CN118033219B

CN118033219B - Power utilization management method and system based on AI current fingerprint

Info

Publication number: CN118033219B
Application number: CN202410431414.3A
Authority: CN
Inventors: 丁继强; 盛锴; 祝培旺; 林德尚; 袁狄平
Original assignee: Guangdong Zhongcheng Zhilian Technology Co ltd
Current assignee: Guangdong Zhongcheng Zhilian Technology Co ltd
Filing date: 2024-04-11
Publication date: 2024-07-02
Anticipated expiration: 2044-04-11

Abstract

The invention belongs to the technical field of electric digital data processing, and discloses an electric management method and system based on AI current fingerprints. According to the invention, the current in the circuit is monitored, when the surge current is monitored, the surge current value is obtained, the working current value is determined according to the surge current value, the current sampling sequence in the circuit is obtained, the current sampling sequence is input into a current analysis model to obtain a current fingerprint code, the current fingerprint code is decomposed based on the decomposition level and the working current value to obtain an electric appliance current fingerprint code, the electric appliance current fingerprint code is monitored, when the electric appliance current fingerprint code is changed into an abnormal electric appliance current fingerprint code, the abnormal electric appliance corresponding to the abnormal electric appliance current fingerprint code is determined, the abnormal electric appliance is subjected to power utilization management, and the power utilization condition of the electric appliances is monitored by the current fingerprint code in the circuit, so that the power utilization management of single electric appliance in the circuit is realized.

Description

Power utilization management method and system based on AI current fingerprint

Technical Field

The invention relates to the technical field of electric digital data processing, in particular to an electric management method and system based on AI current fingerprints.

Background

Along with the continuous development of society, more and more electric equipment enters a family, however, difference exists between the electric equipment and the electric equipment can be worn gradually in the use process, potential safety hazards exist in the electric equipment, the current electric safety of the electric equipment is controlled globally, when abnormal electric equipment exists in a circuit, the safety control of the whole circuit is carried out in order to protect the circuit safety, and then other normal electric equipment can be influenced, the living and production requirements can be seriously influenced, and by adopting the method, the abnormal electric equipment cannot be detected, so that the potential safety hazards cannot be timely isolated, and the electric safety of other electric equipment in the circuit is ensured.

The foregoing is provided merely for the purpose of facilitating understanding of the technical solutions of the present invention and is not intended to represent an admission that the foregoing is prior art.

Disclosure of Invention

The invention mainly aims to provide an AI current fingerprint-based electricity management method and system, and aims to solve the technical problem that the prior art cannot manage electricity of a single electric appliance in a circuit.

In order to achieve the above object, the present invention provides a power consumption management method based on AI current fingerprint, the method comprising the steps of:

monitoring current in a circuit, acquiring a surge current value when the surge current is monitored, and determining a working current value according to the surge current value;

Acquiring a current sampling sequence in a circuit, and inputting the current sampling sequence into a current analysis model to obtain a current fingerprint code;

Decomposing the current fingerprint code based on the decomposition level number and the working current value to obtain a current fingerprint code of the electric appliance;

and monitoring the current fingerprint code of the electrical appliance, determining an abnormal electrical appliance corresponding to the current fingerprint code of the abnormal electrical appliance when the current fingerprint code of the electrical appliance changes into the current fingerprint code of the abnormal electrical appliance, and carrying out power utilization management on the abnormal electrical appliance.

Optionally, the acquiring a current sampling sequence in the circuit includes:

obtaining an analog current signal in a circuit from a current sensor, and converting the analog current signal into a digital current signal through an analog-to-digital converter;

Filtering the digital current signal to obtain a current filtering signal;

And packaging the current filtering signal to obtain a current sampling sequence.

Optionally, the inputting the current sampling sequence into a current analysis model to obtain a current fingerprint code includes:

Inputting the current sampling sequence into a current analysis model, and carrying out dimension ascending on the current sampling sequence based on the sampling time of the current sampling sequence to obtain a three-dimensional current sampling sequence;

Carrying out load identification on the three-dimensional current sampling sequence to obtain a load identification result, and dividing the three-dimensional current sampling sequence into a plurality of data sets according to the load identification result;

And respectively carrying out data processing on the plurality of data sets to obtain the current fingerprint code.

Optionally, before the current sampling sequence is input to the current analysis model to obtain the current fingerprint code, the method further includes:

Acquiring historical current sampling sequence data, and determining characteristic parameters of the historical current sampling sequence data, wherein the historical current sampling sequence data comprises the historical current sampling sequence data and historical current fingerprint codes corresponding to the historical current sampling sequence data;

carrying out noise reduction and normalization treatment on the characteristic parameters to obtain preprocessing data of a historical current sampling sequence;

dividing the preprocessing data of the historical current sampling sequence into a training set and a verification set;

inputting the training set into an initial current analysis model, and carrying out dimension ascending on the historical current sampling sequence preprocessing data in the training set through a kernel function to obtain historical three-dimensional current sampling preprocessing data;

Training the initial current analysis model based on the historical three-dimensional current sampling preprocessing data to obtain an intermediate current analysis model;

Inputting the verification set into the intermediate current analysis model, and obtaining a predicted current fingerprint code based on a penalty factor of the intermediate current analysis model;

Obtaining a current fingerprint code deviation value according to the predicted current fingerprint code and the actual current fingerprint code in the predicted set;

and outputting the intermediate current analysis model as a current analysis model when the current fingerprint code deviation value is within a preset deviation range.

Optionally, after obtaining the current fingerprint code deviation value according to the predicted current fingerprint code and the actual current fingerprint code in the predicted set, the method further includes:

When the current fingerprint code deviation value is out of a preset deviation range, determining an actual deviation value according to the current fingerprint code deviation value and the preset deviation range;

Determining a deviation vector according to the actual deviation value;

Correcting error data based on the deviation vector and the actual deviation value;

Updating the punishment factor of the intermediate current analysis model according to the deviation correcting data to obtain an updated current analysis model, inputting the training set into the updated current analysis model, and returning to the step of carrying out dimension ascending on the historical current sampling sequence preprocessing data in the training set through a kernel function to obtain historical three-dimensional current sampling preprocessing data until the current fingerprint code deviation value is within a preset deviation range.

Optionally, before decomposing the current fingerprint code based on the decomposition level number and the working current value to obtain the current fingerprint code of the electrical appliance, the method further includes:

monitoring the current in the circuit, and determining the change type of the current in the circuit and the current change value in the circuit when the current in the circuit changes;

determining the number of the electric appliances actually working in the circuit according to the change type of the current and the current change value in the circuit;

and obtaining the decomposition level number according to the number of the actually-operated electric appliances.

Optionally, after determining the abnormal electrical appliance corresponding to the current fingerprint code of the abnormal electrical appliance and performing power consumption management on the abnormal electrical appliance, the method further includes:

determining identity information of the abnormal electrical appliance according to the current fingerprint code of the abnormal electrical appliance;

Determining the abnormal electricity utilization type of the abnormal electricity utilization device according to the current fingerprint code of the abnormal electricity utilization device;

generating electrical appliance anomaly information according to the identity information of the anomaly electrical appliance and the anomaly electrical appliance type of the anomaly electrical appliance;

and sending the abnormal information of the electrical appliance to an intelligent terminal, and recording and warning the abnormal behavior of the abnormal electrical appliance.

In addition, in order to achieve the above object, the present invention also provides an AI-current-fingerprint-based electricity management system for executing an AI-current-fingerprint-based electricity management method, the AI-current-fingerprint-based electricity management system comprising:

The current monitoring module is used for monitoring current in the circuit, acquiring a surge current value when the surge current is monitored, and determining a working current value according to the surge current value;

The current analysis module is used for acquiring a current sampling sequence in the circuit, and inputting the current sampling sequence into the current analysis model to obtain a current fingerprint code;

The current decomposition module is used for decomposing the current fingerprint code based on the working current value to obtain a current fingerprint code of the electric appliance;

and the electricity consumption management module is used for monitoring the current fingerprint code of the electrical appliance, determining the abnormal electrical appliance corresponding to the current fingerprint code of the abnormal electrical appliance when the current fingerprint code of the electrical appliance changes into the current fingerprint code of the abnormal electrical appliance, and carrying out electricity consumption management on the abnormal electrical appliance.

Optionally, the electricity management system based on AI current fingerprint further includes:

The model training module is used for acquiring historical current sampling sequence data and determining characteristic parameters of the historical current sampling sequence data, wherein the historical current sampling sequence data comprises the historical current sampling sequence data and historical current fingerprint codes corresponding to the historical current sampling sequence data;

Optionally, the electricity consumption warning module is used for determining the identity information of the abnormal electrical appliance according to the current fingerprint code of the abnormal electrical appliance; determining the abnormal electricity utilization type of the abnormal electricity utilization device according to the current fingerprint code of the abnormal electricity utilization device; generating electrical appliance anomaly information according to the identity information of the anomaly electrical appliance and the anomaly electrical appliance type of the anomaly electrical appliance; and sending the abnormal information of the electrical appliance to an intelligent terminal, and recording and warning the abnormal behavior of the abnormal electrical appliance.

According to the invention, the current in the circuit is monitored, when the surge current is monitored, the surge current value is obtained, the working current value is determined according to the surge current value, the current sampling sequence in the circuit is obtained, the current sampling sequence is input into a current analysis model to obtain a current fingerprint code, the current fingerprint code is decomposed based on the decomposition level and the working current value to obtain an electric appliance current fingerprint code, the electric appliance current fingerprint code is monitored, when the electric appliance current fingerprint code is changed into an abnormal electric appliance current fingerprint code, the abnormal electric appliance corresponding to the abnormal electric appliance current fingerprint code is determined, the abnormal electric appliance is subjected to power utilization management, and the power utilization condition of the electric appliances is monitored by the current fingerprint code in the circuit, so that the power utilization management of single electric appliance in the circuit is realized.

Drawings

FIG. 1 is a schematic structural diagram of an electricity management method device based on AI current fingerprint in a hardware operating environment according to an embodiment of the present invention;

FIG. 2 is a flowchart of a power management method based on AI current fingerprint according to a first embodiment of the present invention;

FIG. 3 is a schematic diagram of a household intelligent circuit breaker arrangement according to an embodiment of the electricity management method based on AI current fingerprints;

FIG. 4 is a diagram of a data set after kernel dimension increase according to an embodiment of the power management method based on AI current fingerprint;

FIG. 5 is a flowchart of a second embodiment of an electricity management method based on AI current fingerprint according to the present invention;

fig. 6 is a block diagram of a power management system based on AI current fingerprint according to a first embodiment of the invention.

The achievement of the objects, functional features and advantages of the present invention will be further described with reference to the accompanying drawings, in conjunction with the embodiments.

Detailed Description

It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the scope of the invention.

Referring to fig. 1, fig. 1 is a schematic structural diagram of an electricity management device based on AI current fingerprint in a hardware operation environment according to an embodiment of the present invention.

As shown in fig. 1, the AI current fingerprint-based electricity management apparatus may include: a processor 1001, such as a central processing unit (Central Processing Unit, CPU), a communication bus 1002, a user interface 1003, a network interface 1004, a memory 1005. Wherein the communication bus 1002 is used to enable connected communication between these components. The user interface 1003 may include a Display, an input unit such as a Keyboard (Keyboard), and the optional user interface 1003 may further include a standard wired interface, a wireless interface. The network interface 1004 may optionally include a standard wired interface, a Wireless interface (e.g., a Wireless-Fidelity (Wi-Fi) interface). The Memory 1005 may be a high-speed random access Memory (Random Access Memory, RAM) or a stable nonvolatile Memory (NVM), such as a disk Memory. The memory 1005 may also optionally be a storage device separate from the processor 1001 described above.

It will be appreciated by those skilled in the art that the structure shown in fig. 1 does not constitute a limitation of the electricity management device based on AI current fingerprints, and may include more or fewer components than shown, or certain components in combination, or a different arrangement of components.

As shown in fig. 1, an operating system, a network communication module, a user interface module, and an electricity management program based on AI current fingerprints may be included in the memory 1005 as one storage medium.

In the electricity management apparatus based on AI current fingerprint shown in fig. 1, the network interface 1004 is mainly used for data communication with a network server; the user interface 1003 is mainly used for data interaction with a user; the processor 1001 and the memory 1005 in the electricity management apparatus based on AI current fingerprint may be provided in the electricity management apparatus based on AI current fingerprint, and the electricity management apparatus based on AI current fingerprint invokes the electricity management program based on AI current fingerprint stored in the memory 1005 through the processor 1001, and executes the electricity management method based on AI current fingerprint provided in the embodiment of the invention.

The embodiment of the invention provides an electricity management method based on AI current fingerprints, and referring to FIG. 2, FIG. 2 is a flow chart of a first embodiment of the electricity management method based on AI current fingerprints.

In this embodiment, the electricity management method based on AI current fingerprint includes the following steps:

step S10: and monitoring the current in the circuit, acquiring a surge current value when the surge current is monitored, and determining a working current value according to the surge current value.

It should be noted that, the execution body of the embodiment is an AI-current-fingerprint-based electricity management device, where the AI-current-fingerprint-based electricity management device has functions of data processing, data communication, program running, and the like, and the AI-current-fingerprint-based electricity management device may be an integrated controller, a control computer, or other devices with similar functions, which is not limited in this embodiment.

It can be understood that the surge current is the current characteristic of the electrical equipment at the moment of switching on, when the power supply is switched on, the peak value of the surge current is far greater than the steady-state input current due to the rapid charging of the input filter capacitor, the surge current suddenly appears and can reach the peak value in a short time, but can quickly drop to the normal working current level, and for a single electrical appliance, a certain relation exists between the surge current generated by the single electrical appliance and the working current when the single electrical appliance is added into a circuit.

In a specific implementation, for the whole electricity consumption condition, an intelligent circuit breaker can be installed on a main circuit or a branch circuit of a circuit, and the intelligent circuit breaker can adaptively perform electricity consumption control according to the current electricity consumption condition in the circuit, so that the circuit is in an open circuit state, and electricity consumption of an electrical appliance on the circuit is controlled. In a normal circuit, since the surge current only exists at a moment when the electric appliance is added into the circuit, the current in the circuit can be monitored, and since the surge current which is short and higher than the normal working current is generated when the electric appliance is connected into the circuit, when the surge current is not generated, the current value in the circuit is in a relatively stable state, so that whether the electric appliance is connected into the circuit or not can be determined according to the current value in the circuit. Correspondingly, when the surge current is monitored, the surge current value can be obtained from the current in the circuit, and then according to the corresponding relation between the surge current value and the working current value:

thus, the actual operating current value of the current consumer can be obtained.

Step S20: and acquiring a current sampling sequence in the circuit, and inputting the current sampling sequence into a current analysis model to obtain a current fingerprint code.

The current sampling sequence is important data indispensable in circuit analysis and power system optimization, current in the circuit can be monitored in real time through a current sensor or measuring equipment during the current sampling sequence, the current is recorded according to a certain time interval to form the current sampling sequence, the current fingerprint code is used for converting the current information into 2-system data codes according to the current information in the circuit, and the 2-system current fingerprint code can be converted into hexadecimal current fingerprint code for facilitating data processing and data storage.

In a specific implementation, the intelligent circuit breaker comprises a current sensor or a measuring device for collecting current data, wherein the intelligent circuit breaker can be arranged according to specific use conditions, and can be arranged on a trunk of a circuit or a branch of the circuit. The difference between the current data of the position where the intelligent circuit breaker is located and the number of the appliances corresponding to the collected current is that no matter where the intelligent circuit breaker is arranged in the circuit, in this embodiment, a home circuit will be described by taking a main circuit position where the intelligent circuit breaker is arranged in the circuit as an example. Referring to fig. 3, fig. 3 is a schematic diagram of a household intelligent circuit breaker arrangement. In the household power utilization circuit, an intelligent breaker can be arranged at a main road, and can be arranged on a branch circuit where each power utilization device is located according to specific use conditions.

Because the household power grid is single-phase power, a current sampling sequence for collecting the single-phase power is a one-dimensional time sequence, each sampling point represents a current value passing through a single-phase wire at a certain moment, the obtained current value is Alternating Current (AC), the size and the direction of the current value change along with time, the load change in the household power grid can be reflected through the current sampling sequence, in the current sampling sequence, each value represents the current size, a positive value represents the current direction the same as a reference direction, a negative value represents the current direction opposite to the reference direction, and based on the current sampling sequence, the information of the waveform, the frequency, the phase and the like of the current can be clearly known, so that the working state of the circuit is determined. After the current sampling sequence is determined, the current sampling sequence can be input into a current analysis model, so that the current analysis model analyzes the current sampling sequence to obtain current fingerprint codes, and the obtained current fingerprint codes are a set of current fingerprint codes of a plurality of electric appliances in the current circuit.

Further, the current sampling sequence in the acquisition circuit includes:

Filtering the digital current signal to obtain a current filtering signal;

In a specific implementation, an analog current signal in a circuit is obtained according to a current sensor in the intelligent circuit breaker, and the analog current signal is converted into a digital current signal through an analog-to-digital converter, wherein the specific process is that the analog current signal is connected to an input pin of the analog-to-digital converter (ADC). The ADC is used to convert a continuous analog current signal into a discrete digital current signal, and the converted digital current signal is represented in binary form, and may be converted into hexadecimal for optimizing the throughput. After the digital current signal is obtained, the converted digital current signal is input into a filter. The purpose of the filter is to remove noise and interference in the signal and to improve the signal-to-noise ratio of the signal. And selecting a proper filtering algorithm according to actual requirements. Common filtering algorithms include arithmetic average filtering, depolarization average filtering, weighted average filtering, moving average filtering, and the like. These algorithms may be selected and adjusted based on the characteristics and requirements of the signal. In this embodiment, a sliding average filtering is taken as an example to describe, first, the size N of a sliding window needs to be determined, the size of the window determines the number of samples in each window, and the formula of the sliding average filtering algorithm is as follows:

Wherein, In order to input a signal to the device,In order to output the signal,Is the current number of samples.

Then, before the formal filtering, it is necessary to initialize the sliding window, calculate the average value in the window at the time of initializing the sliding window, and take the average value in the window as the initial value of the sliding window. In the filtering process, the current window is continuously slid, a new digital current signal sample is added to the tail end of the sliding window every time the new digital current signal sample is received, the forefront sample in the window is removed, namely all samples in the sliding window are subjected to average calculation according to a first-in first-out (FIFO) principle, and a new filtered signal value is obtained. This can be achieved by a simple arithmetic average, i.e. adding all sample values, then dividing by the number of samples, and outputting the calculated sliding average as a filtered signal, which can be expressed as: samples within window areA sliding average value of. This process is repeated continuously, outputting a current filtered signal. After the current filtering signal is obtained, the current filtering signal is packaged according to a time sequence to form a current sampling sequence, and each sampling point represents a current value at a certain moment.

Further, the inputting the current sampling sequence into a current analysis model to obtain a current fingerprint code includes:

In a specific implementation, the current sampling sequence is input into a current analysis model, the current analysis model is a model constructed by a support vector machine algorithm, and the current sampling sequence belongs to linear inseparable data, so that the current sampling sequence is up-dimensioned based on the sampling time of the current sampling sequence and a kernel function in the current analysis model to obtain a three-dimensional current sampling sequence, and the current sampling sequence is changed into the data inseparable. The kernel function may select a polynomial kernel function:。

Wherein, As coefficients of the kernel function,Is a constant term which is used to determine the degree of freedom,Representing the order of the polynomial.

After the current sampling sequence is up-scaled into the three-dimensional current sampling sequence, load identification can be carried out on the three-dimensional current sampling sequence, the load type of the load is determined, a load identification result is obtained, and the three-dimensional current sampling sequence is divided into a plurality of data sets according to the load identification result, and the implementation mode is as follows: projecting the data along the hyperplane direction, returning the data to the plane coordinates, and processing the data, wherein referring to fig. 4, fig. 4 is a schematic diagram of a data set after kernel function dimension increase. For the data set D which is linear and inseparable, the data set D can be linear and segmentable after going through the dimension ascending, the hyperplane which can segment the points is quite large, when the minimum distance between one hyperplane and two sides is maximized, the generalization error of algorithm classification is minimum, namely the optimal hyperplane, and the hyperplane is defined and corresponding data sets are listed, such as:

Wherein, Is a normal vector of the hyperplane,WhereinFor the dimensions of the dimension vector,Is the offset of the hyperplane,Is the coordinate point in the variable set D.

The distance from any point in the sample space to the hyperplane isIn the followingWhen the optimal solution exists, the hyperplane classification is the most accurate.

Wherein,。

After the optimal hyperplane is determined, the data processing is carried out on a plurality of data sets divided by the three-dimensional current sampling sequence to obtain current fingerprint codes, and the current fingerprint codes at the moment are current fingerprint codes in the whole main road and can be regarded as current fingerprint codes obtained after the combination of a plurality of electric appliances.

Step S30: and decomposing the current fingerprint code based on the decomposition level number and the working current value to obtain the current fingerprint code of the electric appliance.

The number of decomposition levels refers to the number of target decomposition levels after decomposing the current fingerprint code, for example, when the number of decomposition levels is 3, the current fingerprint code is decomposed into 3 current fingerprint codes of the electric appliance.

In a specific implementation, when the current fingerprint code is decomposed, the current fingerprint code is decomposed according to the decomposition level number and the working current value, and before the decomposition, the method further comprises the steps of monitoring the current in the circuit, and determining the change type of the current in the circuit and the current change value in the circuit when the current in the circuit changes; determining the number of the electric appliances actually working in the circuit according to the change type of the current and the current change value in the circuit; and obtaining the decomposition level number according to the number of the actually-operated electric appliances. The method comprises the following steps: detecting the current in the circuit, when the electric appliance is connected to the circuit and the power supply is started, generating an instantaneous current which is far greater than the normal working current, namely the surge current, on the branch circuit and the main circuit where the electric appliance is positioned, when the main circuit current is restored to a stable state, determining a current change value according to the current steady state current and the steady state current before the surge current is generated, the change type of the current change value at the moment belongs to the increase type, so that the fact that the number of the current electric appliances in the circuit is increased can be determined, correspondingly, the newly increased number of the electric appliances can be determined according to the number of the surge currents in the circuit, and the number of the electric appliances and the working currents of the electric appliances can be recorded in a database for recording the working currents. Similarly, when the current in the main circuit is monitored, if the current value is found to be smaller, the current difference between the front and the rear can be reduced, the determination of which electric appliance is disconnected from the circuit can be made, the number of electric appliances in the current circuit can be obtained, and the decomposition level can be obtained according to the number of electric appliances.

Step S40: and monitoring the current fingerprint code of the electrical appliance, determining an abnormal electrical appliance corresponding to the current fingerprint code of the abnormal electrical appliance when the current fingerprint code of the electrical appliance changes into the current fingerprint code of the abnormal electrical appliance, and carrying out power utilization management on the abnormal electrical appliance.

It should be noted that, in the normal use configuration of the electric appliance, unique current information is generated according to the current power consumption condition, which includes information such as frequency, phase and current value of current, so that for a single electric appliance, the corresponding current fingerprint is stable, but when the electric appliance is abnormal, the current fingerprint deviates from the current stable state, and is the current fingerprint code of the abnormal electric appliance.

In the specific implementation, when the current fingerprint code of the electrical appliance is monitored, if the current fingerprint code of the electrical appliance is found to be changed into the abnormal electrical appliance code, the abnormal electrical appliance corresponding to the current fingerprint code of the abnormal electrical appliance can be determined, and the abnormal electrical appliance can be subjected to power consumption management. The method specifically comprises the following steps: determining identity information of the abnormal electrical appliance according to the current fingerprint code of the abnormal electrical appliance; determining the abnormal electricity utilization type of the abnormal electricity utilization device according to the current fingerprint code of the abnormal electricity utilization device; generating electrical appliance anomaly information according to the identity information of the anomaly electrical appliance and the anomaly electrical appliance type of the anomaly electrical appliance; the abnormal information of the electrical appliance is sent to an intelligent terminal, such as a mobile phone, an intelligent watch, a bracelet, a tablet personal computer and the like, abnormal behaviors of the abnormal electrical appliance are recorded and alerted, the abnormal behaviors can be stored in a local database or a cloud database, and meanwhile, the abnormal information is sent to the intelligent terminal and used for reminding a user of the abnormal electrical behaviors of the current electrical appliance, and a circuit path of the current abnormal electrical appliance can be disconnected in time.

According to the embodiment, the current in the circuit is monitored, when the surge current is monitored, the surge current value is obtained, the working current value is determined according to the surge current value, the current sampling sequence in the circuit is obtained, the current sampling sequence is input into the current analysis model to obtain the current fingerprint code, the current fingerprint code is decomposed based on the decomposition level and the working current value to obtain the current fingerprint code of the electric appliance, the current fingerprint code of the electric appliance is monitored, when the current fingerprint code of the electric appliance changes into the current fingerprint code of the abnormal electric appliance, the abnormal electric appliance corresponding to the current fingerprint code of the abnormal electric appliance is determined, the power utilization management is carried out on the abnormal electric appliance, and the power utilization condition of the electric appliance is monitored through the current fingerprint code in the circuit, so that the power utilization management of the single electric appliance in the circuit is realized.

Referring to fig. 5, fig. 5 is a flowchart of a second embodiment of a power consumption management method based on AI current fingerprint according to the present invention.

Based on the first embodiment, the electricity management method based on AI current fingerprint of the present embodiment further includes, before the step S20:

Step S201: acquiring historical current sampling sequence data, and determining characteristic parameters of the historical current sampling sequence data, wherein the historical current sampling sequence data comprises the historical current sampling sequence data and historical current fingerprint codes corresponding to the historical current sampling sequence data;

step S202: carrying out noise reduction and normalization treatment on the characteristic parameters to obtain preprocessing data of a historical current sampling sequence;

step S203: dividing the preprocessing data of the historical current sampling sequence into a training set and a verification set;

Step S204: inputting the training set into an initial current analysis model, and carrying out dimension ascending on the historical current sampling sequence preprocessing data in the training set through a kernel function to obtain historical three-dimensional current sampling preprocessing data;

step S205: training the initial current analysis model based on the historical three-dimensional current sampling preprocessing data to obtain an intermediate current analysis model;

step S206: inputting the verification set into the intermediate current analysis model, and obtaining a predicted current fingerprint code based on a penalty factor of the intermediate current analysis model;

step S207: obtaining a current fingerprint code deviation value according to the predicted current fingerprint code and the actual current fingerprint code in the predicted set;

step S208: and outputting the intermediate current analysis model as a current analysis model when the current fingerprint code deviation value is within a preset deviation range.

In a specific implementation, the current analysis mode is obtained by training an initial current analysis model, and when the initial current analysis model is trained, characteristic parameters in the historical current sampling data are determined by acquiring historical current sampling sequence data, wherein the historical current sampling sequence data comprise historical current sampling sequence data and historical current fingerprint codes corresponding to the historical current sampling sequence data, and noise reduction and normalization processing are performed on the characteristic parameters, and specifically can be as follows: removing noise in the characteristic parameters according to the characteristic parameters by a weighted recursive average algorithm, wherein the formula of the weighted recursive average algorithm is as followsWhereinIn order to obtain the result after the filtering,The filter coefficient is a filter coefficient, and the value range is (0.8-0.99). After the filtered characteristic parameters are obtained, the characteristic parameters are preprocessed through a standard deviation normalization method, interference data in the characteristic parameters are removed, and preprocessing data of a historical current sampling sequence are obtained. Dividing the historical current sampling sequence preprocessing data into a training set and a verification set according to a data proportion of 9:1, inputting the training set into an initial current analysis model, and carrying out dimension ascending on the historical current sampling sequence preprocessing data in the training set through a kernel function to obtain historical three-dimensional current sampling preprocessing data, wherein the kernel function is。

Wherein,As coefficients of the kernel function,Is a constant term which is used to determine the degree of freedom,Representing the order of the polynomial.

Training an initial current analysis model based on historical three-dimensional current sampling preprocessing data to obtain an intermediate current analysis model, inputting a verification set into the intermediate current analysis model, and obtaining a predicted current fingerprint code based on a penalty factor of the intermediate current analysis model. And obtaining a current fingerprint code deviation value according to the predicted current fingerprint code and the actual current fingerprint code in the prediction set, comparing the current fingerprint code deviation value with a preset deviation range, and outputting the intermediate current analysis model as a current analysis model when the current fingerprint code deviation value is in the preset deviation range.

Further, after obtaining the current fingerprint code deviation value according to the predicted current fingerprint code and the actual current fingerprint code in the predicted set, the method further includes:

Determining a deviation vector according to the actual deviation value;

In a specific implementation, in order to ensure the accuracy of the results of the amperometric model, the loss function may be usedWherein, the method comprises the steps of, wherein,Is the predicted value of the model and,Is a true value. When the current fingerprint code deviation value is out of the preset deviation range, determining an actual deviation value according to the current fingerprint code deviation value and the preset deviation range, specifically, when the current fingerprint code deviation value is a difference value between the current fingerprint code and an upper critical value and a lower critical value of the preset deviation range, if the current fingerprint code is larger than the upper critical value or smaller than the lower critical value, indicating that the current code value is out of the preset deviation range, at the moment, calculating the current fingerprint code and the upper critical value or the lower critical value of the preset deviation range to obtain the deviation value, wherein the positive and negative values of the deviation value are deviation vectors, and further obtaining deviation correction data according to the actual deviation value of the deviation vector. The deviation rectifying data are the same as the deviation value, but the positive and negative conditions are opposite. Updating punishment factors of the intermediate current analysis model according to the deviation correction data to obtain an updated current analysis model, inputting a training set into the updated current analysis model, and returning to the step of carrying out dimension ascending on historical current sampling sequence pretreatment data in the training set through a kernel function to obtain historical three-dimensional current sampling pretreatment data until a current fingerprint code deviation value is within a preset deviation range.

According to the embodiment, the initial current analysis model is trained, and the model is updated through the loss function, so that the output value of the model can be more fitted with the true value, meanwhile, the current analysis model can be continuously self-learned according to actual use conditions, the current analysis model is obtained through training according to actual conditions, and then an accurate current fingerprint code is output.

Referring to fig. 6, fig. 6 is a block diagram of a first embodiment of the electricity management system based on AI current fingerprinting of the present invention.

As shown in fig. 6, the electricity management system based on AI current fingerprint according to the embodiment of the present invention includes:

The current monitoring module 10 is used for monitoring current in the circuit, acquiring a surge current value when the surge current is monitored, and determining a working current value according to the surge current value;

The current analysis module 20 is used for acquiring a current sampling sequence in the circuit, and inputting the current sampling sequence into the current analysis model to obtain a current fingerprint code;

The current decomposition module 30 is configured to decompose the current fingerprint code based on the working current value to obtain a current fingerprint code of the electrical appliance;

and the electricity consumption management module 40 is used for monitoring the electricity consumption current fingerprint code, determining an abnormal electricity consumption corresponding to the abnormal electricity consumption current fingerprint code when the electricity consumption current fingerprint code is changed into the abnormal electricity consumption current fingerprint code, and carrying out electricity consumption management on the abnormal electricity consumption.

Therefore, the actual working current value of the current electric appliance can be obtained

The intelligent circuit breaker comprises a current sensor or measuring equipment for collecting current data, wherein the intelligent circuit breaker can be arranged according to specific use conditions, and can be arranged on a trunk of a circuit or a branch of the circuit. No matter where the intelligent circuit breaker is arranged in a circuit, the intelligent circuit breaker can collect current data of the position, and the difference is the number of electric appliances corresponding to the collected current. Because the household power grid is single-phase power, a current sampling sequence for collecting the single-phase power is a one-dimensional time sequence, each sampling point represents a current value passing through a single-phase wire at a certain moment, the obtained current value is Alternating Current (AC), the size and the direction of the current value change along with time, the load change in the household power grid can be reflected through the current sampling sequence, in the current sampling sequence, each value represents the current size, a positive value represents the current direction the same as a reference direction, a negative value represents the current direction opposite to the reference direction, and based on the current sampling sequence, the information of the waveform, the frequency, the phase and the like of the current can be clearly known, so that the working state of the circuit is determined. After the current sampling sequence is determined, the current sampling sequence can be input into a current analysis model, so that the current analysis model analyzes the current sampling sequence to obtain current fingerprint codes, and the obtained current fingerprint codes are a set of current fingerprint codes of a plurality of electric appliances in the current circuit.

When the current fingerprint code is decomposed, the current fingerprint code is decomposed according to the decomposition level number and the working current value, and before the decomposition, the current fingerprint code also comprises the steps of monitoring the current in the circuit, and determining the change type of the current in the circuit and the current change value in the circuit when the current in the circuit changes; determining the number of the electric appliances actually working in the circuit according to the change type of the current and the current change value in the circuit; and obtaining the decomposition level number according to the number of the actually-operated electric appliances. The method comprises the following steps: detecting the current in the circuit, when the electric appliance is connected to the circuit and the power supply is started, generating an instantaneous current which is far greater than the normal working current, namely the surge current, on the branch circuit and the main circuit where the electric appliance is positioned, when the main circuit current is restored to a stable state, determining a current change value according to the current steady state current and the steady state current before the surge current is generated, the change type of the current change value at the moment belongs to the increase type, so that the fact that the number of the current electric appliances in the circuit is increased can be determined, correspondingly, the newly increased number of the electric appliances can be determined according to the number of the surge currents in the circuit, and the number of the electric appliances and the working currents of the electric appliances can be recorded in a database for recording the working currents. Similarly, when the current in the main circuit is monitored, if the current value is found to be smaller, the current difference between the front and the rear can be reduced, the determination of which electric appliance is disconnected from the circuit can be made, the number of electric appliances in the current circuit can be obtained, and the decomposition level can be obtained according to the number of electric appliances.

When the current fingerprint code of the electrical appliance is monitored, if the current fingerprint code of the electrical appliance is found to be changed into the abnormal electrical appliance code, the abnormal electrical appliance corresponding to the current fingerprint code of the abnormal electrical appliance can be determined, and the abnormal electrical appliance can be subjected to power consumption management. The method specifically comprises the following steps: determining identity information of the abnormal electrical appliance according to the current fingerprint code of the abnormal electrical appliance; determining the abnormal electricity utilization type of the abnormal electricity utilization device according to the current fingerprint code of the abnormal electricity utilization device; generating electrical appliance anomaly information according to the identity information of the anomaly electrical appliance and the anomaly electrical appliance type of the anomaly electrical appliance; the abnormal information of the electrical appliance is sent to an intelligent terminal, such as a mobile phone, an intelligent watch, a bracelet, a tablet personal computer and the like, abnormal behaviors of the abnormal electrical appliance are recorded and alerted, the abnormal behaviors can be stored in a local database or a cloud database, and meanwhile, the abnormal information is sent to the intelligent terminal and used for reminding a user of the abnormal electrical behaviors of the current electrical appliance, and a circuit path of the current abnormal electrical appliance can be disconnected in time.

Further, the electricity management system based on AI current fingerprint further includes:

The model training module 50 is configured to obtain historical current sampling sequence data, determine a characteristic parameter of the historical current sampling sequence data, and the historical current sampling sequence data includes the historical current sampling sequence data and a historical current fingerprint code corresponding to the historical current sampling sequence data;

The electricity consumption warning module 60 is used for determining the identity information of the abnormal electrical appliance according to the current fingerprint code of the abnormal electrical appliance; determining the abnormal electricity utilization type of the abnormal electricity utilization device according to the current fingerprint code of the abnormal electricity utilization device; generating electrical appliance anomaly information according to the identity information of the anomaly electrical appliance and the anomaly electrical appliance type of the anomaly electrical appliance; and sending the abnormal information of the electrical appliance to an intelligent terminal, and recording and warning the abnormal behavior of the abnormal electrical appliance.

In the specific implementation, when the current fingerprint code of the electrical appliance is monitored, if the current fingerprint code of the electrical appliance is found to be changed into the abnormal electrical appliance code, the abnormal electrical appliance corresponding to the current fingerprint code of the abnormal electrical appliance can be determined, and the abnormal electrical appliance can be subjected to power consumption management. The method specifically comprises the following steps: determining identity information of the abnormal electrical appliance according to the current fingerprint code of the abnormal electrical appliance; determining the abnormal electricity utilization type of the abnormal electricity utilization device according to the current fingerprint code of the abnormal electricity utilization device; generating electrical appliance anomaly information according to the identity information of the anomaly electrical appliance and the anomaly electrical appliance type of the anomaly electrical appliance; the abnormal information of the electrical appliance is sent to an intelligent terminal, such as a mobile phone, a smart watch, a bracelet, a tablet personal computer and the like, abnormal behaviors of the abnormal electrical appliance are recorded and alerted, the abnormal behaviors can be stored in a local database or a cloud database, and meanwhile, the abnormal information is sent to the intelligent terminal for reminding a user of the abnormal electrical behavior of the current electrical appliance, and the circuit path of the current abnormal electrical appliance can be disconnected in time

It should be understood that the foregoing is illustrative only and is not limiting, and that in specific applications, those skilled in the art may set the invention as desired, and the invention is not limited thereto.

It should be understood that, although the steps in the flowcharts in the embodiments of the present application are shown in order as indicated by the arrows, these steps are not necessarily performed in order as indicated by the arrows. The steps are not strictly limited in order and may be performed in other orders, unless explicitly stated herein. Moreover, at least some of the steps in the figures may include multiple sub-steps or stages that are not necessarily performed at the same time, but may be performed at different times, the order of their execution not necessarily occurring in sequence, but may be performed alternately or alternately with other steps or at least a portion of the other steps or stages.

It should be noted that the above-described working procedure is merely illustrative, and does not limit the scope of the present invention, and in practical application, a person skilled in the art may select part or all of them according to actual needs to achieve the purpose of the embodiment, which is not limited herein.

Furthermore, it should be noted that, in this document, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or system that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or system. Without further limitation, an element defined by the phrase "comprising one … …" does not exclude the presence of other like elements in a process, method, article, or system that comprises the element.

The foregoing embodiment numbers of the present invention are merely for the purpose of description, and do not represent the advantages or disadvantages of the embodiments.

From the above description of embodiments, it will be clear to a person skilled in the art that the above embodiment method may be implemented by means of software plus a necessary general hardware platform, but may of course also be implemented by means of hardware, but in many cases the former is a preferred embodiment. Based on such understanding, the technical solution of the present invention may be embodied essentially or in a part contributing to the prior art in the form of a software product stored in a storage medium (e.g. Read Only Memory (ROM)/RAM, magnetic disk, optical disk) and comprising several instructions for causing a terminal device (which may be a mobile phone, a computer, a server, or a network device, etc.) to perform the method according to the embodiments of the present invention.

The foregoing description is only of the preferred embodiments of the present invention, and is not intended to limit the scope of the invention, but rather is intended to cover any equivalents of the structures or equivalent processes disclosed herein or in the alternative, which may be employed directly or indirectly in other related arts.

Claims

1. The electricity consumption management method based on the AI current fingerprint is characterized by comprising the following steps of:

Monitoring the current fingerprint code of the electrical appliance, determining an abnormal electrical appliance corresponding to the current fingerprint code of the abnormal electrical appliance when the current fingerprint code of the electrical appliance changes into the current fingerprint code of the abnormal electrical appliance, and performing power utilization management on the abnormal electrical appliance;

The step of inputting the current sampling sequence into a current analysis model to obtain a current fingerprint code comprises the following steps:

respectively carrying out data processing on a plurality of data sets to obtain current fingerprint codes;

Before the current sampling sequence is input to the current analysis model to obtain the current fingerprint code, the method further comprises the following steps:

Outputting the intermediate current analysis model as a current analysis model when the current fingerprint code deviation value is within a preset deviation range;

The current fingerprint code is decomposed based on the decomposition level number and the working current value, and before the current fingerprint code of the electrical appliance is obtained, the method further comprises the following steps:

2. The AI current fingerprint-based power management method of claim 1, wherein the acquiring a sequence of current samples in a circuit comprises:

Filtering the digital current signal to obtain a current filtering signal;

3. The AI-current-fingerprint-based power consumption management method of claim 1, wherein after obtaining the current fingerprint code deviation value from the predicted current fingerprint code and the actual current fingerprint code in the predicted set, further comprising:

Determining a deviation vector according to the actual deviation value;

4. The electricity management method based on AI current fingerprint as recited in any one of claims 1-3, wherein after determining the abnormal electricity consumer corresponding to the abnormal electricity consumer current fingerprint code and performing electricity management on the abnormal electricity consumer, further comprising:

5. An AI-current-fingerprint-based electricity management system for performing the AI-current-fingerprint-based electricity management method of any of claims 1-4, comprising:

6. The AI current fingerprint-based power management system of claim 5, further comprising:

7. The AI current fingerprint-based power management system of claim 5, further comprising:

The electricity utilization warning module is used for determining the identity information of the abnormal electrical appliance according to the current fingerprint code of the abnormal electrical appliance; determining the abnormal electricity utilization type of the abnormal electricity utilization device according to the current fingerprint code of the abnormal electricity utilization device; generating electrical appliance anomaly information according to the identity information of the anomaly electrical appliance and the anomaly electrical appliance type of the anomaly electrical appliance; and sending the abnormal information of the electrical appliance to an intelligent terminal, and recording and warning the abnormal behavior of the abnormal electrical appliance.