CN113159980A - Optimal regulation and control method for wind-solar absorption-improving thermoelectric combined system - Google Patents

Optimal regulation and control method for wind-solar absorption-improving thermoelectric combined system Download PDF

Info

Publication number
CN113159980A
CN113159980A CN202011519542.1A CN202011519542A CN113159980A CN 113159980 A CN113159980 A CN 113159980A CN 202011519542 A CN202011519542 A CN 202011519542A CN 113159980 A CN113159980 A CN 113159980A
Authority
CN
China
Prior art keywords
stage
model
dirichlet
unit
constructing
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
Application number
CN202011519542.1A
Other languages
Chinese (zh)
Inventor
王雅宾
褚孝国
张�浩
杨政厚
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Huaneng Group Technology Innovation Center Co Ltd
Beijing Huaneng Xinrui Control Technology Co Ltd
Original Assignee
Huaneng Group Technology Innovation Center Co Ltd
Beijing Huaneng Xinrui Control Technology Co Ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Huaneng Group Technology Innovation Center Co Ltd, Beijing Huaneng Xinrui Control Technology Co Ltd filed Critical Huaneng Group Technology Innovation Center Co Ltd
Priority to CN202011519542.1A priority Critical patent/CN113159980A/en
Publication of CN113159980A publication Critical patent/CN113159980A/en
Pending legal-status Critical Current

Links

Images

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
    • G06Q50/00—Systems or methods specially adapted for specific business sectors, e.g. utilities or tourism
    • G06Q50/06—Electricity, gas or water supply
    • 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
    • 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
    • H02J3/38—Arrangements for parallely feeding a single network by two or more generators, converters or transformers
    • H02J3/381—Dispersed generators
    • 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
    • G06F2113/06—Wind turbines or wind farms
    • 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
    • H02J2300/00—Systems for supplying or distributing electric power characterised by decentralized, dispersed, or local generation
    • H02J2300/20—The dispersed energy generation being of renewable origin
    • H02J2300/22—The renewable source being solar energy
    • 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
    • H02J2300/00—Systems for supplying or distributing electric power characterised by decentralized, dispersed, or local generation
    • H02J2300/20—The dispersed energy generation being of renewable origin
    • H02J2300/28—The renewable source being wind energy
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02E—REDUCTION OF GREENHOUSE GAS [GHG] EMISSIONS, RELATED TO ENERGY GENERATION, TRANSMISSION OR DISTRIBUTION
    • Y02E10/00—Energy generation through renewable energy sources
    • Y02E10/70—Wind energy
    • Y02E10/76—Power conversion electric or electronic aspects

Abstract

The optimal regulation and control method for the combined heat and power system for improving wind and light absorption comprises the steps of establishing a combined heat and power model comprising a wind power generation system, a photovoltaic power generation system and a cogeneration system, establishing a Dirichlet fuzzy set, establishing a two-stage combined heat and power system distributed robust model, and solving the two-stage combined heat and power system distributed robust model. The fuzzy set constructed by the invention contains potential actual probability distribution, and a narrower fuzzy set can be constructed by more observation data so as to reduce model conservatism. The construction of the fuzzy set can be converted into a deterministic problem, so that the fuzzy set is connected with a traditional adaptive robust optimization model, and the constructed distributed robust model has inheritance. The model constructed by the invention is based on a thermoelectric system scheduling architecture, follows a self-adaptive mechanism between cooperative regulation and robust optimization, and the obtained result can be directly used for solving the system optimization regulation problem of the current thermoelectric system.

Description

Optimal regulation and control method for wind-solar absorption-improving thermoelectric combined system
Technical Field
The invention belongs to the technical field of optimization of a combined heat and power system, and particularly relates to a combined heat and power system optimization regulation and control method for improving wind and light absorption.
Background
In the winter heating period in northern China, a large number of cogeneration units work in a 'fixed power by heat' operation mode, so that the peak regulation capacity of the system is insufficient, and the situation of wind and light abandonment is severe. The wind power heat supply of abandoned wind can be effectively reduced, and the economic benefit problem and the scheduling platform are not perfect, so that the large-scale popularization and application are difficult. In addition, the increase of the grid-connected amount of the distributed energy greatly influences the safe and stable operation of the power system. Thus, the introduction of the cogeneration system greatly ameliorates the above-described problems. However, in optimizing the scheduling process, the cogeneration system faces many uncertainties, such as the production of renewable energy (wind or solar), market prices, load demands, etc. Obviously, the traditional deterministic optimization method cannot adapt to the combined heat and power optimization scheduling problem considering various uncertainties. Therefore, a reasonable scheduling method is adopted to quantify or weaken the influence of uncertainty on the scheduling strategy, and great attention is paid to people.
For uncertainty in the operation of cogeneration systems, the most basic optimization methods currently include Interval Optimization (IO), Stochastic Programming (SP), and Robust Optimization (RO). In the related technology I, a thermoelectric combined system economic dispatching model with interval uncertainty is established, and then the interval-based economic dispatching model is converted into a deterministic economic dispatching model for solving through probability degree definition. The method is convenient for solving the actual engineering problem, but the decision in the interval number form is not easy to be directly applied. The SP method is another popular method. The method presupposes probability density functions of uncertain parameters, e.g. wind speed models usually follow a weibull distribution. In order to minimize the day-ahead cost and balance the market of the cogeneration system, considering the uncertainty of the wind power and load demand, the second related art proposes a random two-tier optimization problem of renewable energy utilization. The SP method is based on known probability density, but is not suitable for practical situations where it is difficult to find the probability density function of uncertainty. In addition, the scenes generated by the SP method also cause an increase in the amount of calculation. And in the third related technology, under the condition of considering uncertainty of market price, a community energy hub optimal load scheduling model for reducing the total cost is provided. The method takes the boundary of uncertain parameters as a representative, and is difficult to select a proper robust set, so that the result becomes conservative.
In recent years, Distributed Robust Optimization (DRO) has been widely applied to solving uncertainty optimization problems. And combining the advantages of the SP and the RO, wherein the DRO implements a scheduling strategy under the worst uncertain probability distribution in the fuzzy set, and the construction of the fuzzy set is a key factor strategy for influencing scheduling. And a two-stage distribution robust model is established in the fourth related technology, and the optimal quotation strategy of a certain comprehensive wind power plant is deduced. The wind power generation output is used as an uncertain parameter and is characterized by a fuzzy set defining a distribution family. The optimal decision of the bidding strategy is robust to the expectation of the worst case distribution.
Disclosure of Invention
The invention aims to at least solve one of the technical problems in the prior art and provides a combined heat and power system optimization regulation and control method for improving wind and light absorption.
The invention provides a thermoelectric combined system optimization regulation and control method for improving wind and light absorption, which comprises the following steps:
establishing a combined heat and power model comprising a wind power generation system, a photovoltaic power generation system and a combined heat and power generation system;
constructing a Dirichlet fuzzy set;
constructing a two-stage electric heating combined system distributed robust model;
and solving the distributed robust model of the two-stage electric heating combined system.
In some optional embodiments, the building a cogeneration model including a wind power generation system, a photovoltaic power generation system, and a cogeneration system includes:
processing uncertainty of wind speed by using Weibull distribution to construct a wind power generation model;
processing the uncertainty of illumination by utilizing beta distribution to construct a photovoltaic power generation model;
and establishing an electric heat output model of the cogeneration technology according to the electric heat coupling characteristic of the cogeneration technology.
In some optional embodiments, said constructing the dirichlet fuzzy set comprises:
constructing a Dirichlet hybrid model according to the non-supervised classified Dirichlet process;
and constructing the Dirichlet fuzzy set by combining a variation inference principle.
In some optional embodiments, the constructing a dirichlet hybrid model according to the unsupervised classified dirichlet process includes:
the prior dirichlet probability density function is:
Figure BDA0002849030340000031
in the formula, Γ (—) is represented by a Gamma function, and r ═ r (r)1,r2,...,rn)，riAs a prior parameter, satisfies r is more than or equal to 0iLess than or equal to 1 and
Figure BDA0002849030340000032
representing the parameter thetaiMean value, beta, under Dirichlet prior distributioni＝sri，βiRepresentation priorWeighting or implicit observations of the information, s being set at [1,2 ]]Internal;
after obtaining the sample observation value M, the prior dirichlet probability density function is updated by the bayesian process to obtain a posterior distribution:
Figure BDA0002849030340000033
wherein M is the total number of sample observations, MiIndicating the situation xi in M observationsiThe number of occurrences;
after obtaining the data observations, the parameter θ is giveniMean value r of prior probabilitiesiThe situation ξ is estimated from the expected value of the posterior distribution shown in the following expression (3)iThe posterior probability of occurrence, i.e.:
Figure BDA0002849030340000034
the Dirichlet fuzzy set is constructed by combining a variation inference principle, and the construction method comprises the following steps:
obtaining the minimum value and the maximum value of the posterior expected value of the parameter by adopting a parameter estimation method:
Figure BDA0002849030340000035
in some optional embodiments, the constructing the two-stage electric-thermal combined system distributed robust model comprises:
constructing a two-stage objective function, wherein the first stage is to minimize the cost of a pre-dispatching stage, and the second stage is to minimize the cost of a readjusting stage;
and constructing the constraint condition of the two-stage objective function.
In some optional embodiments, the objective function of the first stage is:
Figure BDA0002849030340000036
wherein variables are aggregated
Figure BDA0002849030340000041
Figure BDA0002849030340000042
And
Figure BDA0002849030340000043
is the supply price of the electric energy and the heat energy of the CHP unit c,
Figure BDA0002849030340000044
and
Figure BDA0002849030340000045
is the up/down reserve price of the CHP and CTP units,
Figure BDA0002849030340000046
is the electric energy quotation of the CTP unit g, Sc/BcAnd Sg/BgThe starting and stopping costs of the CHP unit c and the CTP unit g are respectively,
Figure BDA0002849030340000047
and
Figure BDA0002849030340000048
is the output power of the electric energy and the heat energy of the CHP unit c,
Figure BDA0002849030340000049
is the electric energy output power of the CTP unit g,
Figure BDA00028490303400000410
and
Figure BDA00028490303400000411
is the upper/lower spare consumption of the CHP/CTP unit, vc,t/zc,t/vg,t/zg,tIs a binary variable when v isc,t1 denotes time CHPStarting the unit h, then the starting cost ShIs calculated into the total cost, otherwise, the shutdown cost BhIs calculated into the total cost, SgAnd BgIs the start-stop cost of CTP set, phiCHPAnd phiCTPIs a CHP unit and a CTP unit set;
the constraints of the objective function of the first stage are:
Figure BDA00028490303400000412
Figure BDA00028490303400000413
Figure BDA00028490303400000414
Figure BDA00028490303400000415
Figure BDA00028490303400000416
Figure BDA00028490303400000417
Figure BDA00028490303400000418
Figure BDA00028490303400000419
Figure BDA00028490303400000420
Figure BDA00028490303400000421
Figure BDA00028490303400000422
Figure BDA00028490303400000423
Figure BDA00028490303400000424
wherein, BnmIs the admittance between transmission lines nm, deltan,tAnd deltam,tRespectively the phase angle between the busbars n and m,
Figure BDA0002849030340000051
and
Figure BDA0002849030340000052
is a WPP unit w and a PV unitpThe output power of the power converter (c),
Figure BDA0002849030340000053
is the demanded rated power of load l;
Figure BDA0002849030340000054
and
Figure BDA0002849030340000055
is the heat-electricity conversion proportion of the cogeneration unit,
Figure BDA0002849030340000056
and
Figure BDA0002849030340000057
represents an up/down reserve capacity;
Figure BDA0002849030340000058
and
Figure BDA0002849030340000059
is a ramp rate up/down ramp rate limit, uc,tAnd ug,tIs a binary variable;
Figure BDA00028490303400000510
and
Figure BDA00028490303400000511
is the minimum continuous run/off time.
In some optional embodiments, the objective function of the second stage is:
Figure BDA00028490303400000512
wherein, the variable set
Figure BDA00028490303400000513
For any time, it is the uncertain parameter, variable set of readjustment stage
Figure BDA00028490303400000514
Is the decision variable of the second stage and,
Figure BDA00028490303400000515
and
Figure BDA00028490303400000516
the upward/downward adjustment amounts of the CHP unit and the CTP unit respectively,
Figure BDA00028490303400000517
the supply and demand after the rotating reserve is provided are unbalanced;
the constraints of the objective function of the second stage are:
Figure BDA00028490303400000518
Figure BDA00028490303400000519
Figure BDA00028490303400000520
Figure BDA00028490303400000521
Figure BDA00028490303400000522
wherein the constraint (20) is a power balance at real time operation,
Figure BDA00028490303400000523
and
Figure BDA00028490303400000524
and the phase angle between the buses n and m in the second adjustment stage, and the constraint (21) to (24) are constrained by the scheduling result of the first stage, so that each unit is limited and adjusted in real time.
In some optional embodiments, the solving the two-stage electric-thermal combined system distributed robust model comprises:
converting the second-stage double-layer problem into a single-layer problem by using a KKT condition;
and circularly solving the two-stage problem by using a C & CG algorithm.
In some optional embodiments, the two-stage electric-thermal combined system distributed robust model is:
Figure BDA0002849030340000061
in some optional embodiments, the solving the two-stage problem circularly using the C & CG algorithm includes:
decomposing the two-stage problem into a Main Problem (MP) and a Sub Problem (SP), and then carrying out iterative solution;
in each iteration, the optimal solution of the SP is transferred into the MP containing new variables and corresponding constraints;
the MP is described by the following equation (26), the worst uncertainty pi found from the SP of the previous iteration*The total cost is minimized in the case:
Figure BDA0002849030340000062
wherein the content of the first and second substances,
Figure BDA0002849030340000063
the conversion of the second stage double layer problem to a single layer problem using the KKT condition includes:
converting the SP to an equivalent single layer problem:
Figure BDA0002849030340000064
where λ is the dual parameter in equation (25) and j is the indentation of the corresponding constraint.
The fuzzy set constructed by the invention contains potential actual probability distribution, and a narrower fuzzy set can be constructed by more observation data so as to reduce model conservatism. The construction of the fuzzy set can be converted into a deterministic problem, so that the fuzzy set is connected with a traditional adaptive robust optimization model, and the constructed distributed robust model has inheritance. Compared with the traditional method for giving the uncertain set boundary, the model method provided by the invention does not need to know the specific probability distribution of wind power and photoelectricity, can construct the wind power photovoltaic output interval by using a small amount of information under the condition that the historical data of the wind power and the photoelectricity are limited, can realize the conservative control of the robust optimization model by adjusting the uncertainty parameter, and improves the applicability of the model and the method. The model constructed by the invention is based on a thermoelectric system scheduling architecture, follows a self-adaptive mechanism between cooperative regulation and robust optimization, and the obtained result can be directly used for solving the system optimization regulation problem of the current thermoelectric system.
Drawings
FIG. 1 is a flow chart of a method for optimizing and controlling a combined heat and power system for enhancing wind and light absorption according to an embodiment of the invention;
FIG. 2 is a schematic diagram illustrating the CDF confidence as the number of samples increases, in accordance with an embodiment of the present invention;
FIG. 3 is a diagram illustrating conversion of a fuzzy set into an uncertainty set according to another embodiment of the present invention.
Detailed Description
In order to make the technical solutions of the present invention better understood, the present invention will be described in further detail with reference to the accompanying drawings and specific embodiments.
As shown in fig. 1, a method S100 for optimally regulating a cogeneration system to improve wind and light absorption includes steps S110, S120, S130, and S140.
And S110, establishing a combined heat and power model comprising a wind power generation system, a photovoltaic power generation system and a cogeneration system. Step S110 specifically includes the following steps:
and processing uncertainty of wind speed by using Weibull (Weibull) distribution to construct a wind power generation model.
And (3) processing the uncertainty of illumination by utilizing Beta distribution to construct a photovoltaic power generation model.
And establishing a heat and power (CHP) electric heat output model according to the electric heat coupling characteristic of the CHP.
And S120, constructing a Dirichlet fuzzy set.
First, a dirichlet hybrid model is constructed according to the unsupervised classified dirichlet process.
Suppose there are N possible occurrences of an event, each with ξiIndicating that the sample space for this event is Ω ═ ξ1,ξ2,...,ξn}; each instance of the event has a completely uniform probability distribution, with the probability of each instance occurring being equal to (θ)1,θ2,...,θn) And (4) showing. With the aid of the bayesian statistical principle, the prior distribution of dirichlet in the deterministic dirichlet model of the random variable can be expressed by the shown dirichlet distribution, and the prior dirichlet probability density function is expressed by the following formula (1):
Figure BDA0002849030340000081
in the formula, Γ (×) is represented as a Gamma function; r ═ r (r)1,r2,...,rn)，riAs a prior parameter, satisfies r is more than or equal to 0iLess than or equal to 1 and
Figure BDA0002849030340000082
representing the parameter thetaiAverage under Dirichlet prior distribution, for convenience of expression, specifies betai＝sri，βiWeights or implicit observations representing a priori information, s being usually set at 1,2]And (4) the following steps.
After obtaining the sample observation value M, the prior dirichlet probability density function is updated by the bayesian process, and a posterior distribution can be obtained, as shown in the following formula (2):
Figure BDA0002849030340000083
wherein M is the total number of sample observations, MiIndicating the situation xi in M observationsiThe number of occurrences.
After obtaining the data observations, the parameter θ is giveniMean value r of prior probabilitiesiExpected according to the posterior distribution shown in the following formula (3)The value estimates the situation xiiThe posterior probability of occurrence, i.e.:
Figure BDA0002849030340000084
as can be seen from the derivation of the deterministic Dirichlet model shown in equation (3), it is necessary to perform statistical analysis on each riIs set, in other words, the probability of each occurrence of a random event is determined by the prior weight βiAnd (6) determining. However, the existing data information is limited, the samples are insufficient, and the r cannot be correctediGives reasonable guidance when r isiWhen the setting is not reasonable, the accuracy of probability estimation is affected.
In order to avoid the disadvantages of the deterministic dirichlet method, a parameter estimation method is adopted to obtain the minimum value and the maximum value of the parameter posterior expected value, as shown in the following formula (4):
Figure BDA0002849030340000085
the Cumulative Distribution Function (CDF) confidence band contains the true probability Distribution information of the random variable, but does not necessarily contain the estimated range [ ξ ] of the true value of the random variablel,ξu]. Notably, the purpose of fuzzy set design is to encode the information of the CDF confidence interval without assuming any a priori knowledge about the distribution type. As shown in fig. 2, a line (i) represents a lower boundary of a confidence interval of 200 samples, a line (ii) represents a lower boundary of a confidence interval of 2000 samples, a line (iii) represents a lower boundary of a confidence interval of 20000 samples, a line (iv) represents a real cumulative probability distribution, a line (v) represents an upper boundary of a confidence interval of 20000 samples, a line (c) represents a lower boundary of a confidence interval of 2000 samples, and a line (c) represents a lower boundary of a confidence interval of 200 samples. The fuzzy set becomes narrower and narrower by collecting more and more historical data according to the convergence of the confidence interval shown in fig. 2. Taking wind power as an example, when the random variable is the wind power output, the wind power output can be set according to the CDF confidence bandProbability point of (1-gamma)/2, (1+ gamma)/2]Finding out the corresponding real wind power output interval
Figure BDA0002849030340000091
FIG. 3 shows the process of converting a fuzzy set into an uncertain set, wherein the lines
Figure BDA0002849030340000092
Representing the lower boundary of the confidence interval, line
Figure BDA0002849030340000093
Lines representing the true cumulative probability distribution
Figure BDA0002849030340000094
Representing the upper boundary of the confidence interval. As can be seen from fig. 3, the fuzzy set constructed according to the non-exact dirichlet allocation model method is mapped to the boundary of the wind power uncertain set. According to the method, the wind power uncertain interval can be effectively constructed under the conditions that wind power historical data are limited and wind power does not need to be assumed to obey any specific distribution, so that the requirements on wind power historical statistical data are greatly reduced, the universality of the model and the method is improved, and the distributed robust optimization problem is effectively constructed.
Fuzzy sets, in contrast to uncertain sets in conventional robust optimization
Figure BDA0002849030340000095
Is a set of probability distributions (metrics). In the traditional robust optimization, the worst wind power output in an uncertain set is optimized, and the worst probability distribution in a fuzzy set is optimized in the distributed robust optimization.
Compared with the traditional method for giving the uncertain set boundary, the non-precise Dirichlet model method of the embodiment does not need to know the specific probability distribution of wind power and photoelectricity, and can construct the wind power photovoltaic output interval by using a small amount of information under the condition that the historical data of the wind power and the photoelectricity are limited.
S130, constructing a two-stage electric heating combined system distributed robust model. Step S130 specifically includes:
and constructing a two-stage objective function, wherein the first stage is to minimize the cost of the pre-dispatching stage, and the second stage is to minimize the cost of the readjusting stage.
And constructing the constraint condition of the two-stage objective function.
In the first stage, the day-ahead pre-scheduling cost mainly comprises the electricity and heat production cost of a Combined Heat and Power (CHP) unit, the electricity generation cost of a thermal power (CTP) unit, the spare capacity cost and the start/stop cost.
The objective function of the first stage is:
Figure BDA0002849030340000101
wherein variables are aggregated
Figure BDA0002849030340000102
Figure BDA0002849030340000103
And
Figure BDA0002849030340000104
is the supply price of the electric energy and the heat energy of the CHP unit c,
Figure BDA0002849030340000105
and
Figure BDA0002849030340000106
is the up/down reserve price of the CHP and CTP units,
Figure BDA0002849030340000107
is the electric energy quotation of the CTP unit g, Sc/BcAnd Sg/BgThe starting and stopping costs of the CHP unit c and the CTP unit g are respectively,
Figure BDA0002849030340000108
and
Figure BDA0002849030340000109
is the output power of the electric energy and the heat energy of the CHP unit c,
Figure BDA00028490303400001010
is the electric energy output power of the CTP unit g,
Figure BDA00028490303400001011
and
Figure BDA00028490303400001012
is the upper/lower spare consumption of the CHP/CTP unit, vc,t/zc,t/vg,t/zg,tIs a binary variable, e.g. when vc,tWhen the time t indicates that the CHP unit h starts, the starting cost S is 1hShould be calculated into the total cost, otherwise, the shutdown cost BhShould be calculated into the total cost, SgAnd BgIs the start-stop cost of CTP set, phiCHPAnd phiCTPIs a CHP and CTP unit set.
At this stage, since the power generation characteristics of different units in the cogeneration system are different, it is necessary to convert the outputs of the respective units into Direct Current (DC) in a unified manner in order to achieve power convergence in the scheduling stage. Therefore, a DC power flow model is adopted, as shown in the following formula (6), wherein BnmIs the admittance between transmission lines nm, deltan,tAnd deltam,tRespectively the phase angle between the busbars n and m.
Figure BDA00028490303400001013
And
Figure BDA00028490303400001014
is a wind power generation (WPP) set w and a photovoltaic power generation (PV) setpThe output power of (1).
Figure BDA00028490303400001015
Is the demanded rated power of the load l. Constraints (7) and (9) are the power generation capacity limits of the CHP unit and the CTP unit. It is worth mentioning that the CHP unit used by the system operates in the traditional "follow-up thermal load" mode, which is the feasible operating region of the scheduling level in timeThe domain is limited by (8).
Figure BDA00028490303400001016
And
Figure BDA00028490303400001017
is the thermoelectric conversion proportion of the cogeneration unit.
Figure BDA00028490303400001018
And
Figure BDA00028490303400001019
indicating an up/down reserve capacity. Constraints (10), (11), (12), (13) and (14) are ramp rate constraints for CHP and CTP.
Figure BDA00028490303400001020
And
Figure BDA00028490303400001021
is a ramp rate up/down ramp rate limit. u. ofc,tAnd ug,tIs a binary variable describing the on/off state. Minimum continuous run/shut down time to ensure economy and safety of the system
Figure BDA00028490303400001022
And
Figure BDA00028490303400001023
are set at the constraints (15), (16), (17) and (18).
That is, the constraint of the objective function of the first stage is:
Figure BDA00028490303400001024
Figure BDA0002849030340000111
Figure BDA0002849030340000112
Figure BDA0002849030340000113
Figure BDA0002849030340000114
Figure BDA0002849030340000115
Figure BDA0002849030340000116
Figure BDA0002849030340000117
Figure BDA0002849030340000118
Figure BDA0002849030340000119
Figure BDA00028490303400001110
Figure BDA00028490303400001111
Figure BDA00028490303400001112
and in the second stage, modeling is carried out on uncertainty in wind power output, solar photovoltaic output and load requirements by using the fuzzy set generated in the previous step, so as to obtain a distributed robust optimization model. Notably, cogeneration systems can represent a price recipient. Therefore, the present embodiment ignores uncertainty of market price. In this sense, the present embodiment minimizes the real-time readjustment cost of the cogeneration system, which mainly includes the adjustment cost of the CHP unit and the CTP unit and the cost of the grid imbalance.
The objective function for the second stage is:
Figure BDA00028490303400001113
wherein, the variable set
Figure BDA00028490303400001114
At any time, are uncertain parameters of the readjustment phase, which follow the fuzzy set established in the previous step. It should be noted that the heat demand is provided by district heating, and therefore, its uncertainty is much less than at iiVPPThe uncertainty parameter of (1). Therefore, we assume that the thermal load demand of the second stage is equal to the predicted value of the first stage. Variable set
Figure BDA0002849030340000121
Is the decision variable for the second stage.
Figure BDA0002849030340000122
And
Figure BDA0002849030340000123
the up/down adjustment amounts of the CHP unit and the CTP unit are respectively.
Figure BDA0002849030340000124
Meaning the imbalance in supply and demand after the provision of rotating reserve. Since the cogeneration system is a grid connection system, unbalanced electric energy can pass through the price
Figure BDA0002849030340000125
And
Figure BDA0002849030340000126
is the electricity price of the CHP unit and the CTP unit in the second adjusting stage.
The second stage problem is formulated based on the worst case and day-ahead planning of the first stage uncertain fuzzy sets. The constraint (20) is a power balance at real-time runtime, wherein,
Figure BDA0002849030340000127
and
Figure BDA0002849030340000128
is the phase angle between the busbars n and m of the second adjustment phase. And the constraints (21) to (24) are constrained by the scheduling result of the first stage, and each unit is limited and adjusted in real time.
That is, the constraint conditions of the objective function in the second stage are:
Figure BDA0002849030340000129
Figure BDA00028490303400001210
Figure BDA00028490303400001211
Figure BDA00028490303400001212
Figure BDA00028490303400001213
according to the embodiment, the conservative property of the robust optimization model can be controlled by adjusting the uncertainty parameter, and the applicability of the model and the method is improved.
And S140, solving the distributed robust model of the two-stage electric heating combined system. Step S140 specifically includes:
the second stage double layer problem was converted to a single layer problem using karuo-Kuhn-Tucker (KKT) conditions.
And circularly solving the two-stage problem by using a column constraint generation (C & CG) algorithm.
The two-stage model constructed in the previous step is represented in a compact matrix form and a concise notation, that is, the two-stage electric-heat combined system distributed robust model is represented by the following formula (25):
Figure BDA0002849030340000131
the proposed two-stage distributed robust optimization is usually solved using a cut plane based approach. Compared with Benders decomposition technique in the related art, C&The CG algorithm has better convergence. C&The core of the CG algorithm is to decompose the two-stage problem into a Main Problem (MP) and a sub-problem (SP), and then solve iteratively. In each iteration, the optimal solution for the SP is transferred to the MP containing new variables and corresponding constraints. MP is described by the following equation (26), the worst uncertainty pi found from the SP of the previous iteration*The overall cost is minimized.
Figure BDA0002849030340000132
Wherein the content of the first and second substances,
Figure BDA0002849030340000133
since the max-min problem is not easily solved directly by commercial solvers, it is possible to convert SPs to an equivalent single-stage problem. The KKT condition may be used here because the internal minimum problem for SP is linear.
Figure BDA0002849030340000134
Where λ is the dual parameter in equation (25) and j is the indentation of the corresponding constraint.
Notably, the nonlinear complementary constraint can be linearized with a large M method.
The model constructed in the embodiment is based on a thermoelectric system scheduling architecture, and follows a self-adaptive mechanism between cooperative regulation and robust optimization, and the obtained result can be directly used for solving the problem of optimal regulation and control of the current thermoelectric system.
It will be understood that the above embodiments are merely exemplary embodiments taken to illustrate the principles of the present invention, which is not limited thereto. It will be apparent to those skilled in the art that various modifications and improvements can be made without departing from the spirit and substance of the invention, and these modifications and improvements are also considered to be within the scope of the invention.

Claims (10)

1. A thermoelectric combined system optimization regulation and control method for improving wind and light absorption is characterized by comprising the following steps:
establishing a combined heat and power model comprising a wind power generation system, a photovoltaic power generation system and a combined heat and power generation system;
constructing a Dirichlet fuzzy set;
constructing a two-stage electric heating combined system distributed robust model;
and solving the distributed robust model of the two-stage electric heating combined system.
2. The method of claim 1, wherein establishing a cogeneration model comprising a wind power generation system, a photovoltaic power generation system, and a cogeneration system comprises:
processing uncertainty of wind speed by using Weibull distribution to construct a wind power generation model;
processing the uncertainty of illumination by utilizing beta distribution to construct a photovoltaic power generation model;
and establishing an electric heat output model of the cogeneration technology according to the electric heat coupling characteristic of the cogeneration technology.
3. The method of claim 1, wherein said constructing a Dirichlet fuzzy set comprises:
constructing a Dirichlet hybrid model according to the non-supervised classified Dirichlet process;
and constructing the Dirichlet fuzzy set by combining a variation inference principle.
4. The method of claim 3,
the constructing of the dirichlet hybrid model according to the unsupervised classified dirichlet process includes:
the prior dirichlet probability density function is:
Figure FDA0002849030330000011
in the formula, Γ (—) is represented by a Gamma function, and r ═ r (r)1,r2,...,rn)，riAs a prior parameter, satisfies r is more than or equal to 0iLess than or equal to 1 and
Figure FDA0002849030330000012
representing the parameter thetaiMean value, beta, under Dirichlet prior distributioni＝sri，βiWeights or implicit observations representing a priori information, s being set at [1,2 ]]Internal;
after obtaining the sample observation value M, the prior dirichlet probability density function is updated by the bayesian process to obtain a posterior distribution:
Figure FDA0002849030330000021
wherein M is the total number of sample observations, MiIndicating the situation xi in M observationsiThe number of occurrences;
after obtaining the data observations, the parameter θ is giveniMean value r of prior probabilitiesiThe situation ξ is estimated from the expected value of the posterior distribution shown in the following expression (3)iThe posterior probability of occurrence, i.e.:
Figure FDA0002849030330000022
the Dirichlet fuzzy set is constructed by combining a variation inference principle, and the construction method comprises the following steps:
obtaining the minimum value and the maximum value of the posterior expected value of the parameter by adopting a parameter estimation method:
Figure FDA0002849030330000023
5. the method of claim 1, wherein constructing the two-stage electric-thermal combined system distributed robust model comprises:
constructing a two-stage objective function, wherein the first stage is to minimize the cost of a pre-dispatching stage, and the second stage is to minimize the cost of a readjusting stage;
and constructing the constraint condition of the two-stage objective function.
6. The method of claim 5, wherein the objective function of the first stage is:
Figure FDA0002849030330000024
wherein variables are aggregated
Figure FDA0002849030330000025
Figure FDA0002849030330000026
And
Figure FDA0002849030330000027
is the supply price of the electric energy and the heat energy of the CHP unit c,
Figure FDA0002849030330000028
and
Figure FDA0002849030330000029
is the up/down reserve price of the CHP and CTP units,
Figure FDA00028490303300000210
is the electric energy quotation of the CTP unit g, Sc/BcAnd Sg/BgThe starting and stopping costs of the CHP unit c and the CTP unit g are respectively,
Figure FDA00028490303300000211
and
Figure FDA00028490303300000212
is the output power of the electric energy and the heat energy of the CHP unit c,
Figure FDA00028490303300000213
is the electric energy output power of the CTP unit g,
Figure FDA00028490303300000214
and
Figure FDA00028490303300000215
is the upper/lower spare consumption of the CHP/CTP unit, vc,t/zc,t/vg,t/zg,tIs a binary variable when v isc,tWhen the time t indicates that the CHP unit h starts, the starting cost S is 1hIs calculated into the total cost, otherwise, the shutdown cost BhIs calculated into the total cost, SgAnd BgIs the start-stop cost of CTP set, phiCHPAnd phiCTPIs a CHP unit and a CTP unit set;
the constraints of the objective function of the first stage are:
Figure FDA0002849030330000031
Figure FDA0002849030330000032
Figure FDA0002849030330000033
Figure FDA0002849030330000034
Figure FDA0002849030330000035
Figure FDA0002849030330000036
Figure FDA0002849030330000037
Figure FDA0002849030330000038
Figure FDA0002849030330000039
Figure FDA00028490303300000310
Figure FDA00028490303300000311
Figure FDA00028490303300000312
Figure FDA00028490303300000313
wherein, BnmIs the admittance between transmission lines nm, deltan,tAnd deltam,tRespectively the phase angle between the busbars n and m,
Figure FDA00028490303300000314
and
Figure FDA00028490303300000315
is the output power of the WPP unit w and the PV unit p,
Figure FDA00028490303300000316
is the demanded rated power of load l;
Figure FDA00028490303300000317
and
Figure FDA00028490303300000318
is the heat-electricity conversion proportion of the cogeneration unit,
Figure FDA00028490303300000319
and
Figure FDA00028490303300000320
represents an up/down reserve capacity;
Figure FDA00028490303300000321
and
Figure FDA00028490303300000322
is a ramp rate up/down ramp rate limit, uc,tAnd ug,tIs a binary variable;
Tc on/Tc offand
Figure FDA00028490303300000323
is the minimum continuous run/off time.
7. The method of claim 6, wherein the objective function of the second stage is:
Figure FDA0002849030330000041
wherein, the variable set
Figure FDA0002849030330000042
For any time, it is the uncertain parameter, variable set of readjustment stage
Figure FDA0002849030330000043
Is the decision variable of the second stage and,
Figure FDA0002849030330000044
and
Figure FDA0002849030330000045
the upward/downward adjustment amounts of the CHP unit and the CTP unit respectively,
Figure FDA0002849030330000046
the supply and demand after the rotating reserve is provided are unbalanced;
the constraints of the objective function of the second stage are:
Figure FDA0002849030330000047
Figure FDA0002849030330000048
Figure FDA0002849030330000049
Figure FDA00028490303300000410
Figure FDA00028490303300000411
wherein the constraint (20) is a power balance at real time operation,
Figure FDA00028490303300000412
and
Figure FDA00028490303300000413
and the phase angle between the buses n and m in the second adjustment stage, and the constraint (21) to (24) are constrained by the scheduling result of the first stage, so that each unit is limited and adjusted in real time.
8. The method of claim 7, wherein the solving the two-stage combined heat and power system distributed robust model comprises:
converting the second-stage double-layer problem into a single-layer problem by using a KKT condition;
and circularly solving the two-stage problem by using a C & CG algorithm.
9. The method of claim 8, wherein the two-stage combined heat and power system distributed robust model is:
Figure FDA0002849030330000051
10. the method of claim 9,
the cyclic solution of the two-stage problem by using the C & CG algorithm comprises the following steps:
decomposing the two-stage problem into a main problem MP and a sub problem SP, and then carrying out iterative solution;
in each iteration, the optimal solution of the SP is transferred into the MP containing new variables and corresponding constraints;
the MP is described by the following equation (26), the worst uncertainty pi found from the SP of the previous iteration*The total cost is minimized in the case:
Figure FDA0002849030330000052
wherein the content of the first and second substances,
Figure FDA0002849030330000053
the conversion of the second stage double layer problem to a single layer problem using the KKT condition includes:
converting the SP to an equivalent single layer problem:
Figure FDA0002849030330000054
where λ is the dual parameter in equation (25) and j is the indentation of the corresponding constraint.
CN202011519542.1A 2020-12-21 2020-12-21 Optimal regulation and control method for wind-solar absorption-improving thermoelectric combined system Pending CN113159980A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202011519542.1A CN113159980A (en) 2020-12-21 2020-12-21 Optimal regulation and control method for wind-solar absorption-improving thermoelectric combined system

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202011519542.1A CN113159980A (en) 2020-12-21 2020-12-21 Optimal regulation and control method for wind-solar absorption-improving thermoelectric combined system

Publications (1)

Publication Number Publication Date
CN113159980A true CN113159980A (en) 2021-07-23

Family

ID=76882642

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202011519542.1A Pending CN113159980A (en) 2020-12-21 2020-12-21 Optimal regulation and control method for wind-solar absorption-improving thermoelectric combined system

Country Status (1)

Country Link
CN (1) CN113159980A (en)

Similar Documents

Publication Publication Date Title
WO2022100091A1 (en) Centralized control method for scheduling of generalized source storage system
CN108123492B (en) A kind of generation schedule optimization method a few days ago considering thermoelectricity electric heat storage combined adjusting peak
CN106451566B (en) multi-source coordination control method for island intelligent microgrid
CN108229865A (en) A kind of electric heating gas integrated energy system low-carbon economy dispatching method based on carbon transaction
CN113850474B (en) Thermoelectric hydrogen multi-energy flow comprehensive energy system and optimal scheduling method thereof
CN112990523B (en) Hierarchical optimization operation method for regional comprehensive energy system
CN110601260B (en) Light-storage system capacity optimization method for limiting power fluctuation on interconnection line
CN113256045A (en) Park comprehensive energy system day-ahead economic dispatching method considering wind and light uncertainty
CN114169236A (en) Phase-change heat storage type electric heating negative control system control method based on LSTM algorithm
Zhao et al. Day-ahead robust optimal dispatch of integrated energy station considering battery exchange service
CN112836882A (en) Regional comprehensive energy system operation optimization method considering equipment load rate change
CN113078684B (en) Regional energy community planning method based on double-layer optimization
CN112365034B (en) Electric heating comprehensive energy system scheduling method and system
CN111969591B (en) Off-grid hybrid energy system optimization method based on complementation of optical energy storage hydropower station
CN117436773A (en) Independent micro-grid resource planning method and system containing interruptible load
CN110336308B (en) Opportunity constraint-based active power distribution network economic dispatching method
CN116805192A (en) Comprehensive energy system double-layer planning optimization method considering optimal energy rejection rate and application thereof
CN117081143A (en) Method for promoting coordination and optimization operation of park comprehensive energy system for distributed photovoltaic on-site digestion
CN115765034A (en) Photo-thermal-photovoltaic-thermal power combined cooperative control method and system
CN115566731A (en) Flexible power supply planning method based on system multi-energy complementation
CN113159980A (en) Optimal regulation and control method for wind-solar absorption-improving thermoelectric combined system
Chang et al. Research on low-carbon economic operation strategy of renewable energy-pumped storage combined system
CN115051406A (en) Multi-stage coordination control risk scheduling method considering source network uncertainty
CN114188942A (en) Power grid dispatching method comprising large-scale new energy base
CN115882460A (en) Two-stage new energy micro-grid optimization scheduling method considering demand side management

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