CN-121995988-A - Water dispenser water outlet temperature control system based on big data acquisition

CN121995988ACN 121995988 ACN121995988 ACN 121995988ACN-121995988-A

Abstract

The invention discloses a water dispenser water outlet temperature control system based on big data acquisition, which relates to the technical field of intelligent household appliance control and comprises data acquisition, processing, feature fusion, temperature prediction, parameter calculation and control execution modules which are connected in sequence. The system collects multi-source data such as time sequence data of water receiving behaviors of users, ambient temperature, heating current, actual water outlet temperature and the like. And analyzing the user behavior time sequence data to extract the using habit characteristics, analyzing the environment temperature data to extract the trend characteristics, and fusing with the heating current to construct the comprehensive working condition characteristic vector. And inputting the vector and the real-time water temperature into a dynamic temperature prediction model, predicting the future temperature offset, further calculating the target control parameters of the heating element, and issuing and executing. The invention realizes prospective control by predicting the habit of the user and the temperature change trend, reduces the ineffective energy consumption, effectively compensates the thermal inertia of the system, and improves the stability and the response speed of the outlet water temperature.

Inventors

LI JUNJIE
Dong Ruihao

Assignees

南京及时水智能科技有限公司
南京脑一代智能科技有限公司

Dates

Publication Date: 20260508
Application Date: 20260208

Claims (10)

1. Water dispenser goes out water temperature control system based on big data acquisition, its characterized in that, the system includes: the data acquisition module is used for acquiring multi-source real-time operation data of the water dispenser in a preset time period, wherein the multi-source real-time operation data comprise time sequence data of water receiving behaviors of a user, environment temperature data, heating element working current data and actual water outlet temperature monitoring data; the data processing module is used for carrying out behavior pattern analysis processing on the time sequence data of the water receiving behavior of the user, extracting the using habit characteristics of the user, carrying out trend analysis processing on the environmental temperature data and extracting the environmental temperature influence characteristics; The characteristic fusion module is used for carrying out multidimensional characteristic fusion processing on the user using habit characteristics, the environment temperature influence characteristics and the heating element working current data to construct a comprehensive working condition characteristic vector; The temperature prediction module is used for inputting the comprehensive working condition characteristic vector and the actual water outlet temperature monitoring data into a preset dynamic temperature prediction model, performing temperature offset prediction and generating a predicted temperature offset corresponding to a future time point; The parameter calculation module is used for calculating and obtaining target control parameters required by the heating element based on the predicted temperature offset and a preset target outlet water temperature; And the control execution module is used for transmitting the target control parameters to a heating control unit of the water dispenser and driving the heating element to adjust the working state of the heating element.
2. The big data acquisition-based water dispenser outlet water temperature control system according to claim 1, wherein the behavior pattern analysis processing is performed on the time series data of the water receiving behavior of the user, and the extracting of the usage habit characteristics of the user comprises: performing time window segmentation on the time sequence data of the water receiving behaviors of the user to form a water receiving behavior sequence taking a single day or a plurality of continuous days as a unit; Identifying key time nodes in the water receiving behavior sequence, wherein the key time nodes comprise water receiving starting time, water receiving duration and interval duration of two adjacent water receiving times; counting the water receiving frequency distribution of each day in a preset high-frequency use time period to form a daily high-frequency use intensity index; Calculating historical average value and variance of the interval duration of two adjacent water receiving processes as quantitative indexes of the water receiving behavior regularity; And carrying out normalized coding on the daily high-frequency use intensity index and the quantitative index of the water receiving behavior regularity, and generating the user use habit feature vector in a combined way.
3. The big data acquisition-based water dispenser outlet water temperature control system according to claim 2, wherein the trend analysis processing is performed on the environmental temperature data, and extracting environmental temperature influence features includes: Performing periodic smooth filtering on the environmental temperature data to remove short-term noise fluctuation and obtain a smoothed environmental temperature sequence; calculating the temperature fluctuation amplitude and the day-night temperature difference value of the smoothed ambient temperature sequence in the last complete natural day; identifying the overall change trend of the smoothed environmental temperature sequence in a plurality of recent natural days, wherein the overall change trend comprises continuous heating, continuous cooling or periodic oscillation; And carrying out joint coding on the temperature fluctuation amplitude, the day and night temperature difference value and the type identifier of the overall change trend to generate the environment temperature influence characteristic vector.
4. The big data acquisition-based water dispenser outlet water temperature control system of claim 3, wherein the performing multi-dimensional feature fusion processing on the user usage habit feature and the environmental temperature influence feature with the heating element working current data to construct a comprehensive working condition feature vector comprises: Extracting sectional characteristics of the working current data of the heating element to obtain an average current value, a peak current value and a current change rate of each working stage; establishing an association mapping table among the user using habit feature vector, the environment temperature influence feature vector and each segment feature of the heating element working current data; According to the association mapping table, a weight coefficient related to the current characteristic is distributed to each dimension in the user using habit characteristic vector and the environment temperature influence characteristic vector; Carrying out weighted correction on the user using habit feature vector and the environment temperature influence feature vector by using the distributed weight coefficient; and carrying out series connection splicing on the weighted and corrected user habit feature vector, the environment temperature influence feature vector and each sectional feature of the heating element working current data to form the comprehensive working condition feature vector.
5. The big data collection based water dispenser outlet water temperature control system according to claim 4, wherein the inputting the integrated condition feature vector and the actual outlet water temperature monitoring data into a preset dynamic temperature prediction model, performing temperature offset prediction, generating a predicted temperature offset corresponding to a future point in time, comprises: Inputting the comprehensive working condition feature vector into a feature coding layer of the dynamic temperature prediction model to perform high-dimensional nonlinear feature transformation; Inputting the historical sequence of the actual water outlet temperature monitoring data into a time sequence memory layer of the dynamic temperature prediction model, and capturing the time sequence dependency of temperature change; At the fusion prediction layer of the dynamic temperature prediction model, the high-dimensional features transformed by the feature coding layer are interactively integrated with the time sequence dependency relationship captured from the time sequence memory layer; Based on the information after interaction integration, an output layer of the dynamic temperature prediction model generates a prediction of a temperature change trend of one or more preset time points in the future as the predicted temperature offset.
6. The big data acquisition-based water dispenser outlet water temperature control system according to claim 5, wherein the calculating the target control parameters required by the heating element based on the predicted temperature offset and a preset target outlet water temperature includes: Judging whether the actual water outlet temperature at the future moment is higher or lower and the deviation degree relative to the preset target water outlet temperature according to the positive and negative values and the magnitude of the predicted temperature deviation; inquiring a preset control parameter mapping table, wherein the control parameter mapping table stores basic power adjustment amounts corresponding to different temperature offset ranges; Dynamically correcting the basic power adjustment quantity according to the water receiving behavior regularity index and the daily high-frequency use intensity index in the user use habit characteristics to obtain a corrected power adjustment quantity; superposing or subtracting the corrected power adjustment amount and the current working power parameter of the heating element to obtain a target power value in the target control parameter; and calculating the maintenance duration of the target power value by combining the overall change trend in the environmental temperature influence characteristic to form the target control parameter comprising the target power value and the power maintenance duration.
7. The big data collection based water dispenser outlet water temperature control system of claim 6, wherein before the target control parameters are issued to the heating control unit of the water dispenser, further comprising performing safety and efficacy verification on the target control parameters, comprising: Comparing the target power value in the target control parameter with a rated power range allowed by a heating element of the water dispenser, and ensuring that the target power value does not exceed the upper limit of the rated power range and is not lower than the lower limit of a minimum heating function; simulating the energy consumption accumulation amount of the water dispenser from the current state to the predicted water receiving moment of a user in the future under the target control parameters; Comparing the energy consumption accumulation amount with an energy efficiency optimal reference value under the same working condition based on historical big data, and calculating the energy efficiency deviation degree; If the energy efficiency deviation exceeds a preset tolerance threshold, finely adjusting a target power value in the target control parameter according to a preset step, and recalculating the energy efficiency deviation until the tolerance threshold is met; and determining the final parameters after the checksum fine tuning as the target control parameters to be issued.
8. The big data acquisition-based water dispenser outlet water temperature control system according to claim 1, wherein after the acquisition of the multi-source real-time operation data of the water dispenser in the preset time period, the system further comprises preprocessing and quality evaluation of the multi-source real-time operation data, and comprises: Respectively detecting abnormal values of the collected time sequence data of the water receiving behavior of the user, the environmental temperature data, the working current data of the heating element and the actual water outlet temperature monitoring data, and identifying and eliminating error data points obviously exceeding a physical reasonable range; Checking the continuity and the integrity of the time stamp of various data after abnormal values are removed, and supplementing the data with a tiny time gap by adopting a linear interpolation method; Evaluating the data integrity rate of various data after outlier rejection and time stamp supplementation, and triggering the data re-acquisition or feature supplementing process from adjacent time period data if the data integrity rate is lower than a preset standard; And packaging various data which are qualified in evaluation to form a data packet of the multi-source real-time operation data for subsequent processing.
9. The big data acquisition-based water dispenser outlet water temperature control system of claim 1, further comprising periodically updating the dynamic temperature prediction model, comprising: continuously collecting new multi-source real-time operation data and corresponding actual outlet water temperature results to form an incremental training data set; Performing incremental training on the dynamic temperature prediction model deployed currently by using the incremental training data set, and adjusting internal parameters of the model; In the incremental training process, synchronously verifying the prediction precision of the model on the independent test set, and if the prediction precision is reduced, rolling back model parameters and analyzing the data quality; and when the incremental training enables the prediction precision of the model on the test set to be stably improved or the accumulated incremental data quantity reaches a preset threshold value, deploying the updated model as a new dynamic temperature prediction model.
10. The big data acquisition-based water dispenser outlet water temperature control system of claim 1, further comprising status monitoring and adaptive adjustment during system operation, comprising: Real-time monitoring real-time deviation between the actual outlet water temperature monitoring data and the expected temperature calculated based on the predicted temperature offset and the target outlet water temperature; When the real-time deviation continuously exceeds a preset alarm threshold value for a certain period of time, judging that the current temperature control is out of alignment; when the judgment is incorrect, the data acquisition frequency is automatically increased, and the rapid re-checking analysis of the multi-source real-time operation data in the near term is triggered; and according to the result of the rapid re-nucleation analysis, temporarily adjusting the input characteristic weight of the dynamic temperature prediction model or switching to a standby control strategy until the actual outlet water temperature monitoring data is restored to be within an expected control range.

Description

Water dispenser water outlet temperature control system based on big data acquisition Technical Field The invention belongs to the technical field of intelligent household appliance control, and particularly relates to a water dispenser water outlet temperature control system based on big data acquisition. Background The temperature control technology of the current water dispenser mainly depends on a preset program or a simple instant feedback mechanism. Common control modes include timed start-up heating, start-stop control based on a fixed temperature threshold, or proportional-integral-derivative adjustment based on real-time monitoring of water temperature. These methods rely on fixed logic or responses to a current single variable as a control basis. The prior art solutions have drawbacks. The timing heating mode cannot adapt to the actual use rhythm of different users, and energy waste is caused by continuous heat preservation in a non-water period or preparation is insufficient before a water use peak. The feedback control based on the real-time water temperature has inherent hysteresis, and when the ambient temperature is changed severely or the user receives water continuously, the system response is delayed, the fluctuation of the water outlet temperature is obvious, and the stability is difficult to maintain. Furthermore, conventional approaches generally lack the ability to perceive and learn the user's personalized usage habits and long-term trends in the environment, control the process mechanically and passively. The invention aims to solve the problems of high energy consumption and insufficient water outlet temperature stability caused by neglecting the user behavior mode and the environmental dynamic influence in the existing control strategy. The invention also solves the problems of temperature regulation lag and overshoot caused by the thermal inertia of the system in feedback control, and realizes more accurate advanced temperature regulation. Disclosure of Invention The present invention aims to solve at least one of the technical problems existing in the prior art; Therefore, the invention provides a water outlet temperature control system of a water dispenser based on big data acquisition, which comprises: the data acquisition module is used for acquiring multi-source real-time operation data of the water dispenser in a preset time period, wherein the multi-source real-time operation data comprise time sequence data of water receiving behaviors of a user, environment temperature data, heating element working current data and actual water outlet temperature monitoring data; the data processing module is used for carrying out behavior pattern analysis processing on the time sequence data of the water receiving behavior of the user, extracting the using habit characteristics of the user, carrying out trend analysis processing on the environmental temperature data and extracting the environmental temperature influence characteristics; The characteristic fusion module is used for carrying out multidimensional characteristic fusion processing on the user using habit characteristics, the environment temperature influence characteristics and the heating element working current data to construct a comprehensive working condition characteristic vector; The temperature prediction module is used for inputting the comprehensive working condition characteristic vector and the actual water outlet temperature monitoring data into a preset dynamic temperature prediction model, performing temperature offset prediction and generating a predicted temperature offset corresponding to a future time point; The parameter calculation module is used for calculating and obtaining target control parameters required by the heating element based on the predicted temperature offset and a preset target outlet water temperature; And the control execution module is used for transmitting the target control parameters to a heating control unit of the water dispenser and driving the heating element to adjust the working state of the heating element. Further, the performing behavior pattern analysis processing on the time sequence data of the water receiving behavior of the user, extracting the use habit characteristics of the user, includes: performing time window segmentation on the time sequence data of the water receiving behaviors of the user to form a water receiving behavior sequence taking a single day or a plurality of continuous days as a unit; Identifying key time nodes in the water receiving behavior sequence, wherein the key time nodes comprise water receiving starting time, water receiving duration and interval duration of two adjacent water receiving times; counting the water receiving frequency distribution of each day in a preset high-frequency use time period to form a daily high-frequency use intensity index; Calculating historical average value and variance of the interval duration of two adjacent wa