Intelligent monitoring system for air in water dispenser

Abstract

The invention relates to an intelligent monitoring system for air in a water dispenser, which belongs to the technical field of environment monitoring and healthy household appliances and comprises an environment sensing module used for collecting real-time TVOC (total volatile organic compound) concentration in the water dispenser and monitoring water taking behaviors of a user to generate a water taking event data set, and an intelligent control module used for controlling the water taking event data set based on the water taking event data set. Calculating a pollution potential risk index; fusing the pollution potential risk index and the real-time TVOC concentration to generate a purification start decision value; according to the method, the risk level is pre-judged in advance, and precious advance is gained for subsequent purification decision making, so that the peak value of the TVOC concentration is effectively inhibited, and it is ensured that the quality of air in the water dispenser is always below a safety threshold value.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of environmental monitoring and health household appliances, in particular to an intelligent air monitoring system for a water dispenser. BACKGROUND

[0002] A water dispenser generally has an air inlet connected to the outside to balance the air pressure when a user takes water. This structure causes the inflow of external air containing volatile organic compounds, bacteria, mold and other pollutants, resulting in secondary pollution of drinking water and posing a health risk. Existing technologies mostly use a timed purification or continuous purification scheme. The former cannot cope with the instantaneous inflow of pollutants, and the latter causes energy waste and accelerates the aging of components. Therefore, there is a technical contradiction between purification efficiency and operation economy in the field.

[0003] The above information disclosed in the background section is only used to enhance the understanding of the background of the present disclosure, and therefore it can include information that does not constitute prior art known to those of ordinary skill in the art. SUMMARY

[0004] The present application aims to provide an intelligent air monitoring system for a water dispenser to solve the problems raised in the background.

[0005] The technical solution of the present application is as follows: an environmental perception module is used to collect the real-time TVOC concentration inside the water dispenser and monitor the user's water taking behavior to generate a water taking event data set; An intelligent control module is used to calculate a pollution potential risk index based on the water taking event data set, fuse the pollution potential risk index and the real-time TVOC concentration to generate a purification start decision value, and determine the running time of a purification module according to the purification start decision value; An air purification module is used to perform a purification operation within a preset control period according to the running time of the purification module.

[0006] Preferably, the intelligent control module calculates the pollution potential risk index as follows: Based on the water taking event data set, the water taking event frequency and the average duration are counted within a preset sliding time window. The water taking event frequency is compared with a preset maximum safe frequency to obtain a frequency normalization value. The average duration is compared with a preset maximum safe average duration to obtain a duration normalization value. The frequency normalization value and the duration normalization value are weighted and summed according to preset frequency and duration weights to generate the pollution potential risk index.

[0007] Preferably, the intelligent control module generates the purification start decision value as follows: The collected real-time TVOC concentration is compared with a preset TVOC concentration safety target threshold to obtain an instant state risk value; the calculated pollution potential risk index is subjected to nonlinear function processing to obtain a predictive risk value; and the instant state risk value and the predictive risk value are weighted and fused according to a preset instant state weight and a prediction risk weight to generate the purification start decision value.

[0008] Preferably, the nonlinear function processing is that the pollution potential risk index is compared with a preset risk activation threshold, and a Sigmoid activation function is used to process the comparison result.

[0009] Preferably, when the purification start decision value is greater than a preset purification start threshold, the intelligent control module determines the process of the operation of the purification module as follows: The difference between the purification start decision value and the purification start threshold is calculated; the difference is multiplied by a preset proportional gain coefficient to obtain a proportional adjustment duration; the proportional adjustment duration and a preset minimum effective operation duration of the purification module are summed to obtain an initial calculation duration; and the initial calculation duration and the length of the preset control cycle are compared, and the smaller value is taken as the operation duration of the purification module.

[0010] Preferably, when the purification start decision value is not greater than the purification start threshold, the operation duration of the purification module is determined as the preset minimum effective operation duration of the purification module.

[0011] Preferably, the water taking event data set includes the occurrence time and the duration of each water taking event.

[0012] Preferably, the minimum effective operation duration of the purification module is the minimum lighting time required to achieve a stable purification efficiency according to the physical characteristics of the air purification module itself.

[0013] The present application provides an air intelligent monitoring system in a water dispenser by improvement, compared with the prior art, has the following improvements and advantages: 1. The present application realizes the fundamental change from passive response to active defense. The intelligent control module does not wait for the air quality sensing unit to detect that the TVOC concentration exceeds the standard before starting purification. Instead, it converts the user's water-taking behavior, a disturbance source, into a quantitative and forward-looking risk indicator through mathematical modeling. Each component of this formula is clearly defined in the specification, and the construction logic is clear. The normalized values of the two core dimensions, water-taking frequency and average duration, are weighted and summed. Those skilled in the art can implement it based on this. The direct beneficial effect is that the system can predict the risk level in advance before the pollutants really spread and cause water quality deterioration, thus effectively suppressing the peak value of TVOC concentration and ensuring that the air quality inside the water dispenser is always below the safety threshold. 2. An intelligent decision-making and fine execution mechanism that combines prediction and actual measurement is constructed, achieving extreme energy-saving effect. Instead of using predictive risk in isolation, the intelligent control module dynamically fuses this risk with real-time TVOC concentration, responding to the existing pollution stock and reacting to foreseeable pollution increments, achieving a comprehensive and three-dimensional assessment of purification needs. Compared with the existing technology, which runs at a fixed power regardless of pollution, the present application has made a leap from always online to on-demand service, and its energy-saving effect is self-evident. 3. The technical solution model of the present application is simple and easy to deploy reliably in cost-sensitive home appliances. The entire core control method consists of three mathematical formulas with clear logic and small calculation amount. The derivation and composition of these formulas are based on clear control theory and parameter definition, and do not rely on complex machine learning models that require massive data for pre-training. This means that the hardware computing power requirement of the intelligent control module is extremely low, and it can be easily deployed on a general low-cost microcontroller. This technical simplicity and certainty ensures the stability and reliability of system operation, while also significantly reducing product manufacturing costs and R&D barriers, making it highly valuable for industrial application. BRIEF DESCRIPTION OF DRAWINGS

[0014] The present application will be further explained below in conjunction with the accompanying drawings and examples: Figure 1 is a flowchart of the system of the present application. DETAILED DESCRIPTION

[0015] To make the purpose, technical scheme and advantages of the present application clearer, the present application will be further explained in detail below in conjunction with specific examples.

[0016] Example 1: Please refer to Figure 1The application provides an intelligent air monitoring system for a water dispenser, comprising: an environment sensing module, configured to collect real-time TVOC concentration inside the water dispenser and monitor user water taking behavior to generate a water taking event data set; an intelligent control module, configured to calculate a pollution potential risk index based on the water taking event data set, fuse the pollution potential risk index and the real-time TVOC concentration to generate a purification start decision value, and determine a purification module operation duration according to the purification start decision value; an air purification module, configured to perform a purification operation in a preset control period according to the purification module operation duration.

[0017] The monitoring of the user water taking behavior can be achieved by but not limited to the following ways: monitoring pressing events of a physical button linked with a water taking valve, setting a flow sensor in a water path to detect water flow start and stop, or setting an infrared or capacitive sensor in a water taking area to sense placement and removal of a cup; The intelligent air monitoring system for a water dispenser provided by the application solves the inherent contradiction between purification efficiency and operation economy in the prior art. The system builds a complete closed-loop control link from sensing, decision making to execution through precise cooperation of three core modules. The environment sensing module serves as a sensing unit of the system and uninterruptedly captures two types of key information: real-time TVOC concentration representing current pollution state and user water taking behavior data indicating future pollution trend. The intelligent control module serves as a decision making and control core of the system, receives and processes the information, executes a multi-stage algorithm, and outputs accurate control instructions. The air purification module serves as an execution unit of the system and strictly follows the instructions to perform appropriate purification operation in each control period. This architecture enables the system to have predictive intervention capability, optimizes energy consumption under the premise of ensuring air quality safety inside the water dispenser, and thus improves operation economy of the product.

[0018] Embodiment 2 The process of calculating the pollution potential risk index by the intelligent control module is as follows: Based on the water taking event data set, the water taking event frequency and average duration are counted in a preset sliding time window. The water taking event frequency is compared with a preset maximum safe frequency to obtain a frequency normalization value. The average duration is compared with a preset maximum safe average duration to obtain a duration normalization value. The frequency normalization value and the duration normalization value are weighted and summed according to preset frequency weight and duration weight to generate the pollution potential risk index.

[0019] In this embodiment, the calculation of the pollution potential risk index by the intelligent control module is an initial step of realizing predictive intervention, and the core lies in converting user behavior data into a quantitative risk index. The design idea combines the feedforward control theory in the industrial control field and the multi-attribute decision model in risk assessment. The technical motivation is that, compared with the lag response of waiting for the TVOC sensor reading, by directly monitoring the user's water taking behavior as the pollution source, the pollution impact that the system will face can be quantitatively predicted in advance, providing key input for subsequent active compensation control. The process is defined by the following formula one: ; Wherein is the dimensionless pollution potential risk index calculated at the current time , the value range of which is normalized to ; represents the water taking event frequency in the past sliding time window with a length of , for example seconds, and the unit is times / minute; If the current time is t, the window will count and analyze all water taking events occurring from to ; is the average duration of all water taking events calculated in the same window, and the unit is seconds; and are the maximum safe frequency and the maximum safe average duration respectively preset by the system, which are normalized benchmarks, and their values are derived from statistical analysis of user usage habits in typical application scenarios, for example, taking the 95% quantile of the statistical distribution to ensure the universality of the model, and their dimensions are consistent with and ; are the dimensionless frequency weight and duration weight, which satisfy , and their settings are based on experimental data to determine the contribution of different usage modes to the final risk assessment; In actual operation, the intelligent control module converts the discrete events collected by the water taking behavior monitoring unit into a continuous changing and predictive risk index based on this formula; the technical effect of this conversion is that when users take water in a concentrated manner, even if the in-machine TVOC concentration has not risen significantly, value has already risen to trigger an early warning; this is different from the lag mode of passive response to pollution of traditional purification systems, and provides a quantitative basis for subsequent decision-making.

[0020] Embodiment 3 The process of the intelligent control module generating the purification start decision value is as follows: The real-time TVOC concentration is compared with the preset TVOC concentration safety target threshold to obtain the instantaneous risk value; the calculated pollution potential risk index is processed by a nonlinear function to obtain the predictive risk value; based on the preset instantaneous risk weight and predictive risk weight, the instantaneous risk value and predictive risk value are weighted and fused to generate the purification start-up decision value. The nonlinear function processing involves comparing the potential pollution risk index with a preset risk activation threshold and then using the Sigmoid activation function to process the comparison result.

[0021] In this embodiment, the process of the intelligent control module generating the purification start decision value is the core link of the system control logic; its technical solution lies in a fusion control strategy, which aims to balance the two mechanisms of post-event compensation and pre-event prevention. The design draws on the state-space design concept in modern control theory, aiming to construct a single indicator that can comprehensively reflect the risk state of the system. This indicator not only includes the quantification of the current pollution level, but also incorporates the prediction of future pollution trends. In particular, the Sigmoid activation function is introduced for nonlinear processing, with the aim of achieving a sharp change in response when the risk is near the critical threshold, while remaining stable when the risk is low or extremely high, thereby achieving precise amplification of the key risk range. This process is defined by the following formula: ; in For at any time The generated dimensionless purification initiation decision value represents the urgency of the purification requirement. It is the real-time TVOC concentration measured by the air quality sensor unit, and the unit is mg / m³; It is a preset safe target threshold for TVOC concentration, for example, set at 0.6 mg / m³ according to the indoor air quality standard (GB / T18883-2022); It is the pollution potential risk index calculated by Formula 1; and These are dimensionless instantaneous state weights and predicted risk weights, which satisfy... This is used to adjust the relative weight of the two sources of risk in the final decision-making process; It is a dimensionless risk activation threshold, used as an empirical parameter to define... The sensitive interval can be set to 0.5; It is the dimensionless gain coefficient of the Sigmoid function, used to adjust the steepness of the function curve, and its value is determined experimentally. Among them, risk activation threshold The gain coefficient determines the center point at which the system begins to become sensitive to potential risks. then the steepness of the response curve near the center point is controlled: The greater the value, the more intense the system's response when the risk index exceeds the threshold, and vice versa; The application of this formula enables the intelligent control module to make accurate compound judgments; when the water demand of multiple people rises sharply , even if has not yet exceeded the threshold, the predictive risk value on the right side of the formula will significantly increase due to the amplification effect of the Sigmoid function, thereby pushing up to start purification in advance; conversely, if the external environment changes cause to rise, even if is zero, the immediate state risk value on the right side of the formula can still ensure that the system responds; the final technical effect of this fusion decision is to achieve the unity of purification immediacy and operational economy.

[0022] Embodiment 4 When the purification start decision value is greater than the preset purification start threshold value, the intelligent control module determines the process of the purification module running as follows: Calculate the difference between the purification start decision value and the purification start threshold value; multiply the difference by the preset proportional gain coefficient to obtain the proportional adjustment duration; sum the proportional adjustment duration and the preset minimum effective running duration of the purification module to obtain the initial calculation duration; compare the initial calculation duration with the length of the preset control period, and take the smaller value as the purification module running duration; When the purification start decision value is not greater than the purification start threshold value, the purification module running duration is determined as the preset minimum effective running duration of the purification module; The minimum effective running duration of the purification module is the minimum lighting time required to achieve stable purification efficiency according to the physical characteristics of the air purification module itself.

[0023] In this embodiment, the process of the intelligent control module determining the purification module running is the final execution link of the control process; its function is to convert the abstract decision value into specific device control instructions; Its design adopts an effective proportional control idea, aiming to establish a linear mapping relationship from demand intensity to execution intensity, so that the system can proportionally allocate purification resources according to the level of pollution risk; In this scenario, proportional control is chosen instead of more complex PI or PID control, considering the system's need for fast response and stability requirements; only the proportional link is sufficient to achieve the linear correlation between demand intensity and execution intensity, while avoiding the overshoot problem caused by the integral link and the excessive sensitivity to sensor noise caused by the differential link, thereby ensuring control effectiveness while improving system robustness; The generation of the execution instruction is defined by the following formula three: ; Wherein is the runtime length calculated at the beginning of the current decision-making period , which indicates the length of time the air purification module should run in the next control period, in seconds; is the fixed control period preset by the system, for example, 5 seconds, The function ensures that the calculated runtime does not overflow to the next period; is the minimum effective runtime of the purification module, which is determined according to the physical characteristics of the purification component itself, for example, the minimum lighting time required for UVC-LED to reach stable purification efficiency; Setting this minimum effective runtime is to ensure that each purification operation is effective, avoiding the waste of energy and insufficient purification efficiency caused by the purification device, such as UVC-LED, not reaching the best working state due to too short start-up time; is the purification start decision value calculated by formula two; is a dimensionless purification start threshold value that defines the critical point of the system switching from maintenance state to active purification state, and its value is determined by experiments between energy consumption and TVOC peak suppression effect; is a proportional time conversion coefficient, with a dimension of seconds, and the value of the proportional gain coefficient a is calibrated through experiments, aiming to respond quickly and stably to the change of the purification start decision value, avoid system oscillation, and convert the dimensionless decision value deviation to the purification time in seconds; The function corresponds to the logic of the claim, that is, only when is greater than , the proportional adjustment is made; The logic of this calculation is to fine-tune the management of purification resources; when the water dispenser is idle, will be continuously lower than , at this time the purification module will maintain basic operation with the preset minimum effective time Tmin, to save energy while responding to sudden pollution at any time; during the peak period of water use, will be much higher than , the formula will calculate a larger value to drive the purification module to run powerfully and effectively suppress potential pollution; this on-demand dynamic energy management mode significantly reduces the total energy consumption and prolongs the service life of the purification module.

[0024] Example 5 The water taking event data set includes the occurrence time and duration of each water taking event.

[0025] In the present embodiment, the structure of the water intake event dataset provides a data basis for achieving high-precision prediction; the dataset is designed to contain two key dimensions of each independent water intake event: the occurrence time of the event and the duration of the event; the purpose of such design is to provide sufficient input data for the risk assessment model; only recording the number of water intake is insufficient to distinguish the difference in air replacement caused by filling a water bottle and filling a small cup of water, and such difference has a significant impact on secondary pollution; by accurately recording the occurrence time, the system can calculate the water intake frequency; by accurately recording the duration, the system can estimate the cumulative water intake and further calculate the total amount of air entering; therefore, the data structure enables the system to distinguish different user usage patterns, making the calculation of the pollution potential risk index more accurate, and laying a data basis for the success of the entire predictive purification strategy.

[0026] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present application and not to limit the present application. Although the present application has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present application can be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present application, and all should be covered in the scope of the claims of the present application.​

Claims

1. A water dispenser internal air intelligent monitoring system, characterized in that, The method comprises the following steps: An environmental perception module is used to collect real-time TVOC concentration inside the water dispenser and monitor user water taking behavior to generate a water taking event dataset; An intelligent control module is used to calculate a pollution potential risk index based on the water taking event dataset, and fuse the pollution potential risk index with the real-time TVOC concentration to generate a purification start decision value; A purification module running duration is determined according to the purification start decision value; An air purification module is used to perform purification operation within a preset control period according to the purification module running duration.

2. The intelligent air monitoring system in a water dispenser according to claim 1, wherein, The intelligent control module calculates the pollution potential risk index in the following manner: Based on the water taking event dataset, the water taking event frequency and average duration are counted within a preset sliding time window; The water taking event frequency is compared with a preset maximum safe frequency to obtain a frequency normalization value, and the average duration is compared with a preset maximum safe average duration to obtain a duration normalization value; the frequency normalization value and the duration normalization value are weighted and summed according to preset frequency weight and duration weight to generate the pollution potential risk index.

3. The system of claim 1, wherein the system further comprises a sensor for detecting the presence of a user in the vicinity of the water dispenser. The intelligent control module generates the purification start decision value in the following manner: The collected real-time TVOC concentration is compared with a preset TVOC concentration safety target threshold to obtain an instantaneous state risk value; the calculated pollution potential risk index is processed by a nonlinear function to obtain a predictive risk value; The instantaneous state risk value and the predictive risk value are weighted and fused according to preset instantaneous state weight and predictive risk weight to generate the purification start decision value.

4. The intelligent air monitoring system in a water dispenser of claim 3, wherein, The nonlinear function processing is that the pollution potential risk index is compared with a preset risk activation threshold, and the comparison result is processed by a Sigmoid activation function.

5. The intelligent air monitoring system in a water dispenser as claimed in claim 1, wherein, When the purification start decision value is greater than a preset purification start threshold, the intelligent control module determines the purification module running duration in the following manner: The difference between the purification start decision value and the purification start threshold is calculated; the difference is multiplied by a preset proportional gain coefficient to obtain a proportional adjustment duration; the proportional adjustment duration is summed with a preset minimum effective purification module running duration to obtain an initial calculation duration; The initial calculation duration is compared with the length of the preset control period, and the smaller value is taken as the purification module running duration.

6. The intelligent air monitoring system in a water dispenser of claim 5, wherein, When the purification start decision value is not greater than the purification start threshold, the purification module running duration is determined as the preset minimum effective purification module running duration.

7. The intelligent air monitoring system in a water dispenser of claim 1, wherein, The water taking event dataset includes the occurrence time and duration of each water taking event.

8. The intelligent air monitoring system in a water dispenser of claim 5, wherein, The minimum effective purification module running duration is determined according to the physical characteristics of the air purification module itself and is the minimum lighting time required to achieve stable purification efficiency.

Images