Non-intrusive modular edge intelligent load monitoring equipment

Abstract

The invention provides non-intrusive modular edge intelligent load monitoring equipment, and relates to the technical fields of edge artificial intelligence, Internet of Things, detection and sensing, precision metering, electrical parameter measurement and acquisition, digital signal processing, equipment health management and predictive maintenance, algorithm theory, mode recognition, state estimation and the like. The system mainly solves the problems of high data safety, high real-time performance and high reliability of monitoring equipment in multiple aspects of energy consumption analysis, equipment health state monitoring, fire-fighting monitoring, environmental protection monitoring, forbidden load identification and alarm, energy consumption metering and monitoring and the like of scenes such as residential houses, apartments, public facilities, commercial places, industrial and mining enterprises, power transmission, power supply and power transformation, public lighting and the like; and the system has the functions of forbidden blacklist load identification and alarm, load energy consumption monitoring and analysis, load energy consumption control, load equipment health state monitoring, predictive maintenance support and the like.

Description

Technical Field

[0001] The present invention relates to technical fields such as edge artificial intelligence, Internet of Things, detection and sensing, precision metering, electrical parameter measurement and acquisition, digital signal processing, equipment health management and predictive maintenance, algorithm theory, pattern recognition and state estimation, and specifically to non-intrusive modular edge intelligent monitoring technology for systems such as load identification, energy consumption monitoring, equipment health management, and fire monitoring. Background Art

[0002] Monitoring equipment has a wide range of application needs in residential buildings, apartments, public facilities, commercial places, industrial and mining enterprises, power transmission and power supply substation, public lighting, energy consumption analysis, quality inspection, equipment health status monitoring, fire monitoring, environmental monitoring, disabled load identification and alarm, energy consumption metering and monitoring, etc.:

[0003] a) Fires are caused by poor insulation or poor contact of power supply circuits and equipment due to aging, which results in arcs; fires are caused by connecting high-power or flammable and explosive electric bicycle batteries, high-power electric furnaces and other blacklisted prohibited equipment. At present, the fire department does not have effective technical means to conduct autonomous detection, inspection, positioning and alarm for the above-mentioned hidden dangers within its jurisdiction. It is urgently needed to use technologies and equipment that can be easily deployed to perform load identification, arc detection and energy consumption monitoring on multiple sets of energy-consuming equipment by a single set of equipment.

[0004] b) In the context of global energy conservation and emission reduction and striving to meet carbon targets, it is necessary to conduct autonomous inspection, monitoring, positioning, energy consumption metering, energy consumption pattern analysis, alarm and intelligent power restriction control on electrical equipment within the jurisdiction, especially high-energy consumption equipment; the existing user-side equipment such as prepaid meters, two-way metering meters and smart meters cannot meet the relevant needs; there are no relevant technologies and products in China; there is an urgent need for technology and equipment that can identify the load, measure and control the energy consumption of multiple sets of energy-consuming equipment by a single set of equipment.

[0005] c) Existing traditional intrusive monitoring equipment needs to be connected to the system during installation. While it naturally becomes an energy-consuming device itself, it also has a great impact on the working parameters and reliability of the monitored equipment. At the same time, there are many problems such as line modification, installation of a large number of meters, high difficulty in line maintenance, and heavy workload. There is an urgent need for non-intrusive monitoring technology and equipment that can fully exploit and utilize sensor detection data through intelligent processing and a single set of monitoring equipment to monitor multiple loads.

[0006] d) The current algorithm models and computing power for load identification, energy consumption analysis and equipment health monitoring rely on cloud resources, which severely restricts their operating efficiency, data security and system reliability. There is an urgent need for edge intelligent power load identification and energy consumption monitoring technologies and equipment with locally deployed algorithm models and computing power to ensure system operating efficiency, data security and system reliability.

[0007] e) Monitoring technology and equipment are used in different fields such as residential buildings, public buildings, commercial places, industrial and mining enterprises, and their requirements in terms of functions, scales, parameter categories, dynamic ranges, and measurement and control levels vary greatly. It is obviously impossible for integrated equipment to meet the above differentiated needs. The modular design of monitoring equipment can facilitate the maintenance and upgrading of user-side products while ensuring the flexibility of system functions and specifications, rapid development, low cost, and high system reliability; more importantly, it is convenient for product standardization, product replacement, and technology upgrades; it is an inevitable trend in the development of measurement and control technology and equipment. Summary of the invention

[0008] The purpose of the present invention is to provide a non-invasive edge intelligent multifunctional monitoring module device that can meet the above requirements, has the characteristics of rapid construction, easy maintenance and upgrade, and can meet the comprehensive requirements of data security, high real-time performance, high reliability, high maintainability and easy upgrading of various monitoring equipment in current scenarios such as civil residences, apartments, public facilities, commercial places, industrial and mining enterprises, power transmission and power supply transformation, public lighting, etc., and the shortcomings of existing equipment.

[0009] It mainly includes non-intrusive sensing and data acquisition modules based on Internet of Things technology, load control modules, edge artificial intelligence modules, and display and control terminal modules.

[0010] After the present invention adopts the above-mentioned module structure and function, compared with the existing integrated non-intrusive load identification, energy consumption monitoring, fire monitoring and other systems, on the basis of its non-intrusive, load identification, energy consumption monitoring and equipment health management functions and performance, its beneficial effects are: through overall planning, the structure and function of the monitoring system are reasonably divided, and it is modularized, which is convenient for the development and iterative development of monitoring products and improves its comprehensive performance. At the same time, it is convenient for the construction, maintenance and upgrading of the monitoring system, saving the construction and operation and maintenance costs; by adopting edge artificial intelligence platform technology, the data of the monitoring system and its related resources are more secure and confidential, and the monitored objects can be adapted. The scope of monitoring is wider; through data acquisition and sensing modules, high-frequency electromagnetic signals directly related to the status of circuits, connectors, switches, sockets and monitored objects are obtained. Through intelligent analysis, characteristic signals such as arcs, sparks, health status of load equipment, and access and charging of electric bicycle batteries can be accurately extracted, which provides technical means for fire monitoring and real-time alarms, timely predictive maintenance of load equipment, and maximum avoidance or reduction of fire losses; by adopting various types of data acquisition and sensing modules, the deficiency of the original equipment that can only monitor a certain characteristic parameter is overcome, and it has multiple functions to adapt to the monitoring needs of various application scenarios. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] The drawings described herein are used to provide a further understanding of the present invention and constitute a part of the present application, but do not constitute an improper limitation of the present invention. In the drawings:

[0012] Figure 1 It is a schematic diagram of the module structure of the non-intrusive modular edge intelligent load monitoring device of the present invention;

[0013] Figure 2 It is a schematic diagram of the types of sensing and data acquisition modules of the present invention;

[0014] Figure 3 It is a schematic diagram of the working principle of the sensor and data acquisition module of the present invention;

[0015] Figure 4 It is a schematic diagram of the working principle of the load control module of the present invention;

[0016] Figure 5 It is a schematic diagram of the edge artificial intelligence module of the present invention;

[0017] Figure 6 It is a schematic diagram of the deployment method of the present invention in an application scenario.

[0018] Description of reference numerals:

[0019] 1- Temperature and humidity sensor, 2- Gas composition and content sensor, 3- Speed ​​sensor, 4- Vibration sensor, 5- Flow sensor. DETAILED DESCRIPTION

[0020] The present invention will be described in detail below in conjunction with the accompanying drawings and specific embodiments, wherein the illustrative embodiments and descriptions are only used to explain the present invention but are not intended to limit the present invention.

[0021] like Figure 1 As shown, the non-intrusive modular edge intelligent load monitoring device described in this specific embodiment is composed of a sensing and data acquisition module, a load control module, an edge artificial intelligence module, and a display and control terminal module. These modules are interconnected through LAN, WI-FI, etc. to form a load monitoring system, which performs information sensing and data acquisition, exchange and command transmission, and has functions such as load identification, energy consumption monitoring and health management.

[0022] like Figure 2 As shown, in order to reliably perform load identification, energy consumption monitoring and health management, the present invention provides 8 types of sensing and data acquisition modules, including current and voltage, temperature and humidity, gas composition and content, flow, speed, pressure, impact and vibration. The current and voltage sensing and data acquisition module uses non-invasive methods such as inductance and capacitance to sense the current in the wire and the voltage between the conductors, obtains the real-time current and voltage signals input by the power supply circuit when the load is working, and converts them into digital signals, which are then transmitted to the edge artificial intelligence module through the communication interface; the edge artificial intelligence module identifies the type of load and its specifications, calculates energy consumption, and identifies whether there is an abnormal arc in the power supply circuit based on the data provided by the current and voltage sensing and data acquisition module, and generates prohibited load access, excessive energy consumption and abnormal arc alarm signals and instructions to cut off or limit energy consumption. The other seven sensing and data acquisition modules all adopt non-invasive principles to obtain the temperature and humidity, gas composition and content, flow, speed, pressure, impact and vibration information required for load health management, and are selected according to the needs of the load identification and monitoring system; the edge artificial intelligence module comprehensively determines the load health status and change trends based on the data provided by these seven modules, and gives the time and plan for preventive maintenance.

[0023] like Figure 3 As shown in the figure, the sensing and data acquisition module consists of sensors (including inductance and capacitance, temperature and humidity, gas composition and content, flow, speed, pressure, impact and vibration and other sensitive elements), signal conditioning units, bandpass filtering units, signal holding units, microprocessors with A / D and other functions, communication extensions, SD card read / write units, WI-FI, LAN and RS485 and other parts. It mainly performs related signal sensing (non-electrical signal is converted into electrical signal), amplification and other signal conditioning, bandpass filtering, A / D data acquisition, and finally sends the collected data to the edge artificial intelligence module through LAN, RS485 or WI-FI, and stores (or periodically overwrites) the collected data through the SD card read / write unit.

[0024] like Figure 4 As shown, the load control module consists of a microprocessor, WI-FI and LAN communication units, light intensity sensors, drivers, power switches and thyristor current limiters. Its function is to control the power supply on and off of the load or to limit its current through the power switch under the intelligent control of the edge artificial intelligence module, light intensity control, or under the control of the display and control module. The triggering conditions are that the data analysis results provided by the edge artificial intelligence module to the sensor and data acquisition module are that the load is illegally connected, there is an abnormal arc in the load power supply circuit, or the load is current limited after exceeding the specified power; or the management personnel believe that it is necessary to intervene manually.

[0025] like Figure 5As shown in the figure, the edge artificial intelligence module consists of two parts: hardware and software. The hardware mainly consists of a central processing unit, a digital signal processor, an accelerator, a static memory, a dynamic memory, an input and output interface, an SD card memory, a WI-FI / LAN / RS485 communication port, etc. Among them, the digital signal processor is used to perform fast Fourier transform and wavelet transform on the signals provided by the sensor and data acquisition module, and the accelerator is used to accelerate the CPU and model reasoning; the hardware is designed in the form of an industrial core single-board computer. The software is mainly composed of an operating system, an edge intelligent model and a feature library (knowledge base). The edge model mainly extracts features from the time domain, frequency domain and specific domain of the wavelet function of the digital signal provided by the sensor and data acquisition module, the Fourier transform signal and the wavelet transform signal provided by the digital signal processor according to a specific reasoning model, to form the time domain, frequency domain and specific domain of the measured object (load to be identified, arc and energy consumption, load health status, etc.) and their inter-related feature vectors, and ensures that the relationship between the feature vector and the measured object is one-to-one from multiple dimensions such as the properties of the elements (including functions) of the feature vector, numerical accuracy, and correlation. At the same time, the edge model must also standardize the mapping reasoning logic between the feature vector obtained by the test and the feature elements in the feature library, including the tolerance and correction coefficient of the feature vector elements (including functions) and their related numerical values, to ensure correct identification of load types and accurate energy consumption measurement. The feature library (knowledge base) is a collection of several feature vectors. As long as the edge artificial intelligence module has been fully learned and trained, the feature library should contain the feature vectors of various objects that have been or may be encountered, that is, the measured feature vectors can be mapped to the unique elements in the feature library. In this way, it can be guaranteed from a theoretical and technical point of view that there is a one-to-one correspondence between the object to be tested (identified) - the feature vector obtained by the test - the feature element of the feature library, that is, the uniqueness, correctness and energy consumption measurement accuracy of the test and identification results can be guaranteed. In actual work, if there is a deviation in detection and identification, it must be that a new test and identification object (new use, new principle product) has been connected to the energy consumption loop, which can be solved by re-learning and training the edge artificial intelligence module to enrich its feature library data.

[0026] The display and control terminal module consists of two parts: hardware and software. In terms of hardware, mobile phones, IPADs or laptops with mobile communication, WI-FI or other networking functions can meet the requirements. In terms of software, in addition to normal display and browsing of test data and identification object pictures, analog headers, and dynamic change curves, the control interface of the control device associated with the corresponding device must be displayed according to user needs. Users can control and allocate power consumption according to their needs, or even cut off the power supply.

[0027] Taking the identification of power loads (equipment), energy consumption monitoring, health status monitoring and blacklist power load monitoring as an example, the deployment of the non-intrusive modular edge intelligent load monitoring equipment involved in the present invention in application scenarios such as residential areas and workshops is as follows: Figure 6 As shown. Among them, the non-intrusive sensing and data acquisition module is installed at the power supply entrance of the factory workshop and the residential building in the community, and is used to identify the power load of the entire workshop and the entire residential building, and collect energy consumption monitoring data; the load control module is installed at the front end of the power load power input, and is used to execute the control instructions issued by the display and control module or the edge artificial intelligence module; the edge artificial intelligence module can be installed in the workshop, residential building or other places that the manager considers safe; the display and control module is carried by the person in charge or manager of this work, and can grasp the monitoring dynamics and alarm information at any time, and take corresponding measures to deal with it. If the health status of the equipment needs to be monitored, the sensors that collect information related to the health status will be deployed in the corresponding parts. Figure 6 Taking the health monitoring of load 1 (air compressor) as an example, the types of sensors that should be installed are shown.

[0028] The above description is only a preferred embodiment of the present invention, so all equivalent changes or modifications made according to the structure, characteristics and principles described in the scope of the patent application of the present invention are included in the scope of the patent application of the present invention.

Claims

1. A non-intrusive modular edge intelligent load monitoring device, including a sensing and data acquisition module, a load control module, an edge artificial intelligence module, and a display and control terminal module.

2. The non-intrusive modular edge intelligent load monitoring device according to claim 1 is characterized in that It has functions such as disabled blacklist load identification and alarm, load energy consumption monitoring and analysis, load energy consumption control, load equipment health status monitoring and predictive maintenance support, etc.

3. The non-intrusive modular edge intelligent load monitoring device according to claim 1 is characterized in that The energy consumption of the load can be controlled by intelligent control of the edge artificial intelligence module, light intensity control, or under the control of the display and control module, the power supply of the load is controlled on and off through the power switch, or by current limiting control.

4. The non-intrusive modular edge intelligent load monitoring device according to claim 1 is characterized in that The edge artificial intelligence model has the function of extracting features of the time domain, frequency domain and specific domain of the wavelet function of the digital signal provided by the sensor and data acquisition module, the Fourier transform signal and the wavelet transform signal provided by the digital signal processor according to a specific reasoning model, so as to form the feature vectors of the time domain, frequency domain and specific domain of the measured object (load to be identified, arc and energy consumption, load health status, etc.) and their mutual correlation.

5. The non-intrusive modular edge intelligent load monitoring device according to claim 1 is characterized in that A feature library (knowledge base) constructed by the method described in claim 4.

Images