Disclosure of Invention
The technical problem to be solved by the invention is to provide a non-invasive household load identification method for demand side management, which realizes household electrical load identification through steps 1-3 and the like.
In order to solve the technical problems, the technical scheme adopted by the invention is as follows: a non-invasive household load identification method facing demand side management comprises steps 1-3, step 1, collecting and processing mixed load data of a user at a power supply inlet of the user to obtain processed furniture load data;
step 2, performing switching event detection on the processed furniture load data obtained in the step 1 to obtain a load switching event of the detected electric appliance, wherein the load switching event is an operation switching condition or a change of an operation state;
and 3, extracting load characteristics of the load switching event obtained in the step 2 to obtain the load characteristics.
The further technical scheme is as follows: further comprising step 4, step 4: and (4) comparing the load characteristics extracted in the step (3) with the data in the database, and decomposing the power load.
The further technical scheme is as follows: further comprising step 5, step 5: and (4) feeding the power utilization load obtained in the step (4) as a recognition result back to the data center server, and analyzing the power utilization by the data center server according to the power utilization load of the user to obtain the power consumption.
The further technical scheme is as follows: in step 2, X is assumed to be { X ═ XmN is k electrical signals of the current time period.
The further technical scheme is as follows: in the step 2, judging the change condition of the event according to the formula 1;
in the formula 1, H0Check for "no change occurred"; h1Checking for "changed"; x is the number ofmIs as followsThe power values of the m electrical signals, in watts; x is the number ofm+1The power value of m +1 electric signals is unit watt; θ is the change threshold in watts.
The further technical scheme is as follows: in the step 3, the active power waveform at the moment when the electric equipment is turned on is used as the load characteristic.
The further technical scheme is as follows: in the step 3, calculating the opening instant energy of each type of load by using the formula 2;
in equation 2, v (k-1) is the actual measurement of the voltage at monitoring point k-1 in volts; v (k) is an actual measurement of the voltage at monitoring point k in volts; v (k) is the rate of change of instantaneous voltage at monitoring point k, in volts; i (k-1) is an actual measured value of the current at the monitoring point k-1 and is in ampere; i (k) is the actual measurement of current at monitoring point k in amperes; i (k) is the rate of change of instantaneous current at monitoring point k in amperes; k is the number of the monitoring points, and the number of the total monitoring points is the unit; sTTo turn on the instantaneous energy, in watts.
The further technical scheme is as follows: in the step 4, identifying the electric load by formula 3;
in formula 3, AjA characteristic matrix composed of load characteristic quantities of the j-th class of electric loads, a characteristic matrix composed of all characteristics in the whole electric scene, a state matrix composed of the operation states of the M-class of electric loads, and XjIs an element of X, Xj1 is taken to indicate that the type of electrical load is in a working state, xjTaking 0 to represent that the class of electric loads are in a closed state, and d (C, B) represents the distance between the combination of all load characteristics and the total load, and the unit is watt; minimum distance in Watts between the combination of all load characteristics of mind (C, B) and the total load.
The further technical scheme is as follows: in the step 5, the power consumption of each type of load is calculated according to the identification result of each type of power load.
The further technical scheme is as follows: in the step 5, calculating the power consumption of each type of load according to formula 4;
in the formula 4, the first step is,
is the power consumption of load M during the kth run, in watts; n represents the number of measurements in units; n is the total measurement times and the unit times; t is t
nThe time when the household load is started is in unit of minutes; t is t
n-1The time of the first minute for the household load to start, in units of minutes; p
nThe power monitoring value of the nth minute is unit watt; p
n-1The power monitoring value in the previous minute is in watts.
Adopt the produced beneficial effect of above-mentioned technical scheme to lie in:
firstly, a non-invasive household load identification method facing to demand side management comprises the steps of 1-3, wherein in the step 1, mixed load data of a user is collected and processed at a power supply inlet of the user to obtain processed furniture load data; step 2, performing switching event detection on the processed furniture load data obtained in the step 1 to obtain a load switching event of the detected electric appliance, wherein the load switching event is an operation switching condition or a change of an operation state; and 3, extracting load characteristics of the load switching event obtained in the step 2 to obtain the load characteristics. The household electrical load identification is realized through the steps 1 to 3 and the like.
Secondly, the technical scheme can obtain the actual energy consumption level of various loads in the user, and scientific collection and management of energy efficiency data are realized.
Thirdly, the technical scheme can enable a user to timely adjust the power utilization scheme according to the feedback of the data center, so that the power utilization is reasonable, and the effects of energy conservation and emission reduction are achieved.
Fourthly, the technical scheme can enable the user to check the abnormal power utilization equipment as early as possible, and ensure the safe power utilization of the user.
See detailed description of the preferred embodiments.
Detailed Description
The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is obvious that the described embodiments are only a part of the embodiments of the present application, and not all of the embodiments. The following description of at least one exemplary embodiment is merely illustrative in nature and is in no way intended to limit the application, its application, or uses. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present application.
In the following description, numerous specific details are set forth in order to provide a thorough understanding of the present application, but the present application may be practiced in other ways than those described herein, and it will be apparent to those of ordinary skill in the art that the present application is not limited to the specific embodiments disclosed below.
As shown in fig. 1, the invention discloses a non-invasive home load identification method facing demand side management, which comprises steps 1 to 5, and specifically comprises the following steps:
step 1, collecting and processing mixed load data of a user at a power supply inlet of the user to obtain processed furniture load data.
And 2, performing switching event detection on the processed furniture load data obtained in the step 1 to obtain a load switching event of the detected electric appliance, wherein the load switching event is an operation switching condition or a change of an operation state.
And 3, extracting load characteristics of the load switching event obtained in the step 2 to obtain the load characteristics.
And 4, step 4: and (4) comparing the load characteristics extracted in the step (3) with the data in the database, and decomposing the power load.
And 5: and (4) feeding the power load obtained in the step (4) as a recognition result back to the data center server, and carrying out energy consumption analysis by the data center according to the power load of the user so as to provide more efficient and convenient energy efficiency service for the user.
In step 2, X is assumed to be { X ═ XmN is k electrical signals of the current time period, and the change condition of the event is judged by formula 1.
In the formula 1, H0Check for "no change occurred"; h1Checking for "changed"; x is the number ofmThe power value of the mth electric signal is unit watt; x is the number ofm+1The power value of m +1 electric signals is unit watt; θ is the change threshold in watts.
In the step 3, the active power waveform at the moment of starting the electric equipment is used as the load characteristic, and then the energy at the moment of starting each type of load is calculated by using the formula 2.
In equation 2, v (k-1) is the actual measurement of the voltage at monitoring point k-1 in volts; v (k) is an actual measurement of the voltage at monitoring point k in volts; v (k) is the rate of change of instantaneous voltage at monitoring point k, in volts; i (k-1) is an actual measured value of the current at the monitoring point k-1 and is in ampere; i (k) is the actual measurement of current at monitoring point k in amperes; i (k) is the rate of change of instantaneous current at monitoring point k in amperes; k is the number of the monitoring points, and K is the total number of the monitoring points in unit;STTo turn on the instantaneous energy, in watts.
In step 4, the electrical load is identified by equation 3.
In formula 3, AjA characteristic matrix composed of load characteristic quantities of the j-th class of electric loads, a characteristic matrix composed of all characteristics in the whole electric scene, a state matrix composed of the operation states of the M-class of electric loads, and XjIs an element of X, Xj1 is taken to indicate that the type of electrical load is in a working state, xjTaking 0 to represent that the class of electric loads are in a closed state, and d (C, B) represents the distance between the combination of all load characteristics and the total load, and the unit is watt; minimum distance in Watts between the combination of all load characteristics of mind (C, B) and the total load.
In the step 5, the power consumption of each type of load is calculated by formula 4 according to the identification result of each type of power load.
In the formula 4, the first step is,
is the power consumption of load M during the kth run, in watts; n represents the number of measurements in units; n is the total measurement times and the unit times; t is t
nThe time when the household load is started is in unit of minutes; t is t
n-1The time of the first minute for the household load to start, in units of minutes; p
nThe power monitoring value of the nth minute is unit watt; p
n-1The power monitoring value in the previous minute is in watts.
The purpose of the application is:
the technical scheme of the application provides a non-invasive household load identification method for supply and demand interaction aiming at the problems of supply and demand interaction of a smart grid, load information acquisition of residents by the smart grid and data mining.
The invention concept of the application is as follows:
most of the existing non-invasive household load identification technologies are based on high-frequency sampling, the requirement on hardware is high, a data acquisition device on a user side such as an intelligent electric meter is low-frequency sampling, and the method is not beneficial to popularization, so that user load data cannot be fully utilized.
Technical contribution of the present application:
in order to solve the problems, the application provides a non-invasive household load identification method for supply and demand interaction. The technical scheme of this application utilizes the sensor of installing at user entrance to carry out data acquisition, through information such as analysis user's power consumption total current and hybrid power come the operational aspect of every load of discernment user inside, and concrete step is as follows:
step 1: collecting and processing mixed load data of a user at a user power supply inlet:
in order to improve the accuracy of the acquired data, the acquired data needs to be denoised and standardized, and then the acquired data is converted into data which is convenient for model identification, namely load characteristic data such as active power, reactive power and the like.
Step 2: switching event detection is carried out on the processed data, and the purpose is to detect the operation switching condition or the change of the operation state of the electric appliance:
let X be { X ═ XmN is k electrical signals of the current time period, and the change condition of the event, H, can be judged by the hypothesis test of formula 10Defined as "no change" test, H1Defined as the test "changed", theta isA set change threshold.
And step 3: extracting load characteristics through a load switching event detected by the event:
the technical scheme of the application utilizes the active power waveform of the power utilization equipment at the moment of starting as the load characteristic, and then calculates the energy at the moment of starting various loads by utilizing a formula 2, wherein v (K-1) is the actual measurement value of the voltage at a monitoring point K-1, v (K) is the actual measurement value of the voltage at the monitoring point K, V (K) is the change rate of the instantaneous voltage at the monitoring point K, i (K-1) is the actual measurement value of the current at the monitoring point K-1, i (K) is the actual measurement value of the current at the monitoring point K, I (K) is the change rate of the instantaneous current at the monitoring point K, K is the total number of the monitoring points, and S (K) is the total number of the monitoring pointsTTo turn on the instantaneous energy.
And 4, step 4: comparing the extracted load characteristics with data in a database, and decomposing the power load:
the load identification problem of the present application can be described as formula 3, where AjA characteristic matrix composed of load characteristic quantities of the j-th class of electric loads, a characteristic matrix composed of all characteristics in the whole electric scene, a state matrix composed of the operation states of the M-class of electric loads, and Xi1 is taken to indicate that the type of electrical load is in a working state, xiTaking 0 indicates that the class of electrical loads is in the off state, and d (C, B) indicates the distance between the total load and the possible combination of all load characteristics.
And 5: the recognition result of the non-invasive household load recognition system is fed back to the data center, and the data center performs energy consumption analysis according to the information, so that more efficient and convenient energy efficiency service is provided for the user:
the power consumption of each type of load can be calculated by formula 4 according to the identification result of each type of power load, wherein,
for the power consumption of the load M during the kth operation, t
iAnd t
i-1Respectively the time of opening the domestic load and the time of one minute before the opening, P
iAnd P
i-1Power monitoring values for the nth minute and the previous minute, respectively.
The energy consumption analysis of the household load can be carried out through the calculation result, the abnormal electricity utilization behavior of the user can be found, the monitored abnormal condition can be timely fed back to the power user by the data center, the user can adjust the electricity utilization mode and check the abnormal electricity utilization equipment as soon as possible, the electricity utilization mode can be helped to save the electricity charge expense of the user, and the safe electricity utilization of the user can be ensured.
Description of the technical solution:
the embodiment of the application provides a non-invasive household load identification method for supply and demand interaction aiming at the problems of supply and demand interaction of a smart grid, and load information acquisition and data mining of residents by the smart grid.
The non-invasive household load identification method for supply and demand interaction is described in detail below, and specifically includes the following steps 1-5:
step 1, collecting and processing mixed load data of a user at a user power supply inlet:
as shown in fig. 2, for example, a home user is assumed to have a telephone, a television, a refrigerator, an air conditioner, and other power loads, and a monitoring and identifying device is installed at a power inlet of the user, so as to collect and process data through the device.
Step 2, switching event detection is carried out on the processed data, and the purpose is to detect the operation switching condition or the change of the operation state of the electric appliance:
when the state of the electrical load in the home of the user changes, the data collected by the monitoring and identifying device also changes, but the load data has certain fluctuation, so a threshold value theta is set, and when the change value is larger than the threshold value theta, the running state of the electrical load in the home of the user is considered to change.
And 3, extracting load characteristics through the load switching event detected by the event:
as shown in fig. 1, when an event is monitored, load characteristics are extracted, in the present application, an active power waveform at the moment of starting an electrical device is used as the load characteristics, and then, the energy at the moment of starting various loads is calculated by using formula 2.
Step 4, comparing the extracted load characteristics with data in a database, and decomposing the power load:
as shown in fig. 2, the load characteristic library may be obtained through a communication network, and then matched with the extracted load characteristic by equation 3, so as to resolve the power load operation state in the home of the user.
And 5, feeding back the recognition result of the non-invasive household load recognition system to a data center, and carrying out energy consumption analysis by the data center according to the information to provide more efficient and convenient energy efficiency service for users:
as shown in fig. 2, the identification result is obtained and then uploaded to a data center through a communication network, and the data center analyzes the energy consumption and electricity consumption of the user through technologies such as data mining and feeds back the analysis result to the user, so that the user can make adjustments in time, and the electricity is consumed more efficiently and environmentally.
As shown in fig. 3, the recognition effect of the invention is analyzed by taking a refrigerator as an example, and from comparison between the actual power of the refrigerator and the decomposed estimated power, it can be seen that the invention can recognize each state transition process of the refrigerator, the state change of each electrical appliance is regular, when the refrigerator breaks down, the transition power of the refrigerator state changes, and at this time, the data center can find the refrigerator fault in time by analyzing the recognized estimated power change, and inform the user to check in time, so as to ensure the safety of power utilization of the user, and at the same time, can analyze the energy consumption condition of the refrigerator, and similarly, can analyze the electrical appliance condition and the energy consumption condition of the whole family of the user.
After the application runs secretly for a period of time, the feedback of field technicians has the advantages that:
the energy consumption analysis of household loads can be carried out, the abnormal electricity utilization behavior of the user is found, the data center can feed the monitored abnormal condition back to the power user in time, the user can adjust the electricity utilization mode and check the abnormal electricity utilization equipment as soon as possible, the electricity utilization mode can be helped to save electricity charge of the user, and the safe electricity utilization of the user can be ensured.