CN112668791A - Optimization method of combined heat and power system - Google Patents

Abstract

The invention discloses an optimization method of a combined heat and power system, which comprises the following steps: taking the hourly power output power p (i, t) of each unit and the hourly thermal power output power h (i, t) of the thermoelectric unit as optimization decision variables; on the basis of considering wind abandon punishment cost, demand response cost, user thermal comfort compensation cost and electric automobile battery depreciation cost and neglecting the power generation cost of a wind turbine generator, the most economical system power generation and heat supply operation cost is taken as a target function; combining the constraint conditions with the objective function to construct a scheduling model of the cogeneration system; and based on a Benders decomposition algorithm, and adopting an improved particle swarm optimization algorithm to carry out optimization solution on the scheduling model of the combined heat and power system. The flexible heat load considering the heat comfort of the user, the peak-valley time-of-use electricity price demand response and the electric automobile network-accessing ordered charging and discharging load are jointly used as the demand side resource to carry out the optimal scheduling of the combined heat and power system, so that the running economy of the combined heat and power system and the utilization rate of new energy wind power are effectively improved.

Description

Optimization method of combined heat and power system

Technical Field

The invention relates to the technical field of energy utilization, in particular to an optimization method of a combined heat and power system.

Background

Energy crisis and environmental pollution are always important concerns, electric vehicles and wind power are vigorously developed by the nation due to good environmental and economic benefits of the electric vehicles and the wind power, but the electric vehicles and the wind power with high permeability bring new challenges to power system dispatching. In the 'three north' area, the heating period is highly coincident with the wind power large-power generation period, and the power system is restricted by the operation mode of the heating and power cogeneration unit, namely 'fixing the power with heat', so that the peak regulation capacity of the unit is insufficient, a large amount of wind is abandoned, and a large amount of coal is needed for supplying heat, and the environment pollution is caused.

In order to solve the problem of wind abandonment of the cogeneration system, a great deal of research is carried out at home and abroad. The heat storage device is additionally arranged on the power generation side to decouple the rigid coupling relation between heat and electricity so as to enhance the peak regulation capacity of the system and further improve the wind power consumption level. At present, a scheme of increasing the electric load on a load side to absorb redundant wind power by additionally arranging an electric gas conversion device in a system is proposed in documents; the flexibility of the cogeneration unit is improved by coordinating the operation of the electric boiler and the pumped storage or heat storage device, so that more wind power is consumed; the scheme of decoupling the electric-heat coupling operation constraint of the unit by utilizing the heat supply network and the building heat storage existing in the existing system to promote the wind power consumption; based on a peak regulation capacity improvement scheme of decoupling a heat supply network and building heat storage, and taking Jilin province as an example, the potential of the proposed scheme for promoting wind power consumption is estimated; the method provides a mechanism for explaining night electric heating price calculation from a supply side and a demand side and environmental benefits thereof, forms an upper electricity price limit calculation model of the demand side, also provides a combined power generation optimization scheduling model of a heat storage CHP (Heat storage and Power Generator) set and wind power, considering demand response and environmental protection cost, and verifies that the model can effectively promote wind power consumption and reduce pollutant emission through a calculation example.

In the aspect of optimal scheduling of an electric vehicle and a wind power plant, a hierarchical control strategy is proposed in documents to reasonably arrange charging and discharging of the electric vehicle to smooth the output of wind power, and a self-scheduling method for purchasing energy in the day-ahead market of a plug-in electric vehicle aggregator is proposed to provide balance service for a wind power generator.

For the establishment of a thermoelectric system coordinated operation model and the solution of an optimized scheduling problem, a thermoelectric coordinated optimal scheduling model considering the energy storage characteristic of a regional heating network is proposed in documents, and a Benders decomposition algorithm is adopted to solve the model; the gravity search algorithm is used for solving the problem of economic scheduling of cogeneration, and the effectiveness of the proposed algorithm is tested; the method comprehensively considers the abandoned wind cost, the scheduling cost, the energy storage loss cost and the environmental pollution cost, constructs a scheduling model of a regional comprehensive energy system, provides an improved chaos particle swarm optimization algorithm based on particle dimension entropy to solve the model, provides a two-stage scheduling method including heat storage and cogeneration units and demand response resources in the day before wind power consumption, and uses an improved ICA algorithm to solve the scheduling model.

Based on the above analysis and research, most of the current researches for promoting the wind power absorption in the power grid during the heating period are based on configuring energy storage or increasing load side electric load facilities to improve the flexibility of the system, however, the investment and the operation cost of the current energy storage system are higher, and the economical efficiency of the wind power network access is influenced by configuring the energy storage system on a large scale. Therefore, a plurality of students also consider integrating demand side resources into system scheduling to match with wind power internet surfing, and the running economy of the system is improved. However, most of the existing researches are based on matching of a certain single demand side resource with wind power on-line, and the economic efficiency of system operation is improved by considering a plurality of abundant resources on the demand side overall, so that the utilization efficiency of the resources is influenced.

Disclosure of Invention

An object of the present invention is to solve at least the above problems and to provide at least the advantages described later.

The invention also aims to provide an optimization method of the combined heat and power system, which takes the flexible heat load considering the heat comfort of the user, the peak-valley time-of-use electricity price demand response and the electric vehicle network access ordered charging and discharging load as the demand side resource to carry out the optimized scheduling of the combined heat and power system, thereby effectively improving the operating economy of the combined heat and power system and the utilization rate of new energy wind power.

To achieve these objects and other advantages in accordance with the purpose of the invention, there is provided an optimization method of a cogeneration system, comprising the steps of:

s1, taking the hourly power output power p (i, t) of each unit and the hourly thermal power output power h (i, t) of the thermoelectric unit as optimization decision variables;

s2, on the basis of considering wind abandoning punishment cost, demand response cost, user thermal comfort compensation cost and electric automobile battery depreciation cost and neglecting the generating cost of the wind turbine generator, taking the most economical system generating and heating operation cost as an objective function, wherein the objective function is shown as a formula 1:

wherein the content of the first and second substances,

represents the cost of the electricity generated by the conventional unit,

represents the power generation and heat supply cost, rho, of the CHP unit^WΔp_W(i, t) represents the penalty cost of wind power abandonment, C_DR(t) represents the demand response scheduling cost, C_ΔHRepresents the user heating comfort disturbance compensation cost, U_evRepresenting EV battery depreciation cost; t is the total time period number of 1 scheduling period; n is a radical of_gThe number of conventional units; n is a radical of_chpThe number of CHP units; c_DRScheduling costs for demand response per time period; rho^wPunishing price for wind abandon;

s3, combining the constraint conditions with the objective function to construct a scheduling model of the cogeneration system;

and S4, carrying out optimization solution on the scheduling model of the cogeneration system by adopting an improved particle swarm optimization algorithm based on the Benders decomposition algorithm.

Preferably, in the optimization method of the cogeneration system, the wind power abandonment penalty cost ρ is^WΔp_WΔ p in (i, t)_W(i, t) represents the predicted output of wind power at the same momentAnd using the difference between the powers and expressing it by equation 2:

the power utilization method comprises the following steps that pout w (i, t) is predicted output of wind power at the moment t, and pin w (i, t) is utilization power of the wind power at the moment t;

demand response scheduling cost C_DR(t) is represented by the formula 3

Wherein the content of the first and second substances,

and P0L, t is the electricity rate and consumption before the implementation of demand response at time t, s_tAnd P_L,tElectricity rates and electricity amounts after the demand response is implemented for the period t, respectively.

Preferably, in the optimization method of the cogeneration system, the constraint condition includes: system constraint, unit constraint, wind power constraint, demand response constraint, user heating temperature comfort constraint and electric vehicle constraint;

wherein the wind power constraint is represented by formula 4; the user heating temperature comfort constraint is represented by equation 5:

wherein, theta^y(t) is the building interior temperature at time t,

and

the temperature upper and lower limits for the thermal comfort of the user are met.

Preferably, in the optimization method of the cogeneration system, the system constraints include an electric power balance constraint and a heating balance constraint;

additionally, the electric power balance constraints, in turn, include an unspoiled demand response constraint represented by equation 6 and a participated demand response constraint represented by equation 7:

wherein P0L (t) and P_L(t) respectively representing the electrical loads before and after the demand response at the time t; and respectively representing the electric loads of the conventional unit and the CHP unit at the time t; represents the electric load of the EV battery at the time t; n is a radical of_wThe number of wind power plants;

the heating balance constraint is represented by equation 8:

wherein H_load(t) refers to the thermal load of the area and is equal to the sum of the thermal loads of all buildings in the area; h is_chp(i, t) refers to the heat load of the ith CHP unit at the moment t; thermal load h of individual buildings_load(t) is represented by equation 9:

wherein h is_req(t) is the thermal demand of the user at time t; theta_in(t) and θ_in(t-1) indoor temperatures at the time of t and t-1, respectively; c. C_airIs the specific heat capacity of the indoor air; rho_airIs the density of the indoor air; v is the peripheral volume of the building.

Preferably, in the method for optimizing the cogeneration system, the unit constraints include conventional unit constraints and cogeneration unit constraints;

in addition, the conventional unit constraints include upper and lower output limits constraints represented by equation 10, and unit ramp constraints represented by equation 11:

wherein the content of the first and second substances,

and

respectively the minimum and maximum generating power of a conventional unit i;

and

respectively limiting the downward slope climbing limit value and the upward slope climbing limit value of the conventional unit i;

the cogeneration unit constraints in turn include an active power output upper and lower limit constraint represented by equation 12, a unit thermal output constraint represented by equation 13, a unit electrical creep constraint represented by equation 14, and a unit thermal creep constraint represented by equation 15:

max(P_i,min-c_vh_chp(i,t),c_mh_chp(i,t))≤p_chp(i,t)≤P_i,max-c_vh_chp(i, t) equation 12;

0≤h_chp(i,t)≤H_i,maxequation 13;

wherein, P_i,minAnd P_i,maxRespectively the minimum and maximum electric output of the CHP unit i under the pure condensing working condition; c. C_vThe method is characterized in that the reduction amount of the power generation power is generated by the multiple extraction unit heat supply when the total steam inlet amount of the CHP unit is not changed; c. C_mThe elastic coefficients of electric power and thermal power when the CHP unit operates in a backpressure mode; h_i,maxThe maximum thermal power of the CHP unit i;

and

the maximum downward and upward electrical power ramp rates for the CHP unit i.

Preferably, in the optimization method of the cogeneration system, the demand response constraint further includes a power comfort χ represented by formula 16_tConstraints, and the benefit of the demand response interaction represented by equation 17

And (3) constraint:

wherein the content of the first and second substances,

and

respectively the minimum value of the electricity utilization mode comfort level and the interactive benefit comfort level of the user under the demand response; Δ q of_tAnd Δ c_tThe relative increase in demand for electricity in the demand response and the relative increase in electricity rate in the demand response, respectively.

Preferably, in the optimization method of the cogeneration system, the electric vehicle constraints further include EV maximum on-line discharging and charging power limits represented by equations 18 and 19, EV battery maximum SOC limit represented by equation 20, and EV vehicle owner charging demand constraint represented by equation 21:

soc_i≤soc_maxequation 20;

P_ev＝P_c+P_dformula 21;

wherein, soc_maxA maximum state of charge set to prevent overcharging of the battery; p_evThe total unordered charging load before the EV participates in scheduling is obtained; p_cCharging the total network access load after the EV participates in scheduling; p_dAnd discharging the total capacity of the network access after the EV participates in the scheduling.

Preferably, in the optimization method of the cogeneration system, in S4, the optimization solution of the cogeneration system scheduling model by using an improved particle swarm optimization algorithm based on the Benders decomposition algorithm specifically includes: and dividing the optimization solving problem of the scheduling model of the cogeneration system into a power system sub-problem and a thermal system main problem based on a Benders decomposition algorithm, optimizing the decomposed systems by adopting an improved particle swarm optimization algorithm, interacting the systems with optimal output of the power system sub-problem and the thermal system main problem, and feeding back the generated Benders cut constraint condition to the power system sub-problem for iteration until the requirement of inspection is met.

Preferably, in the optimization method of the cogeneration system, the optimization solution of the scheduling model of the cogeneration system is performed by adopting an improved particle swarm optimization algorithm based on a Benders decomposition algorithm, and the optimization solution specifically comprises the following steps:

s4-1, initialization: setting the iteration number k of the Benders decomposition method to be 1 and the initial thermodynamic system variable h for each scheduling time t₁Taking the minimum value of the thermal output of the unit, bringing the minimum value into the power system to carry out first solution to obtain

And the value p of this sub-problem decision variable¹(ii) a Initializing an objective function value derived from a thermodynamic system

Is- ∞;

s4-2, the Lagrangian multiplier for the Benders cut feedback correction constraint is generated and is expressed by the formula 22:

wherein the content of the first and second substances,

a Lagrange multiplier in the feedback correction constraint condition for the main problem of the thermodynamic system returned in the kth iteration;

s4-3, solving a main problem of the thermodynamic system: will be obtained from S4-2

And p obtained by solving sub-problem of power system^kThe lower boundary of the objective function is obtained by adopting an improved particle swarm algorithm to solve

And h^k+1；

S4-4, solving the power system subproblem: the optimal heat output h obtained by the k +1 th thermodynamic system^k+1Bringing the sub-problem of the power system into the known quantity, and solving the sub-problem of the power system by adopting an improved particle swarm optimization algorithm to obtain an upper boundary of an objective function

And p^k+1；

S4-5, feasibility test of the solution obtained by the main problem and the subproblems: the obtained lower and upper boundaries are substituted into the test condition expressed by equation 23 to be tested, and if the test condition is satisfied, the iteration is ended, and

i.e. the optimal solution of the running cost of the original electric heating combined system, p^k+1And h^k+1Namely the optimal electric output and the optimal thermal output of the system; if the test condition is not satisfied, the iteration number k is added by 1 to the original value, and S4-2-S4-5 is repeated.

The invention at least comprises the following beneficial effects:

the optimization method of the combined heat and power system takes flexible heat load, peak-valley time-of-use electricity price demand response and electric vehicle networking ordered charging and discharging load of the heat comfort level of a user into consideration as demand side resources to carry out optimized scheduling of the combined heat and power system, determines the electric heating output condition of each unit in the next scheduling period by solving an optimized scheduling model, arranges the energy of interaction between the electric vehicle and a power grid and the heat value required to be stored or discharged by the building by comprehensively analyzing the heat storage and release capacity of the building in each period, the large-scale adjustable amount of the charging and discharging load of the electric vehicle, the peak-adjustable capacity of the thermoelectric unit and the wind abandoning condition of wind power, follows the heat load under the condition of meeting the heat comfort level of the user in the economic scheduling, and effectively improves the flexibility and the economy of the operation of the power system.

Under the background of large-scale development of wind power and electric vehicles, in order to improve the operation flexibility and economy of a power system, analyze the abundance resources of a demand side, respectively perform peak-valley time-of-use electricity price demand response, and model the flexible heat load and the ordered charging load of the electric vehicle, which take the heat comfort of users into consideration, a scheduling framework is provided, the abundance resources of the demand side are integrated into a cogeneration system to perform unified coordination and optimization of supply and demand resources, and an economic scheduling model of the cogeneration system is established by comprehensively taking the constraint conditions of the cogeneration system into consideration, so that the operation economy of the system and the utilization rate of new energy wind power are improved.

Additional advantages, objects, and features of the invention will be set forth in part in the description which follows and in part will become apparent to those having ordinary skill in the art upon examination of the following or may be learned from practice of the invention.

Drawings

Fig. 1 is a frame structure diagram of an optimization method of a cogeneration system according to the present invention;

FIG. 2 is a diagram illustrating a coupling relationship between the thermoelectric power output of the pumping CHP unit;

FIG. 3 is a flow chart of an optimization solution for the cogeneration system scheduling model;

FIG. 4 is a flowchart of the computation of Monte Carlo simulated EV loads;

FIG. 5 is a load diagram of the electric vehicle in a disordered charging mode;

FIG. 6 is a graph of the load divided over peak-to-valley periods;

FIG. 7 is a load graph before and after peak-to-valley time-of-use electricity price response;

FIG. 8 is a day-ahead wind power prediction force diagram;

fig. 9 is a diagram of the wind power internet access situation under different scenes and at different time intervals;

FIG. 10 is a CHP unit power diagram under each scenario;

FIG. 11 is a CHP unit thermodynamic diagram under each scenario;

FIG. 12 is a conventional machine set electrical force diagram under various scenarios;

fig. 13 is a graph showing the relationship between the heat storage capacity inside the building and the indoor temperature in the situation 3;

FIG. 14 is a diagram of interaction between the electric output of each unit and the EV and the power grid energy under the situation 4;

FIG. 15 is an interaction diagram of EV and grid power under scenario 4;

fig. 16 is a graph showing the relationship between the heat storage capacity of the interior of the building and the indoor temperature in the case of the scenario 4.

Detailed Description

The present invention is further described in detail below with reference to the attached drawings so that those skilled in the art can implement the invention by referring to the description text.

It will be understood that terms such as "having," "including," and "comprising," as used herein, do not preclude the presence or addition of one or more other elements or groups thereof.

As shown in fig. 1, the present invention provides a method for optimizing a cogeneration system, comprising the steps of:

s1, taking the hourly power output power p (i, t) of each unit and the hourly thermal power output power h (i, t) of the thermoelectric unit as optimization decision variables;

s2, on the basis of considering wind abandoning punishment cost, demand response cost, user thermal comfort compensation cost and electric automobile battery depreciation cost and neglecting the generating cost of the wind turbine generator, taking the most economical system generating and heating operation cost as an objective function, wherein the objective function is shown as a formula 1:

wherein the content of the first and second substances,

represents the cost of the electricity generated by the conventional unit,

represents the power generation and heat supply cost, rho, of the CHP unit^WΔp_W(i, t) represents the penalty cost of wind power abandonment, C_DR(t) represents the demand response scheduling cost, C_ΔHRepresents the user heating comfort disturbance compensation cost, U_evRepresenting EV battery depreciation cost; t is the total time period number of 1 scheduling period; n is a radical of_gThe number of conventional units; n is a radical of_chpThe number of CHP units; c_DRScheduling costs for demand response per time period; rho^wPunishing price for wind abandon;

s3, combining the constraint conditions with the objective function to construct a scheduling model of the cogeneration system;

and S4, carrying out optimization solution on the scheduling model of the cogeneration system by adopting an improved particle swarm optimization algorithm based on the Benders decomposition algorithm.

In the above scheme, a combined heat and power cogeneration unit (CHP) is a comprehensive production mode that produces both electric energy and heat energy in a heat supply period according to the principle of energy cascade utilization. Most of heat supply units in China are large air extraction type cogeneration units, and the scheme is to analyze the large air extraction type cogeneration units as heat sources.

The coupling relationship between the electrical output power and the thermal output power of the CHP unit is called as the 'electric heating characteristic'. The characteristic of the thermoelectric operation relationship is shown in fig. 2, in order to meet the heating requirement, the CHP unit usually operates in a "heating and fixing power" mode, which limits the minimum thermal output of the cogeneration unit, and accordingly, the electric output adjustment range of the CHP unit is limited due to the thermoelectric coupling characteristic of the CHP unit. In the figure H_fThe minimum forced heat output of the CHP unit during heating is realized, and the adjustment range of the electrical output of the CHP unit is reduced to P_EAnd P_FMeanwhile, the unit operation area is limited in the area surrounded by the EBFE, and the unit operation flexibility is limited. The reduction of the electric power output adjusting range leads to the reduction of the wind power consumption capacity of the system in the heating season. The method comprises the steps of performing optimal scheduling on a combined heat and power system by taking flexible heat load considering heat comfort of users, peak-valley time-of-use electricity price demand response and electric vehicle networking ordered charging and discharging load as demand side resources, determining the electric heat output condition of each unit in the next scheduling period by solving an optimal scheduling model, and arranging interaction between an electric vehicle and a power grid by comprehensively analyzing the heat storage and release capacity of a building in each time period, the adjustable amount of the charging and discharging load of the large-scale electric vehicle, the adjustable peak capacity of the thermoelectric unit and the wind curtailment condition of wind powerThe energy and the heat value required to be stored or discharged by the building and the heat load are followed under the condition that the heat comfort degree of the user is met in the economic dispatching, so that the flexibility and the economy of the operation of the power system are effectively improved.

The optimization method is a day-ahead scheduling scheme, the scheduling time interval can be set to be 1h, and the scheduling plan is updated once a day. The method comprises the steps of determining the electric heat output condition of each unit in the next dispatching cycle by solving an optimized dispatching model, comprehensively analyzing the heat storage and release capacity of a building in each time interval, the large-scale electric vehicle charging and discharging load adjustable quantity, the peak adjustable capacity of a thermoelectric unit and the wind abandoning condition of wind power, arranging the energy of interaction between an electric vehicle and a power grid and the heat value required to be stored or released by the building, following the heat load under the condition that the economic dispatching meets the heat utilization comfort level of users, preferentially consuming the wind power through the EV ordered charging load under the condition that the charging requirements of the users are met, and considering the current battery charging and discharging efficiency and the battery depreciation cost to realize the virtual energy storage effect by considering the V2G technology of the EV when the wind abandoning still exists. And preferentially meeting the charging load requirements of the electric automobiles of the users on the basis of day-ahead scheduling optimization, sequentially controlling the charging load requirements to match with the consumption of wind power, and scheduling the energy storage system formed by the electric automobiles in a stopped state in the area, wherein the electric automobiles are taken as the resources on the demand side, namely the household automobiles adopting slow charging, the electric automobiles are assumed to be charged every day, the EVs in the area interact with the power grid through an EV centralized controller (Aggregator), the EV centralized controller collects the driving data information of the electric automobiles in the area under jurisdiction, the charging load of the electric automobiles in the area under jurisdiction is predicted, and the adjustable range of the charging load is evaluated.

In a preferred scheme, wind power abandonment penalty cost rho^WΔp_WΔ p in (i, t)_W(i, t) represents the difference between the predicted output of the wind power and the utilization power at the same moment, and is represented by formula 2:

wherein poutw (i, t) is the predicted output of the wind power at the time t, and pin w (i, t) is the utilization power of the wind power at the time t;

demand response scheduling cost C_DR(t) is represented by the formula 3

Wherein the content of the first and second substances,

and P0L, t is the electricity rate and consumption before the implementation of demand response at time t, s_tAnd P_L,tElectricity rates and electricity amounts after the demand response is implemented for the period t, respectively.

In a preferred embodiment, the constraint condition includes: system constraint, unit constraint, wind power constraint, demand response constraint, user heating temperature comfort constraint and electric vehicle constraint;

wherein the wind power constraint is represented by formula 4; the user heating temperature comfort constraint is represented by equation 5:

wherein, theta^y(t) is the building interior temperature at time t,

and

the temperature upper and lower limits for the thermal comfort of the user are met.

In a preferred embodiment, the system constraints include an electrical power balance constraint and a heating balance constraint;

additionally, the electric power balance constraints, in turn, include an unspoiled demand response constraint represented by equation 6 and a participated demand response constraint represented by equation 7:

wherein P0L (t) and P_L(t) respectively representing the electrical loads before and after the demand response at the time t; p is a radical of_g(i, t) and p_chp(i, t) respectively representing the electric loads of the conventional unit and the CHP unit at the moment t; p is a radical of_ev(t) represents the electric load of the EV battery at time t; n is a radical of_wThe number of wind power plants;

the heating balance constraint is represented by equation 8:

wherein H_load(t) refers to the thermal load of the area and is equal to the sum of the thermal loads of all buildings in the area; h is_chp(i, t) refers to the heat load of the ith CHP unit at the moment t; thermal load h of individual buildings_load(t) is represented by equation 9:

wherein h is_req(t) is the thermal demand of the user at time t; theta_in(t) and θ_in(t-1) indoor temperatures at the time of t and t-1, respectively; c. C_airIs the specific heat capacity of the indoor air; rho_airIs the density of the indoor air; v is the peripheral volume of the building.

In a preferred scheme, the unit constraint comprises a conventional unit constraint and a cogeneration unit constraint;

in addition, the conventional unit constraints include upper and lower output limits constraints represented by equation 10, and unit ramp constraints represented by equation 11:

wherein the content of the first and second substances,

and

respectively the minimum and maximum generating power of a conventional unit i;

and

respectively limiting the downward slope climbing limit value and the upward slope climbing limit value of the conventional unit i;

the cogeneration unit constraints in turn include an active power output upper and lower limit constraint represented by equation 12, a unit thermal output constraint represented by equation 13, a unit electrical creep constraint represented by equation 14, and a unit thermal creep constraint represented by equation 15:

max(P_i,min-c_vh_chp(i,t),c_mh_chp(i,t))≤p_chp(i,t)≤P_i,max-c_vh_chp(i, t) equation 12;

0≤h_chp(i,t)≤H_i,maxequation 13;

wherein, P_i,minAnd P_i,maxRespectively the minimum and maximum electric output of the CHP unit i under the pure condensing working condition; c. C_vThe method is characterized in that the reduction amount of the power generation power is generated by the multiple extraction unit heat supply when the total steam inlet amount of the CHP unit is not changed; c. C_mThe elastic coefficients of electric power and thermal power when the CHP unit operates in a backpressure mode; h_i,maxThe maximum thermal power of the CHP unit i;

and

the maximum downward and upward electrical power ramp rates for the CHP unit i.

In the above scheme, in equation 15, when the CHP unit operates in the region C left of the point C shown in fig. 2_xValue of c_vWhen the CHP unit is operated in the area C to the right of the point C shown in FIG. 2_xValue of c_m。

In a preferred embodiment, the demand response constraint further includes power comfort χ represented by equation 16_tConstraints, and the benefit of the demand response interaction represented by equation 17

And (3) constraint:

wherein the content of the first and second substances,

and

respectively the minimum value of the electricity utilization mode comfort level and the interactive benefit comfort level of the user under the demand response; Δ q of_tAnd Δ c_tThe relative increase in demand for electricity in the demand response and the relative increase in electricity rate in the demand response, respectively.

In the scheme, after the peak-valley time-based electricity price is implemented, the electricity consumers can change electricity utilization habits in order to pursue economic benefits, so that the benefits of the users are damaged, and the comfort level of the electricity consumers is introduced to quantify, so that the electricity utilization comfort level of the users is ensured on the basis of carrying out economic optimization on the combined heat and power system.

In a preferred embodiment, the electric vehicle constraints further include EV maximum network access discharge and charging power limits represented by equations 18 and 19, EV battery maximum SOC limit represented by equation 20, and EV vehicle owner charging demand constraint represented by equation 21:

soc_i≤soc_maxequation 20;

P_ev＝P_c+P_dformula 21;

wherein, soc_maxA maximum state of charge set to prevent overcharging of the battery; p_evThe total unordered charging load before the EV participates in scheduling is obtained; p_cCharging the total network access load after the EV participates in scheduling; p_dAnd discharging the total capacity of the network access after the EV participates in the scheduling.

As shown in fig. 3, in a preferred embodiment, in S4, based on the Benders decomposition algorithm, and using an improved particle swarm optimization algorithm to perform optimization solution on the scheduling model of the cogeneration system, specifically, the optimization solution is performed by: and dividing the optimization solving problem of the scheduling model of the cogeneration system into a power system sub-problem and a thermal system main problem based on a Benders decomposition algorithm, optimizing the decomposed systems by adopting an improved particle swarm optimization algorithm, interacting the systems with optimal output of the power system sub-problem and the thermal system main problem, and feeding back the generated Benders cut constraint condition to the power system sub-problem for iteration until the requirement of inspection is met.

In the scheme, the established scheduling model of the cogeneration system is considered to be a large-scale nonlinear programming model, and the traditional optimization algorithm is difficult to solve effectively, so that the idea of the Benders decomposition algorithm is proposed to divide the whole optimization scheduling problem into a power system sub-problem and a thermodynamic system main problem, the decomposed systems are optimized by adopting an improved particle swarm optimization algorithm, the optimal output of the two systems of the interactive system and the generation of the Benders cut constraint feedback sub-problem are iterated until the requirements of inspection are met, and the optimization solution of the scheduling model is further facilitated.

In a preferred scheme, the optimization solution is carried out on the scheduling model of the cogeneration system by adopting an improved particle swarm optimization algorithm based on a Benders decomposition algorithm, and the optimization solution specifically comprises the following steps:

s4-1, initialization: setting the iteration number k of the Benders decomposition method to be 1 and the initial thermodynamic system variable h for each scheduling time t₁Taking the minimum value of the thermal output of the unit, bringing the minimum value into the power system to carry out first solution to obtain

And the value p of this sub-problem decision variable¹(ii) a Initializing an objective function value derived from a thermodynamic system

Is- ∞;

s4-2, the Lagrangian multiplier for the Benders cut feedback correction constraint is generated and is expressed by the formula 22:

wherein the content of the first and second substances,

a Lagrange multiplier in the feedback correction constraint condition for the main problem of the thermodynamic system returned in the kth iteration;

s4-3, solving a main problem of the thermodynamic system: will be obtained from S4-2

And p obtained by solving sub-problem of power system^kThe lower boundary of the objective function is obtained by adopting an improved particle swarm algorithm to solve

And h^k+1；

S4-4, solving the power system subproblem: the optimal heat output h obtained by the k +1 th thermodynamic system^k+1Bringing the sub-problem of the power system into the known quantity, and solving the sub-problem of the power system by adopting an improved particle swarm optimization algorithm to obtain an upper boundary of an objective function

And p^k+1；

S4-5, feasibility test of the solution obtained by the main problem and the subproblems: the obtained lower and upper boundaries are substituted into the test condition expressed by equation 23 to be tested, and if the test condition is satisfied, the iteration is ended, and

i.e. the optimal solution of the running cost of the original electric heating combined system, p^k+1And h^k+1Namely the optimal electric output and the optimal thermal output of the system; if the test condition is not satisfied, the iteration number k is added by 1 to the original value, and S4-2-S4-5 is repeated.

In the above scheme, the Particle Swarm Optimization (PSO) is a swarm intelligence based optimization that considers the solution of each optimization problem as a bird in the search space without mass and volume and extends it to the N-dimensional space. The position and the flight speed of the particle i in the N-dimensional space are each represented as a vector, and all particles have an adaptive value determined by the optimized function.

The algorithm principle is as follows:

updating of particle velocity and position:

v_i,j(t+1)＝wv_i,j(t)+c₁r₁[p_i,j-x_i,j(t)]+c₂r₂[p_g,j-x_i,j(t)]equation 23;

x_i,j(t+1)＝x_i,j(t)+v_i,j(t +1), j ═ 1,2, …, d equation 24;

wherein v is_i,j(t) and v_i,j(t +1) is the current speed of the particle i and the speed of the next stage, respectively; x is the number of_i,j(t) and x_i,j(t +1) is the current position of the particle i and the position of the next stage, respectively; w is the inertial weight; c. C₁And c₂A positive learning factor; r is₁And r₂Is a random number uniformly distributed between 0 and 1.

The improved particle swarm optimization algorithm can effectively jump out a local optimum point when solving the high-dimensional complex problem through the following improvement, the convergence speed and the convergence precision are improved, and the defects of premature convergence and poor convergence of the basic particle swarm optimization algorithm are avoided. The method specifically comprises the following steps:

1. introduction of dynamic inertial weight of inertia

The large inertia factor is beneficial to jumping out of a local minimum point, and the small inertia weight can carry out accurate local search on the current search area. Linear dynamic inertia weight coefficients are employed herein to balance the global search and local refinement capabilities of the PSO algorithm. The formula is as follows:

wherein, w_max、w_minRespectively representing the maximum value and the minimum value of w, m representing the current iteration step number, m_maxThe maximum number of iteration steps is indicated.

2. Trigonometric function type learning factor c₁And c₂Improvement of

To enhance the global search capability of the early particles and the local search capability of the later particles, c is₁And c₂Improvements in or relating to the formula

Where m denotes the current number of iterations, m_maxThe maximum iteration number is expressed, and m is less than or equal to 0.46m in the early stage of search through the formula 26_maxWhen c is greater than₁≥c₂The optimal pbest of the particles can be learnt in multiple directions and individual, and the optimal pbest of the particles can be learnt in less directions; and in the later period, m is more than or equal to 0.46m_maxWhen c is greater than₁≤c₂The particles can be locally drawn to the socially optimal position gbest, and c is preferably selected to make the algorithm convergence better₁+c₂≥4。

3. Introduction of reverse learning mechanism

The reverse learning is to consider the solution of the current position and the solution of the opposite position at the same time, so that the algorithm can be quickly close to a better search area and approach to a global optimal solution more quickly, and the convergence speed is further improved. When the number of times of the particle optimal solution unchanged reaches n times, the reverse learning is carried out for 1 time, so that the overall operation speed of the algorithm is greatly improved compared with that of the reverse learning carried out once for obtaining pbest. For a point in D-dimensional space

The opposite point is defined as:

when the number of times of the unchanged optimal particle solution reaches n times, performing generalized learning on the current population optimal individual according to the following formula:

if particle

If the position of (2) is out of range, the particle position is updated according to the following formula:

wherein the content of the first and second substances,

is a new solution generated by reverse learning,

and carrying out preference on the fitness value of the current optimal gbest.

In conclusion, the improved particle swarm optimization algorithm effectively overcomes the defects of premature convergence and poor convergence of the particle swarm optimization algorithm by introducing dynamic inertia weight and trigonometric function type learning factors, and meanwhile, the particles effectively jump out of local optimal points by introducing a reverse learning mechanism.

Experimental data

In order to verify the optimization effect of the scheduling model established by the application in the combined heat and power system, the installed capacity of the power supply of a certain area is simplified according to the actual power supply structure proportion of the power supply of the three north areas in China.

The region has 8 generator sets (#1, #2, #3, #4, #5, #6, #7, #8), wherein the No. 1-6 generator set is a large extraction type CHP generator set, the No. 1-3 and No. 4-6 generator sets belong to thermal power plants A and B, 2 thermal power plants respectively supply heat to areas I and II, the No. 7 and 8 generator sets are large straight condensing type conventional generator sets, and the system has 1 650MW wind power plant. Unit parameters in a regionAs shown in table 1. Total building peripheral volume of region I, II is 4.96 x 10⁸m³The building related parameters are shown in table 2, and the total heat demand in zones I, II remains 1800MW when the typical day user side temperature is maintained at 20 ℃, with the heat sensing average scale prediction index being between +1 ℃ and-1 ℃ slightly cooler and warmer. The area has 20000 electric vehicles, EV loads are predicted by adopting a Monte Carlo simulation method according to the operation rule of the electric vehicles and the charge-discharge characteristics of batteries, EV operation parameter probability density functions and parameter values are shown in tables 3 and 4, and Uab is 141$/(MW · h).

TABLE 1 Unit parameters

Note: alpha is alpha_i、β_i、γ_iAll units of (a) are t.MW^-2·h^-1。

TABLE 2 building parameters

Building parameters	cair(kJ·kg-1℃-1)	ρair(kg·m-3)	ph($/(MWh))
				Numerical value	1.007	1.2	20

TABLE 3 EV operating parameter probability Density function

TABLE 4 EV parameters

μin	σin	μout	σout	μd	σd	socmax	de/(mile)	Ee/(kW·h)	ηd	pc＝pd/(kW)
											19	0.5	9	0.5	3.37	0.5	0.95	100	40	0.9	7.68

In order to reduce the number of times of charging and discharging the battery back and forth so as to reduce the battery loss cost and meet the travel requirement of a user, the electric automobile with the initial SOC of more than 0.4 can be considered to discharge the power grid so as to fully reduce the SOC value and improve the charging load of the battery in the period of wind curtailment at night. Fig. 4 shows a flowchart of EV charge/discharge load simulation calculation.

According to the behavior of the user, most private car owners will start charging soon after arriving at home on working days, and the loads in the disordered charging mode of the electric car are as shown in fig. 5.

The electric load (excluding EV load) and the predicted value of the day-ahead wind power on a typical day in the calculation example are shown in fig. 6 and 8, and the system does not exchange electric power with other power grids. Wherein, the scheduling period is 1d, and the unit scheduling time length is 1 h. The coal price is 180$/t, and the wind power punishment price is 100$/(MW & h).

The original average electricity price is 65$/(MW & h), the peak and valley electricity prices of the peak-valley time-of-use electricity price are respectively 1.25 times and 0.75 time of the original electricity price, and the ordinary period is unchanged.

The response quantity at each moment is not more than 10% of the load peak value, the total electricity consumption of the user is not changed, and the self-elasticity coefficient and the cross-elasticity coefficient are respectively-0.2 and 0.033. In practical application, the elasticity coefficient value should be investigated according to the behavior characteristics of the local load. The peak-valley average time interval division results are shown in fig. 6, and the daily load curves before and after the peak-valley time-of-use electricity price response are shown in fig. 7.

In order to examine the influence of the abundant resources on the demand side on the operating economy and the wind power consumption of the cogeneration system, four scenarios shown in table 5 are set, and the scheduling result under each scenario is shown in table 6.

Table 5 scene settings

Context	Operation mode (resource abundance on demand side)
		1	Traditional dispatching mode (No peak valley time-of-use electricity price demand response, no consideration of building heat storage characteristics, disordered EV load)
2	Peak-to-valley time of use demand response
		3	Peak-valley time-of-use electricity price demand response and building heat storage characteristics
4	Scheduling strategy (Peak valley time-of-use electricity price demand response, building heat storage characteristic, controlled EV ordered charging and discharging)

TABLE 6 optimized scheduling results for each scenario

From the scheduling results in table 6 above, it can be seen that the peak-valley time-of-day electricity price demand response, the flexible thermal load when the thermal storage characteristics of the building are considered, and the controlled EV grid-connected ordered charging and discharging can promote the wind power consumption of the cogeneration system, so that the system can operate more economically. The peak-valley time-of-use electricity price demand response, the flexible heat load when the heat storage characteristics of the building are considered, and the controlled EV ordered charging and discharging are used as the abundant resources on the demand side and are integrated into the scheduling of the cogeneration system, so that the system can run more economically, and the wind power consumption can be promoted to a greater extent. By adopting the optimization method of the cogeneration system, the cogeneration system has better environmental benefits, namely the power generation amount of the condensing gas is reduced by about 1259MW & h and the wind power consumption is increased by about 1258MW & h compared with the traditional scheduling mode from the viewpoint of the scheduling result, and higher economic benefits are obtained, namely the total power supply and heat supply cost of the system can be reduced by about 185632 dollars compared with the traditional scheduling mode, including the penalty cost of wind abandoning, because more wind power replaces the output of a high-cost coal-electric unit, and the flexible heat load when the heat storage characteristic of a building is considered decouples the thermoelectric rigid constraint of the thermoelectric unit for 'fixing the power with heat', so that the range of the output of the electric power of the thermoelectric unit is enlarged, and the thermoelectric unit operates in a more economic output state of the system.

Scenes 1-4 were analyzed as follows:

by combining the original load curve of fig. 7 and the wind power predicted output curve of fig. 8, it can be known that the wind power output peak time of the "northeast" region exactly corresponds to the electric load valley time, and the wind power output valley time exactly corresponds to the electric load peak time, i.e., the back-peak-shaving characteristic of wind power. As can be seen from fig. 9, the wind curtailment period of the conventional scheduling manner, i.e., scenario 1, mainly occurs at night electricity load valley and wind electricity output peak period (22:00, 23:00, 00:00-07:00), as can be seen from fig. 10 and 12, at this time, the conventional unit operates in the lower limit state of electricity output, and the thermoelectric unit operates in the higher state of electricity output for satisfying the heat supply load and due to the limitation of the thermoelectric coupling constraint of "fixing electricity with heat". The electric output of the conventional unit and the CHP unit can not be adjusted downwards, and the wind has to be abandoned.

When the peak-valley time-of-use electricity price demand response is implemented, namely scene 2 is shown in fig. 10, 11 and 12, compared with scene 1, the wind abandoning situation is improved in 23:00 and 00:00-05:00 time periods, the reason is that the peak-valley time-of-use electricity price demand response changes the load characteristic, so that part of electricity consumption in the peak time period of the electricity load is transferred to the low-valley time period of the electricity load, as shown in fig. 7, the electricity load in the low-valley time period of the load is increased to provide the internet surfing space for wind power, the conventional unit operates in the lower limit state of electricity output, and the electricity output of the thermoelectric unit is the same as scene 1 due to the limitation of thermoelectric coupling constraint of 'fixing electricity by heat'; the reduction in power output of both the conventional unit and the thermoelectric unit during peak load hours is due to the reduction in load during peak load hours as a result of the implementation of the peak-valley time-of-day electricity price demand response. At the moment, because the time-of-use electricity price demand response is limited, the conventional unit operates in the lower limit state of the electric output, the CHP unit cannot adjust the electric output due to the thermo-electric coupling constraint of 'fixing the electricity by heat', and the wind power consumption is increased on the basis of the scenario 1, but still the abandoned wind exists.

When the flexible heat load in the heat storage characteristic of the building is considered on the basis of scenario 2, scenario 3. Compared with scenario 2, the condition of wind abandoning in the time period of 01:00-04:00 is improved, because (as can be seen from fig. 9-13) the building heat release replaces partial heat output of the thermoelectric unit in the time period, the heat output of the thermoelectric unit is reduced, and the electric output is also reduced due to the thermoelectric coupling characteristic, so that an online space is provided for wind power; during the daytime electric load peak period and the wind power off-peak period, under the condition that low-cost wind power is completely on line at the time, the output of a conventional coal-fired unit is reduced, the electric output and the heat output of a CHP unit are improved compared with those of the situation 2, and on the basis of meeting the heat comfort of users, redundant heat after heat load balancing is stored in a building (08:00-11:00) and is used for supplying a part (01:00-04:00) which is insufficient for heat load due to heat output reduction of the CHP unit at night. Fig. 13 shows the situation 3 that the heat storage capacity (positive sign indicates heat storage and negative sign indicates heat release) and the indoor temperature inside the building are scheduled in each time period, it can be seen that the indoor temperature fluctuates between 19 ℃ and 21 ℃, and the fluctuation time is mostly night sleep time periods and is within the temperature range of the thermal comfort degree of the user.

The demand response inhibits the back peak regulation characteristic of the wind power to a certain extent by optimizing a load curve, but part of the wind power still has to abandon the wind in a load valley period due to the 'electricity utilization by heat' operation mode of the CHP unit (as can be known from FIG. 9); when peak-valley time-of-use electricity price demand response is implemented and building heat storage characteristics are considered, namely scenario 3, due to the fact that the building heat storage capacity is limited on the basis of meeting the thermal comfort of users, partial wind abandoning is still performed during the peak period of wind power at the load valley, and the speed is 05:00-07: 00. Based on scenario 3, EV controlled sequential charging and discharging is considered, that is, scenario 4, and fig. 14 illustrates scheduling conditions of each unit and EV in scenario 4. Compared with scenario 3, the wind power consumption can be further improved by controlling the sequential interaction of the EV and the power grid energy while the travel demand of the user is met. This is because EV charging is performed by accessing the network during the wind curtailment period, the power consumption during the low load period is increased, EV charging is performed by discharging to reduce the state of charge of the EV battery during the high load period, so that the EV increases the demand for charging during the low load period (see fig. 5, 9 and 15), fig. 15 is a case where EV and grid energy are exchanged (positive sign indicates EV charging by accessing the network and negative sign indicates EV discharging by accessing the network) in each period scheduled in the scenario 4, and fig. 16 is a case where the building internal heat storage capacity and the indoor temperature in each period are scheduled in each period scheduled in the scenario 4. As shown in fig. 9, 37.2MW of air curtailment amount exists at 06:00, and at this time, the abundant resource on the demand side is utilized to the maximum, but the air curtailment is inevitable due to the requirement of meeting the user's travel demand and the limited charging power of the EVs during network access, and at this time, the increase of the energy storage facility or the release of a larger number of EVs can be considered to take the air curtailment amount in this period. On a day-ahead scheduling scale (ignoring wind power prediction errors), scenario 4 can almost completely take up wind power output.

According to the analysis of the scenes 1-4, the peak-valley time-of-use electricity price demand response can optimize an original load curve for peak clipping and valley filling, and the wind power consumption can be improved; the heat storage characteristic of a building is considered, the thermoelectric coupling characteristic of the CHP unit can be decoupled to a certain extent, and the wind power consumption is improved; the controlled EV ordered charging and discharging can improve the electricity consumption in the wind abandoning period, and the network access discharging in the load peak period can further improve the EV charging demand in the wind abandoning period and improve the wind power consumption. When single demand side resources are considered, the wind power consumption capacity of the wind power generation system is limited. The comprehensive resource abundance of the demand side participates in the system scheduling, so that the wind power consumption can be improved to a greater extent, and the system operation economy is improved.

While embodiments of the invention have been described above, it is not limited to the applications set forth in the description and the embodiments, which are fully applicable in various fields of endeavor to which the invention pertains, and further modifications may readily be made by those skilled in the art, it being understood that the invention is not limited to the details shown and described herein without departing from the general concept defined by the appended claims and their equivalents.

Claims (9)

1. A method for optimizing a cogeneration system, comprising the steps of:

s1, taking the hourly power output power p (i, t) of each unit and the hourly thermal power output power h (i, t) of the thermoelectric unit as optimization decision variables;

s2, on the basis of considering wind abandoning punishment cost, demand response cost, user thermal comfort compensation cost and electric automobile battery depreciation cost and neglecting the generating cost of the wind turbine generator, taking the most economical system generating and heating operation cost as an objective function, wherein the objective function is shown as a formula 1:

wherein the content of the first and second substances,

represents the cost of the electricity generated by the conventional unit,

represents the power generation and heat supply cost, rho, of the CHP unit^WΔp_W(i, t) represents the penalty cost of wind power abandonment, C_DR(t) represents the demand response scheduling cost, C_ΔHRepresents the user heating comfort disturbance compensation cost, U_evRepresenting EV battery depreciation cost; t is the total time period number of 1 scheduling period; n is a radical of_gThe number of conventional units; n is a radical of_chpThe number of CHP units; c_DRScheduling costs for demand response per time period; rho^wPunishing price for wind abandon;

s3, combining the constraint conditions with the objective function to construct a scheduling model of the cogeneration system;

and S4, carrying out optimization solution on the scheduling model of the cogeneration system by adopting an improved particle swarm optimization algorithm based on the Benders decomposition algorithm.

2. The method of optimizing a cogeneration system of claim 1, wherein the wind curtailment penalty cost ρ is^WΔp_WΔ p in (i, t)_W(i, t) represents the difference between the predicted output of the wind power and the utilization power at the same moment, and is represented by formula 2:

wherein poutw (i, t) is the predicted output of the wind power at the time t, and pin w (i, t) is the utilization power of the wind power at the time t;

demand response scheduling cost C_DR(t) is expressed by equation 3:

wherein the content of the first and second substances,

and P0L, t is the electricity rate and consumption before the implementation of demand response at time t, s_tAnd P_L,tElectricity rates and electricity amounts after the demand response is implemented for the period t, respectively.

3. The method for optimizing the cogeneration system of claim 2, wherein said constraints comprise: system constraint, unit constraint, wind power constraint, demand response constraint, user heating temperature comfort constraint and electric vehicle constraint;

wherein the wind power constraint is represented by formula 4; the user heating temperature comfort constraint is represented by equation 5:

wherein, theta^y(t) is the building interior temperature at time t,

and

the temperature upper and lower limits for the thermal comfort of the user are met.

4. The method of optimizing a cogeneration system of claim 3, wherein said system constraints comprise an electrical power balance constraint and a heating balance constraint;

additionally, the electric power balance constraints, in turn, include an unspoiled demand response constraint represented by equation 6 and a participated demand response constraint represented by equation 7:

wherein P0L (t) and P_L(t) respectively representing the electrical loads before and after the demand response at the time t; p is a radical of_g(i, t) and p_chp(i, t) respectively representing the electric loads of the conventional unit and the CHP unit at the moment t; p is a radical of_ev(t) represents the electric load of the EV battery at time t; n is a radical of_wThe number of wind power plants;

the heating balance constraint is represented by equation 8:

wherein H_load(t) refers to the thermal load of the area and is equal to the sum of the thermal loads of all buildings in the area; h is_chp(i, t) refers to the heat load of the ith CHP unit at the moment t; thermal load h of individual buildings_load(t) is represented by equation 9:

wherein h is_req(t) is the thermal demand of the user at time t; theta_in(t) and θ_in(t-1) indoor temperatures at the time of t and t-1, respectively; c. C_airIs the specific heat capacity of the indoor air; rho_airIs the density of the indoor air; v is the peripheral volume of the building.

5. The method of optimizing a cogeneration system of claim 3, wherein said unit constraints comprise conventional unit constraints and cogeneration unit constraints;

in addition, the conventional unit constraints include upper and lower output limits constraints represented by equation 10, and unit ramp constraints represented by equation 11:

wherein the content of the first and second substances,

and

respectively the minimum and maximum generating power of a conventional unit i;

and

respectively limiting the downward slope climbing limit value and the upward slope climbing limit value of the conventional unit i;

the cogeneration unit constraints in turn include an active power output upper and lower limit constraint represented by equation 12, a unit thermal output constraint represented by equation 13, a unit electrical creep constraint represented by equation 14, and a unit thermal creep constraint represented by equation 15:

max(P_i,min-c_vh_chp(i,t),c_mh_chp(i,t))≤p_chp(i,t)≤P_i,max-c_vh_chp(i, t) equation 12;

0≤h_chp(i,t)≤H_i,maxequation 13;

wherein, P_i,minAnd P_i,maxRespectively the minimum and maximum electric output of the CHP unit i under the pure condensing working condition; c. C_vThe method is characterized in that the reduction amount of the power generation power is generated by the multiple extraction unit heat supply when the total steam inlet amount of the CHP unit is not changed; c. C_mThe elastic coefficients of electric power and thermal power when the CHP unit operates in a backpressure mode; h_i,maxThe maximum thermal power of the CHP unit i;

and

the maximum downward and upward electrical power ramp rates for the CHP unit i.

6. The method of optimizing a cogeneration system of claim 3, wherein said demand response constraints further comprise power comfort χ, represented by equation 16_tConstraints, and the benefit of the demand response interaction represented by equation 17

And (3) constraint:

wherein the content of the first and second substances,

and

respectively the minimum value of the electricity utilization mode comfort level and the interactive benefit comfort level of the user under the demand response; Δ q of_tAnd Δ c_tThe relative increase in demand for electricity in the demand response and the relative increase in electricity rate in the demand response, respectively.

7. The method of optimizing a cogeneration system of claim 3, wherein said electric vehicle constraints in turn comprise an EV maximum on-line discharge and charge power limit represented by equation 18 and equation 19, an EV battery maximum SOC limit represented by equation 20, and an EV owner charge demand constraint represented by equation 21:

soc_i≤soc_maxequation 20;

P_ev＝P_c+P_dformula 21;

wherein, soc_maxA maximum state of charge set to prevent overcharging of the battery; p_evThe total unordered charging load before the EV participates in scheduling is obtained; p_cCharging the total network access load after the EV participates in scheduling; p_dAnd discharging the total capacity of the network access after the EV participates in the scheduling.

8. The optimization method of the cogeneration system of claim 1, wherein in S4, based on the Benders decomposition algorithm and using the improved particle swarm optimization algorithm to perform the optimization solution on the scheduling model of the cogeneration system specifically comprises: and dividing the optimization solving problem of the scheduling model of the cogeneration system into a power system sub-problem and a thermal system main problem based on a Benders decomposition algorithm, optimizing the decomposed systems by adopting an improved particle swarm optimization algorithm, interacting the systems with optimal output of the power system sub-problem and the thermal system main problem, and feeding back the generated Benders cut constraint condition to the power system sub-problem for iteration until the requirement of inspection is met.

9. The optimization method of the cogeneration system of claim 8, wherein the optimization solution of the cogeneration system scheduling model is performed by using an improved particle swarm optimization algorithm based on a Benders decomposition algorithm, and specifically comprises the following steps:

s4-1, initialization: setting the iteration number k of the Benders decomposition method to be 1 and the initial thermodynamic system variable h for each scheduling time t₁Taking the minimum value of the thermal output of the unit, bringing the minimum value into the power system to carry out first solution to obtain

And the value p of this sub-problem decision variable¹(ii) a Initializing an objective function value derived from a thermodynamic system

Is- ∞;

s4-2, the Lagrangian multiplier for the Benders cut feedback correction constraint is generated and is expressed by the formula 22:

wherein the content of the first and second substances,

a Lagrange multiplier in the feedback correction constraint condition for the main problem of the thermodynamic system returned in the kth iteration;

s4-3, solving a main problem of the thermodynamic system: will be obtained from S4-2

And p obtained by solving sub-problem of power system^kThe lower boundary of the objective function is obtained by adopting an improved particle swarm algorithm to solve

And h^k+1；

S4-4, solving the power system subproblem: the optimal heat output h obtained by the k +1 th thermodynamic system^k+1Bringing the sub-problem of the power system into the known quantity, and solving the sub-problem of the power system by adopting an improved particle swarm optimization algorithm to obtain an upper boundary of an objective function

And p^k+1；

S4-5, feasibility test of the solution obtained by the main problem and the subproblems: the obtained lower and upper boundaries are substituted into the test condition expressed by equation 23 to be tested, and if the test condition is satisfied, the iteration is ended, and

i.e. the optimal solution of the running cost of the original electric heating combined system, p^k+1And h^k+1Namely the optimal electric output and the optimal thermal output of the system; if the test condition is not satisfied, the iteration number k is added by 1 to the original value, and S4-2-S4-5 is repeated.

Images