Intelligent operation, maintenance and optimization method for household water purifier under architecture of Internet of Things

Abstract

The invention relates to the technical field of furniture equipment operation and maintenance, and discloses a household water purifier intelligent operation and maintenance and optimization method under an Internet of Things architecture, and the method comprises the following steps: S1, multi-dimensional data collection and preprocessing: deploying multiple types of sensors including water quality, operation states, filter element states and selectable environmental parameters on a water purifier, original data are collected and uploaded to the cloud through the Internet of Things module, and the cloud carries out denoising, normalization and missing value filling preprocessing on the data. Multi-dimensional information such as water quality, operation states and environmental parameters is covered through a multi-type sensor array, the problem that evaluation is one-sided due to the fact that a traditional method only monitors a single index (such as TDS) is solved, and on the basis of multi-model collaborative analysis (fault prediction, filter element service life dynamic evaluation and water quality optimization), a strategy is dynamically adjusted in combination with raw water quality and user habits, and the evaluation efficiency is improved. And resource waste or water quality risk caused by fixed cycle replacement of the filter element is avoided.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of furniture equipment operation and maintenance, in particular to an intelligent operation and maintenance and optimization method for household water purifiers under an Internet of Things architecture. BACKGROUND

[0002] Under the Internet of Things architecture, the intelligent operation and maintenance and optimization method for household water purifiers relies on sensor networks, cloud computing platforms, and big data analysis technologies to achieve real-time monitoring and remote management of equipment operating conditions. By deploying water quality sensors, flow meters, filter life monitoring modules, and other devices in the water purifier, key data such as raw water quality, purified water flow, and filter usage can be collected, and wireless communication technologies such as Wi-Fi and NB-IoT can be used to upload data to the cloud platform. Based on the intelligent analysis system in the cloud, equipment operating data can be modeled and mined to achieve fault prediction, filter replacement reminders, energy optimization, and personalized water quality control. At the same time, users can view water quality and equipment status in real time through a mobile app, and maintenance personnel can perform preventive maintenance, service scheduling, and product iteration optimization based on big data, thereby improving equipment reliability, user experience, and overall maintenance efficiency, and building a safe, intelligent, and sustainable household water purification service ecosystem.

[0003] With the rapid development of Internet of Things technology, household water purifiers are gradually evolving towards intelligence and networking. In existing technologies, various sensors such as TDS sensors, flow sensors, and pressure sensors are integrated into water purifiers, and wireless communication modules such as Wi-Fi, Bluetooth, and NB-IoT are used to upload real-time or periodic operating parameters such as water quality indicators, filter usage status, water production, and equipment fault codes to cloud servers to achieve remote monitoring, fault warning, filter life management, and user interaction.

[0004] However, the existing intelligent operation and maintenance and optimization method for household water purifiers under the Internet of Things architecture still has the following shortcomings:

[0005] Insufficient data collection accuracy and comprehensiveness

[0006] The existing equipment is equipped with limited types and quantities of sensors, usually only monitoring a few key indicators such as TDS values and total water production, and lacking effective monitoring of important parameters such as water temperature, pressure fluctuations, raw water hardness, microbial content, and conductivity changes that affect water quality and equipment performance, resulting in insufficient comprehensive and accurate assessment of water purifier operating conditions and water quality changes, making it difficult to achieve fine control and optimization;

[0007] Limited intelligent analysis and decision-making capabilities

[0008] Most current systems only rely on simple threshold judgment (such as water production reaching a set value to prompt replacement of the filter core), lack of intelligent algorithm support based on multi-dimensional data fusion, and cannot accurately predict and diagnose filter core performance degradation trend, water quality dynamic change, and abnormal equipment working condition. Although some systems introduce machine learning models, the training data sources are single and the sample size is small, the model generalization ability is poor, and it is difficult to adapt to individual operation and maintenance needs under different regional and water quality conditions.

[0009] Inaccurate filter core life management

[0010] The actual service life of the filter core is affected by multiple factors such as raw water quality, usage frequency, water temperature, and pressure. However, most existing methods use fixed cycles or simply accumulate water production as the basis for replacing the filter core, which cannot be dynamically adjusted according to actual usage, easily leading to waste due to premature replacement of the filter core or affecting water quality and user health due to the use of expired filter cores.

[0011] Low efficiency of remote operation and service response

[0012] Existing systems are mainly passive alarms, which only notify users through App push or SMS when equipment fails or water quality is abnormal, lack active operation and maintenance scheduling mechanisms and multi-level early warning systems. At the same time, after-sales service providers have difficulty in obtaining comprehensive historical operation data of the equipment, and fault diagnosis relies on on-site troubleshooting, resulting in low service efficiency, high cost, and poor user experience.

[0013] Poor system compatibility and expandability

[0014] Different brands and models of household water purifiers have large differences in hardware configuration, communication protocols, and data formats, making it difficult for systems to interconnect and share data for collaborative analysis, which is not conducive to building a unified intelligent operation and maintenance platform and industry ecosystem. In addition, existing systems have poor expandability and are difficult to quickly adapt to new sensors, functional modules, and optimization algorithms.

[0015] In summary, the existing technology of Internet of Things household water purifier intelligent operation and maintenance and optimization method still has obvious defects in data acquisition, intelligent analysis, filter core management, service response, and system compatibility.

[0016] Therefore, we propose a household water purifier intelligent operation and maintenance and optimization method under the Internet of Things architecture. SUMMARY

[0017] To achieve the above purpose, the present application provides the following technical scheme: a household water purifier intelligent operation and maintenance and optimization method under the Internet of Things architecture, comprising the following steps:

[0018] S1: Multi-dimensional data acquisition and preprocessing: Deploy multiple types of sensors on the water purifier, including water quality, operating status, filter status and optional environmental parameters, to collect raw data and upload it to the cloud through the Internet of Things module. The cloud performs noise reduction, normalization and missing value filling preprocessing on the data.

[0019] S2: Multi-source data fusion and dynamic feature extraction: Based on the preprocessed data, water quality characteristics, operational characteristics, environmental characteristics and time characteristics are extracted, and a comprehensive feature vector is constructed through time series analysis and multivariate statistical methods;

[0020] S3: Intelligent analysis and decision-making based on multi-model collaboration: It adopts a hierarchical intelligent decision-making architecture, and outputs the failure probability, remaining filter life and water quality optimization suggestions through collaborative analysis of the fault prediction sub-model, filter life dynamic assessment sub-model and water quality optimization sub-model. Key decisions are issued after comprehensive judgment by the cloud.

[0021] S4: Proactive Operation and Maintenance Scheduling and User Interaction: Real-time display of water quality and equipment status through the user-end App, generation of multi-level early warning work orders and scheduling of service resources through the operation and maintenance end, and support for remote diagnosis and personalized suggestion push;

[0022] S5: System self-learning and model iteration: Based on user feedback, new fault cases and environmental data, the intelligent analysis model is dynamically updated using online learning algorithms.

[0023] Preferably, the multiple types of sensors include: TDS sensor, conductivity sensor, temperature sensor, pH sensor, inlet / pure water flow sensor, pressure sensor (inlet / concentrate / pure water side), water pump operating parameter sensor, differential pressure sensor between the front and rear ends of each filter element, and optional UV sterilization module status sensor.

[0024] Preferably, the filter cartridge life dynamic evaluation sub-model dynamically calculates the remaining effective life of each filter cartridge based on the front and rear pressure difference change rate, water production accumulation, and the influence weight of raw water quality on the filtration load, rather than a fixed water production threshold.

[0025] Preferably, in the multi-model collaborative intelligent analysis and decision-making, the fault prediction sub-model adopts ensemble learning or deep neural network, the filter cartridge life dynamic assessment sub-model combines dynamic pressure difference threshold and raw water quality parameters, and the water quality optimization sub-model adjusts water pump parameters or filter cartridge combination strategies according to real-time water quality and user objectives.

[0026] Preferably, in the proactive operation and maintenance scheduling and user interaction, the multi-level early warning mechanism includes user reminders, operation and maintenance scheduling and emergency shutdown. The operation and maintenance terminal automatically generates a work order containing the equipment location, fault type and required spare parts, and dispatches the work order to the service personnel through GIS positioning.

[0027] Preferably, the system's self-learning and model iteration dynamically update the intelligent analysis model through an online learning algorithm, and also supports users to manually input special water quality events to assist in model correction.

[0028] Preferably, it also includes multiple types of sensor modules, IoT communication modules, cloud data processing and analysis platforms, user-end interactive apps, and operation and maintenance management backends deployed on the water purifier, wherein the cloud platform integrates intelligent decision-making modules and self-learning modules that coordinate multiple models.

[0029] Compared with existing technologies, this invention provides a method for intelligent operation, maintenance and optimization of household water purifiers under an Internet of Things (IoT) architecture, which has the following beneficial effects:

[0030] 1. The intelligent operation and maintenance and optimization method for household water purifiers under this IoT architecture covers multiple dimensions of information such as water quality, operating status, and environmental parameters through multi-type sensor arrays. This solves the problem of one-sided evaluation caused by traditional methods that only monitor a single indicator (such as TDS). Based on multi-model collaborative analysis (fault prediction + dynamic evaluation of filter life + water quality optimization), and combined with the raw water quality and user habits, the strategy is dynamically adjusted to avoid the waste of resources or water quality risks of replacing filter cartridges at fixed intervals.

[0031] 2. The intelligent operation and maintenance and optimization method for household water purifiers under this IoT architecture, with its proactive multi-level early warning and automatic work order generation mechanism, shortens the fault response time (by an average of more than 60% from the time the user discovers the anomaly to the time the service intervenes), reduces operation and maintenance costs, supports adaptation to different brands / models of water purifier hardware through edge computing nodes, and the model can continuously evolve through online learning to adapt to new water purification technologies and user needs.

[0032] 3. The intelligent operation and maintenance and optimization method of household water purifiers under this Internet of Things architecture allows users to monitor the water quality safety status and equipment health in real time through the App, obtain personalized suggestions, and enhance their trust and stickiness to the product. Detailed Implementation

[0033] The technical solutions in the embodiments of the present invention will be clearly and completely described below. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0034] Example

[0035] An embodiment of an intelligent operation and maintenance and optimization method for home water purifiers under an Internet of Things (IoT) architecture.

[0036] A method for intelligent operation, maintenance, and optimization of household water purifiers under an Internet of Things (IoT) architecture includes the following steps:

[0037] S1: Multi-dimensional data acquisition and preprocessing: Deploy multiple types of sensors on the water purifier, including water quality, operating status, filter status and optional environmental parameters, to collect raw data and upload it to the cloud through the Internet of Things module. The cloud performs noise reduction, normalization and missing value filling preprocessing on the data.

[0038] S2: Multi-source data fusion and dynamic feature extraction: Based on the preprocessed data, water quality characteristics, operational characteristics, environmental characteristics and time characteristics are extracted, and a comprehensive feature vector is constructed through time series analysis and multivariate statistical methods;

[0039] S3: Intelligent analysis and decision-making based on multi-model collaboration: It adopts a hierarchical intelligent decision-making architecture, and outputs the failure probability, remaining filter life and water quality optimization suggestions through collaborative analysis of the fault prediction sub-model, filter life dynamic assessment sub-model and water quality optimization sub-model. Key decisions are issued after comprehensive judgment by the cloud.

[0040] S4: Proactive Operation and Maintenance Scheduling and User Interaction: Real-time display of water quality and equipment status through the user-end App, generation of multi-level early warning work orders and scheduling of service resources through the operation and maintenance end, and support for remote diagnosis and personalized suggestion push;

[0041] S5: System self-learning and model iteration: Based on user feedback, new fault cases and environmental data, the intelligent analysis model is dynamically updated using online learning algorithms.

[0042] Specifically, the various types of sensors include: TDS sensor, conductivity sensor, temperature sensor, pH sensor, inlet / pure water flow sensor, pressure sensor (inlet / concentrate / pure water side), water pump operating parameter sensor, differential pressure sensor before and after each stage of filter cartridges, and optional UV sterilization module status sensor.

[0043] Specifically, the dynamic evaluation sub-model for filter cartridge lifespan dynamically calculates the remaining effective lifespan of each filter cartridge based on the rate of change of pressure difference between the front and rear ends of each filter cartridge, the cumulative water production, and the influence weight of raw water quality on the filtration load, rather than a fixed water production threshold.

[0044] Specifically, in the intelligent analysis and decision-making of multi-model collaboration, the fault prediction sub-model adopts ensemble learning or deep neural networks, the filter life dynamic assessment sub-model combines dynamic differential pressure threshold and raw water quality parameters, and the water quality optimization sub-model adjusts water pump parameters or filter combination strategies according to real-time water quality and user objectives.

[0045] Specifically, in proactive operation and maintenance scheduling and user interaction, the multi-level early warning mechanism includes user reminders, operation and maintenance scheduling and emergency shutdown. The operation and maintenance terminal automatically generates work orders containing equipment location, fault type and required spare parts, and dispatches the work orders to service personnel through GIS location.

[0046] Specifically, the system's self-learning and model iteration dynamically update the intelligent analysis model through online learning algorithms, and also support users to manually enter special water quality events to assist in model correction.

[0047] Preferably, it also includes multiple types of sensor modules, IoT communication modules, cloud data processing and analysis platforms, user-end interactive apps, and operation and maintenance management backends deployed on the water purifier, wherein the cloud platform integrates intelligent decision-making modules and self-learning modules that coordinate multiple models.

[0048] Through the above technical solution, this invention, by using a multi-type sensor array to cover multiple dimensions of information such as water quality, operating status, and environmental parameters, solves the problem of one-sided evaluation caused by traditional methods that only monitor a single indicator (such as TDS). Based on multi-model collaborative analysis (fault prediction + dynamic evaluation of filter life + water quality optimization), and combined with the dynamic adjustment strategy of raw water quality and user habits, it avoids the waste of resources or water quality risks of fixed-cycle filter replacement. The proactive multi-level early warning and automatic work order generation mechanism shortens the fault response time (by an average of more than 60% from the time the user discovers the anomaly to the time the service intervenes), reduces operation and maintenance costs, supports adaptation to different brands / models of water purifier hardware through edge computing nodes, and the model can continuously evolve through online learning to adapt to new water purification technologies and user needs. Users can use the App to monitor the water quality safety status and equipment health in real time, obtain personalized suggestions, and enhance their trust and stickiness to the product.

[0049] Example 1: Intelligent Operation and Maintenance Process of a Typical Household RO Reverse Osmosis Water Purifier

[0050] Equipment Deployment: The following sensors are integrated into the RO reverse osmosis water purifier installed in the user's home: inlet TDS sensor (monitors raw water hardness), differential pressure sensor before and after the RO membrane (monitors the degree of membrane clogging), pure water flow meter (counts daily water production), differential pressure sensors for each filter element (pre-filter PP cotton, activated carbon, RO membrane, post-filter activated carbon), and temperature / pH sensor.

[0051] Data Acquisition and Upload: The water purifier uploads sensor data to the cloud server every 30 minutes via the Wi-Fi module. The data is stored after being processed by noise reduction (such as Kalman filtering to remove flow fluctuation noise) and normalization (such as scaling the pressure difference value to the 0-1 range).

[0052] Dynamic feature extraction: Extract the current RO membrane pressure difference change rate (e.g., an average daily increase of 0.5 kPa over the past 7 days), raw water TDS value (e.g., 150 mg / L, which is of medium hardness), and daily water production (e.g., 15 L) from the cloud, and calculate the RO membrane filtration load index by combining it with historical data.

[0053] Intelligent decision-making:

[0054] The fault prediction model detected abnormal fluctuations in the water pump current (70% probability), determined that there might be a risk of poor circuit contact, and triggered a level 2 warning (pushed to the operation and maintenance terminal to generate a work order);

[0055] The filter life model determines the remaining life of the RO membrane to be 15 days (instead of the traditional fixed threshold of "500L water production") if the RO membrane pressure difference increases beyond the threshold within 7 days (e.g., initial pressure difference 0.5kPa → current 1.2kPa) and the raw water TDS is high (accelerating membrane fouling). This triggers a level 1 warning (reminding users via the App that "the RO membrane is about to expire and it is recommended to schedule a replacement").

[0056] The water quality optimization model detected that the user's water consumption was low at night (energy-saving mode) and automatically reduced the water pump's operating power to reduce noise.

[0057] Operation and maintenance execution: After receiving a level 2 early warning work order, the operation and maintenance terminal automatically matches nearby service personnel and pushes fault codes and historical maintenance records; before the service personnel arrive at the site, they can view the real-time data of the user's equipment (such as the current differential pressure value) through the App, carry the appropriate spare parts (such as RO membrane components), and after on-site replacement, the system automatically updates the filter life model parameters.

[0058] Continuous optimization: After the user replaces the RO membrane, the operation is confirmed in the App. The system records the installation time and initial differential pressure value of the new filter element, and subsequent dynamic evaluation continues to run based on the new baseline.

[0059] Example 2: Regional Water Quality Adaptability Optimization

[0060] When a region enters the rainy season, the cloud system detects an increase in raw water turbidity (such as a short-term increase in conductivity of 30%) by accessing the regional water quality database. It then automatically adjusts the operating strategies of all connected water purifiers in the region: it pushes a notification to users that "the current water quality is relatively turbid, and it is recommended to turn on the pre-filter enhanced filtration mode"; at the same time, it optimizes the filter life model and relaxes the differential pressure threshold of the pre-filter PP cotton filter by 10% (due to its short-term increased load) to ensure a balance between overall filtration efficiency and lifespan.

[0061] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A household water purifier intelligent operation and maintenance and optimization method under an Internet of Things architecture, characterized by: Comprise the following steps: S1: Multi-dimensional data acquisition and preprocessing: Deploy multiple types of sensors including water quality, running state, filter state and optional environmental parameters on the water purifier, collect raw data and upload to the cloud through the Internet of Things module, and preprocess the data on the cloud, including denoising, normalization and missing value filling; S2: Multi-source data fusion and dynamic feature extraction: Based on the preprocessed data, extract water quality features, running features, environmental features and time features, and construct a comprehensive feature vector through time series analysis and multivariate statistical methods; S3: Intelligent analysis and decision-making based on multi-model collaboration: Adopt a hierarchical intelligent decision-making architecture, and output fault probability, filter remaining life and water quality optimization suggestions through collaborative analysis of fault prediction sub-model, filter life dynamic evaluation sub-model and water quality optimization sub-model. Key decisions are made by the cloud and then issued; S4: Active operation and maintenance scheduling and user interaction: Real-time display of water quality and equipment status through user-side App, generation of multi-level warning work orders and dispatch of service resources through operation and maintenance side, support for remote diagnosis and personalized suggestion push; S5: System self-learning and model iteration: Based on user feedback, new fault cases and environmental data, use online learning algorithms to dynamically update intelligent analysis models. 2.The method of claim 1, wherein the method further comprises: The multiple types of sensors include: TDS sensor, conductivity sensor, temperature sensor, pH sensor, inlet / pure water flow sensor, pressure sensor (inlet / concentrated water / pure water side), water pump operating parameter sensor, differential pressure sensor before and after each filter, and optional UV sterilization module state sensor. 3.The method of claim 1, wherein the method further comprises: The filter life dynamic evaluation sub-model dynamically calculates the remaining effective life of each filter based on the differential pressure change rate of each filter, the cumulative water production, and the influence weight of raw water quality on filtration load, rather than a fixed water production threshold.

4. The intelligent operation and maintenance and optimization method of the household water purifier under the Internet of Things architecture according to claim 1, characterized in that: In the multi-model collaborative intelligent analysis and decision-making, the fault prediction sub-model uses ensemble learning or deep neural networks, the filter life dynamic evaluation sub-model combines dynamic differential pressure threshold and raw water quality parameters, and the water quality optimization sub-model adjusts water pump parameters or filter combination strategy according to real-time water quality and user goals. 5.The method of claim 1, wherein the method further comprises: In the active operation and maintenance scheduling and user interaction, the multi-level warning mechanism includes user reminders, operation and maintenance scheduling, and emergency shutdown. The operation and maintenance side automatically generates work orders containing device location, fault type and required spare parts, and sends them to service personnel through GIS positioning. 6.The method of claim 1, wherein the method further comprises: The system self-learning and model iteration dynamically updates intelligent analysis models through online learning algorithms, and supports user manual entry of special water quality events to assist model correction.

7. The intelligent operation and maintenance and optimization method of the household water purifier under the Internet of Things architecture according to claims 1-6, characterized in that: It also includes multiple types of sensor modules deployed on water purifiers, Internet of Things communication modules, cloud data processing and analysis platforms, user-side interaction Apps, and operation and maintenance management backends. The cloud platform integrates multi-model collaborative intelligent decision-making modules and self-learning modules.