A safe electricity pre-warning method based on current fingerprint technology

Abstract

The application discloses a kind of based on current fingerprint technology's safe power consumption early warning method, comprising the following steps: running early using power consumption equipment alarm information, and the transition safety early warning basic model established in combination with the demand of user;Through acquisition terminal continuous acquisition power consumption equipment operating parameter, after using fast FFT calculation, obtain pre-processing safety early warning basic model;The similarity of transition safety early warning basic model and pre-processing safety early warning basic model is compared;Real-time monitoring is carried out to power consumption using final model library comparison alarm.The application is updated iteratively accumulated through a period of time safety monitoring model, based on special current fingerprint identification method can accurately judge what monitored equipment current is in what operating state, the operating law and safe operation mode of field, can find exception in advance, better guarantee personnel and equipment power consumption safety, promote its form good power consumption habit under the premise of meeting user demand through cloud platform, greatly reduce cost.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of intelligent power utilization, in particular to a safe power utilization pre-alarm method based on current fingerprint technology. BACKGROUND

[0002] At present, with the continuous development and change of society, more and more power utilization equipment enters the daily life of thousands of households, and the types of power utilization equipment are more and more complex and diverse, and the working states of each power utilization equipment also exist differences, and even change with time, environment and other factors.

[0003] In the prior art, the main research direction of power utilization safety is to collect the basic power utilization conditions of users through a collection terminal, and model the simple power utilization habits, so that once there is a large deviation in the parameters, the maintenance or users can be timely warned; however, this kind of method can only model and analyze according to the power change of the power utilization equipment, the content is simple and the model is rough, the most real power utilization condition of the user cannot be known, and thus serious consequences such as untimely or false pre-alarm are easily caused, and the equipment is easily in a poor operating environment, causing safety hazards. SUMMARY

[0004] The purpose of the present application is to provide a safe power utilization pre-alarm method based on current fingerprint technology, which can accurately analyze the power utilization habits of users, clearly understand the types of power utilization equipment, and timely and accurately report power utilization alarms.

[0005] The technical solution adopted by the present application is:

[0006] A safe power utilization pre-alarm method based on current fingerprint technology comprises the following steps:

[0007] A: using power utilization equipment alarm information in the early stage of operation, and establishing a transition safety pre-alarm basic model combined with the needs of users;

[0008] B: continuously acquiring power utilization equipment operation parameters through a collection terminal, obtaining the voltage and current values of the power utilization equipment, and using fast FFT calculation to take one sampling period as a basic element to collect the initial data of the power utilization equipment operation site, and record it as a pre-processing safety pre-alarm basic model;

[0009] C: comparing the similarity of the transition safety pre-alarm basic model and the pre-processing safety pre-alarm basic model; if the similarity reaches a set threshold F, then the pre-alarm parameters of this time are combined with the characteristic values in the pre-processing basic model to establish and store as a transition safety basic model G, and if not, the pre-alarm parameters of this time are combined with the characteristic values in the pre-processing basic model to establish and store as a transition basic model T;

[0010] D: According to the eigenvalues in the transition base model G and the current running state eigenvalues, similarity is determined, and whether the similarity reaches a threshold Q is determined. If the threshold Q is reached, the eigenvalues in the transition base model G are selected into the transition model library C, and are sent to the user, and the next step is entered. If the threshold Q is not reached, the current running state eigenvalues are stored in the transition safety base model library;

[0011] E: Similarity of the selected transition model library C and the preprocessed safety warning base model is determined, and whether the similarity reaches a threshold J is determined. If the threshold J is reached, the transition base model G established before is stored, and is recorded as an accurate base model library, and the next step is entered. If the threshold J is not reached, the transition safety base model library is stored.

[0012] F: Whether the number of models in the accurate base model library reaches an upper limit value Z is determined. If yes, the base models in the accurate base model library are stored in a final model library. If not, steps C-E are repeated until the number of models in the accurate base model library reaches the upper limit value Z.

[0013] G: The final model library is used to monitor and compare power consumption in real time and to alarm.

[0014] In step B, the preprocessed safety warning base model establishment specifically includes the following steps: taking one sampling period as a basic element, obtaining the average value N1 of the current value of the power consumption equipment in m1 sampling periods, the average value N2 of the duration of the safety shoulder current domain of the current of the power consumption equipment, the standard deviation N3 of the current effective value of the power consumption equipment in m1 sampling periods, the number N4 of differences between the specific 3rd, 5th, 7th, 9th and 11th harmonics and the user input eigenvalues, the average value N5 of the standard deviation of the active power in m1 sampling periods and the average value N6 of the standard deviation of the reactive power in m1 sampling periods, and the average value N7 of the standard deviation of the power factor in m1 sampling periods.

[0015] The step B needs to meet the following conditions: first, the processed data must be obtained under the normal running condition of the equipment; second, the element model is different for different application occasions; third, the priority of the parameters N1, N2, N3, N4, N5, N6, N7 and the user-defined use equipment habit parameters such as power-on time and power-off time is as follows: the safety shoulder current domain is the highest priority, the standard deviation of the current effective value is the second priority, the power is the third priority, the harmonic is the third priority, the power factor is the fourth priority, and the standard deviation of the sampling current value compared with the user input approximate load current value is the fifth priority. In addition, the individual needs of the user are not included in the priority list.

[0016] The screening of the N1 value in step B specifically includes the following steps: after the current fundamental signal value of the current electric device is obtained by fast FFT calculation, since the sampling period m1 is much larger than the power frequency period 20 ms, the K1 fundamental current values obtained are averaged to obtain K2;

[0017] The fundamental current values of the electric device in the m1 sampling period are set as a set A, A={A1, A2,... Ak,}. If there are discrete values in the set A, these values are discarded in advance before the average value is taken to ensure the effectiveness of the average value. The specific discrete value criterion is that the deviation of the values in the set A from the average value K2 is greater than 20%, and the values greater than 20% are defined as discrete values to be discarded.

[0018] The average value K4 of the K3 fundamental current values is recalculated in the next m1 sampling period, and N average values (N>2) are obtained according to the criterion. Finally, the N average values are averaged to obtain the N1 value, and the N1 value is updated in real time.

[0019] The N2 value in step B, i.e., the obtained safe shoulder current domain, is set according to the running environment of the electric device. The boundary of the safe shoulder current domain is the region where the change rate of the fundamental current value in the power frequency period is less than B compared with the set value, and the range of B is between 0 and 5%.

[0020] The screening process of the N4 value in step B is as follows: for the harmonic signal value obtained by fast FFT calculation, according to the characteristics of the power grid, the main components of the current harmonics in the electric device are odd harmonics, most of which are 3rd, 5th, 7th, and 11th harmonics.

[0021] After the above-mentioned 4 types of harmonics are calculated in m1 sampling periods, the difference between the set value and the professional installation personnel through the mobile phone APP via the cloud platform system is compared, and the number of times N4 is obtained. The number of times involved in N4 is not completely equal, but when the difference rate reaches a1, it is determined to be valid, where a1 fluctuates within 10%.

[0022] The screening process of the N5 and N6 values in step B is as follows: the standard deviations δ1 and δ2 of the active power and reactive power in m1 sampling periods are calculated. When the values of δ1 and δ2 are not within the criterion range, it is determined that the standard deviations are invalid. The specific criterion range is that the standard deviation is within 50 watts, and the standard deviations δP and δq are determined to be valid. Finally, the valid standard deviations are averaged to obtain N5 and N6.

[0023] The screening process of the N7 value in step B is as follows: the power factor is determined according to the characteristics of the electric device nameplate and the set value in the electric field. For example, the power factor of the electric device power grid in the residential area is regulated to be above 0.9, and the power factor standard deviation calculated in the m1 sampling period is above 0.1, which is determined to be invalid data.

[0024] The electric safety device of the present application not only can realize the conventional electric safety monitoring, but more outstandingly realizes the user's self-learning function of the electric habit of the electric equipment. Specifically, the device can be set by the user through a specific APP of the mobile phone before use, including the user's electric habit, the user's personalized early warning, etc. The device receives the user's setting and establishes a basic model combined with the operating parameters. When the electric safety device runs to the site, it starts to establish the first basic model by collecting electric parameters and other information. When there is a warning or alarm in the running site, the device will filter through the basic model and push it to the user through the platform. The user can confirm that the early warning is the user's acceptable electric habit after receiving the early warning information. If the device accepts the current load as legal and within the standard safety range, the electric safety device will adjust the basic model through the information called transition model A. If not, the device also accepts the user's feedback information and establishes another type of basic model called transition model B. In the long run, it can realize very accurate monitoring of the site electricity, promote the user to develop good equipment use habits, improve the safety, and reduce the security risks. BRIEF DESCRIPTION OF DRAWINGS

[0025] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings needed to be used in the embodiments or the prior art description. Obviously, the drawings in the following description only some embodiments of the present application, and for those skilled in the art, other drawings can be obtained without creative labor on the basis of these drawings.

[0026] Figure 1 The flowchart of the present application. DETAILED DESCRIPTION

[0027] The technical solutions in the embodiments of the present application will be described clearly and completely in the following with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only some embodiments of the present application, not all. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the present application.

[0028] As Figure 1 shown, the present application comprises the following steps:

[0029] A: early running alarm information of electric equipment, and a transition safety early warning basic model established in combination with the user's demand;

[0030] B: continuously acquiring the running parameters of the power utilization equipment through the acquisition terminal, obtaining the voltage and current values of the power utilization equipment, and using the fast FFT calculation to obtain an initial data of a running site of the power utilization equipment as a basic element of a sampling period, and recording the initial data as a preprocessing safety early warning basic model;

[0031] In the step B, the preprocessing safety early warning basic model specifically comprises the following steps: taking one sampling period as a basic element, obtaining an average value N1 of current values of the power utilization equipment in m1 sampling periods, an average value N2 of a duration of a safe shoulder current domain of the current of the power utilization equipment, a standard deviation N3 of an effective value of the current of the power utilization equipment in the m1 sampling periods, a number N4 of differences between specific 3rd, 5th, 7th, 9th and 11th harmonics and a user input characteristic value, an average value N5 of a standard deviation of active power in the m1 sampling periods, and an average value N6 of a standard deviation of reactive power in the m1 sampling periods, and an average value N7 of a standard deviation of a power factor in the m1 sampling periods.

[0032] The step B needs to satisfy the following conditions: first, the processed data must be obtained under a normal running condition of the equipment; second, the element model is different for different application occasions; third, priorities of the parameters N1, N2, N3, N4, N5, N6 and N7 and user-defined use habits of the equipment such as power-on time and power-off time are as follows: the safe shoulder current domain has the highest priority, the standard deviation of the effective value of the current has the second priority, the power has the third priority, the harmonic has the third priority, the power factor has the fourth priority, and the standard deviation of the sampling current value compared with the approximate load current value input by the user has the fifth priority; in addition, the individualized needs of the user are not included in the priority list.

[0033] The screening of the N1 value in the step B specifically comprises the following steps: after the fast FFT calculation obtains a current fundamental wave signal value of the current power utilization equipment, since the sampling period m1 is much larger than a power frequency period of 20 ms, the average value K2 of K1 fundamental wave current values is obtained.

[0034] The current fundamental wave value of the power utilization equipment in the m1 sampling period is set as a set A, A={A1, A2,..., Ak}. If there are discrete values in the set A, these values need to be discarded in advance before the average value is obtained to ensure the effectiveness of the average value. Specifically, the discrete value criterion is that the deviation degree of the value in the set A compared with the average value K2 is greater than 20% or more, and the value is discarded as a discrete value.

[0035] The average value K4 of K3 fundamental wave current values is obtained in the next m1 sampling period, and the N (N>2) average values are obtained according to the criterion. Finally, the N average values are averaged to obtain the N1 value, and the N1 value is updated in real time.

[0036] The N2 value in step B is the obtained safety shoulder current domain, which is determined according to the set value of the running environment of the on-site electrical equipment. The boundary of the safety shoulder current domain is the region in which the change rate of the fundamental wave value of the current compared with the set value is less than B in the power frequency cycle, and the range of B is between 0 and 5%.

[0037] The screening process of the N4 value in step B is as follows: for the harmonic signal values obtained by the fast FFT calculation, according to the characteristics of the power grid, the main components of the current harmonics in the electrical equipment are odd harmonics, most of which are 3rd, 5th, 7th and 11th harmonics.

[0038] After the above-mentioned 4 types of harmonics are calculated in m1 sampling periods, the difference between the set value and the number N4 is compared by the professional installation personnel through the mobile phone APP via the cloud platform system; the number involved in N4 is not completely equal in value, but when the difference rate reaches a1, it is determined to be valid, wherein a1 fluctuates within 10%.

[0039] The screening process of the N5 and N6 values in step B is as follows: when the standard deviations δ1 and δ2 of the active power and the reactive power in m1 sampling periods are calculated, if the values of δ1 and δ2 are not within the criterion range, it is determined that the standard deviations are invalid. The specific criterion range is that the standard deviation is within 50 watts, and then the valid standard deviations δP and δq are determined to be valid standard deviations. Finally, the valid standard deviations are calculated to obtain N5 and N6.

[0040] The screening process of the N7 value in step B is as follows: the power factor is determined according to the set value of the characteristics of the electrical equipment nameplate and the on-site electrical equipment, such as the power factor of the electrical equipment grid in the residential area, which is regulated to be above 0.9. If the standard deviation of the power factor calculated in m1 sampling periods is above 0.1, it is determined to be invalid data.

[0041] C: compare the similarity of the transition safety early warning basic model and the pre-processing safety early warning basic model; whether the similarity reaches the set threshold F, if yes, store the early warning parameters of this time and the characteristic values in the pre-processing basic model as the transition safety basic model G, if not, store the early warning parameters of this time and the characteristic values in the pre-processing basic model as the non-transition basic model T;

[0042] D: according to the characteristic values in the transition basic model G and the current running state characteristic values (referring to N1-N7), compare and determine the similarity, whether the similarity reaches the threshold Q, if yes, select the transition model library C from the characteristic values in the transition basic model G, and send it to the user, and enter the next step; if not, store the current running state characteristic values in the non-transition safety basic model library;

[0043] E: The similarity of the transition model library C and the pre-processed safety early warning basic model is selected to determine whether the similarity reaches the threshold J. If the threshold J is reached, the transition basic model G established before is stored and recorded as the accurate basic model library, and the next step is entered. Otherwise, it is not stored in the transition safety basic model library;

[0044] F: The number of models in the accurate basic model library is judged again. If the upper limit value Z is reached, all the basic models in the accurate basic model library are stored in the final model library. Otherwise, steps C-E are repeated until the number of models in the accurate basic model library reaches the upper limit value Z.

[0045] G: The final model library is used to monitor and compare the power consumption in real time and alarm.

[0046] The following specific examples are used to explain the distance. In actual use, the safety power consumption early warning device involved in the present application is based on an embedded software and hardware platform, internally integrated with a 4G mobile communication module, model WH-GM5, having the ability to access a cloud platform and APP remote coordination control. A storage module, model IS61WV5128BLL-10TL, with a capacity of 4MB, has the ability to store characteristic value sample library data. The storage module IS61WV5128BLL-10TL space is mainly divided into two parts. The first part is mainly used to store the basic model and the corresponding transition model A. The second part is mainly used to store the transition model B. The third part is mainly used to store the accurate model. The fourth part is mainly used to store the final model.

[0047] Before starting to enter the field power, a storage area is opened as a monitoring power load characteristic basic model space. This space is mainly used to record the basic model established by the conventional operating parameters of the power load and the basic model established by the special parameters set by the user. The model established by the conventional operating parameters is run for a period of time (default one month, which can be remotely set through a mobile phone). In addition, the user's special demand parameters can be adjusted in real time.

[0048] The power consumption safety device and system involved will continuously update the power consumption equipment operating parameter model.

[0049] After the power consumption safety monitoring device involved is connected to the field, the power consumption equipment is first pre-processed. The device will establish a basic model according to the user's set requirements and parameters and store it.

[0050] After the device is powered on, the running parameters of the power-using equipment are continuously acquired, and the voltage and current values of the power-using equipment are obtained. Then, the average value N1 of the current values of the power-using equipment in m1 sampling periods, the average value N2 of the duration of the safe shoulder current domain of the current of the power-using equipment, the standard deviation N3 of the current effective value of the power-using equipment in m1 sampling periods, the number N4 of differences between the specific 3rd, 5th, 7th, 9th and 11th harmonics and the characteristic value input by the user, the average value N5 of the standard deviation of the active power in m1 sampling periods, the average value N6 of the standard deviation of the reactive power in m1 sampling periods, and the average value N7 of the standard deviation of the power factor in m1 sampling periods are acquired by using fast FFT calculation with one sampling period as a basic element. N1, N2, N3, N4, N5, N6 and N7 are the initial data of the running site of the power-using equipment, which are named as the original basic model. The parameters involved in the original basic model have the following characteristics: first, the processed data must be obtained under the normal running condition of the equipment; second, the element model is different for different application occasions; and third, the main parameters N1, N2, N3, N4, N5, N6 and N7 in the original basic model and the user-defined use habit of the equipment such as power-on time and power-off time are divided into priority levels, in which the safe shoulder current domain has the highest priority, the standard deviation of the current effective value has the second priority, the power has the third priority, the harmonic has the third priority, the power factor has the fourth priority, and the standard deviation of the sampling current value compared with the approximate load current value input by the user has the fifth priority. In addition, the individualized needs of the user are not included in the priority list.

[0051] The original basic model characteristic value processing of N1, N2, N3, N4, N5, N6 and N7 has the following steps.

[0052] The device acquires the N1 value calculated in m1 sampling periods after running in the power-using site.

[0053] The power-using safety monitoring equipment acquires the current voltage fundamental signal and other harmonic signal quantities by using the self-designed precise sampling circuit and processing the voltage and current parameters by using fast FFT in m1 sampling periods. These signal quantities are not directly used without screening, but are subjected to different processing processes to obtain N1, N2, N3, N4, N5, N6 and N7 values.

[0054] After the current fundamental signal value of the current of the power-using equipment is calculated by using fast FFT, the average value K2 of K1 fundamental current values is obtained because the sampling period m1 is much larger than the power frequency period 20 ms.

[0055] The current fundamental value of the electrical equipment in the m1 sampling period is set as set A, A={A1, A2,..., Ak,}. If there are discrete values in set A, these values need to be discarded in advance before taking the average value to ensure the effectiveness of the average value. The specific discrete value criterion is that the deviation of the value in set A from the average value K2 is greater than 20% or more, which is defined as a discrete value to be discarded.

[0056] The average value K4 of the K3 fundamental current values is recalculated in the next m1 sampling period. N average values (N>2) are obtained by this criterion, and N1 value is obtained by averaging the N average values. N1 value is updated in real time.

[0057] The safety shoulder current domain is determined according to the set value of the operating environment of the electrical equipment on site. The boundary of the safety shoulder current domain is the region where the change rate of the fundamental current value compared with the set value is less than B in the power frequency period, and the range of B is between 0 and 5%.

[0058] For the harmonic signal value calculated by fast FFT, according to the characteristics of the power grid, the main components of the current harmonics in the electrical equipment are odd harmonics, most of which are 3rd, 5th, 7th and 11th harmonics.

[0059] After calculating the above four types of harmonics in m1 sampling periods, the difference between the set value and the number N4 is set by the professional installation personnel through the mobile phone APP via the cloud platform system. The number involved in N4 is not completely equal in value, but when the difference rate reaches a1, it is determined to be valid, where a1 fluctuates within 10%.

[0060] The average values N5 and N6 of the standard deviations of active power and reactive power in m1 sampling periods are involved in the safety monitoring system. When the standard deviations δ1 and δ2 of active power and reactive power in m1 sampling periods are calculated, if the values of δ1 and δ2 are not within the criterion range, it is determined that the standard deviations are invalid. The specific criterion range is that the standard deviation is within 50 watts, which is determined to be an effective standard deviation δP and δq. Finally, the effective standard deviations are selected and the average values N5 and N6 are calculated.

[0061] The average value N7 of the standard deviation of power factor in m1 sampling periods is obtained by a method similar to N5 and N6, but the criterion is different. Specifically, the power factor is determined according to the characteristics of the electrical equipment nameplate and the set value of the electrical site. For example, the power factor of the electrical equipment power grid in residential areas is regulated to be above 0.9, and the standard deviation of the power factor calculated in m1 sampling periods is above 0.1, which is determined to be invalid data.

[0062] The safety power consumption early warning device and system calculates N1, N2, N3, N4, N5, N6 and N7 after being installed in the field and running, and then establishes a safety base model Y in combination with the user setting parameters, Y has the following properties: the parameters in the first Y are updated in real time, the value of N1 can change with the running state of the power consumption equipment, for example, the current and power of the commonly used electric rice cooker at home are obviously different in the cooking mode and the heat preservation mode, which requires real-time adjustment of the model; for example, the current and power of the air conditioner in the cooling mode and the heating mode are different, and more complex is that the current, power and other values of the air conditioner are also significantly different at different temperatures in the cooling mode.

[0063] Therefore, the application takes the safety monitoring intelligent electric rice cooker as an example to describe in detail the running process of the safety power consumption early warning device and system based on the current fingerprint technology.

[0064] The application provides a power consumption safety early warning device and system based on current fingerprint technology, which obtains the characteristic values of N1-N7, and establishes a pre-processing safety base model in combination with the user setting parameters, and the device traverses the pre-processing base model when a pre-warning event occurs, and if the similarity to the pre-processing base model reaches a set threshold F, a transition base model G is established in combination with the characteristic values in the pre-processing base model, and the pre-warning event is pushed to the user through the cloud platform; otherwise, if the similarity to the pre-processing base model does not reach the set threshold F, another type of transition base model T is established and stored.

[0065] According to the comparison of the characteristic values in the transition base model G and the current running state characteristic values, the model whose similarity reaches a threshold Q is selected and stored in the transition model library C.

[0066] When the threshold J is reached, the transition base models G established before are stored in the accurate base model library.

[0067] When the number of models in the accurate base model library reaches an upper limit Z, the device stores all the base models in the accurate base model library in the final model library.

[0068] The application solves the accurate safety monitoring and early warning of power consumption equipment, identifies the detailed characteristics of the monitored object based on special current fingerprints, and updates the final model library in real time. The real-time running characteristic values of the monitored power consumption equipment are compared with the characteristic values in the corresponding final model library, the most real state of the current power consumption equipment is obtained, the most reliable and more comprehensive power consumption equipment quality control and safety are realized, and the reliability and accuracy of the pre-warning are greatly improved.

[0069] The above device in actual use, the specific process is as follows: the user inputs the pretreatment parameter according to the use habit, such as power-on time, power content, power length, power-off time, and the effect to be achieved, such as refrigeration, heating and the like.

[0070] The use features include the real-time change values of N1-N7, including the change of the electric load current, power and power factor (such as air conditioner, electric rice cooker and the like) in use and on-off change features, and the specific current fingerprint model is established by collecting the pretreatment parameters, which is called the basic model.

[0071] The electric equipment has specific properties and is relatively fixed, and can be subdivided into various modes, such as resistive load like commonly used electric rice cooker, kettle and the like; inductive load like air conditioning equipment, and the like, and different basic models can be established according to the special requirements or modes input by the user.

[0072] In order to provide the reliability of the electric safety, when the pre-alarm information is contrary to the user's intention, it is necessary to re-enter.

[0073] When the sampled electric equipment electrical information such as excessive current, power factor offset exceeding threshold and the like, the characteristic value needs to be recalculated and the pre-warning mode and pre-warning threshold are corrected.

[0074] The device and system related to the present application will have deviation in accuracy and reliability at the initial stage of the running site, and the update iteration accumulation (1 week normal use period) of the safety monitoring model will continue for a period of time, based on the special current fingerprint recognition method, the current running state of the monitored equipment, the running law of the site and the safe running mode can be accurately judged, the abnormality can be found in advance, the personnel and equipment electric safety can be better guaranteed, the cloud platform can promote the user to form good electric habit under the premise of meeting the user's demand, and the service life of the equipment is prolonged, and the cost is greatly reduced.

[0075] In the description of the present application, it should be noted that for the orientation words, such as the terms "center", "transverse", "longitudinal", "length", "width", "thickness", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside", "clockwise", "counterclockwise" and the like indicate the orientation and positional relationship based on the orientation or positional relationship shown in the drawings, which is only for the convenience of describing the present application and simplifying the description, and does not indicate or imply that the device or element referred to must have a particular orientation, be constructed and operated in a particular orientation, and cannot be understood as limiting the specific protection scope of the present application.

[0076] It should be noted that the terms "first", "second", and the like in the description and in the claims of this application are used for distinguishing between similar elements and not necessarily for describing a particular sequential or chronological order. It is to be understood that the use of such terms is interchangeable under appropriate circumstances such that the descriptive herein is intended to be interpreted in the context of this application. Furthermore, the terms "comprise", "comprising", "include", "including", and the like are to be construed in an inclusive rather than an exclusive sense, that is, in the sense of "including, but not limited to".

[0077] It is to be understood that the above description and the accompanying examples are merely illustrative of preferred embodiments of the application and the application should not be limited to the precise details stated herein. Various modifications and changes can be made to what has been described without departing from the scope of the application. It is therefore desired that what is claimed should be interpreted as being the broadest interpretation possible in accordance with the principles of the application.

Claims

1. A method for early warning of safe electricity use based on current fingerprint technology, characterized in that: Includes the following steps: A: In the early stages of operation, a basic model for transitional safety early warning was established by using alarm information from electrical equipment and combining it with user needs; B: After continuously acquiring the operating parameters of the electrical equipment through the acquisition terminal and obtaining the voltage and current values ​​of the electrical equipment, the initial data of the electrical equipment operation site is collected using one sampling period as the basic element and recorded as the preprocessing safety early warning basic model. C: Compare the similarity between the transitional security early warning basic model and the pre-processing security early warning basic model; check if the similarity reaches the set threshold F. If it does, then use the current early warning parameters and the feature values ​​in the pre-processing security early warning basic model to establish and store the transitional security basic model G. If it does not reach the threshold, then use the current early warning parameters and the feature values ​​in the pre-processing security early warning basic model to establish and store the transitional basic model T. D: Then, based on the feature values ​​in the transitional safety basic model G and the current operating status feature values, a similarity judgment is made to determine whether the similarity reaches the threshold Q. If the threshold Q is reached, the models in the transitional safety basic model G with similarity reaching the threshold Q are selected and stored in the transitional model library C, and sent to the user, and the next step is taken; if the threshold Q is not reached, the current operating status feature values ​​are stored in the transitional safety basic model library. E: Select the similarity between the transition model library C and the preprocessed security early warning basic model and determine whether the similarity reaches the threshold J. If the threshold J is reached, store the previously established transition security basic model G and record it as the accurate basic model library, and proceed to the next step. Otherwise, if the threshold J is not reached, store it in the transition security basic model library. F: Next, check if the number of models in the accurate basic model library has reached the upper limit Z. If so, store all basic models in the accurate basic model library into the final model library; otherwise, repeat step CE until the number of models in the accurate basic model library reaches the upper limit Z. G: Real-time monitoring, comparison, and alarm functions are performed using the final model library.

2. The safe electricity use early warning method based on current fingerprint technology according to claim 1, characterized in that: In step B, the establishment of the preprocessing safety early warning basic model specifically includes the following steps: taking one sampling period as the basic element, obtaining the average value N1 of the current value of the electrical equipment within m1 sampling periods, the average value N2 of the duration of the safe shoulder current domain of the electrical equipment current, the standard deviation N3 of the effective value of the current of the electrical equipment within m1 sampling periods, the number of differences N4 between the specific 3rd, 5th, 7th, 9th and 11th harmonics and the user input characteristic value; the average value N5 of the standard deviation of active power within m1 sampling periods, the average value N6 of the standard deviation of reactive power within m1 sampling periods, and the average value N7 of the standard deviation of power factor within m1 sampling periods.

3. The safe electricity use early warning method based on current fingerprint technology according to claim 2, characterized in that: Step B must meet the following conditions: First, the processed data must be obtained under normal operating conditions of the equipment; second, the meta-model is different for different application scenarios; third, the priority of the parameters N1, N2, N3, N4, N5, N6, N7 and the power-on time and power-off time parameters in the user's customized equipment usage habits is as follows: the safety shoulder current domain has the highest priority, the standard deviation of the current effective value has the second priority, power has the third priority, harmonics have the third priority, power factor has the fourth priority, and the standard deviation of the sampled current value compared with the approximate load current value input by the user has the fifth priority; in addition, the user's personalized needs are not included in the priority list.

4. The safe electricity use early warning method based on current fingerprint technology according to claim 3, characterized in that: The selection of N1 value in step B specifically includes the following steps: After the current fundamental signal value of the current device is obtained by fast FFT calculation, since the sampling period m1 is much larger than the power frequency period of 20ms, the average value K2 of the K1 fundamental current values ​​is taken. The fundamental current values ​​of the electrical equipment within the m1 sampling period are set as A, where A = {A1, A2, ..., Ak}; If there are discrete values ​​in set A, these values ​​need to be discarded before taking the average to ensure the validity of the average. The specific criterion for discrete values ​​is that if the value in set A is compared with the average K2, the deviation is greater than 20% and it is considered a discrete value to be discarded. In the next m1 sampling period, the average value K4 of the K3 fundamental current values ​​is recalculated. Based on this criterion, N average values ​​are obtained, where N>2. Finally, the average of the N average values ​​is calculated again to obtain the value N1, and the value N1 is updated in real time.

5. The safe electricity use early warning method based on current fingerprint technology according to claim 3, characterized in that: In step B, the N2 value, i.e. the safety shoulder current domain, is determined based on the operating environment settings of the on-site electrical equipment. The boundary of the safety shoulder current domain is the region where the rate of change of the fundamental current value relative to the set value within the power frequency cycle is less than B, and the range of B is between 0 and 5%.

6. The safe electricity use early warning method based on current fingerprint technology according to claim 3, characterized in that: The screening process for the N4 value in step B is as follows: For the harmonic signal value obtained by fast FFT calculation, according to the characteristics of the power grid, the main components of the current harmonics in the electrical equipment are odd harmonics, most of which are the 3rd, 5th, 7th and 11th harmonics. After calculating the above four types of harmonics in m1 sampling cycles, the difference N4 between the values ​​set by the professional installer via mobile APP and cloud platform system is first compared with the values ​​set by the cloud platform system. The number of times involved in N4 is not exactly equal, but is only considered valid when the difference rate reaches a1, where a1 fluctuates within the range of 10%.

7. The safe electricity use early warning method based on current fingerprint technology according to claim 3, characterized in that: The screening process for N5 and N6 values ​​in step B is as follows: After calculating the standard deviations δ1 and δ2 of active power and reactive power within m1 sampling periods, if the values ​​of δ1 and δ2 are not within the criterion range, they are determined to be invalid standard deviations. Specifically, the criterion range is that the standard deviation is within 50 watts to be considered a valid standard deviation. δP and δq are valid standard deviations. Finally, the average value of the selected valid standard deviations is calculated to obtain N5 and N6.

8. The safe electricity use early warning method based on current fingerprint technology according to claim 3, characterized in that: The screening process for N7 values ​​in step B is as follows: The power factor is defined based on the characteristics of the electrical equipment nameplate and the value set at the power consumption site. If the power factor specified by the power grid for the electrical equipment in the residential area is above 0.9, then the standard deviation of the power factor calculated within m1 sampling periods is determined to be above 0.1 as invalid data.

Images