CN108306284A - A kind of online load modeling method measured based on local intelligence - Google Patents
A kind of online load modeling method measured based on local intelligence Download PDFInfo
- Publication number
- CN108306284A CN108306284A CN201810010291.0A CN201810010291A CN108306284A CN 108306284 A CN108306284 A CN 108306284A CN 201810010291 A CN201810010291 A CN 201810010291A CN 108306284 A CN108306284 A CN 108306284A
- Authority
- CN
- China
- Prior art keywords
- load
- pressure side
- model
- induction motor
- user
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
- H02J3/00—Circuit arrangements for ac mains or ac distribution networks
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
- H02J2203/00—Indexing scheme relating to details of circuit arrangements for AC mains or AC distribution networks
- H02J2203/20—Simulating, e g planning, reliability check, modelling or computer assisted design [CAD]
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
- H02J3/00—Circuit arrangements for ac mains or ac distribution networks
- H02J3/003—Load forecast, e.g. methods or systems for forecasting future load demand
Landscapes
- Engineering & Computer Science (AREA)
- Power Engineering (AREA)
- Tests Of Circuit Breakers, Generators, And Electric Motors (AREA)
- Supply And Distribution Of Alternating Current (AREA)
Abstract
The present invention discloses a kind of online load modeling method measured based on local intelligence, includes the following steps：Step 1, the daily load data based on intelligent meter data recording system acquisition carry out user clustering, and selected part typical user carries out intrusive load monitoring in every class user, and load monitoring device is intelligent socket；Step 2, the real-time load information based on intelligent socket acquisition carries out online survey and debates, and obtains load model, other users in such is substituted using the load model of every Lei Zhong typical users, to obtain the integrated load model of all users in real time；Step 3, the polymerization model of underlying user static load and induction motor load is established respectively；Step 4, the influence for considering distribution network, by static load and induction motor load polymerization model from low-pressure side gradually to high-pressure side equivalence, the final integrated load model obtained under power distribution network 220/110kV busbares.Such method can improve the accuracy of load model, reflect the actual characteristic of load.
Description
Technical field
It is the invention belongs to technical field of power system operation control, more particularly to a kind of based on the online of local intelligence measurement
Load modeling method.
Background technology
The accuracy of load model has very important influence to electric power system design, calculating and security and stability analysis.
As the continuous expansion of system scale, coming into operation for new technology new equipment bring the challenge of bigger to load modeling work.For a long time
Since, domestic and international researcher has carried out a large amount of Modeling for Electric Loads work, has generally formed two kinds of load modeling methods,
That is Component Based and Measurement-based approach, these methods achieve certain effect, while there is also respective limitations.
The basic thought of Component Based is the load composition for counting all types of users, determines static load and induction motor load
Proportion, synthesis obtain overall load model, its shortcoming is that statistical work time and effort consuming, model time variation is poor.Total body examination is debated
The basic thought of method is to regard load group as an entirety, according to collection in worksite measurement data, is carried out to load model parameters whole
Body recognizes, and the measurement work that method is debated in total body examination is complicated, and model accuracy is difficult to ensure.
The load model that traditional load modeling method obtains generally is connected to 220kV or 110kV busbares.In general, voltage
Lower grade, more clear closer to end load ingredient, while the sample chosen is more, and the accuracy of load identification is also higher.
Past, due to the limitation of technological means, it is difficult to be deep into power distribution network underlying user and carry out load measurement and modeling work.But with
The construction of extensive intelligent grid, the fast development of the information technologies such as calculating, communication, sensing so that electric load information is online
Monitoring is possibly realized.Load on-line monitoring equipment can acquire power system customer information on load and carry out online survey and debate in real time, be
Power department provides real-time, accurate information on load, and can carry out real-time load modeling using these measurement informations studies
Work.
Invention content
The purpose of the present invention is to provide a kind of online load modeling method measured based on local intelligence, based on distribution
The load model of all elements in net bottom layer voltage class carries out layering aggregation, equivalent upwards step by step, finally obtains 220kV/
110kV busbar integrated load models；Compared to traditional load modeling method, the present invention can improve the accuracy of load model, instead
Reflect the actual characteristic of load.
In order to achieve the above objectives, solution of the invention is：
A kind of online load modeling method measured based on local intelligence, is included the following steps：
Step 1, the daily load data based on intelligent meter data recording system acquisition carry out user clustering, often selected part in class user
Typical user carries out intrusive load monitoring, and load monitoring device is intelligent socket；
Step 2, the real-time load information based on intelligent socket acquisition carries out online survey and debates, and obtains load model, using every
The load model of Lei Zhong typical users substitutes other users in such, to obtain the integrated load model of all users in real time；
Step 3, the polymerization model of underlying user static load and induction motor load is established respectively；
Step 4, the influence for considering distribution network, by static load and induction motor load polymerization model by low-pressure side by
It walks to high-pressure side equivalence, the final integrated load model obtained under power distribution network 220/110kV busbares.
In above-mentioned steps 1, the particular content that the daily load data based on intelligent meter data recording system acquisition carry out user clustering is：
For the daily load data of given N number of user, K user is randomly selected, the initial of one user group of each user representative gathers
Remaining other users are assigned to according to electricity consumption Euclidean distance apart from nearest cluster centre, form K use altogether by class center
Family group；The cluster centre of each user group is recalculated, i.e., all user power consumptions in user group are averaged, again by institute
There is user to be assigned to apart from nearest cluster centre, constantly repeated according to this process, until each user group cluster centre not
It changes again or clustering criteria function reaches the condition of convergence.
In above-mentioned steps 2, intelligent socket is by filtering sampling module, data processing module, communication module, execution module and electricity
Five part of source module forms, and filtering sampling module acquires voltage, electric current and the frequency parameter of current intelligent socket on-load, will be high
Voltage and high current signal are converted to low voltage signal, are analyzed for data processing module；Data processing module is from hardware architecture
Plan as a whole the operation of whole system, it is internal to complete following software function：Parameter calculating, data communication, enhanced protection, instruction are held
Row；Communication module is one of the channel that intelligent socket is interacted with control server；Execution module is responsible for executing data processing module
Control instruction, and implementing result is fed back, while supporting circuit break-make and infrared adjusting both of which；Power module will
220V alternating currents are converted to direct current 5V and 3.3V, are used for system operation.
In above-mentioned steps 3：The polynomial form that static load model is recommended using IEEE Task Force is as follows：
In formula, a, b, c are active power coefficient, and α, β, γ are reactive power coefficient, and U is the virtual voltage of load, U0For
The rated voltage of load, P0、Q0The active power and reactive power of load respectively under rated voltage, P, Q are respectively load consumption
Practical active power and reactive power；
The polymerization model method for establishing static load is：By load active power and reactive power according to constant-impedance, permanent electricity
Stream and invariable power component are weighted according to coefficient respectively, such as following formula：
In formula, P01,P02,…,P0nFor the rated active power of single static load, Q01,Q02,…,Q0nFor single static
The rated reactive power of load, P0SAnd Q0SRespectively it polymerize the rated active power and reactive power of static load, a1,a2,…,
an；b1,b2,…,bnFor the active power coefficient of single static load, α1,α2,…, αn；β1,β2,…,βnIt is respectively single quiet
The reactive power coefficient of state load, aS、bS、cSTo polymerize the active power coefficient of static load, αS、βS、γSIt is static negative for polymerization
The reactive power coefficient of lotus.
In above-mentioned steps 3：Induction motor load uses three rank machine-electricity transient models, as follows：
In formula, T0'=(Xr+Xm)/ω0Rr, X=Xs+Xm, X '=Xs+XmXr/(Xm+Xr), T0' it is that the transient state open circuit time is normal
Number, X are rotor open circuit reactance, and X ' is short-circuit reactance when rotor is motionless,For transient internal voltage,For motor voltage,For electricity
Motivation electric current, ω are motor speed, ω0For motor rated speed, TEFor electromagnetic torque, TMFor machine torque, H is inertia
Time constant, RrFor rotor resistance, XrFor rotor reactance, RsFor stator resistance, XsFor stator reactance, XmFor excitation reactance；
The rated capacity for polymerizeing induction motor load is the sum of the rated capacity of separate unit induction motor load, i.e.,：
In formula, k is induction motor load quantity；The Equivalent Circuit Parameter for polymerizeing induction motor load is equivalent circuit
In each branch admittance weighted mean, i.e.,：
In formula, proportionality coefficient ρi=SNi/SNM, ZiFor the electric branch circuit impedance of separate unit induction motor load, ZMFor polymerization
The electric branch circuit impedance of induction motor load, to stator branch ZM=Rs+jXs, to field excitation branch line ZM=jXm, to rotor branch
ZM=Rr/s+jXr；Polymerization induction motor load inertia time constant be：
In formula, HiFor the inertia time constant of separate unit induction motor load, HMTo polymerize the inertia of induction motor load
Time constant.
In above-mentioned steps 4：The influence for considering distribution network, by static load and induction motor load polymerization model by low
Side is pressed gradually to high-pressure side equivalent time, to enable：
In formula, ZDFor distribution network impedance, YSFor low-pressure side static load Equivalent admittance, λ1, λ2, λ3It is quiet with low-pressure side
State load Equivalent admittance and the relevant variable of distribution network impedance；
By static load polymerization model by the lateral high-pressure side equivalent time of low pressure, high-pressure side static load Equivalent admittance is：
In formula, PH、QHRespectively flow into the active power and reactive power of high-voltage side bus, UHFor high side bus voltage；
By induction motor load polymerization model by the lateral high-pressure side equivalent time of low pressure, high-pressure side induction motor load etc.
Value parameter computational methods are：
In formula, T '0H=(XrH+XmH)/ω0HRrH, T '0HFor high-pressure side induction conductivity transient state open circuit time constant, ω0HFor
High-pressure side induction conductivity rated speed, THE、TMHRespectively high-pressure side induction conductivity electromagnetic torque, machine torque, HHFor height
Press side induction conductivity inertia time constant, RrH、XrHRespectively high-pressure side induction conductivity rotor resistance, rotor reactance, RsH、
XsHRespectively high-pressure side induction conductivity stator resistance, stator reactance, XmHFor high-pressure side induction conductivity excitation reactance.
After adopting the above scheme, compared with prior art, the beneficial effects of the present invention are：
(1) electric power can be improved closer to actual distribution network load in the integrated load model that the modeling method carried obtains
Simulation accuracy；
(2) modeling method carried can realize the dynamic update of load model, can react the real-time fortune of power system load
Row state.
Description of the drawings
Fig. 1 is the flow chart of the present invention；
Fig. 2 is the day power curve of typical electrical equipment；
Fig. 3 is that 3 machine, 9 node connects simple power distribution network example figure；
Fig. 4 is 110kV node voltage simulation curve figures；
Fig. 5 is 110kV node active power simulation curve figures.
Specific implementation mode
Below with reference to attached drawing, technical scheme of the present invention is described in detail.
As shown in Figure 1, the present invention provides a kind of online load modeling method measured based on local intelligence, including walk as follows
Suddenly：
Step 1, the daily load data based on intelligent meter data recording system acquisition carry out user clustering, often selected part in class user
Typical user carries out intrusive load monitoring, and load monitoring device is intelligent socket；
Step 2, the real-time load information based on intelligent socket acquisition carries out online survey and debates, and obtains load model, using every
The load model of Lei Zhong typical users substitutes other users in such, to obtain the integrated load model of all users in real time；
Step 3, the polymerization model of underlying user static load and induction motor load is established respectively；
Step 4, the influence for considering distribution network, by static load and induction motor load polymerization model by low-pressure side by
It walks to high-pressure side equivalence, the final integrated load model obtained under power distribution network 220/110kV busbares.
In the step 1, the user clustering method based on user's daily load data is：For the day of given N number of user
Load data randomly selects K user, the initial cluster center of one user group of each user representative.By other remaining use
Family is assigned to according to electricity consumption Euclidean distance apart from nearest cluster centre, forms K user group altogether.Recalculate each user
The cluster centre of group, i.e., be averaged all user power consumptions in user group, and all users are assigned to distance recently again
Cluster centre, constantly repeated according to this process, until the cluster centre of each user group no longer changes or clustering criteria
Function reaches the condition of convergence.
In the step 2, intrusive load identification tool is intelligent socket, mainly by filtering sampling module, data processing
Module, communication module, execution module and five part of power module composition.Filtering sampling module acquires current intelligent socket on-load
Voltage, electric current and frequency parameter, high voltage and high current signal are converted into low voltage signal, for data processing module point
Analysis.Data processing module is the core of intelligent socket, plans as a whole the operation of whole system from hardware architecture, and internal completion is as follows
Software function：Parameter calculating, data communication, enhanced protection, instruction execution.Communication module is that intelligent socket is handed over control server
One of mutual channel, can support WiFi, Zigbee, and LoRa patterns are optional.Execution module is responsible for executing the control of data processing module
System instruction, and implementing result is fed back, while supporting circuit break-make and infrared adjusting both of which.Power module is entire
The energy source of intelligent socket module, 220V alternating currents are converted to direct current 5V and 3.3V by it, are used for system operation.
In the step 3：The polynomial form that static load model is recommended using IEEE Task Force is as follows：
In formula, a, b, c are active power coefficient, and α, β, γ are reactive power coefficient, and U is the virtual voltage of load, U0For
The rated voltage of load, P0、Q0The active power and reactive power of load respectively under rated voltage, P, Q are respectively load consumption
Practical active power and reactive power.
The polymerization model method for establishing static load is：By load active power and reactive power according to constant-impedance, permanent electricity
Stream and invariable power component are weighted according to coefficient respectively, such as following formula：
In formula, P01,P02,…,P0nFor the rated active power of single static load, Q01,Q02,…,Q0nFor single static
The rated reactive power of load, P0SAnd Q0SRespectively it polymerize the rated active power and reactive power of static load, a1,a2,…,
an；b1,b2,…,bnFor the active power coefficient of single static load, α1,α2,…, αn；β1,β2,…,βnIt is respectively single quiet
The reactive power coefficient of state load, aS、bS、cSTo polymerize the active power coefficient of static load, αS、βS、γSIt is static negative for polymerization
The reactive power coefficient of lotus.
Induction motor load uses three rank machine-electricity transient models, as follows：
In formula, T0'=(Xr+Xm)/ω0Rr, X=Xs+Xm, X '=Xs+XmXr/(Xm+Xr), T0' it is that the transient state open circuit time is normal
Number, X are rotor open circuit reactance, and X ' is short-circuit reactance when rotor is motionless,For transient internal voltage,For motor voltage,For
Motor current, ω are motor speed, ω0For motor rated speed, TEFor electromagnetic torque, TMFor machine torque, H is used
Property time constant, RrFor rotor resistance, XrFor rotor reactance, RsFor stator resistance, XsFor stator reactance, XmFor excitation reactance.
The rated capacity for polymerizeing induction motor load is the sum of the rated capacity of separate unit induction motor load, i.e.,：
In formula, k is induction motor load quantity.The Equivalent Circuit Parameter for polymerizeing induction motor load is equivalent circuit
In each branch admittance weighted mean, i.e.,：
In formula, proportionality coefficient ρi=SNi/SNM, ZiFor the electric branch circuit impedance of separate unit induction motor load, ZMFor polymerization
The electric branch circuit impedance of induction motor load, to stator branch ZM=Rs+jXs, to field excitation branch line ZM=jXm, to rotor branch
ZM=Rr/s+jXr.Polymerization induction motor load inertia time constant be：
In formula, HiFor the inertia time constant of separate unit induction motor load, HMTo polymerize the inertia of induction motor load
Time constant.
In the step 4：The influence for considering distribution network, by static load and induction motor load polymerization model by low
Side is pressed gradually to high-pressure side equivalent time, to enable：
In formula, ZDFor distribution network impedance, YSFor low-pressure side static load Equivalent admittance, λ1, λ2, λ3It is quiet with low-pressure side
State load Equivalent admittance and the relevant variable of distribution network impedance.
By static load polymerization model by the lateral high-pressure side equivalent time of low pressure, high-pressure side static load Equivalent admittance is：
In formula, PH、QHRespectively flow into the active power and reactive power of high-voltage side bus, UHFor high side bus voltage.
By induction motor load polymerization model by the lateral high-pressure side equivalent time of low pressure, high-pressure side induction motor load etc.
Value parameter computational methods are：
In formula, T '0H=(XrH+XmH)/ω0HRrH, T '0HFor high-pressure side induction conductivity transient state open circuit time constant, ω0H
For high-pressure side induction conductivity rated speed, THE、TMHRespectively high-pressure side induction conductivity electromagnetic torque, machinery
Torque, HHFor high-pressure side induction conductivity inertia time constant, RrH、XrHRespectively high-pressure side induction conductivity rotor resistance, turn
Sub- reactance, RsH、XsHRespectively high-pressure side induction conductivity stator resistance, stator reactance, XmHFor high-pressure side induction conductivity excitation
Reactance.
Below in conjunction with embodiment, present invention is further described in detail, but the present invention is not limited to given implementations
Example.
The continuous three days daily load data of 200 users for choosing certain city are research sample, and it is small to be divided into 1 between data sampling
When, data scale 200*3*24=14400.User clustering is carried out using the clustering method carried, when cluster numbers are 3, is gathered
Class effect is optimal, wherein belong to one, two, the number of users of three classes be respectively 91,86,23.Selected part is typical in per class user
User carries out intrusive load monitoring, acquires the electric information of different electrical equipments in real time by intelligent socket, and be uploaded to control
Preset all types of load power informations carry out sample training and characteristic matching in control server, with server, and identification is negative
Lotus type.According to the load type that identification obtains, different type load chooses typical load model parameter.Fig. 2 show intelligence
The day power curve of a few quasi-representative electrical equipments of socket acquisition, sample frequency 1Hz.
Circuit as shown in Figure 3 is built in PSCAD/EMTDC, and a simple distribution is connect under WSCC9 node systems node 6
Net.In Fig. 3, the synthetic load under 220V busbares indicates the Load aggregation model of above-mentioned three classes user respectively.Under 10kV busbares
Load indicates that large scale industry load, 110kV/10kV transformer reactances take XT1=0.03p.u., 10kV/220V transformer reactance
Take XT2=0.02p.u..Load 1, Load 2 and Load in the polymerization model parameter such as table 1 of above-mentioned three quasi-representatives resident
3, industrial load polymerization model parameter is as shown in Load 4.
1 load parameter of table
| Rs | Xs | Xm | Rr | Xr | A | B | H | |
| Load 1 | 0.023 | 0.126 | 3.39 | 0.0136 | 0.126 | 0.85 | 0 | 1.07 |
| Load 2 | 0.032 | 0.096 | 2.69 | 0.032 | 0.096 | 1 | 0 | 0.50 |
| Load 3 | 0.083 | 0.095 | 2.1 | 0.046 | 0.095 | 1 | 0 | 0.47 |
| Load 4 | 0.018 | 0.117 | 3.6 | 0.009 | 0.117 | 1 | 0 | 1.40 |
| Z% | I% | P% | ηs | RD | XD | kM | SN | |
| Load 1 | 0.33 | 0.32 | 0.35 | 0.85 | 0.002 | 0.042 | 0.35 | 15 |
| Load 2 | 0.20 | 0.50 | 0.30 | 0.85 | 0.001 | 0.04 | 0.2 | 12 |
| Load 3 | 0.20 | 0.55 | 0.25 | 0.85 | 0.001 | 0.04 | 0.45 | 8 |
| Load 4 | 0.10 | 0.85 | 0.05 | 0.85 | 0.003 | 0.04 | 0.75 | 25 |
Parameter includes upper and lower two row in table, the first behavior induction-motor load basic parameter, in the second row, Z%, I% and P%
Constant-impedance, constant current and constant power load model ratio in static load, η are indicated respectivelysFor static load power factor, RD、XDFor
Distribution line impedance, kMFor motor ratio, SNFor the rated capacity (MW) of load.
The model accuracy of comparison institute's extracting method of the present invention and the whole discrimination method of tradition：
(1) Load aggregation from bottom to top carried, step by step equivalent method of the invention are taken, 110kV busbares are integrated negative
Lotus carries out fining modeling, and detailed process is：1) Load 1, Load 2 and 3 load models of Load are rolled over through distribution line respectively
It calculates to 220V busbares；2) polymerization model of three type loads under 220V busbares is established；3) polymerization model under 220V busbares is passed through
10kV/220V transformers are converted to 10kV busbares；4) 4 load models of Load are converted through distribution line to 10kV busbares；5)
Establish the Load aggregation model under 10kV busbares；6) by the polymerization model under 10kV busbares through 110kV/10kV transformers convert to
110kV busbares obtain the integrated load model under 110kV busbares；
(2) electrical quantity for only acquiring 110kV busbares, using particle cluster algorithm, to the distribution network load mould under 110kV busbares
Type carries out whole identification.
Take the Composite Load Model Parameters that above two method obtains as shown in table 2.
The Composite Load Model Parameters that 2 two kinds of load modeling methods of table obtain
| Rs | Xs | Xm | Rr | Xr | H | A | B | SM | |
| Modeling method from bottom to top | 0.034 | 0.152 | 3.166 | 0.020 | 0.113 | 1.05 | 0.96 | 0 | 20.4 |
| Whole discrimination method | 0.027 | 0.116 | 3.302 | 0.019 | 0.116 | 1.25 | 1 | 0 | 22.1 |
| kM | P0 | a | b | c | Q0 | α | β | γ | |
| Modeling method from bottom to top | 0.40 | 25.15 | 0.22 | 0.52 | 0.26 | 16.45 | 0.22 | 0.52 | 0.26 |
| Whole discrimination method | 0.43 | 24.22 | 0.24 | 0.47 | 0.29 | 16.01 | 0.25 | 0.46 | 0.29 |
Short-time grounding fault is set when 2s at No. 5 nodes, and Fig. 4 and Fig. 5 show and different load modeling method taken to obtain
Voltage, the active power curves of power distribution network 110kV nodes when the load model arrived.The result shows that passing through entirety compared to traditional
Obtained integrated load model is recognized, the load model obtained by fining modeling, polymerization from bottom to top is closer to reality
Electricity distribution network model.
Above example is merely illustrative of the invention's technical idea, and protection scope of the present invention cannot be limited with this, every
According to technological thought proposed by the present invention, any change done on the basis of technical solution each falls within the scope of the present invention
Within.
Claims (6)
1. a kind of online load modeling method measured based on local intelligence, it is characterised in that include the following steps：
Step 1, the daily load data based on intelligent meter data recording system acquisition carry out user clustering, and selected part is typical in every class user
User carries out intrusive load monitoring, and load monitoring device is intelligent socket；
Step 2, the real-time load information based on intelligent socket acquisition carries out online survey and debates, and load model is obtained, using in every class
The load model of typical user substitutes other users in such, to obtain the integrated load model of all users in real time；
Step 3, the polymerization model of underlying user static load and induction motor load is established respectively；
Step 4, the influence for considering distribution network, by static load and induction motor load polymerization model from low-pressure side gradually to
High-pressure side is equivalent, the final integrated load model obtained under power distribution network 220/110kV busbares.
2. a kind of online load modeling method measured based on local intelligence as described in claim 1, it is characterised in that：It is described
In step 1, the particular content that the daily load data based on intelligent meter data recording system acquisition carry out user clustering is：For given N
The daily load data of a user, randomly select K user, and the initial cluster center of one user group of each user representative will remain
Remaining other users are assigned to according to electricity consumption Euclidean distance apart from nearest cluster centre, form K user group altogether；Again it counts
The cluster centre of each user group is calculated, i.e., all user power consumptions in user group is averaged, again distributes all users
To apart from nearest cluster centre, constantly repeated according to this process, until the cluster centre of each user group no longer changes,
Or clustering criteria function reaches the condition of convergence.
3. a kind of online load modeling method measured based on local intelligence as described in claim 1, it is characterised in that：It is described
In step 2, intelligent socket is by filtering sampling module, data processing module, five part of communication module, execution module and power module
Composition, filtering sampling module acquires voltage, electric current and the frequency parameter of current intelligent socket on-load, by high voltage and high current
Signal is converted to low voltage signal, is analyzed for data processing module；Data processing module plans as a whole whole system from hardware architecture
Operation, it is internal to complete following software function：Parameter calculating, data communication, enhanced protection, instruction execution；Communication module is
One of the channel that intelligent socket is interacted with control server；Execution module is responsible for executing the control instruction of data processing module, and
Implementing result is fed back, while supporting circuit break-make and infrared adjusting both of which；Power module turns 220V alternating currents
It is changed to direct current 5V and 3.3V, is used for system operation.
4. a kind of online load modeling method measured based on local intelligence as described in claim 1, it is characterised in that：It is described
In step 3：The polynomial form that static load model is recommended using IEEE Task Force is as follows：
In formula, a, b, c are active power coefficient, and α, β, γ are reactive power coefficient, and U is the virtual voltage of load, U0For load
Rated voltage, P0、Q0The active power and reactive power of load respectively under rated voltage, P, Q are respectively the reality of load consumption
Active power and reactive power；
The polymerization model method for establishing static load is：By load active power and reactive power according to constant-impedance, constant current and
Invariable power component is weighted according to coefficient respectively, such as following formula：
In formula, P01,P02,…,P0nFor the rated active power of single static load, Q01,Q02,…,Q0nFor single static load
Rated reactive power, P0SAnd Q0SRespectively it polymerize the rated active power and reactive power of static load, a1,a2,…,an；
b1,b2,…,bnFor the active power coefficient of single static load, α1,α2,…,αn；β1,β2,…,βnRespectively single static is negative
The reactive power coefficient of lotus, aS、bS、cSTo polymerize the active power coefficient of static load, αS、βS、γSFor polymerization static load
Reactive power coefficient.
5. a kind of online load modeling method measured based on local intelligence as described in claim 1, it is characterised in that：It is described
In step 3：Induction motor load uses three rank machine-electricity transient models, as follows：
In formula, T0'=(Xr+Xm)/ω0Rr, X=Xs+Xm, X '=Xs+XmXr/(Xm+Xr), T0' it is transient state open circuit time constant, X is
Rotor open circuit reactance, X ' are short-circuit reactance when rotor is motionless,For transient internal voltage,For motor voltage,For electronic electromechanics
Stream, ω is motor speed, ω0For motor rated speed, TEFor electromagnetic torque, TMFor machine torque, H is that inertia time is normal
Number, RrFor rotor resistance, XrFor rotor reactance, RsFor stator resistance, XsFor stator reactance, XmFor excitation reactance；
The rated capacity for polymerizeing induction motor load is the sum of the rated capacity of separate unit induction motor load, i.e.,：
In formula, k is induction motor load quantity；The Equivalent Circuit Parameter for polymerizeing induction motor load is each in equivalent circuit
The weighted mean of branch admittance, i.e.,：
In formula, proportionality coefficient ρi=SNi/SNM, ZiFor the electric branch circuit impedance of separate unit induction motor load, ZMTo polymerize induced electricity
The electric branch circuit impedance of engine load, to stator branch ZM=Rs+jXs, to field excitation branch line ZM=jXm, to rotor branch ZM=Rr/
s+jXr；Polymerization induction motor load inertia time constant be：
In formula, HiFor the inertia time constant of separate unit induction motor load, HMTo polymerize the inertia time of induction motor load
Constant.
6. a kind of online load modeling method measured based on local intelligence as described in claim 1, it is characterised in that：It is described
In step 4：The influence for considering distribution network, by static load and induction motor load polymerization model from low-pressure side gradually to height
Side equivalent time is pressed, is enabled：
In formula, ZDFor distribution network impedance, YSFor low-pressure side static load Equivalent admittance, λ1, λ2, λ3It is negative with low-pressure side static state
Lotus Equivalent admittance and the relevant variable of distribution network impedance；
By static load polymerization model by the lateral high-pressure side equivalent time of low pressure, high-pressure side static load Equivalent admittance is：
In formula, PH、QHRespectively flow into the active power and reactive power of high-voltage side bus, UHFor high side bus voltage；
By induction motor load polymerization model by the lateral high-pressure side equivalent time of low pressure, high-pressure side induction motor load equivalence ginseng
Number calculating method is：
In formula, T0H'=(XrH+XmH)/ω0HRrH, T0H' it is high-pressure side induction conductivity transient state open circuit time constant, ω0HFor high pressure
Side induction conductivity rated speed, THE、TMHRespectively high-pressure side induction conductivity electromagnetic torque, machine torque, HHFor high-pressure side
Induction conductivity inertia time constant, RrH、XrHRespectively high-pressure side induction conductivity rotor resistance, rotor reactance, RsH、XsHPoint
Not Wei high-pressure side induction conductivity stator resistance, stator reactance, XmHFor high-pressure side induction conductivity excitation reactance.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810010291.0A CN108306284B (en) | 2018-01-05 | 2018-01-05 | Online load modeling method based on local intelligent measurement |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810010291.0A CN108306284B (en) | 2018-01-05 | 2018-01-05 | Online load modeling method based on local intelligent measurement |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108306284A true CN108306284A (en) | 2018-07-20 |
| CN108306284B CN108306284B (en) | 2021-04-27 |
Family
ID=62868283
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810010291.0A Active CN108306284B (en) | 2018-01-05 | 2018-01-05 | Online load modeling method based on local intelligent measurement |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108306284B (en) |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109713662A (en) * | 2018-12-20 | 2019-05-03 | 清华大学 | A kind of method of power system load model identified parameters to low pressure node equivalent |
| CN109827310A (en) * | 2019-01-31 | 2019-05-31 | 河海大学 | A kind of residual air-conditioning load group polymerization model method for building up |
| CN110048409A (en) * | 2019-04-19 | 2019-07-23 | 郑州电力高等专科学校 | A kind of electric load structural recognition method based on load starting transient characterisitics |
| CN112257212A (en) * | 2020-12-21 | 2021-01-22 | 中国电力科学研究院有限公司 | Modeling method and system of comprehensive load model considering distributed power generation system |
| CN112290538A (en) * | 2020-09-30 | 2021-01-29 | 天津大学 | Load model parameter online correction method based on aggregation-identification double-layer framework |
| CN113761700A (en) * | 2020-06-05 | 2021-12-07 | 国家电网有限公司华东分部 | Load modeling and online correction method and system based on dynamic clustering |
| CN115018172A (en) * | 2022-06-16 | 2022-09-06 | 国网湖南省电力有限公司 | Real-time judgment method for induction motor proportion in power grid comprehensive load model |
| CN118094961A (en) * | 2024-04-24 | 2024-05-28 | 广东电网有限责任公司中山供电局 | Load model modeling method, storage medium and system for electric power system |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101192754A (en) * | 2007-12-11 | 2008-06-04 | 中国电力科学研究院 | A realization method for integrated load model of distribution network in power system |
| US20090192953A1 (en) * | 2008-01-24 | 2009-07-30 | Hua-Guang Zhang | Load Fuzzy Modeling Device And Method For Power System |
| CN103001230A (en) * | 2012-11-16 | 2013-03-27 | 天津大学 | Non-invasive power load monitoring and decomposing current mode matching method |
| CN103049651A (en) * | 2012-12-13 | 2013-04-17 | 航天科工深圳（集团）有限公司 | Method and device used for power load aggregation |
| CN105404784A (en) * | 2015-12-07 | 2016-03-16 | 河南许继仪表有限公司 | Non-invasive power load decomposition method |
-
2018
- 2018-01-05 CN CN201810010291.0A patent/CN108306284B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101192754A (en) * | 2007-12-11 | 2008-06-04 | 中国电力科学研究院 | A realization method for integrated load model of distribution network in power system |
| US20090192953A1 (en) * | 2008-01-24 | 2009-07-30 | Hua-Guang Zhang | Load Fuzzy Modeling Device And Method For Power System |
| CN103001230A (en) * | 2012-11-16 | 2013-03-27 | 天津大学 | Non-invasive power load monitoring and decomposing current mode matching method |
| CN103049651A (en) * | 2012-12-13 | 2013-04-17 | 航天科工深圳（集团）有限公司 | Method and device used for power load aggregation |
| CN105404784A (en) * | 2015-12-07 | 2016-03-16 | 河南许继仪表有限公司 | Non-invasive power load decomposition method |
Cited By (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109713662A (en) * | 2018-12-20 | 2019-05-03 | 清华大学 | A kind of method of power system load model identified parameters to low pressure node equivalent |
| CN109827310A (en) * | 2019-01-31 | 2019-05-31 | 河海大学 | A kind of residual air-conditioning load group polymerization model method for building up |
| CN110048409A (en) * | 2019-04-19 | 2019-07-23 | 郑州电力高等专科学校 | A kind of electric load structural recognition method based on load starting transient characterisitics |
| CN110048409B (en) * | 2019-04-19 | 2021-01-29 | 郑州电力高等专科学校 | Power load structure identification method based on load starting transient characteristics |
| CN113761700A (en) * | 2020-06-05 | 2021-12-07 | 国家电网有限公司华东分部 | Load modeling and online correction method and system based on dynamic clustering |
| CN112290538A (en) * | 2020-09-30 | 2021-01-29 | 天津大学 | Load model parameter online correction method based on aggregation-identification double-layer framework |
| CN112257212A (en) * | 2020-12-21 | 2021-01-22 | 中国电力科学研究院有限公司 | Modeling method and system of comprehensive load model considering distributed power generation system |
| CN112257212B (en) * | 2020-12-21 | 2021-05-14 | 中国电力科学研究院有限公司 | Modeling method and system of comprehensive load model considering distributed power generation system |
| CN115018172A (en) * | 2022-06-16 | 2022-09-06 | 国网湖南省电力有限公司 | Real-time judgment method for induction motor proportion in power grid comprehensive load model |
| CN115018172B (en) * | 2022-06-16 | 2024-06-14 | 国网湖南省电力有限公司 | Real-time judging method for proportion of induction motor in power grid comprehensive load model |
| CN118094961A (en) * | 2024-04-24 | 2024-05-28 | 广东电网有限责任公司中山供电局 | Load model modeling method, storage medium and system for electric power system |
| CN118094961B (en) * | 2024-04-24 | 2024-08-20 | 广东电网有限责任公司中山供电局 | Load model modeling method, storage medium and system for electric power system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108306284B (en) | 2021-04-27 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108306284A (en) | A kind of online load modeling method measured based on local intelligence | |
| CN101777765B (en) | On-line load simulation method of power system | |
| CN104362622B (en) | The load model of a kind of band small power station unit area power grid and modeling method | |
| CN109286188B (en) | 10kV power distribution network theoretical line loss calculation method based on multi-source data set | |
| CN110070282A (en) | A kind of low-voltage platform area line loss analysis of Influential Factors method based on Synthesis Relational Grade | |
| CN107979086A (en) | Voltage sag reason recognition methods based on EM algorithms and gradient boosted tree | |
| CN104392056B (en) | A kind of directly driven wind-powered unit modeling method and device | |
| CN110048409B (en) | Power load structure identification method based on load starting transient characteristics | |
| CN109636009A (en) | It is a kind of to establish the method and system for determining the neural network model of grid line loss | |
| CN115622053B (en) | Automatic load modeling method and device for considering distributed power supply | |
| CN103793556B (en) | Comprehensively go out transformer station's load modeling method of specific electric load | |
| CN109494719A (en) | A kind of mesolow mixing power distribution network stratification impedance analysis method | |
| CN107093897A (en) | A kind of distribution net platform region low pressure loss analysis method | |
| CN103632031B (en) | A kind of rural area based on load curve decomposition load type load modeling method | |
| CN102856896A (en) | On-line analytical method for direct-current transmission loss | |
| CN102118026A (en) | ETAP-based harmonic and negative sequence power flow method for large-scale power grid | |
| CN106356843B (en) | Synthesis load modeling method for reflecting network voltage characteristic | |
| CN204515103U (en) | Classical insulation test macro | |
| CN105701265B (en) | A kind of double-fed fan motor unit modeling method and device | |
| Jereminov et al. | Impact of load models on power flow optimization | |
| CN112784870A (en) | Typical distribution area classification method and system based on clustering algorithm | |
| CN105514981B (en) | A kind of photovoltaic DC-to-AC converter on-position optimization method based on BIM | |
| CN113054747B (en) | Non-invasive detection method for nonlinear load of power distribution station | |
| CN104483570A (en) | Empirical method for harmonic control loss reduction effect of power distribution network | |
| CN115392347A (en) | User phase relation identification method based on generation countermeasure network |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| TA01 | Transfer of patent application right | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20201209 Address after: Room 428, building a, 218 Hongkou Road, Kunlun Street, Liyang City, Changzhou City, Jiangsu Province Applicant after: Liyang Research Institute of Southeast University Applicant after: SOUTHEAST University Address before: 210096, No. four archway, Xuanwu District, Jiangsu, Nanjing 2 Applicant before: SOUTHEAST University |
|
| GR01 | Patent grant | ||
| GR01 | Patent grant |