CN114629242A

CN114629242A - Non-invasive power load detection and decomposition method

Info

Publication number: CN114629242A
Application number: CN202210265304.5A
Authority: CN
Inventors: 张良均; 刘名军; 张奥多; 周龙; 赵云龙; 李振林; 郭信佑; 刘晓玲
Original assignee: Guangdong Teddy Intelligent Technology Co ltd
Current assignee: Guangdong Teddy Intelligent Technology Co ltd
Priority date: 2022-03-17
Filing date: 2022-03-17
Publication date: 2022-06-14

Abstract

The application provides a non-intrusive-based power load detection and decomposition method, which comprises the following steps: the power state of the user's consumer is obtained through the load detection data, and the vacancy value in the obtained consumer state is detected, and the method further comprises the following steps: analyzing the vacancy value and the gear change, and filling data after vacancy detection; performing correlation analysis on the detected vacancy value and the transient state of the corresponding equipment, monitoring according to the feasibility evaluation result and the obtained load detection data of the electric equipment in the current state, and performing load characteristic analysis on the electric equipment; and performing corresponding power protection through the result obtained by analysis, and performing electric equipment protection through analyzing the risk level. The invention can perform early warning and protect the power equipment when the power equipment has the vacancy value.

Description

Non-invasive power load detection and decomposition method

[ technical field ] A method for producing a semiconductor device

The invention relates to the technical field of intelligent equipment, in particular to a non-invasive power load detection and decomposition method.

[ background of the invention ]

When non-invasive load detection is carried out, due to the fact that the equipment is non-invasive, in the process of switching, starting and closing of each piece of equipment, due to the characteristics of the power acquisition equipment, the data acquired by the monitoring equipment are null values due to the transient work of some pieces of power equipment, and data loss is caused. When data is missing, it is not beneficial for the monitoring equipment to analyze the current power situation, and if the danger of power instant happens at the moment, the monitoring equipment is difficult to rapidly carry out work such as electrical protection. Which may result in a power outage or equipment damage. According to the method, risk evaluation is carried out under the condition that the vacancy value occurs in the data collection, various conditions of the electric equipment can not be estimated by using data in the time period of the occurrence of the vacancy value, analysis carried out by using the data before and after the vacancy value time period is not complete and reliable, and whether safety accidents occur or not in the time period of the occurrence of the vacancy value can not be specifically analyzed according to the data, so that the data in the time period of the occurrence of the vacancy value can be collected by using new load detection equipment besides the data detected by the non-invasive load detection equipment, and the vacancy value collected by the data of the non-invasive load detection equipment is replaced. And comprehensively judging and evaluating the risk condition of the electric equipment by using the data in the vacancy value time period and combining the load data of the time before and after the vacancy value. And corresponding limiting measures are required to be carried out on the electricity consumption of the user according to the evaluation result so as to avoid danger.

[ summary of the invention ]

The invention provides a non-intrusive power load detection and decomposition method, which mainly comprises the following steps:

acquiring the power state of the electric equipment of the user through the load detection data; detecting the obtained vacancy value in the state of the electric equipment; performing correlation analysis on the detected vacancy values and the transient states of the corresponding equipment; obtaining vacancy values of different power equipment and a load detection data change period according to correlation analysis to predict; performing feasibility evaluation based on the predicted result; monitoring according to the feasibility evaluation result and the acquired load detection data of the electric equipment in the current state; analyzing according to the monitored state load of the electric equipment; performing corresponding power protection according to the result obtained by analysis;

further optionally, the obtaining the power state of the electric device of the user through the load detection data includes:

load detection data of the electric equipment is obtained through the non-invasive load detection equipment, and the electricity utilization conditions of different electric equipment are analyzed according to the load detection data; acquiring the type of the electric equipment according to the identification of the load detection data of the electric equipment and obtaining the power utilization gear of the corresponding electric equipment; the method comprises the following steps: clustering and analyzing the load detection data to infer the state of the electric equipment;

the method for clustering, analyzing and inferring the state of the electric equipment by using the load detection data further comprises the following steps:

the method comprises the steps of analyzing electric equipment according to electric data actually acquired by load detection, judging the type of the electric equipment and the gear change condition of the electric equipment by integrating steady-state and transient-state electric power data in a period of time in which the electric equipment operates, and analyzing according to load detection data. And recording matched electric equipment under the condition of current electricity utilization change, and recording load detection data characteristics of the current equipment. In actual operation, the corresponding electric equipment is judged according to the existing electric characteristics. And finding out the clustering center of the electricity utilization characteristics of the electric appliances by using the MeanShift algorithm for the electricity utilization characteristic data of the wavelengths of the electric appliances, and recording. When the non-invasive load detection equipment is used by a plurality of different electric appliances, the cluster analysis is also carried out according to different electricity utilization characteristics, the characteristic range of each different electric appliance is found out according to different electricity utilization characteristics, and the type of the electric appliance at the moment is judged according to the existing electric appliance characteristic record.

Further optionally, the detecting the vacancy value in the acquired state of the electric device includes:

comprehensively analyzing the vacancy condition of the load detection data by using the acquired load detection data; comparing the time point corresponding to the vacancy value with the current running state of the power equipment, analyzing the specific state of the power equipment under the condition that vacancy occurs, and judging whether the data of the power equipment under the state can not be collected or not because of the data caused by self characteristics or the data can not be collected due to the occurrence of an emergency condition; the method comprises the following steps: analyzing a vacancy value and gear change; judging whether the vacancy value exists or not by using the acquired data; filling data after vacancy detection;

the analysis of the vacancy value and the gear change further comprises the following steps:

in actual detection data of the non-invasive load detection device, the acquired electric equipment data cannot acquire data in the current time period due to the characteristics of the detection device or the special condition of the electric equipment, so that a specific measured value cannot be acquired in a certain time period, and finally, the situation that data vacancy occurs in a certain time period exists in the acquired data.

The method for judging whether the vacancy value exists by using the acquired data further comprises the following steps:

the method comprises the steps of performing vacancy detection through acquired data, acquiring time point data in a character string form, reading the character string and performing date conversion, performing addition and subtraction on the time points, judging whether the interval between the two time points is more than 1 second, if so, determining that the data vacancy occurs in a corresponding time period, and if so, determining that the data vacancy does not occur. For the existence of the vacancy value, firstly, the single electric equipment needs to be judged. The calculation is performed by:

step 1: and acquiring the electricity utilization data of each electricity utilization device through the load detection device.

Step 2: and presetting a blank array H, writing the electricity consumption data of each single device into an Excel table, reading a character column for recording time in the Excel table, and presetting the currently acquired total time data of the device as A, wherein the line number of the A is changed but the column number is only one column.

And step 3: and (4) entering circulation, presetting i as a data table of the ith electric equipment, and presetting a total of N electric equipment. I.e. 1,2.. N.

And 4, step 4: in the loop of step 3, j is preset as the jth time data of the ith device, and the loop is entered. Acquiring time data of the jth character string type of the ith electric equipment, namely j is 1,2.. length (A) -1, modifying the time data of the character string type into a numerical value type, solving a difference value of two adjacent time data, and taking an absolute value. And recording the calculated value to the jth position of the ith row of the array H whether the calculated value is greater than 1 or not.

And 5: and when the step 4 is completed in a circulating way, obtaining the value of the ith row of the array H.

Step 6: and (4) after the circulation of the step (3) is completed, finally obtaining a two-dimensional array H, wherein H has N rows, and the number of columns is the time data volume of the equipment with the maximum acquisition time point.

And 7: and substituting the array into a loop statement to detect the vacancy values of each device in different time periods, judging whether the ith row in the H array has a number larger than 1, and if not, judging whether the corresponding position has the vacancy value or not.

Carry out data filling after vacancy detects, still include:

in the time period of the occurrence of the vacancy value, the electric equipment can cause the load detection equipment to be incapable of acquiring real-time data due to different conditions, so that the vacancy value of the data is caused, and the vacancy value is filled by calculating a new value by using the data as the actual data exists in the previous period of time when the vacancy time occurs and the later period of time when the vacancy time ends. Since the time of the vacancy is not fixed, a case of only one second of vacancy or a case of 10 minutes of vacancy may be differentiated. When the vacancy value is filled, the running state of the electric equipment at the current time point needs to be integrated, data filling needs to be carried out according to the electricity utilization rule under the corresponding gear of the current time point, so that the vacancy is filled through the existing data, whether the electricity utilization state of the electric equipment is changed in the vacancy time period needs to be judged, corresponding data filling is carried out according to the state change in the vacancy time period of the electric equipment, and the filled data is analyzed, wherein the filling accuracy is high.

Further optionally, the correlation analysis of the detected occurrence of the vacancy values with corresponding device transients includes:

judging whether a two-dimensional array H obtained by calculation in the vacancy value exists by using the obtained data, and comparing the H of the two-dimensional array with the original detection data of the electric equipment; judging whether the transient state of the electric equipment occurs or not in the time period of the occurrence of the vacancy value; performing relevance analysis by judging the transient time point of the electric appliance and the time period of the data of the load detection, calculating the relevance degree between the transient time point and the time period of the data of the load detection, and comprehensively analyzing the relevance between the time period of the vacancy value and the time point of the transient state occurrence through the calculated relevance degree; recording each state change position of the collected data, recording the start time and the end time of each time period of the occurrence of the vacancy value, and comparing the start time and the end time with the transient time points respectively; the influence of the transient states of different electric equipment on the load detection equipment is correspondingly correlated, so that when the correlation analysis is used, the correlation analysis needs to be carried out by firstly synthesizing the transient states and the vacancy values of all the equipment, after the correlation analysis is carried out, the correlation analysis is carried out by controlling the single equipment, and the results of the two correlation analyses are comprehensively compared to carry out comprehensive judgment so as to obtain different correlation coefficients under different electric equipment, thereby obtaining the correlation size between the transient state and the vacancy value of each electric equipment; the method comprises the following steps: preparing data based on the vacancy value and the transient occurrence time of the electric equipment; preparing data under two conditions of gear switching and gear switching-free within the vacancy value time period; performing relevance analysis on the occurrence condition of the vacancy values and the transient state;

the data preparation is carried out based on the vacancy value and the transient occurrence time of the electric equipment, and the method further comprises the following steps:

the data collected by the independent operation of different devices and the operation state change condition of the devices at the current time are marked through the load detection device, so that a time point which contains the vacancy value and obtains the occurrence time period of the vacancy value and the occurrence transient state of the electric appliance is obtained. And writing the time point of each electrical appliance transient occurrence into a table. And judging whether the vacancy values exist or not according to the acquired data, acquiring the starting time point of the vacancy values of the electric appliances and the current operating state of the electric appliances under the position of the detected vacancy values of each electric appliance, judging, comparing the starting time point and the ending time point of the vacancy values with the duration time of each operating state of the electric appliances to obtain the operating state of the electric appliances when the vacancy values appear, and recording the starting time point, the ending time point and the state change conditions of the vacancy values into a table. Two situations can therefore occur: the first condition is as follows: the time period during which the empty value lasts is within the time of one state of the electrical consumer, and no change of the operating state occurs. Case two: two operation states of the electric equipment occur in the time period of the continuous vacancy value, namely, the operation states of the electric equipment are switched in the time period of the occurrence of the vacancy value.

The data preparation of the two conditions of gear switching and gear switching-free in the vacancy value time period further comprises the following steps:

and performing data sorting on each vacant position in the data obtained by load detection, wherein in the first situation of performing data preparation based on the vacancy value and the transient occurrence time of the electric equipment, the vacancy value starting time point, the vacancy value ending time point, the current gear of the electric equipment when the vacancy value occurs and the name of the electric equipment need to be written into an Excel table. In the second situation of data preparation based on the vacancy value and the transient occurrence time of the electric equipment, the vacancy value starting time point, the gear switching time point, the gear before switching, the vacancy value ending time point, the gear after switching and the name of the electric equipment need to be recorded in an Excel table. Thus, two Excel tables are obtained, which are the device state data obtained in two cases.

The occurrence condition of the vacancy value and the transient state are subjected to correlation analysis, and the method further comprises the following steps:

and preparing data in two conditions of gear switching and gear switching without the gear within the vacancy value time period, reading the data in the first condition, and comparing the initial time point of occurrence of the vacancy value with the time point of the latest gear switching. And quantitatively analyzing the time of the occurrence of the vacancy value and the time of gear switching, comparing each vacancy value time point with the latest gear switching time point, drawing a vacancy value scatter diagram, performing linear regression analysis on the scatter diagram, obtaining a correlation function relation between a transient time point and a vacancy value starting time point, ending time point and median of the time point, and judging whether the linear regression is reasonable or not through correlation. And comprehensively judging the association degree through the relational parameters obtained by the correlation analysis. And setting each gear switching time point as a function value, setting each vacancy value occurrence time point, each end time point and each median of the time points as independent variables, and performing regression analysis on the independent variables and the function values respectively. Similarly, the data in case two is read. And taking the vacancy value starting time point of different data under the same transient state, taking the ending time point and the median of the gear switching time point as independent variables, and taking the gear switching time point as the median to perform relevance analysis. And calculating to obtain a functional relation. And the relational case in both cases is compared.

Further optionally, the obtaining of the vacancy values of different power devices and the load detection data change period according to the correlation analysis for prediction includes:

performing relevance analysis according to the occurrence condition of the vacancy value and the transient state to obtain a functional relation of a form of y ═ ax + b and vacancy value data under two conditions, performing deep analysis on the basis, and performing multiple linear regression analysis on the time transient state of the nearest distance by respectively taking the vacancy value starting time period, ending time period and time median as three independent variables; the method comprises the following steps: a regression analysis process;

the regression analysis process further comprises:

and calculating a functional relation between three time point parameters of the vacancy value of each electric equipment and the state change time point of the electric equipment through multivariate nonlinear regression analysis, and performing regression prediction on the vacancy value of the electric equipment through the functional relation. The undetermined relation of a functional expression y (i) ═ b0+ b1 x1(i) + b2 x2(i) + b3 x3(i) is established, the seconds corresponding to each state change time point of each electric device are read as y values, and the seconds corresponding to each vacancy value starting time point, each vacancy value ending time point and the median of the vacancy time period are read as the values of independent variables x1, x2 and x3 respectively. And calculating the constants b0, b1, b2 and b3 by using a multiple linear regression solving function to obtain specific parameter values of the regression function relational expression. And carrying out the substitution through a functional relation to obtain the occurrence condition and the duration of the vacancy values under different gear conditions.

Further optionally, the performing feasibility assessment based on the predicted result comprises:

the regression analysis process compares the calculated regression function relational expression with the vacancy value before occurrence, obtains a regression function characteristic value for prediction, and comprehensively judges the prediction result to carry out feasibility evaluation on the power utilization condition of the predicted vacancy value position; the method comprises the following steps: combining the feasibility evaluation process of a secondary load tester;

the feasibility assessment process combined with the secondary load tester further comprises the following steps:

and predicting the filled part by acquiring the detected data and the functional relation according to the data of the detection time which is one second before the occurrence of the vacancy value and one second after the end of the vacancy value, and acquiring the data of the output total power, the useful power, the current voltage, the corresponding cycle and the harmonic wave to perform comprehensive analysis. And finding out the predicted power utilization data of the position corresponding to the vacancy value for evaluation. And acquiring power utilization state data of non-null values before and after the null value in the load detection data for evaluation. In a time period in which the non-invasive load detection equipment cannot acquire data, if the duration is short, the secondary load detector is not started for the moment to measure the load detection data in the vacant time period. And presetting that the time of the vacancy exceeds 5 seconds, and automatically starting a secondary load detector to replace and detect the load of the electric equipment. And when the data of the non-invasive load detection equipment is recovered to be normal and continues to detect the data, the secondary load detection equipment records the load detection data collected in the vacant time period and automatically closes the detection. And comparing a regression function relation formula made by the vacancy value time period of the load detection equipment with detection data of the secondary load detector in the vacancy time period, evaluating the feasibility of the regression function relation formula, predicting the actual state of the electric equipment according to the comparison result, and analyzing the difference and the prediction result to evaluate.

Further optionally, the monitoring according to the feasibility evaluation result and the obtained load detection data of the electric device in the current state includes:

monitoring the load detection data characteristics of each electric equipment through the collected load detection data, storing the collected abnormal electricity utilization characteristic data into an abnormal electricity utilization characteristic database, comprehensively judging whether the electric equipment of the current user has electricity utilization potential safety hazards or not by acquiring and extracting the electricity utilization characteristic data in real time, judging the monitoring of the load of the electric equipment through an evaluation result according to a feasibility evaluation flow, and properly adjusting the electricity consumption of the user; the method comprises the following steps: monitoring the risk of state evaluation;

the monitoring of the risk of the state assessment further comprises:

according to the feasibility evaluation flow combined with the secondary load tester, load detection data of the secondary load detector under the condition of the vacancy value of the load detection equipment and load detection data of the time period of the abnormal power utilization condition are obtained, the characteristics of the load detection data under the abnormal power utilization condition are subjected to risk analysis, the state of the electric equipment in the vacancy value time period is predicted, the risk of the electric equipment under the abnormal state is judged, the data characteristics and the risk condition under the vacancy condition are recorded into an abnormal power utilization characteristic database according to the current situation, and the situation is used for comparing the situation of the data of the subsequent monitoring load detection.

Further optionally, the analyzing according to the monitored state load of the electric device includes:

analyzing various possibilities of the load of the monitored electric equipment, and comprehensively judging whether the electric equipment has a short circuit or is on fire according to the monitored data and the evaluation result of the occurrence of the vacancy value; analyzing whether the electric equipment catches fire or is short-circuited according to the load condition of the electric equipment measured by the secondary load detector during the vacancy value period; load detection data before and after the vacancy value and data of a secondary load detector during the vacancy period are comprehensively judged, load detection is respectively carried out on the same electric equipment under different fault conditions, comparison is carried out according to the load detection result and a normal state, possible fault states of the electric equipment in the time period of occurrence of the vacancy value when the vacancy value occurs are analyzed according to the occurrence position of the vacancy value and the load detection data before and after the occurrence of the vacancy value, and the emergent conditions of the electric equipment are analyzed and predicted according to the possible fault states; the method comprises the following steps: analyzing the load characteristics of the electric equipment;

the load characteristic analysis of the electric equipment further comprises the following steps:

and carrying out load characteristic analysis according to the load detection data of the electric equipment in the same operation state in the time periods without the vacancy values and with the vacancy values. And comparing and analyzing the load detection data in the vacant time period with the load difference in the non-vacant time period, judging the possibility of the fault of the electric equipment used by the user in the using process, and performing abnormal early warning on the electric equipment according to the possibility. Comparing a section of load detection data which is closest to the data time difference of the vacancy value in the obtained load detection data to obtain an absolute value, calculating a change ratio of the absolute value and the data under the condition of no vacancy, performing risk evaluation according to the difference ratio, wherein the larger the difference ratio is, the higher the risk level is, so that the risk level of each electric device is evaluated, and the load characteristics under the current risk evaluation are used for judging whether the electric devices of the current user normally operate. If the difference proportion of the current in the vacancy value in the load detection data of the electric equipment exceeds 20%, increasing a risk grade every time the difference proportion exceeds 10% so as to evaluate the risk grade, properly reducing the transmission power of the electric equipment according to the evaluation condition of the risk grade so as to manage and control the risk, and storing the load detection data of the current electric equipment at the current risk grade into a database.

Further optionally, the performing the corresponding power protection according to the result of the analysis includes:

according to the load characteristic analysis of the electric equipment, carrying out power limiting evaluation according to the risk level obtained by analysis, obtaining the load detection data condition of the electric equipment in the period of the vacancy value to carry out power limiting calculation, carrying out adjustment calculation according to the risk level of the condition of the electric equipment and the difference proportion of the corresponding load detection data exceeding the normal state, feeding back the calculated result to an electric meter for providing electric power, and properly limiting the current on a live wire and inputting the current and the voltage of a user so as to regulate and control the current; the method comprises the following steps: protecting the electric equipment by analyzing the risk grade;

carry out consumer protection through analysis risk level, still include:

and judging the dangerous condition of the electric equipment in the electric equipment through the risk grade calculated during the vacancy value, and judging whether the electric equipment has an equipment short circuit or equipment fire. And predicting and analyzing the obtained risk grade and the load difference proportion between the corresponding vacancy value time period and the normal condition to obtain the electric equipment with the possibility of fire or fault in the electric equipment, protecting the electric equipment with the possibility of fire or fault by limiting the output power of the electric equipment, and ensuring the electric safety of a user by power failure if necessary. Judging the possibility of fire or short circuit fault of the electric equipment according to the difference proportion calculated by the load detection data of the electric equipment in the empty value time period, respectively obtaining the difference proportion and risk grade of the data corresponding to the current, the voltage, the power and the cycle harmonic wave, evenly distributing the weights of the current, the voltage, the power, the cycle wave and the harmonic wave, respectively multiplying and summing the weights by the corresponding difference proportion to obtain the comprehensive difference proportion, judging whether the difference proportion exceeds 20% of the non-empty value, if the difference proportion exceeds the corresponding range, limiting the power of the electric equipment according to the proportion, and if the difference proportion exceeds 100%, carrying out power-off operation.

The technical scheme provided by the embodiment of the invention has the following beneficial effects:

the invention predicts the period and the time length of the vacancy value by predicting the generation time and the situation of the vacancy value. And carrying out weighting processing of interpolation on the empty values. And the completion of the vacancy value is better compensated. The purpose of better monitoring the power equipment and the real-time power is achieved. And comparing and analyzing the data obtained by interpolation weighting with the data obtained by measurement by using a secondary load detector, so that the potential safety hazard of the electric equipment is relatively accurately evaluated, and the evaluation result is subjected to corresponding power transmission limiting operation to ensure the electric safety of a user and the operation safety of the electric equipment. Through comparison of data and analysis of the running states of the data and the electric equipment during the vacancy value period, the electricity utilization condition of the electric equipment of a user is evaluated, and output power of the electric equipment which is normally evaluated is adjusted, so that energy consumption is saved. Meanwhile, the electric equipment with abnormal electricity utilization is evaluated and monitored, and potential safety hazards in the future are avoided by power limiting or power failure operation, so that the potential safety hazards of electricity utilization of users are reduced.

[ description of the drawings ]

Fig. 1 is a flow chart of a non-intrusive power load detection and decomposition method according to the present invention.

Fig. 2 is a sequence diagram between a user and a power supply system in a non-intrusive power load detection and decomposition system according to the present invention.

Fig. 3 is a sequence diagram between the load detection device and the load analysis system in the non-intrusive current load detection and decomposition method according to the present invention.

[ detailed description ] A

In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention will be described in detail with reference to the accompanying drawings and specific embodiments.

Fig. 1 is a flow chart of a non-intrusive power load detection and decomposition method according to the present invention. Fig. 2 is a sequence diagram between a user and a power supply system in a non-intrusive power load detection and decomposition system according to the present invention. Fig. 3 is a sequence diagram between the load detection device and the load analysis system in the non-intrusive current load detection and decomposition method according to the present invention. As shown in fig. 1, the non-intrusive power load detection and decomposition method according to the present embodiment may specifically include:

and step 101, acquiring the power state of the electric equipment of the user through the load detection data.

Load detection data of the electric equipment is obtained through the non-invasive load detection equipment, and the electricity utilization conditions of different electric equipment are analyzed according to the load detection data. The type of the electric consumer is obtained according to the identification of the load detection data of the electric consumer, and the electric consumption gear of the corresponding electric consumer is obtained.

And performing cluster analysis by using the load detection data to infer the state of the electric equipment.

The method comprises the steps of analyzing electric equipment according to electric data actually acquired by load detection, judging the type of the electric equipment and the gear change condition of the electric equipment by integrating steady-state and transient-state electric data when the electric equipment runs in a time period, and analyzing according to the load detection data. And recording matched electric equipment under the condition of current electricity utilization change, and recording load detection data characteristics of the current equipment. In actual operation, the corresponding electric equipment is judged according to the existing electric characteristics. And finding out the clustering center of the electricity utilization characteristics of the electric equipment by using the MeanShift algorithm to the electricity utilization characteristic data of the electric equipment wavelength, and recording. When the non-invasive load detection equipment is used by a plurality of different electric appliances, the cluster analysis is also carried out according to different electricity utilization characteristics, the characteristic range of each different electric appliance is found out according to different electricity utilization characteristics, and the type of the electric appliance at the moment is judged according to the existing electric appliance characteristic record. For example: in a non-invasive load detection test, the electricity utilization characteristics of an electric appliance of an incandescent lamp in working and the gear position of 1 gear and the electricity utilization characteristic data of a water dispenser in a heat preservation state are obtained, detection equipment records the electricity utilization characteristics of the incandescent lamp and the data of the water dispenser in the heat preservation state respectively, and when the detection equipment detects in a new household to obtain the electricity utilization data and analyzes the electricity utilization data, the electricity utilization data similar to the electricity utilization characteristics of the incandescent lamp and the water dispenser in the heat preservation state recorded before appears, the current user is judged to be using the incandescent lamp and the water dispenser, the gear position of the incandescent lamp is 1 gear, and the water dispenser is in the heat preservation state.

And 102, detecting the vacancy value in the acquired state of the electric equipment.

And comprehensively analyzing the vacancy condition of the load detection data by using the acquired load detection data. Comparing the time point corresponding to the vacancy value with the running state of the current power equipment, analyzing the specific state of the power equipment under the condition that vacancy occurs, and judging whether the data cannot be collected or not because of the data caused by the characteristics of the power equipment under the state or not or whether the data cannot be collected due to the occurrence of emergency conditions or not by the power equipment under the state.

And analyzing the vacancy value and the gear change.

In actual detection data of the non-invasive load detection device, the acquired electric equipment data cannot acquire data in the current time period due to the characteristics of the detection device or the special condition of the electric equipment, so that a specific measured value cannot be acquired in a certain time period, and finally, the situation that data vacancy occurs in a certain time period exists in the acquired data. For example: load detection equipment is at the data collection in-process, the consumer has appeared the gear change or has appeared that mains current is too big and has leaded to the load data that can't collect electrical apparatus, consequently need to use the data of acquireing to carry out the vacancy value and detect, carry out the integrated analysis with the vacancy value that detects, and detect the vacancy of some period of time that the vacancy value appears, obtain wherein the time interval size that the vacancy appears through using simple logic calculation, and be used for taking notes current consumer's gear change condition according to the time interval that calculates.

And judging whether the vacancy value exists or not by using the acquired data.

Step 2: and presetting a null array H, writing the electricity consumption data of each single device into an Excel table, reading a character column for recording time in the Excel table, and presetting the currently acquired total time data of the device as A, wherein the line number of the A is changed but the column number is only one column.

And step 3: and (4) entering a cycle, presetting i as a data table of the ith electric equipment, and presetting a total of N electric equipment. I.e. 1,2.. N.

Step 6: and (3) after the circulation of the step 3 is completed, finally obtaining a two-dimensional array H, wherein the H has N rows, and the number of columns is the time data volume of the equipment with the maximum acquisition time point.

And 7: and substituting the array into a loop statement to detect the vacancy values of each device in different time periods, judging whether the ith row in the H array has a number larger than 1, and if not, judging whether the corresponding position has the vacancy value or not. For example: the first powered device is vacant at the 30 th second of starting use, that is, the number of seconds corresponding to two adjacent data is not continuous. This position is thus detected as the position where the missing value occurs.

And filling data after vacancy detection.

In the time period of the occurrence of the vacancy value, the electric equipment can cause the load detection equipment to be incapable of acquiring real-time data due to different conditions, so that the vacancy value of the data is caused, and the vacancy value is filled by calculating a new value by using the data because actual data exist in the previous period of the occurrence of the vacancy time and the later period of the end of the vacancy time. Because the time of the vacancy is not fixed, the situation of only one second of vacancy or the situation of 10 minutes of vacancy can be different. When the vacancy value is filled, the running state of the electric equipment at the current time point needs to be integrated, data filling needs to be carried out according to the electricity utilization rule under the corresponding gear of the current time point, so that the vacancy is filled through the existing data, whether the electricity utilization state of the electric equipment is changed in the vacancy time period needs to be judged, corresponding data filling is carried out according to the state change in the vacancy time period of the electric equipment, and the filled data is analyzed, wherein the filling accuracy is high. For example: if the power consumption equipment is vacant for 50 seconds in the operation process, whether the gear change exists in the power consumption equipment in the vacant time is judged, and if the gear change occurs in the 30 th second, the gear change needs to be divided into the first 30 seconds and the second 20 seconds to be respectively calculated and filled in the vacancy. Therefore, linear interpolation is used for filling the gap value temporarily, and if the state transformation exists in the gap time period, the linear interpolation algorithm is required to be used for filling the gap in a segmented mode. When the vacancy value is filled, the state change of the comprehensive electric equipment is needed to be calculated.

And 103, performing correlation analysis on the detected vacancy values and the transient state of the corresponding equipment.

And judging whether a two-dimensional array H obtained by calculation in the vacancy value exists by using the obtained data, and comparing the H of the two-dimensional array with the original detection data of the electric equipment. And judging whether the transient state of the electric equipment occurs or not in the time period of the occurrence of the vacancy value. And performing relevance analysis by judging the transient time point of the electric appliance and the time period of the data of the load detection when the vacancy value occurs, calculating the relevance degree between the transient time point and the time period of the load detection, and comprehensively analyzing the relevance between the time period of the vacancy value and the time point of the transient occurrence through the calculated relevance degree. Each state change location of the collected data is recorded and also the start time and the end time of the time period for each occurrence of a vacancy value are recorded and compared with the transient time points, respectively. And the influence of the transient states of different electric equipment on the load detection equipment is also correspondingly correlated, so that when the correlation analysis is used, the correlation analysis needs to be carried out by firstly synthesizing the transient states and the vacancy values of all the equipment, after the correlation analysis is carried out, the correlation analysis is carried out by controlling a single equipment, and the results of the two correlation analyses are comprehensively compared to carry out comprehensive judgment, so that different correlation coefficients under different electric equipment are obtained, and the correlation size between the transient state and the vacancy value of each electric equipment is obtained.

And preparing data based on the vacancy value and the transient occurrence time of the electric equipment.

The data collected by the independent operation of different devices and the operation state change condition of the devices at the current time are marked through the load detection device, so that a time point which contains the vacancy value and obtains the occurrence time period of the vacancy value and the occurrence transient state of the electric appliance is obtained. And writing the time point of each electrical appliance transient occurrence into a table. And judging whether the vacancy values exist or not according to the acquired data, acquiring the starting time point of the vacancy values of the electric appliances and the current operating state of the electric appliances under the position of the detected vacancy values of each electric appliance, judging, comparing the starting time point and the ending time point of the vacancy values with the duration time of each operating state of the electric appliances to obtain the operating state of the electric appliances when the vacancy values appear, and recording the starting time point, the ending time point and the state change conditions of the vacancy values into a table. Two situations can therefore occur: the first condition is as follows: the time period during which the vacancy value continues is within the time of a state of the consumer, and no change of the operating state occurs. Case two: two operation states of the electric equipment occur in the time period of the continuous vacancy value, namely, the electric equipment is switched in the operation state in the time period of the occurrence of the vacancy value.

And data preparation of two conditions of gear shifting and gear shifting is carried out in the vacancy value time period.

And performing data sorting on each vacant position in the data obtained by load detection, wherein in the first situation of performing data preparation based on the vacancy value and the transient occurrence time of the electric equipment, the vacancy value starting time point, the vacancy value ending time point, the current gear of the electric equipment when the vacancy value occurs and the name of the electric equipment need to be written into an Excel table. In the second case of data preparation based on the vacancy value and the transient occurrence time of the electric equipment, the vacancy value starting time point, the gear switching time point, the gear before switching, the vacancy value ending time point, the gear after switching and the name of the electric equipment need to be recorded in an Excel table. Thus, two Excel tables are obtained, which are the device state data obtained in two cases. For example: when the time period of the running state of the electric equipment in the 1 st gear is 30 seconds to 90 seconds, and the occurrence time of the vacancy value is 50 seconds to 55 seconds, the time point of the 50 th second, the time point of the 55 seconds, the current gear and the name of the electric equipment are written into a first table, otherwise, when the gear of the electric equipment is changed from the 1 st gear to the 2 nd gear in the 90 seconds, and the occurrence time of the vacancy value is 80 seconds to 100 seconds, the time point of the 80 th second, the time point of the 90 seconds, the end time point of the vacancy value, the gear 1 before switching, the gear 2 after switching and the name of the electric equipment are written into a second table.

And performing correlation analysis on the occurrence condition of the vacancy value and the transient state.

And preparing data under two conditions of gear switching and gear switching without gear within the vacancy value time period, reading the data under the first condition, and comparing the initial time point of occurrence of the vacancy value with the time point of latest gear switching. And quantitatively analyzing the time of the occurrence of the vacancy value and the time of gear switching, comparing each vacancy value time point with the latest gear switching time point, drawing a vacancy value scatter diagram, performing linear regression analysis on the scatter diagram, obtaining a correlation function relation between a transient time point and a vacancy value starting time point, ending time point and median of the time point, and judging whether the linear regression is reasonable or not through correlation. And comprehensively judging the association degree through the relational parameters obtained by the correlation analysis. And setting each gear switching time point as a function value, setting each vacancy value occurrence time point, each end time point and each median of the time points as independent variables, and performing regression analysis on the independent variables and the function values respectively. Similarly, the data in case two is read. And taking the vacancy value starting time point of different data under the same transient state, taking the ending time point and the median of the gear switching time point as independent variables, and taking the gear switching time point as the median to perform relevance analysis. And calculating to obtain a functional relation. And the relational case in both cases is compared. For example: in the case where the detection device starts collecting data from 14 hours 55 minutes 44 seconds, and the state changes at 14 hours 56 minutes 43 seconds, the value of the function y is the number of seconds at which the state is switched, i.e. the number of seconds taken at 14 hours 56 minutes 43 seconds minus 14 hours 55 minutes 44 seconds, by reading the time point of the state change and subtracting the time point at which the measurement is started. If the occurrence time of the null value is within a period of 14 hours 55 minutes 58 seconds to 14 hours 56 minutes 10 seconds, values of 14 minutes 58 seconds, 14 hours 56 minutes 10 seconds, and 14 hours 55 minutes 44 seconds subtracted from the median of the null value are respectively used as the independent variable x, and therefore, scatter charts of the null value time point and the state change time point are obtained. And performing multiple linear regression calculation on the point-in-time data through a scatter diagram to obtain a functional relation y between the point-in-time of the state change and the point-in-time of the occurrence of the vacancy value.

And 104, obtaining vacancy values of different power equipment and the change cycle of the load detection data according to the correlation analysis to predict.

And performing relevance analysis according to the occurrence condition of the vacancy value and the transient state to obtain a functional relation of a form of y ═ ax + b and vacancy value data under two conditions, performing in-depth analysis on the basis, and performing multiple linear regression analysis on the time transient state of the nearest distance by respectively taking the starting time period, the ending time period and the time median of the vacancy value as three independent variables.

And (5) regression analysis process.

And calculating a functional relation between three time point parameters of the vacancy value of each electric equipment and the state change time point of the electric equipment through multivariate nonlinear regression analysis, and performing regression prediction on the vacancy value of the electric equipment through the functional relation. The undetermined relation of a functional expression y (i) ═ b0+ b1 x1(i) + b2 x2(i) + b3 x3(i) is established, the seconds corresponding to each state change time point of each electric device are read as y values, and the seconds corresponding to each vacancy value starting time point, each vacancy value ending time point and the median of the vacancy time period are read as the values of independent variables x1, x2 and x3 respectively. And calculating the constants b0, b1, b2 and b3 by using a multiple linear regression solving function to obtain specific parameter values of the regression function relational expression. And carrying out the substitution through a functional relation to obtain the occurrence condition and the duration of the vacancy values under different gear conditions. For example: the current gear change time point of the equipment occurs at 40 seconds, the gear change when the current gear change time of the equipment is 35 seconds to 37 seconds and is closest to the gear change when the current gear change time is 40 seconds to 40 seconds, x1 is 35 seconds, x2 is 37 seconds, x3 is 36 seconds, and the like, the electricity utilization data of each electric equipment are independently analyzed by the multiple linear regression, finally, the functional relation of each electric equipment about the gear change and the vacancy value period is obtained, and the abnormal condition of the electric equipment is predicted by substituting the corresponding parameter values according to the data in the vacancy value time period.

And 105, carrying out feasibility evaluation based on the predicted result.

And the regression analysis process compares the calculated regression function relational expression with the vacancy value before the occurrence, obtains the regression function characteristic value for prediction, and comprehensively judges the prediction result to evaluate the feasibility of the power utilization condition of the predicted vacancy value position.

And (4) combining the feasibility evaluation process of a secondary load tester.

And predicting the filled part by acquiring the detected data and the functional relation according to the data of the detection time which is one second before the occurrence of the vacancy value and one second after the end of the vacancy value, and acquiring the data of the output total power, the useful power, the current voltage, the corresponding cycle and the harmonic wave to perform comprehensive analysis. And finding out the predicted power utilization data of the position corresponding to the vacancy value for evaluation. And acquiring power utilization state data of non-null values before and after the null value in the load detection data for evaluation. In a time period in which the non-invasive load detection equipment cannot acquire data, if the duration is short, the secondary load detector is not started for the moment to measure the load detection data in the vacant time period. And presetting that the time of the vacancy exceeds 5 seconds, and automatically starting a secondary load detector to replace and detect the load of the electric equipment. And when the data of the non-invasive load detection equipment is recovered to be normal and continues to detect the data, the secondary load detection equipment records the load detection data collected in the vacant time period and automatically closes the detection. And comparing a regression function relation formula made by the vacancy value time period of the load detection equipment with detection data of the secondary load detector in the vacancy time period, evaluating the feasibility of the regression function relation formula, predicting the actual state of the electric equipment according to the comparison result, and analyzing the difference and the prediction result to evaluate. For example: and under the condition that the load detection equipment has an empty value, evaluating all power utilization data of the existing detection data in the non-intrusive load detection equipment, performing regression analysis, detecting the electric equipment by using a secondary load detector within the time period of the empty value, summing difference values of the regression function relational expression and actual data of the secondary load detector, evaluating the feasibility of the prediction condition of the regression function according to the calculated difference, wherein if the difference is smaller, the feasibility is higher, and the negative feasibility is smaller.

And step 106, monitoring according to the feasibility evaluation result and the acquired load detection data of the electric equipment in the current state.

Load detection data characteristics of each electric equipment are monitored through collected load detection data, collected abnormal electricity utilization characteristic data are stored in an abnormal electricity utilization characteristic database, whether electricity utilization potential safety hazards exist in the electric equipment of a current user is comprehensively judged by acquiring and extracting the electricity utilization characteristic data in real time, according to a feasibility evaluation flow, the electric equipment load is monitored through evaluation results, and the electricity consumption of the user is properly adjusted.

The risk of the state assessment is monitored.

According to the feasibility evaluation flow combined with the secondary load tester, load detection data of the secondary load detector under the condition of the vacancy value of the load detection equipment and load detection data of the time period of the abnormal power utilization condition are obtained, the characteristics of the load detection data under the abnormal power utilization condition are subjected to risk analysis, the state of the electric equipment in the vacancy value time period is predicted, the risk of the electric equipment under the abnormal state is judged, the data characteristics and the risk condition under the vacancy condition are recorded into an abnormal power utilization characteristic database according to the current situation, and the situation is used for comparing the situation of the data of the subsequent monitoring load detection. For example: when a user uses the electric equipment, the data of load detection has a vacancy value, the duration is long, the secondary load detector is automatically started and collects the electric data, and when the non-invasive load detection equipment can acquire the detection data again, the secondary load detector uploads the data in the vacancy time period and automatically closes. The power utilization condition of the equipment under the condition of the vacancy value is monitored, and risk analysis is carried out on the power utilization equipment according to load data in the vacancy value time period. And after the risk level is obtained, carrying out difference analysis on the electricity utilization data under the condition and the data characteristics under the normal condition, obtaining the electricity utilization characteristics under the abnormal condition, and recording the electricity utilization characteristics into a database. If load data similar to the characteristic reappears in subsequent monitoring, the risk situation is analyzed through data comparison.

And 107, analyzing according to the monitored state load of the electric equipment.

The method comprises the steps of analyzing various possibilities of load of monitored electric equipment, and comprehensively judging whether the electric equipment is short-circuited or on fire according to monitored data and an evaluation result of the occurrence of a vacancy value. And analyzing whether the electric equipment catches fire or is short-circuited according to the load condition of the electric equipment measured by the secondary load detector during the vacancy value. The method comprises the steps of comprehensively evaluating load detection data before and after an empty value and data of a secondary load detector during the empty period, respectively detecting the load of the same electric equipment under different fault conditions, comparing the load detection result with a normal state, analyzing possible fault states of the electric equipment in the time period of the occurrence of the empty value when the empty value occurs according to the occurrence position of the empty value and the load detection data before and after the occurrence of the empty value, and analyzing and predicting the emergency situation of the electric equipment.

And analyzing the load characteristic of the electric equipment.

And carrying out load characteristic analysis according to the load detection data of the electric equipment in the same operation state in the time periods without the vacancy values and with the vacancy values. And comparing and analyzing the load detection data in the vacant time period with the load difference in the non-vacant time period, judging the possibility of the fault of the electric equipment used by the user in the using process, and performing abnormal early warning on the electric equipment according to the possibility. Comparing a section of load detection data which is closest to the data time difference of the vacancy value in the obtained load detection data, calculating a change ratio of the absolute value and the data without vacancy, performing risk assessment according to the difference ratio, wherein the larger the difference ratio is, the higher the risk level is, so that the risk level of each electric device is assessed, and the load characteristics under the current risk assessment are used for judging whether the electric device of the current user normally operates. If the difference proportion of the current in the vacancy value in the load detection data of the electric equipment exceeds 20%, increasing a risk grade every time the difference proportion exceeds 10% so as to evaluate the risk grade, properly reducing the transmission power of the electric equipment according to the evaluation condition of the risk grade so as to manage and control the risk, and storing the load detection data of the current electric equipment at the current risk grade into a database. For example: when load data detected by a secondary load detector of electric equipment of a user in an vacancy value time period is compared with data in a time period closest to a vacancy value in non-invasive load detection in the same state, the specific load data of the two detected data are subtracted, the difference proportion is calculated, if the current is 20A in the non-vacancy value time period, the data current detected in the vacancy value period is 30A, an absolute value is calculated by subtracting and is divided by 20A, the difference proportion is 50 percent and exceeds 20 percent, the electric equipment is assessed to be risky, the risk grade of the current electric equipment is judged to be 3, and therefore risk control is carried out on the electric equipment by reducing transmission power, and the load data under the vacancy value condition of the current electric equipment is recorded into a database.

And step 108, performing corresponding power protection according to the analyzed result.

And performing power limiting evaluation according to the load characteristic analysis of the electric equipment and the risk level obtained by analysis, acquiring the load detection data condition of the electric equipment in the period of the vacancy value to perform power limiting calculation, performing adjustment calculation according to the risk level of the condition of the electric equipment and the difference proportion of the corresponding load detection data exceeding the normal condition, feeding back the calculated result to an ammeter for providing electric power, and properly limiting the current on the live wire and inputting the current and the voltage of the user to regulate and control.

And protecting the electric equipment by analyzing the risk level.

And judging the dangerous condition of the electric equipment in the electric equipment according to the risk grade calculated during the vacancy value, and judging whether the electric equipment has an equipment short circuit or an equipment fire. And predicting and analyzing the obtained risk grade and the load difference proportion between the corresponding vacancy value time period and the normal condition to obtain the electric equipment with the possibility of fire or fault in the electric equipment, protecting the electric equipment with the possibility of fire or fault by limiting the output power of the electric equipment, and ensuring the electric safety of a user by power failure if necessary. Judging the possibility of fire or short circuit fault of the electric equipment according to the difference proportion calculated by the load detection data of the electric equipment in the empty value time period, respectively obtaining the difference proportion and risk grade of the data corresponding to the current, the voltage, the power and the cycle harmonic wave, evenly distributing the weights of the current, the voltage, the power, the cycle wave and the harmonic wave, respectively multiplying and summing the weights by the corresponding difference proportion to obtain the comprehensive difference proportion, judging whether the difference proportion exceeds 20% of the non-empty value, if the difference proportion exceeds the corresponding range, limiting the power of the electric equipment according to the proportion, and if the difference proportion exceeds 100%, carrying out power-off operation. For example: the difference calculation is carried out on the data of the non-invasive load detection and the data of the secondary load detector in the vacancy value time period to obtain that the current difference proportion is 10%, the voltage difference proportion is 80%, the power difference proportion is 50%, the cycle difference proportion is 5% and the harmonic difference proportion is 10%, then the difference proportion is weighted and summed to obtain that the average difference percentage is 31%, therefore, the power limiting protection is carried out on the electric equipment which is likely to have fire or fault through the limitation of input power, the real-time difference proportion calculation is carried out on the obtained new detection data, and the difference proportion is stabilized within 20%, so that the safety accidents caused by the fire or fault are reduced.

The above description is only an embodiment of the present invention, and not intended to limit the scope of the present invention, and all modifications of equivalent structures and equivalent processes, which are made by using the contents of the present specification and the accompanying drawings, or directly or indirectly applied to other related technical fields, are included in the scope of the present invention.

Programs for implementing the information governance of the present invention may be written in computer program code for carrying out operations of the present invention in one or more programming languages, including an object oriented programming language such as Java, python, C + +, or a combination thereof, as well as conventional procedural programming languages, such as the C language or similar programming languages.

The program code may execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any type of network, including a Local Area Network (LAN) or a Wide Area Network (WAN), or the connection may be made to an external computer (for example, through the Internet using an Internet service provider).

In the embodiments provided in the present invention, it should be understood that the disclosed system, apparatus and method may be implemented in other ways. For example, the above-described device embodiments are merely illustrative, and for example, the division of the units is only one logical functional division, and other divisions may be realized in practice.

The units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one position, or may be distributed on multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of the embodiment.

In addition, functional units in the embodiments of the present invention may be integrated into one processing unit, or each unit may exist alone physically, or two or more units are integrated into one unit. The integrated unit can be realized in a form of hardware, or in a form of hardware plus a software functional unit.

The integrated unit implemented in the form of a software functional unit may be stored in a computer readable storage medium. The software functional unit is stored in a storage medium and includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device) or a processor (processor) to execute some steps of the methods according to the embodiments of the present invention.

And the aforementioned storage medium includes: various media capable of storing program codes, such as a usb disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk, or an optical disk.

Claims

1. A non-intrusive based power load detection and decomposition method, the method comprising:

the method for acquiring the power state of the electric equipment of the user through the load detection data specifically comprises the following steps: clustering and analyzing the load detection data to infer the state of the electric equipment; detecting the vacancy value in the acquired state of the electric equipment, wherein detecting the vacancy value in the acquired state of the electric equipment specifically comprises: analyzing the vacancy value and the gear change, judging whether the vacancy value exists or not by using the acquired data, and filling data after vacancy detection; performing correlation analysis on the detected vacancy value and the corresponding device transient state, wherein the correlation analysis on the detected vacancy value and the corresponding device transient state specifically includes: data preparation is carried out based on the vacancy value and the transient state occurrence time of the electric equipment, the data preparation comprises data preparation of two conditions of gear switching and gear switching-free within a vacancy value time period, and then relevance analysis is carried out on the occurrence condition and the transient state of the vacancy value; obtaining vacancy values and load detection data change periods of different power equipment according to correlation analysis for prediction, and obtaining vacancy values and load detection data change periods of different power equipment according to correlation analysis for prediction, specifically comprising: a regression analysis process; performing feasibility evaluation based on the predicted result, wherein the feasibility evaluation based on the predicted result specifically comprises the following steps: combining the feasibility evaluation process of a secondary load tester; monitoring according to the feasibility evaluation result and the acquired load detection data of the electric equipment in the current state, wherein monitoring according to the feasibility evaluation result and the acquired load detection data of the electric equipment in the current state specifically comprises: monitoring the risk of state evaluation; analyzing according to the monitored state load of the electric equipment, wherein the analyzing according to the monitored state load of the electric equipment specifically comprises: analyzing the load characteristics of the electric equipment; and performing corresponding power protection through the result obtained by analysis, wherein the power protection comprises the following steps: and protecting the electric equipment by analyzing the risk grade.

2. The method of claim 1, wherein the obtaining of the power status of the consumer's electrical equipment from the load detection data comprises:

the method for clustering, analyzing and inferring the state of the electric equipment by using the load detection data specifically comprises the following steps:

analyzing the electric equipment according to the electric data actually acquired by load detection, judging the type of the electric equipment and the gear change condition of the electric equipment by integrating the steady-state and transient-state electric data within the operation time period of the electric equipment, and analyzing according to the load detection data; recording matched electric equipment under the condition of current electricity utilization change, and recording load detection data characteristics of the current equipment; judging the corresponding electric equipment according to the existing electric characteristics in the actual application; finding out a clustering center of the electricity utilization characteristics of the electric equipment by using a MeanShift algorithm for the electricity utilization characteristic data of the wavelength of the electric equipment, and recording; when the non-invasive load detection equipment is used by a plurality of different electric appliances, the cluster analysis is also carried out according to different electricity utilization characteristics, the characteristic range of each different electric appliance is found out according to different electricity utilization characteristics, and the type of the electric appliance at the moment is judged according to the existing electric appliance characteristic record.

3. The method of claim 1, wherein the detecting for the vacancy value in the acquired powered device state comprises:

the analysis of the vacancy value and the gear change specifically comprises the following steps:

in actual detection data of the non-invasive load detection equipment, the acquired electric equipment data cannot acquire data in the current time period due to the characteristics of the detection equipment or the special condition of the electric equipment, so that a specific measured value cannot be acquired in a certain time period, and finally, the situation that data vacancy occurs in a certain time period exists in the acquired data;

the determining whether the vacancy value exists by using the acquired data specifically includes:

performing vacancy detection through the acquired data, wherein the acquired time point data is acquired in a character string form, and the time points are subjected to addition and subtraction operation by reading the character string and performing date conversion, and by judging whether the interval between the two time points is more than 1 second or not, if the interval is more than one second, the data vacancy condition occurs in the corresponding time period, and if the interval is equal to one second, the data vacancy condition does not occur; aiming at whether the vacancy value exists or not, firstly, judging the single electric equipment; the calculation is performed by:

step 1: acquiring power utilization data of each power utilization device through load detection equipment;

step 2: presetting a null array H, writing the electricity consumption data of each single device into an Excel table, reading a character column for recording time in the Excel table, and presetting the total time data of the currently acquired device as A, wherein the number of lines of the A is changed but the number of columns is only one;

and step 3: entering a cycle, presetting i as a data table of the ith electric equipment, and presetting a total of N electric equipment; i 1,2.. N;

and 4, step 4: in the cycle of the step 3, presetting j as the jth time data of the ith device, and entering the cycle; acquiring time data of a jth character string type of an ith electric device, namely j is 1,2.. length (A) -1, modifying the time data of the character string type into a numerical value type, solving a difference value of two adjacent time data, and taking an absolute value; recording the calculated numerical value to the jth position of the ith row of the array H no matter whether the numerical value is larger than 1 or not;

and 5: when the circulation of the step 4 is completed, obtaining the value of the ith row of the array H;

step 6: after the cycle of the step 3 is completed, a two-dimensional array H is finally obtained, wherein the H has N rows, and the number of columns is the time data volume of the device with the maximum acquisition time point;

and 7: substituting the array into a loop statement to detect vacancy values of each device in different time periods, and judging whether the ith row in the H array has a number greater than 1, if not, the corresponding position has a vacancy value, otherwise, the ith row does not have a vacancy value;

the data filling is performed after the vacancy detection, and the method specifically comprises the following steps:

in the time period of the occurrence of the vacancy value, the electric equipment can cause the load detection equipment to be incapable of acquiring real-time data due to different conditions, so that the vacancy value of the data is caused, and the vacancy value is filled by calculating a new value by using the data because actual data exists in the front period of the occurrence of the vacancy time and the later period of the end of the vacancy time; because the vacancy time is not fixed, the situation that only one second is vacant or the situation that 10 minutes is vacant can be different; when the vacancy value is filled, the running state of the electric equipment at the current time point needs to be integrated, data filling needs to be carried out according to the electricity utilization rule under the corresponding gear of the current time point, so that the vacancy is filled through the existing data, whether the electricity utilization state of the electric equipment is changed in the vacancy time period needs to be judged, corresponding data filling is carried out according to the state change in the vacancy time period of the electric equipment, and the filled data is analyzed, wherein the filling accuracy is high.

4. The method of claim 1, wherein the correlating the detected occurrence of the vacancy values with corresponding device transients comprises:

judging whether a two-dimensional array H obtained by calculation in the vacancy value exists by using the obtained data, and comparing the H of the two-dimensional array with the original detection data of the electric equipment; judging whether the transient state of the electric equipment occurs or not in the time period of the occurrence of the vacancy value; performing relevance analysis by judging the transient time point of the electric appliance and the data vacancy value time period of the load detection, calculating the relevance degree between the transient time point and the data vacancy value time period, and comprehensively analyzing the relevance between the vacancy value time period and the transient state occurrence time point through the calculated relevance degree; recording each state change position of the collected data, and recording the start time and the end time of each time period of the occurrence of the vacancy value, and comparing the start time and the end time with the transient time point respectively; the influence of the transient states of different electric equipment on the load detection equipment is correspondingly correlated, so that when the correlation analysis is used, the correlation analysis needs to be carried out by firstly synthesizing the transient states and the vacancy values of all the equipment, after the correlation analysis is carried out, the correlation analysis is carried out by controlling the single equipment, and the results of the two correlation analyses are comprehensively compared to carry out comprehensive judgment so as to obtain different correlation coefficients under different electric equipment, thereby obtaining the correlation size between the transient state and the vacancy value of each electric equipment; the method comprises the following steps: preparing data based on the vacancy value and the transient occurrence time of the electric equipment; preparing data under two conditions of gear switching and gear switching-free within the vacancy value time period; carrying out correlation analysis on the occurrence condition of the vacancy values and the transient state;

the data preparation based on the vacancy value and the transient occurrence time of the electric equipment specifically comprises the following steps:

marking the data collected by the independent operation of different devices and the operation state change condition of the devices at the current time by using the load detection device, thereby obtaining a time point which contains a vacancy value and acquires a time period of occurrence of the vacancy value and a transient state of the electric appliance; writing the time point of each electrical appliance transient state into a table; judging whether the vacancy values exist or not according to the acquired data, acquiring the starting time point of the vacancy values of the electric appliances and the current running state of the electric appliances to judge when each detected electric appliance is at the position of the occurrence of the vacancy values, comparing the starting time point and the ending time point of the vacancy values with the duration time of each running state of the electric appliances to obtain the running state of the electric appliances when the vacancy values occur, and recording the starting time point, the ending time point and the state change conditions of the vacancy values into a table; two situations can therefore occur: the first condition is as follows: the continuous time period of the vacancy value is within the time of one state of the electric equipment, and the change of the running state does not occur; case two: two running states of the electric equipment appear in the continuous time period of the vacancy value, namely the electric equipment is switched in the running state in the time period of the occurrence of the vacancy value;

data preparation of two conditions of gear switching and gear switching-free within the vacancy value time period specifically comprises the following steps:

performing data sorting on each vacant position in the data obtained by load detection, wherein in the first situation of performing data preparation based on the vacancy value and the transient occurrence time of the electric equipment, the vacancy value starting time point, the vacancy value ending time point, the current gear of the electric equipment when the vacancy value occurs and the name of the electric equipment are written into an Excel table; in the second situation of data preparation based on the vacancy value and the transient occurrence time of the electric equipment, the vacancy value starting time point, the gear switching time point, the gear before switching, the vacancy value ending time point, the gear after switching and the name of the electric equipment need to be recorded into an Excel table; thus, two Excel tables are obtained and are respectively the equipment state data obtained under two conditions;

the correlation analysis is carried out on the occurrence condition of the vacancy values and the transient state, and the correlation analysis specifically comprises the following steps:

preparing data under two conditions of gear switching and gear switching without gear within the vacancy value time period, reading the data under the first condition, and comparing the initial time point of occurrence of the vacancy value with the time point of latest gear switching; the time of the occurrence of the vacancy value and the time of gear switching are quantitatively analyzed, each vacancy value time point is compared with the latest gear switching time point to draw a vacancy value scatter diagram, linear regression analysis is carried out on the scatter diagram, a correlation function relation between a transient time point and a vacancy value starting time point, ending time point and median of the time point is obtained, and whether the linear regression is reasonable or not is judged through correlation; comprehensively judging the correlation degree through the relational parameters obtained by the correlation analysis; setting each gear switching time point as a function value, setting each vacancy value occurrence time point, ending time point and median of the time points as independent variables respectively, and performing regression analysis with the function values respectively; similarly, reading the data under the second condition; taking the vacancy value starting time point of different data under the same transient state, taking the ending time point and the median of the gear switching time point as independent variables, and taking the gear switching time point as the median to perform relevance analysis; calculating to obtain a functional relation; and the relational case in both cases is compared.

5. The method of claim 1, wherein the deriving the vacancy values and the load detection data variation periods of different power equipment for prediction according to correlation analysis comprises:

the regression analysis process specifically comprises the following steps:

calculating a function relation between three time point parameters of the vacancy value of each piece of electric equipment and the state change time point of the electric equipment through multivariate nonlinear regression analysis, and performing regression prediction on the vacancy value of the electric equipment through the function relation; establishing a pending relation of a functional formula y (i) ═ b0+ b1 x1(i) + b2 x2(i) + b3 x3(i), reading the seconds corresponding to each state change time point of each electric device as a y value, reading the seconds corresponding to each vacancy value start time point, each vacancy value end time point and the seconds corresponding to the median of the vacancy time period as the values of independent variables x1, x2 and x3 respectively; calculating constants b0, b1, b2 and b3 by using a multiple linear regression solving function to obtain specific parameter values of a regression function relational expression; and carrying out the substitution through a functional relation to obtain the occurrence condition and the duration of the vacancy values under different gear conditions.

6. The method of claim 1, wherein the performing a feasibility assessment based on the predicted outcome comprises:

the feasibility evaluation process combined with the secondary load tester specifically comprises the following steps:

predicting the filled part by the data of the detection time which is one second before the occurrence of the vacancy value and one second after the end of the vacancy value through acquiring the detected data and the functional relation, and acquiring the data of the output total work, the useful work, the current voltage, the corresponding cycle and the harmonic wave to carry out comprehensive analysis; finding out the predicted power utilization data of the position corresponding to the vacancy value for evaluation; acquiring power utilization state data of non-null values before and after null value in the load detection data for evaluation; in a time period in which the non-invasive load detection equipment cannot acquire data, if the duration is short, the secondary load detector is not started for the time to measure the load detection data in the vacant time period; presetting that when the vacancy duration exceeds 5 seconds, the secondary load detector is automatically started to perform replacement detection on the load of the electric equipment; when the data of the non-invasive load detection equipment is recovered to be normal and continues to detect the data, the secondary load detection equipment records the load detection data collected in the vacant time period and automatically closes the detection; and comparing a regression function relation formula made by the vacancy value time period of the load detection equipment with detection data of the secondary load detector in the vacancy time period, evaluating the feasibility of the regression function relation formula, predicting the actual state of the electric equipment according to the comparison result, and analyzing the difference and the prediction result to evaluate.

7. The method of claim 1, wherein the monitoring according to the feasibility assessment result and the obtained load detection data of the current state of the electric equipment comprises:

the monitoring of the risk of the state assessment specifically includes:

8. The method of claim 1, wherein the analyzing based on the monitored status load of the powered device comprises:

analyzing various possibilities of the load of the monitored electric equipment, and comprehensively judging whether the electric equipment has a short circuit or is on fire according to the monitored data and the evaluation result of the occurrence of the vacancy value; analyzing whether the electric equipment catches fire or is short-circuited according to the load condition of the electric equipment measured by the secondary load detector during the vacancy value period; comprehensively evaluating load detection data before and after the vacancy value and data of a secondary load detector during the vacancy period, respectively detecting the load of the same electric equipment under different fault conditions, comparing the load detection result with a normal state according to the load detection result, analyzing possible fault states of the electric equipment in the time period of occurrence of the vacancy value when the vacancy value occurs according to the occurrence position of the vacancy value and the load detection data before and after the occurrence of the vacancy value, and analyzing and predicting the emergency situation of the electric equipment; the method comprises the following steps: analyzing the load characteristics of the electric equipment;

the load characteristic analysis of the electric equipment specifically comprises the following steps:

carrying out load characteristic analysis according to load detection data of the electric equipment in the same running state in the time periods without the vacancy values and with the vacancy values; comparing and analyzing the load detection data in the vacant time period with the load difference in the non-vacant time period, judging the possibility of failure of the electric equipment used by the user in the using process, and performing abnormal early warning on the electric equipment according to the possibility; comparing a section of load detection data which is closest to the data time difference of the vacancy value in the obtained load detection data to obtain an absolute value, calculating a change ratio of the absolute value and the data under the condition of no vacancy, performing risk evaluation according to the difference ratio, wherein the larger the difference ratio is, the higher the risk level is, so that the risk level of each electric device is evaluated, and the load characteristics under the current risk evaluation are used for judging whether the electric devices of the current user normally operate or not; if the difference proportion of the current in the vacancy value in the load detection data of the electric equipment exceeds 20%, increasing a risk grade every time the difference proportion exceeds 10% so as to evaluate the risk grade, properly reducing the transmission power of the electric equipment according to the evaluation condition of the risk grade so as to manage and control the risk, and storing the load detection data of the current electric equipment at the current risk grade into a database.

9. The method of claim 1, wherein the analyzing results for respective power protection comprises:

according to the load characteristic analysis of the electric equipment, carrying out power limiting evaluation according to the risk level obtained by analysis, obtaining the load detection data condition of the electric equipment in the period of the vacancy value to carry out power limiting calculation, carrying out adjustment calculation according to the risk level of the condition of the electric equipment and the difference proportion of the corresponding load detection data exceeding the normal state, feeding back the calculated result to an electric meter for providing electric power, and properly limiting the current on a live wire and inputting the current and the voltage of a user so as to regulate and control the current;

the electric equipment protection is carried out by analyzing the risk grade, and the method specifically comprises the following steps:

judging the dangerous condition of the electric equipment in the electric equipment through the risk grade calculated during the vacancy value, and judging whether the electric equipment has an equipment short circuit or an equipment fire; the electric equipment with the possibility of fire or fault is obtained through prediction analysis according to the obtained risk level and the load difference proportion between the corresponding vacancy value time period and the load under the normal condition, the electric equipment with the possibility of fire or fault is protected by limiting the output power of the electric equipment, and the electric safety of a user is ensured through power failure under the necessary condition; judging the possibility of fire or short circuit fault of the electric equipment according to the difference proportion calculated by the load detection data of the electric equipment in the empty value time period, respectively obtaining the difference proportion and risk grade of the data corresponding to the current, the voltage, the power and the cycle harmonic wave, evenly distributing the weights of the current, the voltage, the power, the cycle wave and the harmonic wave, respectively multiplying and summing the weights by the corresponding difference proportion to obtain the comprehensive difference proportion, judging whether the difference proportion exceeds 20% of the non-empty value, if the difference proportion exceeds the corresponding range, limiting the power of the electric equipment according to the proportion, and if the difference proportion exceeds 100%, carrying out power-off operation.