CN-122024392-A - Intelligent monitoring system integrating load identification and electric fire early warning

CN122024392ACN 122024392 ACN122024392 ACN 122024392ACN-122024392-A

Abstract

The invention relates to the technical field of electrical safety monitoring, in particular to an intelligent monitoring system integrating load identification and electrical fire early warning. The cloud terminal comprises a data acquisition module, a data processing module, a transmission module and a cloud server. The cloud end realizes the accurate identification of load equipment such as an air conditioner and a refrigerator based on an electrical appliance characteristic database and a machine learning algorithm, performs time sequence analysis and multidimensional diagnosis by combining history and real-time data on the basis, judges whether hidden troubles such as electric leakage, aging or overlarge harmonic waves exist in the equipment or not, and performs grading early warning according to the development trend of the hidden troubles. The problems that the traditional electric fire monitoring cannot locate specific hidden danger equipment and has low early warning accuracy are solved, the span from line-level alarm to equipment-level accurate early warning is realized, and hidden danger investigation efficiency and early warning prospective are remarkably improved.

Inventors

DAI JIE
MENG YATING
WANG ZONGLI

Assignees

苏州万户安物联网科技有限公司

Dates

Publication Date: 20260512
Application Date: 20260303

Claims (7)

1. The intelligent monitoring system integrating load identification and electric fire early warning is characterized by comprising a data acquisition module, a data processing module, a transmission module and a cloud server, wherein the data processing module is connected with the data acquisition module, the transmission module is connected with the data processing module, and the cloud server is connected with the transmission module; the data acquisition module is used for acquiring electricity consumption data of the monitoring point, wherein the electricity consumption data comprises current waveform data acquired by a current transformer, residual current data acquired by a residual current transformer and temperature data acquired by a temperature sensor; The data processing module is used for carrying out signal conditioning, characteristic extraction and compression packaging processing on the power utilization data received from the data acquisition module; the transmission module is used for sending the data packet processed by the data processing module to the cloud server through the communication module; The cloud server is used for analyzing received data to realize load identification and electric fire early warning, wherein the load identification is used for carrying out feature matching and pattern identification on current waveform data in the electricity utilization data, distinguishing load equipment and being capable of identifying parallel running states of various loads in the same loop, and the electric fire early warning comprises the step of carrying out time sequence comparison analysis on the identified load equipment by combining historical running data and real-time electricity utilization data, diagnosing whether hidden danger exists or not, and classifying the hidden danger according to dynamic development trend and threshold deviation degree of the hidden danger.
2. The intelligent monitoring system for integrating load identification and electrical fire pre-warning of claim 1, wherein, The data acquisition module is also used for acquiring voltage waveform data through the voltage module.
3. The intelligent monitoring system for integrating load identification and electrical fire pre-warning of claim 1, wherein, The communication module in the transmission module is a communication module supporting 485 communication, 4G/5G communication or NB-IoT communication.
4. The intelligent monitoring system for integrating load identification and electrical fire pre-warning of claim 1, wherein, The cloud server comprises an analysis unit, a grading unit and a self-learning unit, wherein the analysis unit is connected with the transmission module, the grading unit is connected with the analysis unit, and the self-learning unit is connected with the analysis unit; The analysis unit is used for matching the load characteristic vector extracted from the electricity utilization data with a multidimensional characteristic template through a machine learning classification algorithm so as to realize the accurate identification and classification of load equipment; The grading unit is used for receiving the identified load equipment information, calling a corresponding safety analysis model aiming at the load equipment, and integrating the leakage characteristics, the temperature rise curve and the harmonic spectrum data of the load equipment to perform fusion diagnosis and trend prediction to generate grading early warning information; The self-learning unit is used for optimizing model parameters and expansion feature templates of the machine learning classification algorithm according to the continuously accessed anonymized electricity consumption data.
5. The intelligent monitoring system for integrating load identification and electrical fire pre-warning of claim 4, The classifying unit judges that the hidden danger is serious if the hidden danger parameter exceeds the safety threshold and is in an accelerated deterioration trend, judges that the hidden danger is moderate if the hidden danger parameter exceeds the safety threshold and changes smoothly, and judges that the hidden danger is general if the hidden danger parameter is close to but does not exceed the safety threshold.
6. The intelligent monitoring system for integrating load identification and electrical fire pre-warning of claim 1, wherein, The intelligent monitoring system integrating load identification and electric fire early warning further comprises a data display module, wherein the data display module is connected with the cloud server; the data display module is used for receiving and displaying the analysis result and the grading early warning information generated by the cloud server and supporting the user to confirm or process feedback of the early warning information; The data display module comprises at least one of a PC end management platform, a mobile phone APP and a mobile phone applet.
7. The intelligent monitoring system for integrating load identification and electrical fire pre-warning of claim 1, wherein, The data acquisition module adopts a non-invasive installation mode, and the current transformer is an open type transformer and can be sleeved on a tested line.

Description

Intelligent monitoring system integrating load identification and electric fire early warning Technical Field The invention relates to the technical field of electrical safety monitoring, in particular to an intelligent monitoring system integrating load identification and electrical fire early warning. Background Electrical fire is one of the common fire types, and the key of prevention is to monitor and pre-warn electrical hidden trouble in time. Currently, common electrical fire monitoring devices mainly monitor overall electrical parameters in a power distribution loop, such as total current, residual current, temperature, etc., and alarm when a set threshold is exceeded. The method can reflect the overall abnormality of the circuit, but has obvious defects that once an alarm is given, a user cannot easily judge which specific electric equipment (such as an air conditioner, a refrigerator, a washing machine and the like) fails, so that the investigation is difficult, meanwhile, the judgment is easy to be interfered by normal load fluctuation only by relying on a threshold value, false alarm or missing alarm is caused, particularly early hidden danger (such as slow insulation aging, gradual increase of harmonic waves and the like) of the gradual development inside a single equipment cannot be identified, and the early warning accuracy and the prospective are limited. In order to realize the distinction and monitoring of electric equipment, various load identification methods have been developed in the prior art, and mainly can be divided into two main categories, namely invasive and non-invasive. Non-invasive load monitoring (NILM) is a typical and widely studied technological path therein that decomposes and identifies individual sub-loads by analyzing the total current, voltage waveforms and their derivative features. Specific implementation means include, but are not limited to, simple comparison based on steady state characteristics (such as steady state current amplitude, active/reactive power), identification by using transient characteristics (such as current surge waveform when the equipment is started and stopped), extraction of unique 'electric fingerprint' of the electric appliance by analyzing high-frequency harmonic components of the current waveform, and classification and state identification of complex load mixed waveforms by applying a machine learning algorithm (such as a support vector machine and a hidden Markov model). The main application scenarios of these technologies focus on home energy management, electricity usage behavior analysis and energy conservation optimization, aiming at answering the question of "when, what kind of electric appliances use how much electricity". However, the load identification technology is mainly used for energy consumption metering and management optimization, and the technical objective and the electrical safety early warning have essential differences. These techniques fail to systematically blend with deep electrical safety hazard diagnostic functions. Specifically, after the load type is identified, the existing scheme does not construct a multi-dimensional security risk evaluation system for the specific equipment, so that the electric fire early warning still stays at a fuzzy circuit level, and advanced early warning and hierarchical management of the specific equipment cannot be realized accurately. Disclosure of Invention The invention aims to provide an intelligent monitoring system integrating load identification and electric fire early warning, which solves the problem of integrating a load accurate identification technology with a device-level electric potential safety hazard deep diagnosis and grading early warning mechanism. In order to achieve the above purpose, the invention provides an intelligent monitoring system integrating load identification and electric fire early warning, which comprises a data acquisition module, a data processing module, a transmission module and a cloud server, wherein the data processing module is connected with the data acquisition module, the transmission module is connected with the data processing module, and the cloud server is connected with the transmission module; the data acquisition module is used for acquiring electricity consumption data of the monitoring point, wherein the electricity consumption data comprises current waveform data acquired by a current transformer, residual current data acquired by a residual current transformer and temperature data acquired by a temperature sensor; The data processing module is used for carrying out signal conditioning, characteristic extraction and compression packaging processing on the power utilization data received from the data acquisition module; the transmission module is used for sending the data packet processed by the data processing module to the cloud server through the communication module; The cloud server is used for analyzing received data