CN117474270A - BP-based electric bus excitation-response characteristic accurate quantification method - Google Patents
BP-based electric bus excitation-response characteristic accurate quantification method Download PDFInfo
- Publication number
- CN117474270A CN117474270A CN202311501128.1A CN202311501128A CN117474270A CN 117474270 A CN117474270 A CN 117474270A CN 202311501128 A CN202311501128 A CN 202311501128A CN 117474270 A CN117474270 A CN 117474270A
- Authority
- CN
- China
- Prior art keywords
- electric bus
- charging
- discharge
- electric
- model
- 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.)
- Pending
Links
- 230000004044 response Effects 0.000 title claims abstract description 70
- 230000005284 excitation Effects 0.000 title claims abstract description 43
- 238000000034 method Methods 0.000 title claims abstract description 41
- 238000011002 quantification Methods 0.000 title claims abstract description 12
- 238000013528 artificial neural network Methods 0.000 claims abstract description 67
- 238000007599 discharging Methods 0.000 claims abstract description 67
- 238000012549 training Methods 0.000 claims abstract description 22
- 238000010276 construction Methods 0.000 claims description 12
- 230000002238 attenuated effect Effects 0.000 claims description 11
- 230000005611 electricity Effects 0.000 claims description 9
- 230000006870 function Effects 0.000 claims description 9
- 238000003062 neural network model Methods 0.000 claims description 9
- NAWXUBYGYWOOIX-SFHVURJKSA-N (2s)-2-[[4-[2-(2,4-diaminoquinazolin-6-yl)ethyl]benzoyl]amino]-4-methylidenepentanedioic acid Chemical compound C1=CC2=NC(N)=NC(N)=C2C=C1CCC1=CC=C(C(=O)N[C@@H](CC(=C)C(O)=O)C(O)=O)C=C1 NAWXUBYGYWOOIX-SFHVURJKSA-N 0.000 claims description 3
- 238000001514 detection method Methods 0.000 claims description 3
- 230000002222 downregulating effect Effects 0.000 claims description 3
- 230000003828 downregulation Effects 0.000 claims description 3
- 238000004880 explosion Methods 0.000 claims description 3
- 230000001105 regulatory effect Effects 0.000 claims description 3
- 238000012360 testing method Methods 0.000 claims description 3
- 230000003827 upregulation Effects 0.000 claims description 3
- 230000033228 biological regulation Effects 0.000 abstract description 5
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0631—Resource planning, allocation, distributing or scheduling for enterprises or organisations
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
- G06F30/27—Design optimisation, verification or simulation using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/06—Energy or water supply
-
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2111/00—Details relating to CAD techniques
- G06F2111/04—Constraint-based CAD
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2113/00—Details relating to the application field
- G06F2113/04—Power grid 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]
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Human Resources & Organizations (AREA)
- Theoretical Computer Science (AREA)
- Economics (AREA)
- Physics & Mathematics (AREA)
- Strategic Management (AREA)
- General Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- Entrepreneurship & Innovation (AREA)
- Marketing (AREA)
- Evolutionary Computation (AREA)
- Health & Medical Sciences (AREA)
- Tourism & Hospitality (AREA)
- Water Supply & Treatment (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Development Economics (AREA)
- Quality & Reliability (AREA)
- Public Health (AREA)
- Game Theory and Decision Science (AREA)
- General Health & Medical Sciences (AREA)
- Primary Health Care (AREA)
- Artificial Intelligence (AREA)
- Educational Administration (AREA)
- Operations Research (AREA)
- Medical Informatics (AREA)
- Software Systems (AREA)
- Computer Hardware Design (AREA)
- Geometry (AREA)
- General Engineering & Computer Science (AREA)
- Power Engineering (AREA)
- Supply And Distribution Of Alternating Current (AREA)
Abstract
The invention discloses an accurate quantification method for excitation-response characteristics of an electric bus based on BP, and belongs to the field of power load regulation and control. Acquiring charging and running information of an electric bus, constructing a charging and discharging running constraint model of the electric bus, wherein the running constraint is mainly determined by the current residual SOC and the residual departure time of the electric bus, providing a charging and discharging loss model of the electric bus based on a BP neural network, training the acquired charging information and running information of the electric bus, and setting coefficients in the constructed charging and discharging loss model; the BP neural network-based electric bus excitation-response characteristic model is provided, and the accurate electric bus excitation-response characteristic can be quantitatively obtained by combining the electric bus charge-discharge operation constraint and the BP neural network-based charge-discharge loss model. The invention improves the accuracy of the excitation-response characteristics of the electric buses and provides a strong technical support for the electric buses to participate in the regulation and control of the power grid.
Description
Technical Field
The invention belongs to the field of power load regulation and control, and particularly relates to an accurate quantification method for excitation-response characteristics of an electric bus based on BP.
Background
The electric buses participate in power grid dispatching and need to be charged and discharged, so that certain loss can be generated. In addition, the normal operation plan of the electric bus cannot be influenced when the electric bus participates in the regulation and control of the power grid. Thus, in practice, rather than powering an electric bus, the electric bus participates in the response; only when the incentive price reaches a certain value, the electric bus participates in the response without affecting the operation plan of the electric bus. The response capability of the power grid is further evaluated by grasping the excitation-response characteristics of the power grid, and the power grid is supported to effectively schedule the power grid.
However, in the research of the excitation-response characteristics of the electric buses in the existing method, the influence of loss coefficients on the characteristics is less concerned, the coefficients are generally given according to experience and are not accurate enough, and the excitation-response characteristics of the given electric buses are not accurate enough. According to the invention, based on the BP neural network, the history sample of the electric bus is trained and learned, so that a more accurate loss coefficient can be given, and the accuracy of excitation-response characteristics of the electric bus is improved.
Disclosure of Invention
Aiming at the problems, the invention provides an accurate quantification method for the excitation-response characteristics of an electric bus based on BP, which is used for solving the problem that the excitation-response characteristics of the existing electric bus are not accurate enough.
The technical scheme adopted by the invention is as follows: the accurate quantification method of the excitation-response characteristics of the electric buses based on BP comprises the steps of electric bus charging and operation information acquisition, electric bus charging and discharging operation constraint model construction, electric bus charging and discharging loss model construction based on BP neural network, and electric bus excitation-response characteristic model construction based on BP neural network, and the specific steps are as follows:
s01: firstly, collected electric bus charging information: the method comprises the steps of current SOC of an electric bus, initial SOC of the electric bus, attenuated SOC of the electric bus, charging power of the electric bus, discharging power of the electric bus, charging efficiency of the electric bus and discharging efficiency of the electric bus; then, the collected electric bus operation information mainly includes: departure time, incentive price, up-regulation response power, down-regulation response power;
s02: constructing a constraint model of the charge and discharge operation of the electric bus: the electric bus participates in the power grid response on the premise that operation constraint conditions are required to be met, namely, the electric bus participates in the power grid response on the premise that the operation plan of the electric bus is not influenced, the operation constraint is determined by the current residual SOC and the residual departure time of the electric bus, a built electric bus charging and discharging operation constraint model comprises a charging constraint model and a discharging constraint model, the charging operation constraint model aims at guaranteeing that the electric bus does not exceed the attenuated SOC after participating in the charging response, battery charging explosion is prevented, and the discharging operation constraint model aims at guaranteeing that the residual SOC after participating in the discharging response of the electric bus does not influence the next departure requirement;
s03: building an electric bus charge-discharge loss model based on BP neural network: the electric bus can generate charging loss or discharging loss in the charging and discharging process, and the electric bus responds to charging and discharging only when the excitation price is larger than the charging and discharging loss, so that a BP neural network model is adopted to train collected charging information and operation information of the electric bus, coefficients in a built BP neural network-based electric bus charging and discharging loss model are set, and the accuracy of the charging and discharging loss model is improved;
s04: according to the BP neural network-based electric bus charge-discharge loss model constructed in the step S03, the charge-discharge loss of the electric bus is accurately described, the BP neural network-based electric bus excitation-response characteristic model is constructed in combination with the electric bus charge-discharge operation constraint in the step S02, the accurate electric bus excitation-response characteristic is quantitatively obtained, and further the electric bus is effectively scheduled to participate in the power grid operation.
Further, the data collected by the electric bus charging and operation information collection method proposed in step S01 is as follows by using formulas (1) and (2):
wherein: electric bus charging information acquired at moment X (t) is t, S h (t) is the current SOC of the electric bus h acquired at the moment t,for the collected electric bus h initial SOC, < > for>SOC attenuated for the collected electric bus h, < ->Charging power of electric bus h collected at t moment, < > for>Electric bus h discharge power collected at t moment, < > for>Charging efficiency for collected electric buses h, < >>For the collected electric bus h discharge efficiency, Y (t) is electric bus operation information collected at t moment, < >>Is the departure time f of the electric bus h after the moment t ex (t) is the incentive price collected at the moment t, P up (t) the up-regulating response power acquired at the moment t, P dn And (t) the down-regulating response power acquired at the moment t.
Further, the step of the electric bus charging and discharging constraint model constructed in the step S02 is as follows:
s21: constructing an electric bus discharging operation constraint model, as shown in formulas (3) - (5), wherein formula (3) represents the residual after discharging delta tminGreater than the discharge threshold->When the electric bus h meets the discharging operation constraint condition; residual SOC of electric bus after discharging>Calculated according to formula (4), discharge threshold +.>Calculating according to formula (5);
in the method, in the process of the invention,for the SOC remaining after the electric bus h discharges Δtmin at time t, +.>Is the discharge threshold value of the electric bus h, S h And (t) is the SOC of the electric bus h at the time t. />For the discharge efficiency of the electric bus h, +.>For the discharge power of the electric bus h, +.>Charging efficiency for electric bus h, +.>Is the charging power of the electric bus h,for the next departure time of the electric bus after time t,/for the next departure time of the electric bus after time t>For electric buses to meet the time +.>Minimum SOC required when transmitting.
S22: constructing a charging operation constraint model of the electric bus, wherein as shown in formulas (6) - (8), the formula (6) represents the SOC of the electric bus after charging delta tminLess than the charging threshold->When the electric bus h meets the charging operation constraint condition, the SOC of the electric bus after charging is +.>Calculated according to equation (7), the charging threshold value of the electric bus +.>Calculated according to equation (8).
In the method, in the process of the invention,SOC after charging delta tmin at t for electric bus h, +.>The charging threshold value of the electric bus h is S h (t) is the SOC of the electric bus h at the time t, < >>Charging efficiency for electric bus h, +.>Charging power for electric bus h, +.>The maximum SOC value of the electric bus h after attenuation.
Further, the electric bus charge and discharge loss model based on the BP neural network constructed in the step S3 comprises an electric bus charge and discharge loss model, and the construction steps are as follows:
s31, constructing an electric bus charging and discharging loss model: electric bus discharge loss model constructedAs shown in formula (9), the constructed electric bus charging loss model>As shown in formula (10), the charge-discharge loss model of the electric bus is mainly equal to the initial maximum SOC of the electric bus +.>Maximum SOC (state of charge) of electric bus after attenuation>SOCS of electric bus h at t moment h (t) determining that the number of the cells to be processed is,
wherein:for the discharge loss of the electric bus h at time t, < >>Is the attenuated SOC of the electric bus h,is the initial SOC of the electric bus h, S h (t) is the SOC of the electric bus h at the time t, < >>For the discharge efficiency of the electric bus h, +.>Charging efficiency for electric bus h, +.>For the discharge power of the electric bus h, +.>Charging power k for electric bus h 1 Is the discharge attenuation coefficient k of the electric bus 2 Is the discharge residual SOC coefficient, k of the electric bus 3 Is the depth of discharge coefficient, k of the electric bus 4 The charging attenuation coefficient k of the electric bus 5 Charging residual SOC coefficient k for electric bus 6 The charging depth coefficient of the electric bus;
s32, adopting a BP neural network model to set coefficients in a charge and discharge loss model of the electric bus, and extracting characteristics related to discharge loss by training and learning charge and discharge data samples and partial operation data samples of the electric bus, so as to realize accurate setting of the discharge loss coefficients of the electric bus, wherein 4 types of inputs and 3 outputs are respectively provided, and the 4 types of inputs are respectively: x (t) is the charge and discharge operation data of the electric bus at the moment t,The next departure time f of the electric bus h after the time t ex (t) incentive price, P at time t dn (t) is the down-regulated response power at time t; the 3 outputs are respectively: discharge attenuation coefficient k of electric bus 1 Discharge residual SOC coefficient k of electric bus 2 Depth of discharge coefficient k for electric buses 3 ；
The method for setting the coefficients in the electric bus charge-discharge loss model comprises the following steps of: collecting training samples of charge and discharge information and part of operation information of an electric bus, training the collected samples, carrying out network test on a trained discharge loss coefficient setting model, obtaining a BP neural network training function, a connection weight and a threshold value for discharge loss coefficient setting in an off-line training detection neural network model, extracting sample characteristics, constructing the electric bus discharge loss coefficient setting model based on the BP neural network, wherein the hidden layer of the neural network adopts a tan sig function, and the output of the hidden layer is shown as a formula (11):
wherein: x is X i ' t is the input of the discharge loss coefficient setting neural network, i is the number of layers of the neural network training, and 5 layers of training is needed to be carried out on the sample; h is a j Setting the output of the hidden layer of the neural network for the discharge loss coefficient; w (W) ij Setting a connection weight value from an input layer to an output layer of the neural network for the discharge loss coefficient; b j Setting a threshold value of the neural network for the discharge loss coefficient; j ranges from 1 to n; n is the number of the hidden layer units of the discharge loss coefficient setting neural network. The output layer of the discharging loss coefficient setting neural network adopts purelin function, and the output of the neural network is shown as formula (13)
Wherein: k (k) m Setting the output of the neural network for the discharge loss coefficient; w (w) jm Setting a connection weight value from a hidden layer unit to an output layer unit of the neural network for the discharge loss coefficient, wherein m ranges from 1 to 3;
obtaining accurate discharge attenuation coefficient k of electric bus 1 Discharge residual SOC coefficient k of electric bus 2 Depth of discharge coefficient k for electric buses 3 Then substituting the obtained product into a formula (9) to obtain the accurate electric motorA bus discharge loss model; by adopting the same method, an accurate electric bus charging loss model is obtained.
In a further step S04, firstly, an electric bus charging and discharging participation degree model is built, a formula (14) is an electric bus discharging participation degree model, a formula (15) is an electric bus charging participation degree model,
wherein: f (f) ex (t) is the excitation electricity price at the time t,for the discharge loss of the electric bus h at time t, < >>The charge loss of the electric bus h at the time t is obtained according to formulas (3), (6), (14) and (15), and the charge-discharge excitation-response characteristic curve of the electric bus can be obtained according to the excitation electricity price f when the electric bus meets formulas (3) and (14) ex (t) participate in the discharge response; when the electric bus satisfies the formulas (6) and (15), the electric bus can be started according to the exciting electricity price f ex (t) participate in the charging response.
The invention has the advantages and beneficial effects that: according to the method, based on the BP neural network, the history sample of the electric bus is trained and learned, so that a more accurate loss coefficient can be given, the accuracy of excitation-response characteristics of the electric bus is improved, and a strong technical support is provided for the electric bus to participate in power grid regulation.
Drawings
FIG. 1 is a flow chart of the present invention;
FIG. 2 is a graph of a discharge loss factor tuning model based on a BP neural network;
fig. 3 is a graph of excitation-response characteristics of an electric bus.
Detailed Description
The invention is further illustrated by the following examples according to the drawings of the specification:
example 1
As shown in fig. 1, the accurate quantification method of the excitation-response characteristics of the electric bus based on the BP comprises the steps of electric bus charging and operation information acquisition, construction of an electric bus charging and discharging operation constraint model, construction of an electric bus charging and discharging loss model based on the BP neural network, and construction of an electric bus excitation-response characteristic model based on the BP neural network, wherein the steps are as follows:
s01: firstly, collected electric bus charging information: the method comprises the steps of current SOC of an electric bus, initial SOC of the electric bus, attenuated SOC of the electric bus, charging power of the electric bus, discharging power of the electric bus, charging efficiency of the electric bus and discharging efficiency of the electric bus; then, the collected electric bus operation information mainly includes: departure time, incentive price, up-regulation response power, down-regulation response power;
s02: constructing a constraint model of the charge and discharge operation of the electric bus: the electric bus participates in the power grid response on the premise that operation constraint conditions are required to be met, namely, the electric bus participates in the power grid response on the premise that the operation plan of the electric bus is not influenced, the operation constraint is determined by the current residual SOC and the residual departure time of the electric bus, a built electric bus charging and discharging operation constraint model comprises a charging constraint model and a discharging constraint model, the charging operation constraint model aims at guaranteeing that the electric bus does not exceed the attenuated SOC after participating in the charging response, battery charging explosion is prevented, and the discharging operation constraint model aims at guaranteeing that the residual SOC after participating in the discharging response of the electric bus does not influence the next departure requirement;
s03: building an electric bus charge-discharge loss model based on BP neural network: the electric bus can generate charging loss or discharging loss in the charging and discharging process, and the electric bus responds to charging and discharging only when the excitation price is larger than the charging and discharging loss, so that a BP neural network model is adopted to train collected charging information and operation information of the electric bus, coefficients in a built BP neural network-based electric bus charging and discharging loss model are set, and the accuracy of the charging and discharging loss model is improved;
s04: according to the BP neural network-based electric bus charge-discharge loss model constructed in the step S03, the charge-discharge loss of the electric bus is accurately described, the BP neural network-based electric bus excitation-response characteristic model is constructed in combination with the electric bus charge-discharge operation constraint in the step S02, the accurate electric bus excitation-response characteristic is quantitatively obtained, and further the electric bus is effectively scheduled to participate in the power grid operation.
The data collected by the electric bus charging and operation information collecting method provided in step S01 of the present embodiment are as follows by using formulas (1) and (2):
wherein: electric bus charging information acquired at moment X (t) is t, S h (t) is the current SOC of the electric bus h acquired at the moment t,for the collected electric bus h initial SOC, < > for>SOC attenuated for the collected electric bus h, < ->Charging power of electric bus h collected at t moment, < > for>Electric bus h discharge power collected at t moment, < > for>Charging efficiency for collected electric buses h, < >>For the collected electric bus h discharge efficiency, Y (t) is electric bus operation information collected at t moment, < >>Is the departure time f of the electric bus h after the moment t ex (t) is the incentive price collected at the moment t, P up (t) the up-regulating response power acquired at the moment t, P dn And (t) the down-regulating response power acquired at the moment t.
The steps of the electric bus charging and discharging constraint model constructed in the step S02 of the embodiment are as follows:
s21: constructing an electric bus discharging operation constraint model, as shown in formulas (3) - (5), wherein formula (3) represents the SOC remained after discharging ΔtminGreater than the discharge threshold->When the electric bus h meets the discharging operation constraint condition; residual SOC of electric bus after discharging>Calculated according to formula (4), discharge threshold +.>Calculating according to formula (5);
in the method, in the process of the invention,for the SOC remaining after the electric bus h discharges Δtmin at time t, +.>Is the discharge threshold value of the electric bus h, S h And (t) is the SOC of the electric bus h at the time t. />For the discharge efficiency of the electric bus h, +.>For the discharge power of the electric bus h, +.>Charging efficiency for electric bus h, +.>Is the charging power of the electric bus h,for the next departure time of the electric bus after time t,/for the next departure time of the electric bus after time t>For electric buses to meet the time +.>Minimum SOC required when transmitting.
S22: constructing a charging operation constraint model of the electric bus, wherein as shown in formulas (6) - (8), the formula (6) represents the SOC of the electric bus after charging delta tminLess than the charging threshold->When the electric bus h meets the charging operation constraint condition, the SOC of the electric bus after charging is +.>Calculated according to equation (7), the charging threshold value of the electric bus +.>Calculated according to equation (8).
In the method, in the process of the invention,SOC after charging delta tmin at t for electric bus h, +.>The charging threshold value of the electric bus h is S h (t) is the SOC of the electric bus h at the time t, < >>Charging efficiency for electric bus h, +.>Charging power for electric bus h, +.>The maximum SOC value of the electric bus h after attenuation.
The electric bus charge and discharge loss model based on the BP neural network constructed in the step S3 of the embodiment comprises a bus charge and discharge loss model, and the construction steps are as follows:
s31, constructing an electric bus charging and discharging loss model: electric bus discharge loss model constructedAs shown in formula (9), the constructed electric bus charging loss model>As shown in formula (10), the charge-discharge loss model of the electric bus is mainly equal to the initial maximum SOC of the electric bus +.>Maximum SOC (state of charge) of electric bus after attenuation>SOCS of electric bus h at t moment h (t) determining that the number of the cells to be processed is,
wherein:for the discharge loss of the electric bus h at time t, < >>Is the attenuated SOC of the electric bus h,is the initial SOC of the electric bus h, S h (t) is the SOC of the electric bus h at the time t, < >>For the discharge efficiency of the electric bus h, +.>Charging efficiency for electric bus h, +.>For the discharge power of the electric bus h, +.>Charging power k for electric bus h 1 Is the discharge attenuation coefficient k of the electric bus 2 Is the discharge residual SOC coefficient, k of the electric bus 3 Is the depth of discharge coefficient, k of the electric bus 4 The charging attenuation coefficient k of the electric bus 5 Charging residual SOC coefficient k for electric bus 6 The charging depth coefficient of the electric bus;
s32, adopting a BP neural network model to set coefficients in a charge and discharge loss model of the electric bus, wherein the built discharge loss coefficient setting model based on the BP neural network is shown in fig. 2, and the characteristics related to discharge loss are extracted by training and learning charge and discharge data samples and partial operation data samples of the electric bus, so that the accurate setting of the discharge loss coefficient of the electric bus is realized, 4 types of inputs and 3 outputs are shared, and the 4 types of inputs are respectively: x (t) is the charge and discharge operation data of the electric bus at the moment t,The next departure time f of the electric bus h after the time t ex (t) incentive price, P at time t dn (t) is the down-regulated response power at time t; the 3 outputs are respectively: discharge attenuation coefficient k of electric bus 1 Discharge residual SOC coefficient k of electric bus 2 Depth of discharge coefficient k for electric buses 3 ；
The method for setting the coefficients in the electric bus charge-discharge loss model comprises the following steps of: collecting training samples of charge and discharge information and part of operation information of an electric bus, training the collected samples, carrying out network test on a trained discharge loss coefficient setting model, obtaining a BP neural network training function, a connection weight and a threshold value for discharge loss coefficient setting in an off-line training detection neural network model, extracting sample characteristics, constructing the electric bus discharge loss coefficient setting model based on the BP neural network, wherein the hidden layer of the neural network adopts a tan sig function, and the output of the hidden layer is shown as a formula (11):
wherein: x is X i ' t is the input of the discharge loss coefficient setting neural network, i is the number of layers of the neural network training, and 5 layers of training is needed to be carried out on the sample; h is a j Setting the output of the hidden layer of the neural network for the discharge loss coefficient; w (W) ij Setting a connection weight value from an input layer to an output layer of the neural network for the discharge loss coefficient; b j Setting a threshold value of the neural network for the discharge loss coefficient; j ranges from 1 to n; n is the number of the hidden layer units of the discharge loss coefficient setting neural network. The output layer of the discharging loss coefficient setting neural network adopts purelin function, and the output of the neural network is shown as formula (13)
Wherein: k (k) m Setting the output of the neural network for the discharge loss coefficient; w (w) jm Setting a connection weight value from a hidden layer unit to an output layer unit of the neural network for the discharge loss coefficient, wherein m ranges from 1 to 3;
obtaining accurate discharge attenuation coefficient k of electric bus 1 Discharge residual SOC coefficient k of electric bus 2 Depth of discharge coefficient k for electric buses 3 Then substituting the model into a formula (9) to obtain an accurate electric bus discharge loss model; by adopting the same method, an accurate electric bus charging loss model is obtained.
In step S04 of this embodiment, firstly, an electric bus charge and discharge participation model is constructed, formula (14) is the electric bus discharge participation model, formula (15) is the electric bus charge participation model,
wherein: f (f) ex (t) is the excitation electricity price at the time t,for the discharge loss of the electric bus h at time t, < >>The charge loss of the electric bus h at the time t is then obtained according to formulas (3), (6), (14) and (15), and as shown in fig. 3, when the electric bus satisfies formulas (3) and (14), the charge-discharge excitation-response characteristic curve of the electric bus can be obtained according to the excitation electricity price f ex (t) participate in the discharge response; when the electric bus satisfies the formulas (6) and (15), the electric bus can be started according to the exciting electricity price f ex (t) participate in the charging response.
Claims (5)
1. The accurate quantification method of the excitation-response characteristics of the electric buses based on BP comprises the steps of electric bus charging and operation information acquisition, construction of an electric bus charging and discharging operation constraint model, construction of an electric bus charging and discharging loss model based on BP neural network, and construction of the excitation-response characteristics model of the electric buses based on BP neural network, and is characterized by comprising the following steps:
s01: firstly, collected electric bus charging information: the method comprises the steps of current SOC of an electric bus, initial SOC of the electric bus, attenuated SOC of the electric bus, charging power of the electric bus, discharging power of the electric bus, charging efficiency of the electric bus and discharging efficiency of the electric bus; then, the collected electric bus operation information mainly includes: departure time, incentive price, up-regulation response power, down-regulation response power;
s02: constructing a constraint model of the charge and discharge operation of the electric bus: the electric bus participates in the power grid response on the premise that operation constraint conditions are required to be met, namely, the electric bus participates in the power grid response on the premise that the operation plan of the electric bus is not influenced, the operation constraint is determined by the current residual SOC and the residual departure time of the electric bus, a built electric bus charging and discharging operation constraint model comprises a charging constraint model and a discharging constraint model, the charging operation constraint model aims at guaranteeing that the electric bus does not exceed the attenuated SOC after participating in the charging response, battery charging explosion is prevented, and the discharging operation constraint model aims at guaranteeing that the residual SOC after participating in the discharging response of the electric bus does not influence the next departure requirement;
s03: building an electric bus charge-discharge loss model based on BP neural network: the electric bus can generate charging loss or discharging loss in the charging and discharging process, and the electric bus responds to charging and discharging only when the excitation price is larger than the charging and discharging loss, so that a BP neural network model is adopted to train collected charging information and operation information of the electric bus, coefficients in a built BP neural network-based electric bus charging and discharging loss model are set, and the accuracy of the charging and discharging loss model is improved;
s04: according to the BP neural network-based electric bus charge-discharge loss model constructed in the step S03, the charge-discharge loss of the electric bus is accurately described, and the BP neural network-based electric bus excitation-response characteristic model is constructed by combining the electric bus charge-discharge operation constraint in the step S02, namely, the accurate electric bus excitation-response characteristic is quantitatively obtained, so that the electric bus is effectively scheduled to participate in the operation of a power grid.
2. The accurate quantification method of the excitation-response characteristics of the electric bus based on the BP of claim 1, wherein the data acquired by the electric bus charging and operation information acquisition method proposed in the step S01 are as follows by using formulas (1) and (2):
wherein: electric bus charging information acquired at moment X (t) is t, S h (t) is the current SOC of the electric bus h acquired at the moment t,for the collected electric bus h initial SOC, < > for>SOC attenuated for the collected electric bus h, < ->Charging power of electric bus h collected at t moment, < > for>Electric bus h discharge acquired at t momentPower (I)>Charging efficiency for collected electric buses h, < >>For the collected electric bus h discharge efficiency, Y (t) is electric bus operation information collected at t moment, < >>Is the departure time f of the electric bus h after the moment t ex (t) is the incentive price collected at the moment t, P up (t) the up-regulating response power acquired at the moment t, P dn And (t) the down-regulating response power acquired at the moment t.
3. The BP-based electric bus excitation-response characteristic accurate quantification method of claim 2, wherein the electric bus charging and discharging constraint model constructed in the step S02 comprises the following steps:
s21: constructing an electric bus discharging operation constraint model, as shown in formulas (3) - (5), wherein formula (3) represents the residual after discharging delta tminGreater than the discharge threshold->When the electric bus h meets the discharging operation constraint condition; residual ∈of electric bus after discharge>Calculated according to formula (4), discharge threshold +.>Calculating according to formula (5);
in the method, in the process of the invention,for the SOC remaining after the electric bus h discharges Δtmin at time t, +.>Is the discharge threshold value of the electric bus h, S h And (t) is the SOC of the electric bus h at the time t. />For the discharge efficiency of the electric bus h, +.>For the discharge power of the electric bus h, +.>Charging efficiency for electric bus h, +.>Charging power for electric bus h, +.>For the next departure time of the electric bus after time t,/for the next departure time of the electric bus after time t>For electric buses to meet the time +.>Minimum SOC required when transmitting.
S22: constructing a charging operation constraint model of the electric bus, wherein the formula (6) represents that the electric bus is charged delta tmin as shown in the formulas (6) - (8)Less than the charging threshold->When the electric bus h meets the charging operation constraint condition, the electric bus after charging is +.>Calculated according to equation (7), the charging threshold value of the electric bus +.>Calculated according to equation (8).
In the method, in the process of the invention,SOC after charging delta tmin at t for electric bus h, +.>The charging threshold value of the electric bus h is S h (t) is the SOC of the electric bus h at the time t, < >>Charging efficiency for electric bus h, +.>Charging power for electric bus h, +.>The maximum SOC value of the electric bus h after attenuation.
4. The BP-based electric bus excitation-response characteristic accurate quantification method is characterized in that an electric bus charge-discharge loss model based on a BP neural network constructed in the step S3 comprises an electric bus charge-discharge loss model, and the construction steps are as follows:
s31, constructing an electric bus charging and discharging loss model: electric bus discharge loss model constructedAs shown in formula (9), the constructed electric bus charging loss model>As shown in formula (10), the charge and discharge loss model of the electric bus is mainly equal to the initial maximum +.>Maximum of electric bus after attenuation>SOCS of electric bus h at t moment h (t) determining that the number of the cells to be processed is,
wherein:for the discharge loss of the electric bus h at time t, < >>SOC after attenuation of the electric bus h, < >>Is the initial SOC of the electric bus h, S h (t) is the SOC of the electric bus h at the time t, < >>For the discharge efficiency of the electric bus h, +.>Charging efficiency for electric bus h, +.>For the discharge power of the electric bus h, +.>Charging power k for electric bus h 1 Is electric powerDischarge attenuation coefficient k of bus 2 Is the discharge residual SOC coefficient, k of the electric bus 3 Is the depth of discharge coefficient, k of the electric bus 4 The charging attenuation coefficient k of the electric bus 5 Charging residual SOC coefficient k for electric bus 6 The charging depth coefficient of the electric bus;
s32, adopting a BP neural network model to set coefficients in a charge and discharge loss model of the electric bus, and extracting characteristics related to discharge loss by training and learning charge and discharge data samples and partial operation data samples of the electric bus, so as to realize accurate setting of the discharge loss coefficients of the electric bus, wherein 4 types of inputs and 3 outputs are respectively provided, and the 4 types of inputs are respectively: x (t) is the charge and discharge operation data of the electric bus at the moment t,The next departure time f of the electric bus h after the time t ex (t) incentive price, P at time t dn (t) is the down-regulated response power at time t; the 3 outputs are respectively: discharge attenuation coefficient k of electric bus 1 Discharge residual SOC coefficient k of electric bus 2 Depth of discharge coefficient k for electric buses 3 ；
The method for setting the coefficients in the electric bus charge-discharge loss model comprises the following steps of: collecting training samples of charge and discharge information and part of operation information of an electric bus, training the collected samples, carrying out network test on a trained discharge loss coefficient setting model, obtaining a BP neural network training function, a connection weight and a threshold value for discharge loss coefficient setting in an off-line training detection neural network model, extracting sample characteristics, constructing the electric bus discharge loss coefficient setting model based on the BP neural network, wherein the hidden layer of the neural network adopts a tan sig function, and the output of the hidden layer is shown as a formula (11):
wherein: x is X i ' t is the input of the discharge loss coefficient setting neural network, i is the number of layers of the neural network training, and 5 layers of training is needed to be carried out on the sample; h is a j Setting the output of the hidden layer of the neural network for the discharge loss coefficient; w (W) ij Setting a connection weight value from an input layer to an output layer of the neural network for the discharge loss coefficient; b j Setting a threshold value of the neural network for the discharge loss coefficient; j ranges from 1 to n; n is the number of the hidden layer units of the discharge loss coefficient setting neural network. The output layer of the discharging loss coefficient setting neural network adopts purelin function, and the output of the neural network is shown as formula (13)
Wherein: k (k) m Setting the output of the neural network for the discharge loss coefficient; w (w) jm Setting a connection weight value from a hidden layer unit to an output layer unit of the neural network for the discharge loss coefficient, wherein m ranges from 1 to 3;
obtaining accurate discharge attenuation coefficient k of electric bus 1 Discharge residual SOC coefficient k of electric bus 2 Depth of discharge coefficient k for electric buses 3 Then substituting the model into a formula (9) to obtain an accurate electric bus discharge loss model; by adopting the same method, an accurate electric bus charging loss model is obtained.
5. The BP-based electric bus excitation-response characteristic accurate quantification method according to claim 4, wherein in step S04, an electric bus charge and discharge participation model is first constructed, formula (14) is an electric bus discharge participation model, formula (15) is an electric bus charge participation model,
wherein: f (f) ex (t) is the excitation electricity price at the time t,for the discharge loss of the electric bus h at time t, < >>The charge loss of the electric bus h at the time t is obtained according to formulas (3), (6), (14) and (15), and the charge-discharge excitation-response characteristic curve of the electric bus can be obtained according to the excitation electricity price f when the electric bus meets formulas (3) and (14) ex (t) participate in the discharge response; when the electric bus satisfies the formulas (6) and (15), the electric bus can be started according to the exciting electricity price f ex (t) participate in the charging response. />
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311501128.1A CN117474270A (en) | 2023-11-10 | 2023-11-10 | BP-based electric bus excitation-response characteristic accurate quantification method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311501128.1A CN117474270A (en) | 2023-11-10 | 2023-11-10 | BP-based electric bus excitation-response characteristic accurate quantification method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117474270A true CN117474270A (en) | 2024-01-30 |
Family
ID=89629077
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311501128.1A Pending CN117474270A (en) | 2023-11-10 | 2023-11-10 | BP-based electric bus excitation-response characteristic accurate quantification method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117474270A (en) |
Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103507656A (en) * | 2013-10-10 | 2014-01-15 | 同济大学 | Range-extending electric vehicle energy management method and system capable of achieving on-line self-regulation |
| US20150329003A1 (en) * | 2014-05-15 | 2015-11-19 | Ford Global Technologies, Llc | Electric vehicle operation to manage battery capacity |
| CN106712065A (en) * | 2017-02-23 | 2017-05-24 | 沈阳工业大学 | Prediction method of indexes of influence of electric card charging on power distribution network active loss |
| CN110428105A (en) * | 2019-08-01 | 2019-11-08 | 国网江苏省电力有限公司苏州供电分公司 | A kind of electric bus charge and discharge Optimization Scheduling a few days ago |
| CN111523722A (en) * | 2020-04-20 | 2020-08-11 | 武汉大学 | Intelligent charging station optimization selection system based on deep reinforcement learning |
| CN112491036A (en) * | 2020-10-28 | 2021-03-12 | 四川大学 | Multi-subject interactive game power distribution network peak regulation operation control method and system |
| CN112906983A (en) * | 2021-03-22 | 2021-06-04 | 吉林大学 | Electric bus charging scheme optimization method considering time-of-use electricity price influence |
| CN113486504A (en) * | 2021-06-28 | 2021-10-08 | 上海电机学院 | Battery management control method based on scheduling cost |
| CN113702836A (en) * | 2021-07-23 | 2021-11-26 | 国家电网有限公司西北分部 | Lithium ion battery state-of-charge estimation method based on EMD-GRU |
| KR20210148759A (en) * | 2020-06-01 | 2021-12-08 | 한국전력공사 | Charge/Discharge Scheduling Management System, Charge/Discharge Scheduling Server Apparatus and Charge/Discharge Scheduling Method for Electric Vehicle with SOC control |
| CN114389294A (en) * | 2022-01-06 | 2022-04-22 | 国网江苏省电力有限公司电力科学研究院 | Dimension reduction equivalent-oriented mass electric vehicle centralized control method and system |
| WO2023082555A1 (en) * | 2021-11-12 | 2023-05-19 | 上汽通用五菱汽车股份有限公司 | Electric vehicle remaining performance evaluation method and device and computer readable storage medium |
-
2023
- 2023-11-10 CN CN202311501128.1A patent/CN117474270A/en active Pending
Patent Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103507656A (en) * | 2013-10-10 | 2014-01-15 | 同济大学 | Range-extending electric vehicle energy management method and system capable of achieving on-line self-regulation |
| US20150329003A1 (en) * | 2014-05-15 | 2015-11-19 | Ford Global Technologies, Llc | Electric vehicle operation to manage battery capacity |
| CN106712065A (en) * | 2017-02-23 | 2017-05-24 | 沈阳工业大学 | Prediction method of indexes of influence of electric card charging on power distribution network active loss |
| CN110428105A (en) * | 2019-08-01 | 2019-11-08 | 国网江苏省电力有限公司苏州供电分公司 | A kind of electric bus charge and discharge Optimization Scheduling a few days ago |
| CN111523722A (en) * | 2020-04-20 | 2020-08-11 | 武汉大学 | Intelligent charging station optimization selection system based on deep reinforcement learning |
| KR20210148759A (en) * | 2020-06-01 | 2021-12-08 | 한국전력공사 | Charge/Discharge Scheduling Management System, Charge/Discharge Scheduling Server Apparatus and Charge/Discharge Scheduling Method for Electric Vehicle with SOC control |
| CN112491036A (en) * | 2020-10-28 | 2021-03-12 | 四川大学 | Multi-subject interactive game power distribution network peak regulation operation control method and system |
| CN112906983A (en) * | 2021-03-22 | 2021-06-04 | 吉林大学 | Electric bus charging scheme optimization method considering time-of-use electricity price influence |
| CN113486504A (en) * | 2021-06-28 | 2021-10-08 | 上海电机学院 | Battery management control method based on scheduling cost |
| CN113702836A (en) * | 2021-07-23 | 2021-11-26 | 国家电网有限公司西北分部 | Lithium ion battery state-of-charge estimation method based on EMD-GRU |
| WO2023082555A1 (en) * | 2021-11-12 | 2023-05-19 | 上汽通用五菱汽车股份有限公司 | Electric vehicle remaining performance evaluation method and device and computer readable storage medium |
| CN114389294A (en) * | 2022-01-06 | 2022-04-22 | 国网江苏省电力有限公司电力科学研究院 | Dimension reduction equivalent-oriented mass electric vehicle centralized control method and system |
Non-Patent Citations (7)
| Title |
|---|
| MOGHADAM MRV等: "Distributed Frequency Control via Randomized Response of Electric Vehicles in Power Grid", 《IEEE TRANSACTIONS ON SUSTAINABLE ENERGY》, vol. 7, no. 1, 18 November 2015 (2015-11-18), pages 312 - 324, XP011594241, DOI: 10.1109/TSTE.2015.2494504 * |
| SINGH J 等: "Cost Benefit Analysis for V2G Implementation of Electric Vehicles in Distribution System", 《IEEE TRANSACTIONS ON INDUSTRY APPLICATIONS》, vol. 56, no. 5, 14 April 2020 (2020-04-14), pages 5963 - 5973, XP011809533, DOI: 10.1109/TIA.2020.2986185 * |
| 张书盈等: "考虑分时电价和电池损耗的电动汽车集群V2G响应成本分析", 《电力***及其自动化学报》, vol. 29, no. 11, 15 November 2017 (2017-11-15), pages 39 - 46 * |
| 王震等: "基于预测电池SOC的充电控制策略研究", 《电子质量》, no. 02, 20 February 2017 (2017-02-20), pages 1 - 3 * |
| 章锐等: "电动公交车负荷参与新型城网调峰的能力评估", 《中国电机工程学报》, vol. 42, no. 1, 31 August 2022 (2022-08-31), pages 82 - 94 * |
| 陈丽娟等: "计及电池损耗的电动公交车参与V2G的优化调度策略", 《电力***自动化》, vol. 44, no. 11, 10 June 2020 (2020-06-10), pages 52 - 60 * |
| 马俊等: "不同工况对锂离子电池SOC预估精度的影响", 《电源技术》, vol. 40, no. 12, 20 December 2016 (2016-12-20), pages 2335 - 2338 * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110133507B (en) | Battery remaining capacity estimation method based on NARX-UKF algorithm | |
| CN111680848A (en) | Battery life prediction method based on prediction model fusion and storage medium | |
| CN107290688A (en) | A kind of lithium battery SOC methods of estimation based on adaptive fuzzy Kalman filtering | |
| CN112434463B (en) | Energy management system for vehicle hybrid power supply | |
| CN105974327A (en) | Lithium battery pack SOC prediction method based on neural network and UKF | |
| CN111366848A (en) | Battery health state prediction method based on PSO-ELM algorithm | |
| CN111352032A (en) | Lithium battery dynamic peak power prediction method | |
| CN112526883B (en) | Vehicle energy management method based on intelligent networking information | |
| CN117104074B (en) | Method for controlling charge and discharge and charge and discharge control system | |
| CN111814826B (en) | Rapid detection and rating method for residual energy of retired power battery | |
| CN115639481B (en) | Battery data preprocessing system and method based on big data prediction SOC | |
| CN110658459A (en) | Lithium ion battery state of charge estimation method based on bidirectional cyclic neural network | |
| Lu et al. | Modeling discharge characteristics for predicting battery remaining life | |
| CN112163372A (en) | SOC estimation method of power battery | |
| CN114865668A (en) | Energy storage scheduling support evaluation method | |
| CN115753132A (en) | Development method for low-temperature driving range test regulations of electric automobile | |
| CN114646888A (en) | Assessment method and system for capacity attenuation of power battery | |
| CN110232432B (en) | Lithium battery pack SOC prediction method based on artificial life model | |
| Wang et al. | Research on multi-parameter evaluation of electric vehicle power battery consistency based on principal component analysis | |
| CN117474270A (en) | BP-based electric bus excitation-response characteristic accurate quantification method | |
| CN117491872A (en) | Reconfigurable battery module fault multistage diagnosis method | |
| CN116482543A (en) | UPS battery health state evaluation method, device and storage medium | |
| CN116679213A (en) | SOH estimation method for electric vehicle power battery based on integrated deep learning | |
| Li et al. | State of health prediction for battery based on ensemble learning | |
| CN109617153A (en) | A kind of mining lithium battery management system |
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 |