, automatically triggers the alarm mechanism, and promptly notifies relevant personnel to handle it, thereby improving the emergency response capability of the system; through real-time monitoring of water level and water quality data, evaluates the execution effect of the adjustment strategy, and ensures the effectiveness and timeliness of the adjustment strategy. Significance achieved: Through dynamic adjustment strategies, waterways can be adjusted in real time to ensure that the water level is within a safe range, thereby improving the stability and reliability of the system; automatically triggers the alarm mechanism to ensure timely response in emergency situations, reduce losses, and improve the safety and reliability of the system; through real-time monitoring and feedback, continuously optimizes the adjustment strategy, improves the adaptive ability and operating efficiency of the system, and ensures the long-term stable operation of the waterway.

[0033] In summary, this embodiment uses data analysis and prediction to provide early warning of water level changes, optimize resource allocation, and improve operational efficiency; through waterway status assessment, ensure the safe operation and water quality safety of the waterway of the drinking water equipment, and improve the overall health level of the system; through dynamic adjustment strategy generation and execution, adjust the waterway in real time to ensure that the water level is within a safe range and improve the stability and reliability of the system; through real-time monitoring and feedback, continuously optimize the adjustment strategy, improve the system's adaptability and operational efficiency, and ensure the long-term stable operation of the waterway.

[0034] Example 4: Figure 4 As shown, based on Example 3, the process of dynamically evaluating a waterway provided by the embodiment of the present invention includes the following steps: S2021: Obtain the design capacity data of the water tank as the basic parameter for evaluation; collect the current water level data in real time as the dynamic input for evaluation; perform fuzzy processing on the current water level data and convert it into fuzzy language variables, such as "low water level", "medium water level", "high water level", etc.; perform fuzzy processing on the design capacity of the water tank and convert it into fuzzy language variables, such as "small capacity", "medium capacity", "large capacity", etc.; S2022: Based on expert knowledge and historical data, a fuzzy rule base is constructed, such as "if the water level is low and the capacity is small, the waterway capacity is low", etc.; the fuzzified water level and capacity are input into the fuzzy reasoning engine, and the fuzzy evaluation result of the waterway capacity is obtained through reasoning through the fuzzy rule base; the fuzzy evaluation result is defuzzified and converted into a specific numerical value as the dynamic evaluation result of the waterway capacity; S2023: Obtain historical and real-time water level data as the basic input for evaluation; combine the data from water quality sensors to obtain water quality status as an auxiliary input for evaluation; define the nodes of the Bayesian network, such as water level, water quality, and waterway capacity, as variables for evaluation; Among them, water level and waterway capacity have a direct causal relationship: water level H directly affects waterway capacity C, Description: The higher the water level, the greater the waterway capacity; the lower the water level, the smaller the waterway capacity; Mathematical expression: C=f(H), where f is a function that describes the relationship between water level and waterway capacity; example: when the water level is "high water level", the waterway capacity is "high capacity"; when the water level is "low water level", the waterway capacity is "low capacity"; Water quality and waterway capacity, indirect causal relationship: water quality Q indirectly affects waterway capacity C by affecting waterway stability S; description: the better the water quality, the higher the waterway stability and the more stable the waterway capacity; the worse the water quality, the lower the waterway stability and the more unstable the waterway capacity; Mathematical expression: S=g(Q), where g is a function that describes the relationship between water quality and waterway stability; then C=h(S), where h is a function that describes the relationship between waterway stability and waterway capacity; example: when the water quality is "high quality", the waterway stability is "high stability" and the waterway capacity is "high capacity"; when the water quality is "poor quality", the waterway stability is "low stability" and the waterway capacity is "low capacity"; S2024: Construct a conditional probability table to describe the conditional probability relationship between each node; input water level and water quality data into the Bayesian network inference engine, and obtain the stability assessment result of the waterway through probabilistic reasoning; output the stability assessment result of the waterway, such as "stable", "unstable", etc., and generate corresponding decision support information based on the assessment results, such as "need to increase water intake", "need to clean water quality", etc.

[0035] The working principle and beneficial effects of the above technical solution are as follows: the present embodiment first obtains the design capacity data of the water tank as the basic parameter for evaluation; collects the current water level data in real time as the dynamic input for evaluation; fuzzifies the current water level data and converts it into fuzzy language variables, such as "low water level", "medium water level", "high water level", etc.; fuzzifies the design capacity of the water tank and converts it into fuzzy language variables, such as "small capacity", "medium capacity", "large capacity", etc.; secondly, based on expert knowledge and historical data, a fuzzy rule base is constructed, such as "if the water level is low and the capacity is small, the waterway capacity is low", etc.; the fuzzified water level and capacity are input into the fuzzy inference engine, and the fuzzy evaluation result of the waterway capacity is obtained through reasoning through the fuzzy rule base; the fuzzy evaluation result is converted into the fuzzy rule base. The estimation results are defuzzified and converted into specific numerical values ​​as the dynamic evaluation results of waterway capacity; then historical and real-time water level data are obtained as the basic input for the evaluation; the water quality status is obtained in combination with the data of the water quality sensor as the auxiliary input for the evaluation; the nodes of the Bayesian network, such as water level, water quality and waterway capacity, are defined as evaluation variables; finally, a conditional probability table is constructed to describe the conditional probability relationship between each node; the water level and water quality data are input into the Bayesian network inference engine, and the stability evaluation results of the waterway are obtained through probabilistic reasoning; the stability evaluation results of the waterway, such as "stable", "unstable", etc., are output, and based on the evaluation results, corresponding decision support information is generated, such as "need to increase the water inlet", "need to clean the water quality", etc. Step S2021 of the above scheme, data acquisition and fuzzification processing, provides basic data support for the evaluation by acquiring the design capacity and real-time water level data of the water tank; converts the specific water level data into fuzzy language variables (such as "low water level", "medium water level", "high water level"), making the evaluation process more flexible and adaptable, and able to handle uncertainty; converts the design capacity of the water tank into fuzzy language variables (such as "small capacity", "medium capacity", "large capacity"), providing a basis for fuzzy reasoning. Significance: Through fuzzification processing, it is possible to better cope with the uncertainty of water level changes and improve the accuracy and robustness of the evaluation; it provides the necessary input for fuzzy reasoning to ensure the continuity and consistency of the evaluation process. Step S2022, fuzzy reasoning and defuzzification, builds a fuzzy rule base based on expert knowledge and historical data, making the evaluation process more intelligent and scientific; through the fuzzy reasoning engine, the fuzzy rule base is used for reasoning to obtain the fuzzy evaluation results of the waterway capacity; the fuzzy evaluation results are converted into specific numerical values, making the evaluation results more intuitive and easy to understand. Significance: Through fuzzy reasoning, multiple factors can be taken into consideration to improve the comprehensiveness and accuracy of the evaluation; defuzzification processing makes the evaluation results more specific and practical, which is convenient for subsequent decision support.Step S2023: Data acquisition and Bayesian network construction, obtain historical and real-time water level data, combine with water quality sensor data, provide comprehensive data support for evaluation; define the nodes of the Bayesian network (such as water level, water quality and waterway capacity, etc.), and provide a basis for probabilistic reasoning. Significance: Through the Bayesian network, the relationship between multiple variables can be comprehensively considered to improve the complexity and comprehensiveness of the evaluation; and provide the necessary structure and data support for probabilistic reasoning. Step S2024: Probabilistic reasoning and decision support, describe the conditional probability relationship between each node, and provide a basis for probabilistic reasoning; through the Bayesian network reasoning engine, use the conditional probability table to perform probabilistic reasoning to obtain the stability evaluation results of the waterway; based on the evaluation results, generate corresponding decision support information, such as "need to increase water intake" and "need to clean water quality". Significance: Through probabilistic reasoning, the stability of the waterway can be evaluated more scientifically, and the accuracy and reliability of the evaluation can be improved; the generated decision support information provides guidance for actual operations and improves the efficiency and effectiveness of waterway management.

[0036] In summary, the waterway dynamic evaluation process of this embodiment not only realizes the comprehensive acquisition and processing of data, but also improves the accuracy and scientificity of the evaluation through fuzzy reasoning and probabilistic reasoning. The decision support information finally generated provides powerful guidance for practical operations and improves the efficiency and effectiveness of waterway management. Through the above hierarchical fuzzy reasoning and probabilistic reasoning process, the dynamic evaluation of waterway capacity and status is realized; the specific steps include input data preparation, fuzzification processing, fuzzy rule base construction, fuzzy reasoning engine, defuzzification processing, stability evaluation model construction, probabilistic reasoning process and dynamic evaluation result output. The steps are interconnected to form a complete dynamic evaluation process to ensure the safe operation and efficient management of waterways.

[0037] Example 5: Figure 5 As shown, based on Example 4, the process of defining a node of a Bayesian network provided in an embodiment of the present invention includes the following steps: S20231: Node definition of Bayesian network, water level primary variable, affecting waterway capacity, water quality secondary variable, indirectly affecting waterway capacity through waterway stability, waterway stability intermediate variable, connecting water quality and waterway capacity, waterway capacity tertiary variable, and final output result; S20232: In a directed acyclic graph, nodes represent different entities or states, and edges represent dependencies or causal relationships between nodes. Each edge has a direction, indicating a unidirectional relationship from one node to another. This edge indicates that node H has a direct impact on node C, and the path indicates that node Q indirectly affects node C through node S. S20233: Establish a conditional probability table. The conditional probability table of waterway stability is based on water quality, which describes the probability distribution of waterway stability under different water quality conditions. The conditional probability table of waterway capacity is based on water level H and waterway stability, which describes the probability distribution of waterway capacity under different water levels and waterway stability conditions.

[0038] The working principle and beneficial effects of the above technical solution are as follows: In this embodiment, the nodes of the Bayesian network are first defined, the water level is a primary variable, which affects the waterway capacity, the water quality is a secondary variable, which indirectly affects the waterway capacity through the waterway stability, the waterway stability is an intermediate variable, which connects the water quality and the waterway capacity, and the waterway capacity is a tertiary variable, and the final output is the result; secondly, in a directed acyclic graph, the nodes represent different entities or states, the edges represent the dependency or causal relationship between the nodes, and each edge has a direction, representing a unidirectional relationship from one node to another; this edge indicates that node H has a direct impact on node C, and the path indicates that node Q indirectly affects node C through node S; finally, a conditional probability table is established, the conditional probability table of waterway stability based on water quality, which describes the probability distribution of waterway stability under different water quality conditions; the conditional probability table of waterway capacity based on water level H and waterway stability, which describes the probability distribution of waterway capacity under different water level and waterway stability conditions. Step S20231 of the above scheme defines the nodes of the Bayesian network. By defining the nodes and their hierarchical relationships, the direct and indirect influence relationships between the variables are clarified; the complex system is decomposed into multiple levels of nodes for easy understanding and analysis. Significance: Help users understand the role and relationship of each variable in the system as a whole; decompose complex problems into multiple sub-problems for easy gradual solution and optimization. Step S20232 In a directed acyclic graph, nodes represent different entities or states, and edges represent the dependency or causal relationship between nodes. The relationship between nodes and edges is graphically represented by DAG, which intuitively displays the dependency and causal relationship between variables; the direction of each edge is clarified to ensure that the dependency is unidirectional and avoid circular dependency. Significance: Through graphical display, users can intuitively understand the dependency between variables; ensure that the system logic is clear, which is convenient for the subsequent establishment and reasoning of conditional probability tables. Step S20233 Establish a conditional probability table, through which the dependency between nodes is quantified and a specific probability distribution is provided; provide basic data for the reasoning of the Bayesian network, making probability-based reasoning possible. Significance: The conditional probability table can more accurately predict the state and changes of each variable. It provides a probability-based basis for decision-making, helping users make more reasonable decisions in uncertainty.

[0039] In summary, this embodiment clarifies the hierarchy and relationship of each variable by defining nodes, laying the foundation for subsequent graphical representation and probabilistic modeling; graphically represents the relationship between nodes and edges through DAG, intuitively displays the dependency and causal relationship between variables, and ensures clear logic of waterway control; quantifies the dependency between nodes by establishing a conditional probability table, provides basic data for Bayesian network reasoning, and makes accurate prediction and decision support possible. Through these steps, complex waterway control is decomposed into multiple levels of nodes and dependencies, which facilitates the analysis and understanding of waterway control; by quantifying the dependency between nodes, a probability-based reasoning basis is provided, making it possible to make accurate predictions and reasonable decisions in uncertainty; and provides a probability-based basis for decision-making, helping users make better decisions in complex and uncertain environments.

[0040] Example 6: Figure 6 As shown, based on Example 4, the process of obtaining the stability evaluation result of the waterway provided in the embodiment of the present invention includes the following steps: S20241: normalize the water level and water quality data to the interval [0, 1], and input the normalized real-time water level and water quality data into the Bayesian network inference engine; S20242: The Bayesian network inference engine calculates the posterior probability of waterway stability based on the input data and the existing conditional probability table; and searches for the corresponding probability value in the conditional probability table based on the input water level and water quality data; S20243: Use Bayes' theorem to calculate the posterior probability of waterway stability, and output the evaluation result of waterway stability based on the calculated posterior probability; if the waterway stability is evaluated as unstable, take measures.

[0041] The working principle and beneficial effects of the above technical solution are as follows: This embodiment first normalizes the water level and water quality data to the interval [0, 1], and inputs the normalized real-time water level and water quality data into the Bayesian network inference engine; secondly, the Bayesian network inference engine calculates the posterior probability of waterway stability based on the input data and the existing conditional probability table; according to the input water level and water quality data, the corresponding probability value in the conditional probability table is searched; finally, the posterior probability of waterway stability is calculated using Bayesian theorem, and the evaluation result of waterway stability is output according to the calculated posterior probability; if the waterway stability is evaluated as unstable, measures are taken. Step S20241 of the above solution is data normalization and input, which normalizes the water level and water quality data to the interval [0, 1] to ensure that data of different dimensions are compared on the same scale to avoid reasoning errors caused by differences in data magnitude; the normalized data is input into the Bayesian network inference engine to provide an accurate data basis for probability calculation. Significance: Normalization can eliminate the impact of data magnitude differences on the inference results and improve the accuracy of Bayesian network inference; ensure that the format and range of the input data are consistent, so that the Bayesian network model can perform unified inference calculations. Step S20242: Conditional probability query and posterior probability calculation. According to the input water level and water quality data, the corresponding probability value in the conditional probability table (CPT) is searched to obtain the prior probability of waterway stability under specific conditions; the posterior probability of waterway stability is calculated using Bayesian theorem, and a more accurate evaluation result is obtained by combining the prior probability and observation data. Significance: Through real-time query and calculation, the waterway stability can be dynamically evaluated, and the stability state under the current water level and water quality conditions can be timely reflected; the probability-based evaluation results are provided to provide a scientific basis for decision-making and enhance the reliability and effectiveness of decision-making. Step S20243: Output the evaluation results and take measures. According to the calculated posterior probability, the evaluation results of the waterway stability are output to clarify the current state of the waterway stability; according to the evaluation results, corresponding measures are taken, such as increasing the water intake or cleaning the water quality, to maintain or improve the waterway stability. Significance: By real-time assessment and taking measures, we can respond to changes in waterway stability in a timely manner and avoid potential risks and losses; measures based on the assessment results will help optimize waterway management and improve the efficiency and safety of water resource utilization.

[0042] In summary, this embodiment realizes the real-time, dynamic and scientific evaluation of waterway stability, and provides strong decision-making support for waterway management. The specific significance includes: the evaluation results based on probabilistic reasoning make the decision more scientific and reasonable; through real-time monitoring and dynamic evaluation, the reliability and stability of the waterway management system are enhanced; measures based on the evaluation results help to optimize the utilization of water resources and improve the efficiency and benefits of waterway management. The Bayesian network reasoning engine can provide accurate and reliable evaluation results in a complex waterway management environment, providing strong support for the maintenance and optimization of waterway stability.

[0043] Example 7: Figure 7 As shown, based on Example 6, the process of outputting the evaluation result of waterway stability provided by the embodiment of the present invention includes the following steps: S202431: Calculate the probability of observed data under given waterway stability conditions, reflecting the possibility of observed data under different stability conditions; S202432: Calculate the marginal probability of the observed data, which is the joint probability of the observed data under all possible waterway stability states; include all states of waterway stability and calculate the joint probability of the observed data; first calculate the probability of the observed data under stable conditions and multiply it by the prior probability of stability; then calculate the probability of the observed data under unstable conditions and multiply it by the prior probability of instability; finally, add these two results to get the marginal probability of the observed data; S202433: According to Bayes' theorem, the prior probability, likelihood function and marginal probability are combined to calculate the posterior probability of waterway stability, which reflects the update of waterway stability after observing the current data; Among them, the likelihood function is multiplied by the prior probability to obtain an intermediate result; the intermediate result is divided by the marginal probability to obtain the posterior probability.

[0044] The working principle and beneficial effects of the above technical solution are as follows: This embodiment first calculates the probability of the observed data under given waterway stability conditions, reflecting the possibility of the observed data under different stability states; secondly, the marginal probability of the observed data is calculated, which is the joint probability of the observed data under all possible waterway stability states; it includes all states of waterway stability and calculates the joint probability of the observed data; first calculates the probability of the observed data under stable conditions, multiplied by the prior probability of stability; then calculates the probability of the observed data under unstable conditions, multiplied by the prior probability of instability; finally, the two results are added together to obtain the marginal probability of the observed data; finally, according to Bayes' theorem, the prior probability, likelihood function and marginal probability are combined to calculate the posterior probability of waterway stability, reflecting the update of waterway stability after observing the current data. Step S202431 of the above solution calculates the probability of the observed data under given waterway stability conditions, generates a likelihood function by calculating the probability of the observed data under known waterway stability conditions; the likelihood function reflects the possibility of the observed data under different stability states; the conditional probability table is used to query the probability of the observed data under different waterway stability states to provide basic data for calculation. Significance: By calculating the likelihood function, the impact of the observed data on the waterway stability is quantified, providing a basis for subsequent Bayesian updates; the accurate calculation of the likelihood function helps to improve the accuracy and reliability of the Bayesian network model. Step S202432 calculates the marginal probability of the observed data, that is, the joint probability of the observed data under all possible waterway stability states. This step involves weighted summation of the probabilities of the observed data under different stability states; by first calculating the probability of the observed data under stable conditions, multiplying it by the stable prior probability; then calculating the probability of the observed data under unstable conditions, multiplying it by the unstable prior probability; finally, adding these two results to obtain the marginal probability of the observed data. Significance: The calculation of the marginal probability comprehensively considers all possible waterway stability states, ensuring that the impact of the observed data can be fully reflected when calculating the posterior probability; by calculating the marginal probability, the robustness of the Bayesian network model is improved, ensuring that the model can operate stably under different conditions. Step S202433 combines the prior probability, likelihood function and marginal probability according to Bayesian theorem to calculate the posterior probability of waterway stability. By combining the prior probability, likelihood function and marginal probability according to Bayesian theorem, the posterior probability of waterway stability is calculated; the belief in waterway stability is updated; the posterior probability is calculated by substituting the known prior probability, likelihood function and marginal probability into the Bayesian theorem formula. Significance: The posterior probability reflects the updated belief in waterway stability after observing the current data; the updating process enables the model to dynamically adapt to new observation data and improve the accuracy of prediction; by calculating the posterior probability, a scientific basis is provided for waterway management, supporting decision makers to make more reasonable decisions under different conditions.

[0045] In summary, this embodiment can systematically evaluate the stability of the waterway, and dynamically update the stability of the waterway according to the observed data; calculate the likelihood function, quantify the impact of the observed data, and improve the accuracy of the model; calculate the marginal probability, comprehensively consider all possibilities, and improve the robustness of the model; apply the Bayesian theorem, update the posterior probability, support decision-making, and improve the prediction accuracy. Together, they constitute a complete Bayesian network reasoning process to ensure the scientificity and reliability of the waterway stability assessment.

[0046] Example 8: Figure 8 As shown, based on Example 7, the process of adjusting the dynamic waterway control adjustment strategy provided by the embodiment of the present invention includes the following steps: S301: preprocessing the input dynamically adjusted waterway data, including data cleaning, normalization and feature extraction; extracting key water level change trends, water quality parameter changes and sensor response time from the original water level data through feature extraction, and using the features as inputs to the machine learning model; S302: Select a deep neural network model for initialization, and input the preprocessed features into the deep neural network model for training; during the training process, the deep neural network model optimizes the parameters of the deep neural network model by learning the association between water level changes in historical data and control strategies; S303: Decomposing the dynamic waterway control adjustment strategy into three levels: bottom-level strategy learning, middle-level strategy learning, and high-level strategy learning; in the strategy learning process of each level, evaluating the actual waterway control effect and generating a feedback signal; optimizing the dynamic waterway control adjustment strategy through continuous iterative learning; Among them, the bottom-level strategy learning: mainly focuses on the real-time response to the water level change trend. Through training models, it can quickly adjust the water channel control parameters according to the changes in the current water level data, such as the start and stop of the water pump, the opening and closing of the valve, etc.; middle-level strategy learning: on the basis of the bottom-level strategy, consider the changes in water quality parameters; through training models, it can dynamically adjust the operating status of water quality cleaning equipment according to the real-time changes in water quality data to ensure that the water quality is always within a safe range; high-level strategy learning: on the basis of the middle-level strategy, comprehensively consider the water level change trend and the long-term change trend of water quality parameters; through training models, it can predict future water level and water quality changes, and adjust the water channel control strategy in advance to deal with potential risks and challenges.

[0047] Among them, in step S303, the optimization of the dynamic waterway control adjustment strategy is a multi-level and multi-dimensional complex process, involving the comprehensive role of bottom-level, middle-level and high-level strategy learning; In the formula, Indicates time t The comprehensive optimization value of the dynamic waterway control adjustment strategy reflects the effectiveness of the overall strategy; k Represents the hierarchical index of policy learning ( k =1) is the underlying strategy, ( k =2) is the middle-level strategy, ( k =3) is a high-level strategy; Indicates k The importance weight coefficient of layer strategy learning reflects the contribution weight of this layer in the overall strategy; Indicates time t Time k The basic effectiveness value of the layer strategy reflects the execution effect of the layer strategy at the current moment; Indicates k The water level change trend influencing factor of the layer strategy reflects the sensitivity of the layer strategy to water level changes; Indicates time t Time k The intensity of the water level change trend of the layer strategy reflects the dynamic change characteristics of the current water level; Indicates k The water quality parameter influencing factors of the layer strategy reflect the sensitivity of the layer strategy to water quality changes; Indicates time t Time k The water quality parameter deviation value of the layer strategy reflects the degree of deviation between the current water quality and the preset safety range; Indicates time t The comprehensive feedback adjustment coefficient reflects the dynamic adjustment ability of the overall strategy; Indicates time t The sensor response time influencing factor reflects the real-time impact of the sensor on the strategy adjustment; Indicates time t The sensor response time deviation value reflects the dynamic change of the sensor response time; Indicates time t The influencing factors of external environment changes reflect the impact of the external environment (such as temperature, pressure, etc.) on strategy adjustment; Indicates time t The intensity of changes in the external environment reflects the fluctuation characteristics of the external environment at the current moment.

[0048] Hierarchical strategy learning part: It reflects the combined effect of bottom-level, middle-level, and high-level strategy learning. The basic performance value of each level The trend of water level change and water quality parameter deviations Interact with each other, combining their respective weight coefficients , and , get the optimization contribution value of this level; Low-level policy representation and Mainly based on short-term response to real-time water level changes and water quality parameters; mid-level strategy representation and On the basis of the bottom layer, the medium-term trend of water quality parameters is further considered; the high-level strategy indicates and Comprehensively consider the long-term trends of water levels and water quality and make predictive adjustments; Comprehensive feedback adjustment part: The sensor response time and the feedback effect of external environment changes are comprehensively considered; and Reflect the real-time impact of sensor response time on strategy adjustment; and Reflect the continuous impact of changes in the external environment on strategy adjustments; Dynamically adjust the intensity of comprehensive feedback according to the current status to ensure flexibility and adaptability of strategy adjustments.

[0049] Hierarchical means realize unified modeling of bottom-level, middle-level and high-level strategy learning, reflecting the multi-level characteristics of strategy optimization; multidimensional means include multiple dimensions such as water level changes, water quality parameters, sensor response time and external environment changes, ensuring the comprehensiveness and accuracy of the formula; dynamic means realize dynamic feedback adjustment of strategy optimization, ensuring the real-time and adaptability of the strategy; bottom-level, middle-level and high-level strategy learning are closely integrated, which not only reflects the independence of strategies at each level, but also realizes the dynamic optimization of the overall strategy through the comprehensive feedback adjustment part, ensuring that the waterway control adjustment strategy always remains efficient and accurate in complex and changing scenarios.

[0050] The working principle and beneficial effects of the above technical solution are as follows: This embodiment first preprocesses the input dynamic waterway adjustment data, including data cleaning, normalization and feature extraction; through feature extraction, key water level change trends, water quality parameter changes and sensor response time and other features are extracted from the original water level data, and the features are used as the input of the machine learning model; secondly, the deep neural network model is selected for initialization, and the preprocessed features are input into the deep neural network model for training; during the training process, the deep neural network model optimizes the parameters of the deep neural network model by learning the relationship between water level changes in historical data and control strategies; finally, the dynamic waterway control adjustment strategy is decomposed into three levels: bottom-level strategy learning, middle-level strategy learning and high-level strategy learning; in the strategy learning process at each level, the deep neural network model is initialized according to the actual water level. The waterway control effect is evaluated and a feedback signal is generated; the dynamic waterway control adjustment strategy is optimized through continuous iterative learning; among them, the bottom-level strategy learning: mainly focuses on the real-time response to the water level change trend, and through training the model, it can quickly adjust the waterway control parameters according to the changes in the current water level data, such as the start and stop of the water pump, the opening and closing of the valve, etc.; the middle-level strategy learning: on the basis of the bottom-level strategy, the changes in water quality parameters are considered; through training the model, it can dynamically adjust the operating status of the water quality cleaning equipment according to the real-time changes in the water quality data to ensure that the water quality is always within a safe range; the high-level strategy learning: on the basis of the middle-level strategy, the water level change trend and the long-term change trend of the water quality parameters are comprehensively considered; through training the model, it can predict the future water level and water quality changes, and adjust the waterway control strategy in advance to cope with potential risks and challenges. Step S301 of the above scheme is data preprocessing and feature extraction, which removes noise and outliers to ensure the accuracy and reliability of input data; unifies data of different dimensions to the same scale to avoid the impact of data deviation on model training; extracts key water level change trends, water quality parameter changes, sensor response time and other features from the original data to provide high-quality input data for model training. Significance achieved: Through data preprocessing and feature extraction, the accuracy and consistency of input data are guaranteed, laying a solid foundation for subsequent model training; it helps to improve the training efficiency and prediction accuracy of the model, so as to better realize dynamic waterway control. Step S302 is the initialization and training of the deep neural network model. The deep neural network model is selected to better capture the complex relationship between water level changes and control strategies by using its powerful nonlinear fitting ability; by learning the relationship between water level changes and control strategies in historical data, the parameters of the deep neural network model are optimized so that it can more accurately predict future water level changes and water quality parameters.Significance achieved: The application of deep neural network model enables the system to learn complex control strategies from a large amount of historical data, improving the intelligent level of waterway control; it not only helps to respond to water level changes in real time, but also predicts future water level and water quality changes, takes control measures in advance, and ensures the stable operation of the waterway system. Step S303 hierarchical strategy learning and optimization, through training model, enables it to quickly adjust waterway control parameters such as water pump start and stop, valve opening and closing, etc. according to the changes in current water level data; on the basis of the bottom-level strategy, further consider the changes in water quality parameters, through training model, enable it to dynamically adjust the operating status of water quality cleaning equipment according to the real-time changes in water quality data, and ensure that the water quality is always within a safe range; on the basis of the middle-level strategy, comprehensively consider the water level change trend and the long-term change trend of water quality parameters, through training model, enable it to predict future water level and water quality changes, and adjust the waterway control strategy in advance to cope with potential risks and challenges. Significance achieved: The hierarchical strategy learning method enables the system to comprehensively optimize the waterway control strategy from multiple dimensions; the bottom-level strategy ensures the real-time response of the water level, the middle-level strategy focuses on the safety of water quality, and the high-level strategy provides the prediction and response capabilities of long-term trends; the multi-level control strategy not only improves the system's response speed and control accuracy, but also can cope with complex and changeable waterway environments, ensuring the stable operation of the waterway system and water quality safety.

[0051] In summary, the adjustment process of the dynamic waterway control adjustment strategy in this embodiment not only realizes the real-time response and prediction of water level and water quality changes, but also comprehensively optimizes the waterway control strategy through a hierarchical strategy learning method. It not only improves the intelligence level of the waterway system, but also ensures the stable operation and water quality safety of the waterway system, and has important practical application value. Through the hierarchical learning process, the comprehensive optimization of the dynamic waterway control adjustment strategy is achieved. From the bottom-level response to water level changes, to the middle-level water quality parameter adjustment, to the high-level long-term trend prediction, each level of learning provides a solid foundation for the final control strategy; the hierarchical learning method not only improves the accuracy and efficiency of waterway control, but also can cope with complex and changeable waterway environments and ensure the stable operation of the waterway system.

[0052] Example 9: Fig. 9 As shown, on the basis of Embodiment 1 to Embodiment 8, the waterway detection and control circuit provided by the embodiment of the present invention includes: The data acquisition module is responsible for placing various types of sensors on the water tank, realizing multi-dimensional collection of the water level in the water tank through the sensors, fusing and calibrating the collected water level data, and obtaining the pre-processed water level data; Among them, various types of sensors include ultrasonic sensors, pressure sensors, capacitive sensors and water quality sensors; ultrasonic sensors are suitable for water level measurement, pressure sensors are suitable for depth measurement, capacitive sensors are suitable for liquid level detection, and water quality sensors are used to monitor water quality parameters such as pH value and dissolved oxygen; the selection of each sensor should be based on its functional characteristics and applicable scenarios; The dynamic adjustment module is responsible for obtaining the pre-processed water level data and inputting it into the dynamic waterway control adjustment strategy. It dynamically adjusts the waterway according to the water level change trend and waterway status. At the same time, it compares the water level data with the preset multi-level water level threshold. When the water level data is lower than the safety threshold in the multi-level water level threshold, the alarm mechanism is automatically triggered. The water quality data is obtained from the water level data and compared with the preset safety range. When it exceeds the safety range, the water quality is automatically cleaned. The iterative learning module is responsible for inputting the data after dynamic adjustment of the waterway into the machine learning model, learning the data and dynamic waterway control adjustment strategy, and adjusting the dynamic waterway control adjustment strategy through continuous iterative learning.

[0053] The working principle and beneficial effects of the above technical solution are as follows: the data acquisition module of this embodiment arranges various types of sensors on the water tank, realizes multi-dimensional collection of the water level in the water tank through the sensors, fuses and calibrates the collected water level data, and obtains the pre-processed water level data; among them, the various types of sensors include ultrasonic sensors, pressure sensors, capacitive sensors and water quality sensors; ultrasonic sensors are suitable for water level measurement, pressure sensors are suitable for depth measurement, capacitive sensors are suitable for liquid level detection, and water quality sensors are used to monitor water quality parameters such as pH value, dissolved oxygen, etc.; the selection of each sensor shall be based on its functional characteristics and applicable scenarios; dynamic The adjustment module obtains the pre-processed water level data and inputs it into the dynamic waterway control adjustment strategy. According to the water level change trend and waterway status, the waterway is dynamically adjusted. At the same time, the water level data is compared with the preset multi-level water level threshold. When the water level data is lower than the safety threshold in the multi-level water level threshold, the alarm mechanism is automatically triggered. The water quality data is obtained from the water level data and compared with the preset safety range. When it exceeds the safety range, the water quality is automatically cleaned. The iterative learning module inputs the data after the dynamic adjustment of the waterway into the machine learning model, learns the data and the dynamic waterway control adjustment strategy, and adjusts the dynamic waterway control adjustment strategy through continuous iterative learning. The data acquisition module of the above scheme realizes the multi-dimensional collection of water level and water quality parameters by deploying various types of sensors (such as ultrasonic sensors, pressure sensors, capacitive sensors and water quality sensors) on the water tank; the collected water level data is fused and calibrated to ensure the accuracy and consistency of the data, and the pre-processed water level data is obtained. Significance achieved: Multi-dimensional data collection and fusion calibration ensure the accuracy and reliability of water level and water quality data, and provide high-quality input data for subsequent dynamic adjustment and control; it helps to improve the accuracy and stability of the waterway control system and ensure the normal operation of the waterway system. The dynamic adjustment module inputs the pre-processed water level data into the dynamic waterway control adjustment strategy, and dynamically adjusts the waterway control parameters according to the water level change trend and waterway status; compares the water level data with the preset multi-level water level thresholds, and automatically triggers the alarm mechanism when the water level data is lower than the safety threshold to ensure that the water level is within the safety range; obtains water quality data from the water level data and compares it with the preset safety range. When the water quality parameters exceed the safety range, the water quality is automatically cleaned to ensure water quality safety. Significance achieved: The dynamic adjustment module realizes real-time monitoring and control of water level and water quality changes, ensuring that the water level and water quality are always within the safety range; it not only improves the safety and reliability of the waterway system, but also can respond to changes in water level and water quality in a timely manner, reducing potential risks and losses.The iterative learning module inputs the data after the dynamic adjustment of the waterway into the machine learning model, learns the data and the dynamic waterway control adjustment strategy, and optimizes the dynamic waterway control adjustment strategy through continuous iterative learning; through the training of the machine learning model, the waterway control strategy is continuously optimized to improve the response speed and control accuracy of the system. Significance achieved: Through the application of the machine learning model, the iterative learning module enables the system to learn the optimal control strategy from historical data, improving the intelligent level of waterway control; it not only helps to respond to changes in water level and water quality in real time, but also predicts future trends, takes control measures in advance, and ensures the stable operation of the waterway system and water quality safety.

[0054] To sum up, the water channel detection and control circuit of this embodiment not only realizes multi-dimensional monitoring and control of water level and water quality, but also comprehensively optimizes the water channel control strategy through dynamic adjustment and iterative learning; it not only improves the safety and reliability of the water channel system, but also ensures that the water level and water quality are always within a safe range, and has important practical application value.

[0055] Embodiment 10: Based on Embodiment 9, the application of the water channel detection and control circuit provided in the embodiment of the present invention includes: arranging ultrasonic sensors and capacitive sensors in the water tank of the water dispenser; the ultrasonic sensor is used to measure the water level height, and the capacitive sensor is used to detect the liquid level change; the data of the two sensors are integrated and calibrated to obtain accurate water level data; the pre-processed water level data is input into the dynamic water channel control adjustment strategy; according to the water level change trend and the water channel state, the water channel is dynamically adjusted to ensure that the water level is always within a safe range; when the water level data is lower than the safety threshold in the multi-level water level threshold, the water replenishment mechanism is automatically triggered to ensure that the water dispenser always has sufficient water supply; Water quality sensors are installed in the water tank of the water dispenser; the water quality sensors are used to monitor water quality parameters, such as pH value, dissolved oxygen, etc.; the collected water quality data are integrated and calibrated to obtain accurate water quality data; the pre-processed water quality data is input into the dynamic waterway control adjustment strategy; the waterway is dynamically adjusted according to the water quality change trend and waterway status to ensure that the water quality always meets the safety standards; when the water quality data exceeds the preset safety range, the cleaning program is automatically started to ensure that the water in the water dispenser is always clean and hygienic; The data after dynamic adjustment of the water path is input into the machine learning model of the water dispenser controller; through continuous iterative learning, the data and dynamic water path control adjustment strategy are learned; according to the learning results, the dynamic water path control adjustment strategy is adjusted to improve the intelligence level of the water dispenser and ensure its long-term stable operation.

[0056] This embodiment can also be applied to urban water supply systems to monitor the water level in the urban water supply system in real time to ensure the normal operation of the water supply network; monitor the water quality to ensure that the water quality meets the drinking water standards; dynamically adjust the water supply strategy according to changes in water level and water quality, optimize the operating efficiency of the water supply network, and ensure the water needs of urban residents; improve the intelligence level of the urban water supply system, ensure water supply safety and stability, reduce the occurrence of water supply accidents, and improve the quality of life of residents.

[0057] The working principle and beneficial effects of the above technical solution are as follows: This embodiment provides high-precision water level measurement and liquid level change detection; ensures the accuracy and reliability of water level data; adjusts the waterway in real time according to the water level change trend and the waterway status to ensure that the water level is always within the safe range; when the water level is lower than the safety threshold, automatically triggers water replenishment to ensure that the water dispenser always has sufficient water supply. Significance: Through real-time monitoring and automatic control, ensure that the water level is always within the safe range to avoid dry burning or other safety hazards caused by too low water level; the automatic water replenishment mechanism ensures that users have water available at any time, improving the convenience and satisfaction of use. Water quality monitoring and cleaning, real-time monitoring of water quality parameters, such as pH value, dissolved oxygen, etc., to ensure that the water quality meets safety standards; ensure the accuracy and reliability of water quality data; according to the water quality change trend and the waterway status, adjust the waterway in real time to ensure that the water quality always meets safety standards; when the water quality exceeds the safety range, automatically start the cleaning program to ensure that the water in the water dispenser is always clean and hygienic. Significance: Through real-time monitoring and automatic cleaning, ensure that the water quality always meets the drinking water standards and protect the health and safety of users; regular cleaning can reduce the accumulation of scale and bacteria and extend the service life of the water dispenser. Dynamic adjustment and optimization Through continuous iterative learning, optimize the water channel control strategy and improve the intelligence level of the water dispenser; according to the learning results, dynamically adjust the water channel to ensure the long-term stable operation of the water dispenser. Significance: Through machine learning models, continuously optimize the control strategy to make the water dispenser have a higher level of intelligence and adapt to various complex environments; dynamically adjust the water channel strategy to ensure that the water dispenser maintains an efficient and stable operating state in long-term use. Applied to urban water supply systems to ensure the normal operation of the water supply network and the compliance of water quality with drinking water standards; according to changes in water level and water quality, optimize the operating efficiency of the water supply network to ensure the water demand of urban residents; through machine learning models, optimize the water supply strategy to ensure safe and stable water supply. Significance: Through real-time monitoring and dynamic adjustment, ensure the safety and stability of the urban water supply system and reduce the occurrence of water supply accidents; optimize the water supply strategy to ensure the water demand of urban residents and improve the quality of life of residents.

[0058] To sum up, the application of the water path detection and control circuit of the water dispenser of this embodiment not only improves the intelligence level and safety of the equipment, but also ensures the long-term stable operation of the water quality and water supply system, which has important practical significance and broad application prospects.

[0059] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention belong to the scope of equivalent technologies of the present invention, the present invention is also intended to include these modifications and variations.

Claims

1. A waterway detection and control method, characterized in that: The following steps are involved: Various types of sensors are placed on the water tank to collect the water level in multiple dimensions. The collected water level data are fused and calibrated to obtain pre-processed water level data. Obtain pre-processed water level data and input it into the dynamic waterway control adjustment strategy to dynamically adjust the waterway according to the water level change trend and waterway status; The data after dynamic adjustment of the waterway is input into the machine learning model to learn the data and dynamic waterway control adjustment strategy. Through continuous iterative learning, the dynamic waterway control adjustment strategy is adjusted.

2. The waterway detection and control method according to claim 1, characterized in that: The process of placing various types of sensors on the water tank includes the following steps: Obtain the geometric characteristics of the water tank, including the height, diameter, wall thickness and inlet and outlet positions of the water tank. According to the height and diameter of the water tank, preliminarily determine the number and distribution of monitoring points, divide the water tank into three areas: upper, middle and lower, and set several monitoring points in each area; Analyze the inlet and outlet locations of the water tank to determine the water flow path and retention area; set up multiple monitoring points based on the geometric characteristics of the water tank and the water flow path, and each monitoring point should contain at least one type of sensor; The pre-processed water level data is input into the dynamic waterway control adjustment strategy, and the position and parameters of the sensor are dynamically adjusted according to the water level change trend and waterway status; Water level change trend prediction formula: In the formula, represents the predicted water level change, represents the observed water level change, represents the last predicted water level change, represents the smoothing coefficient; Sensor position adjustment: In the formula, represents the adjusted sensor position, Indicates the original sensor position, Indicates the position adjustment amount.

3. The waterway detection and control method according to claim 1, characterized in that: At the same time, the water level data is compared with the preset multi-level water level thresholds. When the water level data is lower than the safety threshold in the multi-level water level threshold, the alarm mechanism is automatically triggered; water quality data is obtained from the water level data and compared with the preset safety range. When it exceeds the safety range, the water quality is automatically cleaned.

4. The waterway detection and control method according to claim 1, characterized in that: The process of dynamically adjusting the waterway includes the following steps: Perform time series analysis on water level data to extract water level change trends and periodic fluctuation target features; use the target feature time series prediction model to predict future water level change trends and generate prediction curves; According to the design capacity of the water tank and the current water level, the waterway capacity is dynamically evaluated through fuzzy reasoning; based on the stability evaluation method, the waterway is dynamically evaluated through probabilistic reasoning; combined with the data from the water quality sensor, it is evaluated whether the water quality meets the safety standards and whether water cleaning is required; According to the water level change trend and waterway status, multi-level water level thresholds are set, including safety thresholds and warning thresholds; adaptive threshold setting is used in combination with historical water level data and real-time water level data to dynamically adjust the thresholds; based on the comparison results between the current water level and the threshold, the corresponding dynamic waterway control adjustment strategy is generated; at the same time, when the water level is lower than the safety threshold, the alarm mechanism is automatically triggered; according to the generated dynamic waterway control adjustment strategy, the waterway is adjusted in real time; and the implementation effect of the adjustment strategy is evaluated by real-time monitoring of water level and water quality data.

5. The waterway detection and control method according to claim 4, characterized in that: The process of dynamic assessment of waterways includes the following steps: Obtain the design capacity data of the water tank as the basic parameter for evaluation; collect the current water level data in real time as the dynamic input for evaluation; perform fuzzy processing on the current water level data and convert it into fuzzy language variables; perform fuzzy processing on the design capacity of the water tank and convert it into fuzzy language variables; Based on expert knowledge and historical data, a fuzzy rule base is constructed; the fuzzified water level and capacity are input into the fuzzy reasoning engine, and the fuzzy evaluation results of the waterway capacity are obtained through reasoning through the fuzzy rule base; the fuzzy evaluation results are defuzzified and converted into specific values ​​as the dynamic evaluation results of the waterway capacity; Obtain historical and real-time water level data as the basic input for evaluation; combine the data from water quality sensors to obtain water quality status as an auxiliary input for evaluation; define the nodes of the Bayesian network as variables for evaluation; Construct a conditional probability table to describe the conditional probability relationship between nodes; input water level and water quality data into the Bayesian network inference engine, and obtain the stability assessment results of the waterway through probabilistic reasoning; Output the stability assessment results of the waterway and generate corresponding decision support information based on the assessment results.

6. The waterway detection and control method according to claim 5, characterized in that: The process of defining the nodes of a Bayesian network consists of the following steps: The node definition of the Bayesian network is: water level is a primary variable, affecting the waterway capacity; water quality is a secondary variable, indirectly affecting the waterway capacity through waterway stability; waterway stability is an intermediate variable, connecting water quality and waterway capacity; waterway capacity is a tertiary variable, and the final output is the result.

7. The waterway detection and control method according to claim 6, characterized in that: In a directed acyclic graph, nodes represent different entities or states, and edges represent dependencies or causal relationships between nodes. Each edge has a direction, indicating a unidirectional relationship from one node to another; this edge indicates that node H has a direct impact on node C, and the path indicates that node Q indirectly affects node C through node S.

8. The waterway detection and control method according to claim 7, characterized in that: A conditional probability table is established. The conditional probability table of waterway stability is based on water quality, which describes the probability distribution of waterway stability under different water quality conditions. The conditional probability table of waterway capacity is based on water level H and waterway stability, which describes the probability distribution of waterway capacity under different water levels and waterway stability conditions.

9. A waterway detection and control circuit, characterized in that: Include: The data acquisition module is responsible for placing various types of sensors on the water tank, realizing multi-dimensional collection of the water level in the water tank through the sensors, fusing and calibrating the collected water level data, and obtaining the pre-processed water level data; The dynamic adjustment module is responsible for obtaining the pre-processed water level data and inputting it into the dynamic waterway control adjustment strategy. It dynamically adjusts the waterway according to the water level change trend and waterway status. At the same time, it compares the water level data with the preset multi-level water level threshold. When the water level data is lower than the safety threshold in the multi-level water level threshold, the alarm mechanism is automatically triggered. The water quality data is obtained from the water level data and compared with the preset safety range. When it exceeds the safety range, the water quality is automatically cleaned. The iterative learning module is responsible for inputting the data after dynamic adjustment of the waterway into the machine learning model, learning the data and dynamic waterway control adjustment strategy, and adjusting the dynamic waterway control adjustment strategy through continuous iterative learning.

10. An application of a waterway detection and control method, characterized in that: Ultrasonic sensors and capacitive sensors are arranged in the water tank of the water dispenser; the ultrasonic sensor is used to measure the water level height, and the capacitive sensor is used to detect the liquid level change; the data of the two sensors are fused and calibrated to obtain the water level data; the pre-processed water level data is input into the dynamic water channel control adjustment strategy; the water channel is dynamically adjusted according to the water level change trend and the water channel status.

Images