Resident community energy metering device with multivariate data access and fusion

Abstract

The invention discloses a multivariate data access and fusion resident community energy metering device, and the device comprises a data collection module which is used for collecting multi-energy data in a resident community through a plurality of communication interfaces, the energy comprises electricity, water, gas and heat, and the communication interfaces support RS485, NB-IoT, LoRa, Zigbee and Wi-Fi; an edge calculation module connected with the data acquisition module and used for preprocessing the acquired multi-energy data, including data format standardization, time sequence reconstruction and preliminary anomaly detection; the data fusion module is connected with the edge calculation module and is used for performing multi-dimensional modeling and data fusion on the multi-energy data based on a deep learning algorithm; compatible access of multi-energy equipment such as electricity, water, gas and heat is realized through modular design, real-time processing, intelligent analysis and user-friendly interaction of energy data are realized by adopting edge computing and deep learning technologies, and technical support is provided for development of smart communities and green buildings.

Description

Technical Field

[0001] The present invention belongs to the technical field of energy management, and in particular relates to a residential community energy metering device for multi-data access and integration. Background Art

[0002] As global energy shortages and environmental problems become increasingly severe, the rational management and efficient use of energy have become an important issue in social development. Especially in residential communities, due to the variety of energy consumption (such as electricity, water, gas, and heat) and the complexity of various energy equipment, traditional energy metering and management methods face the following major problems:

[0003] 1. Difficulty in accessing multi-energy devices

[0004] Existing energy metering devices are often designed for a single energy source (such as electricity meters or water meters), making it difficult to achieve unified access and management of multiple energy sources. The communication protocols and interface standards of various energy devices are not unified (such as RS485, LoRa, NB-IoT, etc.), resulting in poor device compatibility and high deployment costs.

[0005] 2. Limited data analysis capabilities

[0006] Current energy metering devices are mostly at the level of data collection and simple storage, lacking intelligent data analysis capabilities. Existing technologies make it difficult to conduct in-depth modeling and fusion analysis of multi-energy data, and are unable to provide high value-added services such as energy consumption trend prediction, anomaly detection, and energy-saving recommendations.

[0007] 3. Inadequate user experience

[0008] Traditional energy metering devices mainly display numerical values ​​and lack an intuitive visual interface, making it difficult for users to quickly understand energy consumption. The feedback mechanism for user behavior is weak, and it is impossible to promptly remind users of abnormal energy consumption or provide energy-saving suggestions.

[0009] 4. Insufficient system scalability and security

[0010] The energy management device lacks modular design, making it difficult to expand the system's functions or upgrade modules as needed. The security measures for data transmission and storage are not in place, making it vulnerable to cyber attacks or data leaks.

[0011] 5. Dual requirements of environment and policy

[0012] With the introduction of the concepts of green buildings and smart communities, the country has put forward higher requirements for energy management, and innovative technical means are needed to support refined management and energy conservation and emission reduction. Current technologies are insufficient in meeting the requirements of energy conservation and environmental protection policies, and it is difficult to support the development of smart energy systems. Summary of the invention

[0013] In order to solve the problems raised in the above-mentioned background technology, the present invention provides a residential community energy metering device with multi-data access and integration. Through modular design, it realizes compatible access to multiple energy equipment such as electricity, water, gas, and heat, and adopts edge computing and deep learning technology to realize real-time processing, intelligent analysis and user-friendly interaction of energy data, providing technical support for the development of smart communities and green buildings.

[0014] To achieve the above object, the present invention provides the following technical solutions:

[0015] A residential community energy metering device with multi-data access and integration, the device comprising:

[0016] A data acquisition module is used to collect multi-energy data in residential communities through a variety of communication interfaces, including electricity, water, gas and heat. The communication interfaces support RS485, NB-IoT, LoRa, Zigbee and Wi-Fi;

[0017] An edge computing module, connected to the data acquisition module, for preprocessing the collected multi-energy data, including data format standardization, time series reconstruction, and preliminary anomaly detection;

[0018] A data fusion module, connected to the edge computing module, for performing multi-dimensional modeling and data fusion on multi-energy data based on a deep learning algorithm;

[0019] A communication module, connected to the data acquisition module and the data fusion module, for uploading the processed data to a remote server or a local display device through multi-protocol transmission;

[0020] Power modules, including main power supply and solar backup modules, are used to provide power support for continuous operation in power outage environments;

[0021] The user interaction module is connected to the data fusion module and is used to provide users with data visualization, abnormal energy consumption alarms and personalized energy-saving suggestions.

[0022] The data acquisition module has a built-in multi-mode sensor, which can simultaneously collect multiple energy data and support real-time monitoring and high-frequency data sampling.

[0023] Among them, the edge computing module integrates ARM processor and FPGA coprocessor to realize efficient localized data processing; the data fusion module adopts Transformer deep learning algorithm to perform layered data fusion of time, space and device characteristics, and supports energy consumption trend prediction and anomaly detection; the edge computing module and data fusion module support dynamic load optimization function, and realize real-time load adjustment by analyzing energy usage.

[0024] Among them, the communication module has a built-in 5G communication unit for achieving high-speed and low-latency data transmission.

[0025] Among them, the user interaction module supports displaying energy usage to users through mobile and web applications, and provides energy optimization suggestions; the user interaction module supports generating personalized reports through big data analysis results, including energy usage rankings, energy-saving suggestions and equipment abnormality prompts.

[0026] The power module includes a solar panel, a lithium battery pack and an automatic switching circuit, which is used to seamlessly switch to the backup power supply when the main power supply fails.

[0027] The device further includes a system security module for performing AES and RSA double encryption on data transmission and performing multi-level authority management on user access.

[0028] The device has the function of diagnosing equipment faults, can monitor the operating status of the device in real time, and send out an alarm signal when an abnormality occurs.

[0029] Among them, the device adopts a modular design, and the data acquisition module, edge computing module, data fusion module, communication module and power module can be independently disassembled and upgraded; the device realizes automatic protocol switching and unified access to multiple energy equipment through the intelligent gateway integration function; the edge computing module of the device adopts a time synchronization mechanism to ensure the timing consistency of multi-source data.

[0030] The multi-protocol transmission includes one or more of MQTT, HTTP / REST and CoAP.

[0031] Compared with the prior art, the present invention has the following beneficial effects:

[0032] The present invention proposes a residential community energy metering device with multi-data access and integration. Through modular design and intelligent data processing, it solves the problems of poor data access compatibility, low analysis accuracy and insufficient user experience in the prior art, and has the following significant beneficial effects:

[0033] 1. Unified access and management of multiple energy data

[0034] The data acquisition module that supports multiple communication protocols (such as RS485, NB-IoT, LoRa, Zigbee, Wi-Fi) is used to achieve unified access to multiple energy devices such as electricity, water, gas, and heat. The automatic protocol switching function of the intelligent gateway reduces manual configuration and improves system compatibility and deployment efficiency. It avoids the access difficulties caused by differences in device protocols in traditional energy management systems. It realizes unified management and efficient data collection of multiple energy devices.

[0035] 2. Efficiency and accuracy of data processing

[0036] The edge computing module combines ARM and FPGA coprocessors to provide efficient data preprocessing capabilities, including data format standardization, noise filtering, time series reconstruction, and missing data completion. The data fusion module uses the Transformer deep learning model, which can perform multi-dimensional modeling and analysis of time, space, and device characteristics. It significantly improves the efficiency of data processing and the accuracy of analysis, ensuring the real-time and reliability of data. It reduces noise and errors in the data, providing high-quality data support for subsequent energy optimization and anomaly detection.

[0037] 3. Intelligent energy analysis and optimization

[0038] The data fusion module has a self-learning function, which can optimize model parameters based on historical data to improve the accuracy of energy consumption assessment and trend prediction. The system provides personalized energy-saving suggestions, combines user energy consumption behavior and historical records, and recommends optimization solutions (such as adjusting the equipment usage period). It helps users achieve accurate energy consumption management and energy-saving optimization. It improves users' trust in the energy management system and their stickiness to use it.

[0039] 4. Real-time anomaly detection and response

[0040] The edge computing module and data fusion module have real-time anomaly detection functions, which can quickly identify energy consumption anomalies (such as equipment failure or water and gas leakage). Through the communication module, abnormal information can be notified to users in a variety of ways (WeChat, DingTalk, SMS) at the first time. The system's response speed to emergencies is improved, and the risk of energy waste and equipment damage is reduced. Users can deal with abnormal situations in a timely manner and reduce possible economic losses.

[0041] 5. Highly flexible modular design

[0042] The modular hardware design enables the data acquisition module, edge computing module, data fusion module, communication module and power module to be independently disassembled and upgraded. The system supports on-demand replacement or expansion of modules (such as replacing the communication module to adapt to 5G network upgrades). The system has good scalability and adaptability, reducing the user's long-term maintenance costs. It meets the personalized needs of different communities or projects and increases the versatility of the equipment.

[0043] 6. Reliability of data security and rights management

[0044] The system uses AES and RSA double encryption to ensure the confidentiality and integrity of data during transmission. The multi-level permission management mechanism effectively prevents unauthorized access and improves the security of data access. It protects user privacy and system data security, and is particularly suitable for scenarios with high data security requirements. It prevents malicious attacks and data leakage, and improves the stability and reliability of the system.

[0045] 7. Excellent User Experience

[0046] The user interaction module provides users with intuitive real-time data display, historical trend analysis, personalized report generation and other functions through mobile App and Web interface. Dynamic charts and energy-saving suggestions enhance the readability and practicality of data display. It provides a convenient interactive experience and reduces user learning costs. It improves user participation through real-time alarms and optimization suggestions, and promotes the implementation of energy-saving behaviors.

[0047] 8. Energy saving and environmental protection benefits

[0048] Through intelligent analysis and optimization suggestions, it helps users reduce energy usage during peak hours and improve energy efficiency. It reduces energy waste and indirectly reduces carbon emissions. It has significant energy-saving benefits for community users and public building users. It supports the national energy conservation and emission reduction policy and provides technical support for the development of smart communities and green buildings. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] Figure 1 It is a schematic diagram of the structure and operation flow of the invention; DETAILED DESCRIPTION

[0050] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0051] See also Figure 1 , a residential community energy metering device with multi-data access and integration, the device includes:

[0052] 1. Data acquisition module, used to collect multi-energy data in residential communities through multiple communication interfaces, the energy includes electricity, water, gas and heat, the communication interface supports RS485, NB-I oT, LoRa, Zigbee and Wi-Fi; the data acquisition module includes: a multi-mode sensor interface, a data sampling circuit and a communication control unit, which can adapt to the output signal of standard energy metering equipment, the acquisition frequency can be set to 1Hz to 1kHz, and has an automatic abnormal data marking function; the communication control unit has a built-in multi-protocol parsing library to realize automatic switching and protocol parsing of RS485, NB-I oT, LoRa, Zigbee and Wi-Fi;

[0053] Specifically, the data acquisition module consists of:

[0054] Hardware architecture design

[0055] (1) Multimode sensor interface

[0056] Provides multiple standardized interfaces, such as analog voltage input, analog current input, and digital interface (I2C, SPI, UART). Supports connection to electric meters, water meters, gas meters, heat meters and other sensing devices, and adapts to multiple signal types (voltage, current, pulse signals, etc.).

[0057] (2) Data sampling circuit

[0058] High-speed analog-to-digital converter (ADC): used to sample analog signals and convert them into digital signals, with optional 12-bit or 16-bit resolution and support for high sampling rates (1kHz). Digital filter: performs noise removal on the collected signals to improve data quality. Data buffer: used to temporarily store sampled data to avoid data loss.

[0059] (3) Communication control unit

[0060] Embedded MCU (STM32 or ESP32): used to control the interfaces and data processing logic of multiple communication protocols. Support multiple communication modules: such as RS485 module, NB-IoT module, LoRa module, Zigbee module and Wi-Fi module, the module is connected to MCU via SPI or UART.

[0061] Software architecture design

[0062] (1) Multi-protocol parsing library

[0063] Implement a parsing library that supports multiple protocols, including the following functions:

[0064] RS485 protocol: Adopt Modbus RTU standard to analyze energy data and status data.

[0065] NB-I oT protocol: uses UDP or TCP protocol stack to communicate with cellular network base stations and upload collected data.

[0066] LoRa protocol: supports point-to-point communication or LoRaWAN protocol, parsing long-distance low-power data.

[0067] Zigbee protocol: supports IEEE 802.15.4 standard and builds local area network communication.

[0068] Wi-Fi protocol: Based on TCP / IP stack, parse data and upload it to the cloud.

[0069] Protocol switching logic: Each interface runs independently and can dynamically load the corresponding protocol stack. Automatically switch protocols according to the identification of the sensor or communication module.

[0070] (2) Automatic abnormal data marking

[0071] Based on statistical analysis and rule engine: detect data outside the normal range (by setting upper and lower thresholds). Analyze sudden changes or noise peaks in sampled signals.

[0072] Abnormal tag storage: Use additional fields to record abnormal status, such as "normal", "above threshold", "below threshold". Store the marked data in the local buffer or upload it through the communication module.

[0073] 2. An edge computing module connected to the data acquisition module is used to pre-process the collected multi-energy data, including data format standardization, time series reconstruction, and preliminary anomaly detection; the edge computing module integrates an ARM Cortex-A72 processor and an FPGA coprocessor, the ARM processor is used to run the data pre-processing algorithm, and the FPGA coprocessor is used to accelerate time series reconstruction and anomaly detection; pre-processing includes: unifying data units, removing noise signals, and interpolating missing data;

[0074] Specifically, the edge computing module consists of:

[0075] Hardware architecture design

[0076] ARM Cortex-A72 processor: used to run general data processing tasks, including data format standardization, unit conversion, and anomaly detection algorithms. Provides efficient computing power and supports complex data preprocessing logic.

[0077] FPGA coprocessor: Accelerates computationally intensive tasks such as time series reconstruction and noise filtering. Leverages parallel computing power to handle time series operations on large amounts of energy data.

[0078] Storage and interface: Cache (DDR4 memory): used to store collected data and intermediate processing results. Data transmission interface (such as SPI, I2C, UART): communicate with the data acquisition module.

[0079] Software architecture design

[0080] (1) Data format standardization: unify the formats and units of multi-energy data, such as:

[0081] Electricity: Convert sampled voltage / current to power / energy.

[0082] Water: Convert volume flow rate to standard units (e.g., cubic meters per hour).

[0083] Gas: Convert pressure data to flow.

[0084] Thermal: Calculates thermal power and converts to kWh.

[0085] Select the corresponding unit conversion rules according to the energy type and sensor identification. Use the ARM processor to run the unit conversion algorithm to standardize the raw data into a unified format. The standardized data is stored in the memory or local cache.

[0086] (2) Time series reconstruction: Synchronize data from multiple sources and different sampling frequencies into a unified time series.

[0087] FPGA implements timestamp alignment logic: each piece of data is attached with a collection timestamp. Data of different frequencies are aligned through interpolation or sampling resampling. Data reconstruction algorithm (running on FPGA): Linear interpolation: fills the gaps between samples. Sliding window average: smooths time series data. Missing value completion: predicts and completes the breakpoint data using adjacent points or historical data.

[0088] (3) Preliminary anomaly detection: Detecting abnormal points in energy data, such as mutations, extreme values, or noise.

[0089] FPGA is used for real-time detection: Setting dynamic threshold range (based on historical mean ± standard deviation). Detecting data points exceeding the threshold and marking them as abnormal.

[0090] ARM is used for anomaly analysis: anomalies (equipment failure, sensor error) are classified through statistical analysis or rule engines. Anomalies are marked and stored in a specific buffer or an alarm mechanism is triggered.

[0091] (4) Noise removal: remove high-frequency noise or random interference signals from the data.

[0092] FPGA hardware filter: Use a low-pass filter (such as Butterworth filter) to remove high-frequency noise. Use a median filter to handle instantaneous interference.

[0093] ARM processor checksum storage: Check the filtered data to ensure it is smooth and undistorted.

[0094] (5) Data caching and output: The processed data is cached in local storage and uploaded to the data fusion module regularly. ARM is responsible for scheduling data transmission to the cache. The data in the cache is transmitted to the data fusion module through the interface.

[0095] 3. A data fusion module, connected to the edge computing module, for performing multi-dimensional modeling and data fusion on multi-energy data based on a deep learning algorithm; the data fusion module adopts a Transformer deep learning model, whose input is multi-dimensional data related to time, space and equipment characteristics, and whose output is the fused energy usage assessment result; the data fusion module has a self-learning function and can continuously optimize the modeling results according to the newly added data;

[0096] Specifically, the data fusion module consists of:

[0097] Hardware and software architecture

[0098] (1) Hardware support: GPU or NPU (neural network processing unit): used to accelerate the reasoning and training of deep learning models. High-speed memory (DDR4 / DDR5): to store model parameters and temporary calculation results. Network interface: to support data input of edge computing modules and remote acquisition of training data sets.

[0099] (2) Software support: Deep learning frameworks such as TensorFlow, PyTorch, or ONNX Runtime to run Transformer models. Data management module: responsible for receiving and standardizing processed multi-energy data to ensure uniform input format.

[0100] Model design and implementation

[0101] (1) Input data design

[0102] The input dimensions include:

[0103] Temporal characteristics: The time series of collected data represents the temporal variation pattern of energy use.

[0104] Spatial characteristics: The geographical distribution or regional attributes of energy equipment, such as floor, room number, etc.

[0105] Device characteristics: device type, collection frequency, data unit, etc.

[0106] Data structure: Time series data (power, flow): processed as fixed-length window data (such as the past 24 hours).

[0107] Spatial and device characteristics: treated as categorical variables or embedded vectors.

[0108] (2) Model architecture design (Transformer model)

[0109] The encoder part includes:

[0110] Input: time series (as main input), spatial and device characteristics (processed by embedding). Multi-head self-attention mechanism: captures the correlation between different time points in the time series. Position encoding: adds the time position information of the time series.

[0111] Fusion layer: The spatial and device characteristic embedding vectors are combined with the time series features through a linear transformation.

[0112] Decoder part:

[0113] Output: Fusion energy usage assessment results (such as predicted total energy consumption, abnormal distribution probability).

[0114] Model output: Overall energy consumption trend (e.g., forecast for the next 24 hours). Usage percentage of different energy types. Abnormal probability distribution.

[0115] (3) Model optimization and training

[0116] Training method: Supervised training using historical data.

[0117] Loss function: A weighted loss that combines regression (error between predicted and actual values) and classification (outlier detection).

[0118] Optimizer: Use Adam or its variants to optimize model parameters.

[0119] Self-learning mechanism: The model continuously records new data (such as the latest energy usage records) while running.

[0120] Periodic incremental training: Fine-tune the model based on new data to avoid overfitting.

[0121] 4. A communication module, connected to the data acquisition module and the data fusion module, for uploading the processed data to a remote server or a local display device through multi-protocol transmission; the communication module includes a built-in 5G communication unit and an Ethernet interface, supports multi-protocol transmission (MQTT, HTTP / REST, CoAP); and is equipped with a data caching mechanism, which can temporarily store data and automatically resume uploading when the network is interrupted;

[0122] 5. Power module, including main power supply and solar backup module, used to provide power support for continuous operation in a power outage environment; the power module includes a solar panel (power range 10W-50W), a lithium battery pack (capacity 5Ah-20Ah) and an automatic switching circuit to achieve seamless switching between main power and solar power supply;

[0123] 6. User interaction module, connected to the data fusion module, used to provide users with data visualization, abnormal energy usage alarm and personalized energy-saving suggestions; the user interaction module displays energy usage through mobile applications and web interfaces, uses interactive dynamic charts, supports real-time monitoring and historical trend analysis; is equipped with energy consumption ranking, abnormal reminder and energy-saving suggestion functions, and generates personalized reports based on big data analysis results;

[0124] Specifically, the user interaction module consists of:

[0125] (1) Hardware and system architecture

[0126] Server and cloud platform: Use cloud services (such as Alibaba Cloud, Tencent Cloud, and Huawei Cloud) to host the system. Data storage uses relational databases (such as MySQL) and distributed storage (such as OSS object storage). The computing engine uses local services (such as Alibaba Cloud Funct i on Compute) for data analysis and report generation.

[0127] Mobile and Web interfaces: Front-end development: Use Vue.js or React to build interfaces, supporting responsive design. Back-end communication: Connect with RESTful API via HTTPS, using widely supported interface specifications.

[0128] (2) Data visualization

[0129] Real-time update of energy usage data. Supports multi-dimensional display (such as daily energy consumption, peak and valley electricity price distribution). Data flow architecture: Use WebSocket protocol to achieve real-time data push. The chart library uses ECharts to render dynamic charts.

[0130] (3) Abnormal energy usage alarm

[0131] According to the set rules, abnormal energy consumption behavior is detected and message reminders are triggered. Optional multi-channel alarm methods are provided (WeChat, DingTalk, SMS).

[0132] (4) Personalized energy-saving suggestions

[0133] Generate recommendations based on residents’ energy usage habits, such as reminding users to use electricity during off-peak hours. Check usage habits of high-energy-consuming devices.

[0134] (5) Historical trend analysis

[0135] Supports viewing energy consumption trends by day, week, month, and year. Provides sub-item analysis (such as electricity, water, gas, and heat are displayed separately).

[0136] (6) Energy consumption ranking and report generation

[0137] Energy consumption ranking: compare the energy consumption of users with other residents in the community. Report generation: generate Chinese energy consumption reports, support WeChat / DingTalk sharing.

[0138] 7. System security module, which is used to perform AES and RSA double encryption on data transmission and multi-level authority management on user access; equipment fault diagnosis function, which can monitor the operation status of the device in real time and send out alarm signals when abnormalities occur;

[0139] 8. Modular design enables the data acquisition module, edge computing module, data fusion module, communication module and power module to be independently disassembled and upgraded; the device realizes automatic protocol switching and unified access to multiple energy devices through the intelligent gateway integration function;

[0140] Specifically, the modular design method is:

[0141] (1) Modular hardware design

[0142] Independent hardware interface: Each module is interconnected through a standardized interface to ensure pluggability and interchangeability. Common industrial-grade interfaces are used: Data transmission: SPI, I2C, UART, Ethernet interface. Power transmission: Standard DC power supply interface to ensure independent power supply of the module.

[0143] (2) Modular architecture example

[0144] Data acquisition module: responsible for the collection and preliminary processing of energy data, independent power supply and interface.

[0145] Edge computing module: receives and processes data from the acquisition module, and supports independent expansion of computing capabilities.

[0146] Data fusion module: processes fusion tasks from the edge computing module and can be replaced with a higher performance model.

[0147] Communication module: You can choose cellular communication (such as 5G module) or Ethernet communication module for replacement.

[0148] Power module: Supports power modules with different power requirements and can be replaced with high-efficiency power supplies or backup batteries.

[0149] (3) Module disassembly and upgrade

[0150] Each module is individually packaged in a removable slot (such as DIN rail mounting or plug-in connector). When upgrading, only the corresponding module needs to be replaced without replacing the entire system.

[0151] (4) Module interconnection protocol

[0152] Use standard communication protocols: Internal communication: I2C for low-speed configuration, SPI for high-speed data transmission. External communication: Ethernet interface or wireless communication module.

[0153] Data transmission protocol specification: Data acquisition module and edge computing module: Protobuf or JSON is used for data encapsulation. Heartbeat signals are used between modules to detect connection status and ensure module independence.

[0154] The working process of the residential community energy metering device with multi-data access and integration is as follows:

[0155] Step 1: Device startup and initialization

[0156] The device is powered on and the power module provides stable operating power.

[0157] The system runs a self-test program to check the connection status of the data acquisition module, edge computing module, data fusion module, communication module, and user interaction module. Verify whether the sensors and interfaces are working properly.

[0158] The communication module connects to the network and establishes communication with the cloud server via 5G or Ethernet. It detects the network status and prepares to transmit data.

[0159] Step 2: Data Collection

[0160] Connect multiple energy devices, and the data acquisition module establishes communication with electricity meters, water meters, gas meters, heat meters and other devices in the residential community. Automatically identify device type and protocol. Real-time data acquisition, regularly collect energy data from devices, such as: Electricity: voltage, current, power. Water: flow, total water consumption. Gas: pressure, flow. Heat: temperature, thermal power. Data sampling frequency is set according to demand.

[0161] Initial processing: standardize the format of the collected data and mark the initial outliers.

[0162] Data transmission, sending formatted data to the edge computing module through a unified protocol.

[0163] Step 3: Data preprocessing (edge ​​computing module)

[0164] The format is unified, and the multi-energy data sent by the data acquisition module is further processed to ensure consistent units and unified formats.

[0165] Time series reconstruction aligns different energy data in time to generate a unified time series.

[0166] Noise filtering, applies low-pass filters and median filters to remove noise and short-term anomalies in the data.

[0167] Missing data completion: missing data in the collection process is completed through interpolation or machine learning algorithms.

[0168] Preliminary anomaly detection uses a rule engine or simple model to detect obviously abnormal data and mark it as "suspicious".

[0169] Data output, the preprocessed data is transmitted to the data fusion module.

[0170] Step 4: Data fusion and analysis

[0171] Multidimensional data modeling, using the Transformer deep learning model, taking time series, spatial features, and device characteristics as input.

[0172] Energy consumption assessment, analyzing the current consumption of different energy sources (such as electricity, water, gas, and heat). Calculate the overall energy consumption and generate energy consumption trend forecasts.

[0173] Abnormal pattern recognition, combining historical data and real-time data to identify abnormal energy usage patterns (such as equipment failures).

[0174] Energy-saving suggestion generation: based on energy usage behavior analysis, optimization suggestions are generated (such as adjusting air conditioning temperature or reducing electricity consumption during peak hours).

[0175] Data output, the analysis results are transmitted to the communication module, ready to be uploaded to the server or sent to the user terminal.

[0176] Step 5: Data transmission and caching (communication module)

[0177] Data transmission: Use MQTT, HTTP / REST or CoAP protocol to upload data to the remote server. Choose 5G or Ethernet for transmission according to network conditions.

[0178] Cache management: When the network is interrupted, data is temporarily stored in a local cache (such as memory or Flash storage). When the network is restored, the unuploaded data is automatically extracted from the cache and the transmission is completed.

[0179] Server storage: After receiving data, the server stores it in the database to provide support for user query and analysis.

[0180] Step 6: User Interaction and Feedback

[0181] Data display: users can view energy data through mobile apps or web terminals, and display the energy consumption of electricity, water, gas and heat in real time. Dynamic charts show energy consumption trends (such as the past 7 days or 30 days).

[0182] Abnormal alarm: When abnormal energy consumption (such as sudden power surge) is detected, an alarm is triggered: an alarm message is sent to the user's WeChat, DingTalk or SMS. The alarm details and suggested handling methods are displayed on the App interface.

[0183] Energy saving suggestions: Provide personalized energy saving suggestions, such as reminding users to use electricity during off-peak hours and recommend shutting down devices that have not been used for a long time.

[0184] Historical analysis: users can view historical energy consumption data by day, week, month, and year. Generate personalized energy consumption reports, including energy consumption overview, energy saving suggestions, and abnormal records.

[0185] Step 7: System maintenance and upgrade

[0186] Module upgrade: users can replace modules according to their needs (such as upgrading communication modules to support faster networks). The system automatically detects the new module and completes initialization.

[0187] Fault monitoring: The system monitors the operating status of each module in real time. When a module abnormality is detected (such as the edge computing module temperature is too high), an alarm is triggered.

[0188] User authority management: different users (such as administrators and ordinary users) access different functions, and the corresponding interface and operation options are displayed according to the authority.

[0189] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A residential community energy metering device with multi-data access and integration, characterized in that: The device includes: A data acquisition module is used to collect multi-energy data in residential communities through multiple communication interfaces, including electricity, water, gas and heat. The communication interfaces support RS485, NB-IoT, LoRa, Zigbee and Wi-Fi; An edge computing module, connected to the data acquisition module, for preprocessing the collected multi-energy data, including data format standardization, time series reconstruction, and preliminary anomaly detection; A data fusion module, connected to the edge computing module, for performing multi-dimensional modeling and data fusion on multi-energy data based on a deep learning algorithm; A communication module, connected to the data acquisition module and the data fusion module, for uploading the processed data to a remote server or a local display device through multi-protocol transmission; Power modules, including main power supply and solar backup modules, are used to provide power support for continuous operation in power outage environments; The user interaction module is connected to the data fusion module and is used to provide users with data visualization, abnormal energy consumption alarms and personalized energy-saving suggestions.

2. The residential community energy metering device with multi-dimensional data access and integration according to claim 1 is characterized by: The data acquisition module has a built-in multi-mode sensor, which can simultaneously collect multiple energy data and support real-time monitoring and high-frequency data sampling.

3. The residential community energy metering device with multi-dimensional data access and integration according to claim 1 is characterized by: The edge computing module integrates an ARM processor and an FPGA coprocessor to achieve efficient localized data processing; the data fusion module adopts the Transformer deep learning algorithm to perform layered data fusion on time, space and device characteristics, and supports energy consumption trend prediction and anomaly detection; the edge computing module and the data fusion module support dynamic load optimization function, and realize real-time load adjustment by analyzing energy usage.

4. The residential community energy metering device with multi-data access and integration according to claim 1 is characterized by: The communication module has a built-in 5G communication unit for achieving high-speed, low-latency data transmission.

5. The residential community energy metering device with multi-dimensional data access and integration according to claim 1 is characterized by: The user interaction module supports displaying energy usage to users through mobile and web applications and provides energy optimization suggestions; The user interaction module supports generating personalized reports through big data analysis results, including energy usage rankings, energy-saving suggestions, and equipment abnormality prompts.

6. The residential community energy metering device with multi-dimensional data access and integration according to claim 1 is characterized by: The power supply module includes a solar panel, a lithium battery pack and an automatic switching circuit, which is used for seamlessly switching to a backup power supply when the main power supply fails.

7. The residential community energy metering device with multi-data access and integration according to claim 1 is characterized by: The device further includes a system security module for performing AES and RSA double encryption on data transmission and performing multi-level authority management on user access.

8. The residential community energy metering device with multi-dimensional data access and integration according to claim 1 is characterized by: The device has an equipment fault diagnosis function, can monitor the operating status of the device in real time, and send out an alarm signal when an abnormality occurs.

9. The residential community energy metering device with multi-dimensional data access and integration according to claim 1 is characterized by: The device adopts a modular design, and the data acquisition module, edge computing module, data fusion module, communication module and power module can be independently disassembled and upgraded; the device realizes automatic protocol switching and unified access to multiple energy devices through the intelligent gateway integration function; the edge computing module of the device adopts a time synchronization mechanism to ensure the timing consistency of multi-source data.

10. The residential community energy metering device with multi-dimensional data access and integration according to claim 1 is characterized by: The multi-protocol transmission includes one or more of MQTT, HTTP / REST and CoAP.

Images