CN113705935A - Multi-objective intelligent optimization-based load economic distribution method for cogeneration unit - Google Patents

Multi-objective intelligent optimization-based load economic distribution method for cogeneration unit Download PDF

Info

Publication number
CN113705935A
CN113705935A CN202111122814.9A CN202111122814A CN113705935A CN 113705935 A CN113705935 A CN 113705935A CN 202111122814 A CN202111122814 A CN 202111122814A CN 113705935 A CN113705935 A CN 113705935A
Authority
CN
China
Prior art keywords
unit
load
cogeneration
coal consumption
particle
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
CN202111122814.9A
Other languages
Chinese (zh)
Inventor
叶道正
李怡
伍华贵
郭伟康
周颖驰
连晖
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.)
Fujian Hongshan Thermoelectricity Co ltd
Original Assignee
Fujian Hongshan Thermoelectricity 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 Fujian Hongshan Thermoelectricity Co ltd filed Critical Fujian Hongshan Thermoelectricity Co ltd
Priority to CN202111122814.9A priority Critical patent/CN113705935A/en
Publication of CN113705935A publication Critical patent/CN113705935A/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
    • G06Q10/00—Administration; Management
    • G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/004—Artificial life, i.e. computing arrangements simulating life
    • G06N3/006—Artificial life, i.e. computing arrangements simulating life based on simulated virtual individual or collective life forms, e.g. social simulations or particle swarm optimisation [PSO]
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N7/00—Computing arrangements based on specific mathematical models
    • G06N7/01—Probabilistic graphical models, e.g. probabilistic networks
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
    • G06Q10/063—Operations research, analysis or management
    • G06Q10/0631—Resource planning, allocation, distributing or scheduling for enterprises or organisations
    • G06Q10/06313—Resource planning in a project environment
    • 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
    • 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
    • Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
    • Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
    • Y02P90/80—Management or planning
    • Y02P90/82—Energy audits or management systems therefor

Abstract

The invention discloses a multi-objective intelligent optimization-based load economic distribution method for a cogeneration unit. Step 1: carrying out system analysis on the energy consumption cost influence decision of the cogeneration unit; step 2: determining an objective function based on the system analysis of step 1; and step 3: confirming the constraint conditions of the generating load and the heating load of the unit; and 4, step 4: resolving the constraint conditions by using a simulated annealing particle swarm algorithm according to the constraint conditions in the step 3; and 5: and performing load economic distribution of the cogeneration unit based on the calculation result of the step 4. The method is used for solving the problem that when a mathematical model of the cogeneration unit is constructed, all independent variables and constraint conditions of the unit cannot be fully considered, or the goals of minimum total cost and minimum pollutants cannot be considered.

Description

Multi-objective intelligent optimization-based load economic distribution method for cogeneration unit
Technical Field
The invention belongs to the technical field of cogeneration units; in particular to a multi-objective intelligent optimization-based load economic distribution method for a cogeneration unit.
Background
Due to the needs of energy structure adjustment and environmental protection, cogeneration units play an increasingly important role in energy supply. The cogeneration unit does not cool the steam which has produced the electric energy, but supplies heat to the user, thus not only saving energy, but also increasing economic benefit. In addition, the cogeneration unit uses a high-pressure boiler or an ultrahigh-pressure boiler, and can perform desulfurization and denitration treatment in the boiler, thereby reducing the emission of pollutants.
In the prior art, when a mathematical model of a cogeneration unit is constructed, all independent variables and constraint conditions of the unit cannot be fully considered, or the aims of minimum total cost and minimum pollutants cannot be considered. When the model is optimized, either the accuracy of the solution is not high enough, or the found solution is easy to fall into a local optimal solution.
Disclosure of Invention
The invention provides a multi-objective intelligent optimization-based load economic distribution method for a cogeneration unit, which is used for solving the problems that all independent variables and constraint conditions of the cogeneration unit cannot be fully considered or the objectives of minimum total cost and minimum pollutants cannot be considered when a mathematical model of the cogeneration unit is constructed.
The invention is realized by the following technical scheme:
the method for economically distributing the load of the cogeneration unit based on multi-objective intelligent optimization comprises the following steps:
step 1: carrying out system analysis on the energy consumption cost influence decision of the cogeneration unit;
step 2: determining an objective function based on the system analysis of step 1;
and step 3: confirming the constraint conditions of the generating load and the heating load of the unit;
and 4, step 4: resolving the constraint conditions by using a simulated annealing particle swarm algorithm according to the constraint conditions in the step 3;
and 5: and performing load economic distribution of the cogeneration unit based on the calculation result of the step 4.
Further, the system analysis of the step 1 specifically includes researching the influence of the change of factors of main equipment, a system, operation, fuel and environment of the unit on the economy of the unit, and deeply analyzing the influence of each factor on the target value of the power supply coal consumption rate;
the influence factors specifically comprise design performance of the main equipment and the main auxiliary equipment, operation performance of the main equipment and the main auxiliary equipment, unit aging degree, fuel performance change, main operation parameters of the unit, start-stop process of the unit, ambient temperature and other operation and equipment factors;
various factors influencing the coal consumption and economic performance indexes of the unit are analyzed, energy consumption differences caused by the various factors are quantitatively analyzed, and the relation between the total energy consumption, the generating power and the heat supply capacity of the unit is determined.
Further, the step 2 of determining the objective function specifically includes determining standard total coal consumption, wherein the coal consumption of the cogeneration unit mainly includes heat supply coal consumption and power generation coal consumption, and the coal consumption b required when the unit generates powerP() As indicated by the general representation of the,
bPi(Pi)＝λ1Pi+λ2Dh1(i)+λ3Dh2(i)+λ4Pi 2+λ5Dh1(i)2+λ6Dh2(i)2+λ7 (1)
wherein λ isj J 1,2, 7 is a coefficient required for generating power by the unit, DhIs the total heat supply load, P is the power generation load;
coal consumption b required by unit for heat supplyD() As indicated by the general representation of the,
bDi(Dh)＝χ1Pi+χ2Dh1(i)+χ3Dh2(i)+χ4Pi 2+χ5Dh1(i)2+χ6Dh2(i)2+χ7 (2)
wherein, χjJ 1,2, 7 is a coefficient required by the unit for supplying heat;
total coal consumption B of unit for power generation and heat supplyi() As indicated by the general representation of the,
Bi(P,Dh)＝bp×Pi/103+bD×Dh(i)×H(i)106 (3)
wherein, H () is the enthalpy value of the extraction steam, and P is the power generation load;
in summary, the standard lowest overall coal consumption function is,
Figure BDA0003277611270000021
further, secondly, determining the pollutant discharge amount;
function f of emission of pollutants of computer seti() As indicated by the general representation of the,
fi(P,Dh)＝γ1Pi+γ2Dh1(i)+γ3Dh2(i)+γ4Pi 2+γ5Dh1(i)2+γ6Dh2(i)2+γ7 (5)
wherein, γjJ 1,2, 7 is a characteristic coefficient of the unit when the unit discharges pollutants;
the objective function for the lowest emission of pollutants is,
Figure BDA0003277611270000022
further, the electrical load adjustment time is the adjustment time after receiving the grid command, and the objective function is as follows:
T(Pi)＝maxti(Pi) (7)
wherein, ti(Pi) Means that the ith unit completes the charge PiRequired time, min, ti(Pi)＝|Pi-Pnow,i|/vi；Pnow,iIs the load, MW, currently borne by the ith unit; v. ofiThe rate of movement of the electrical load in the unit, MW/min;
when calculating the comprehensive economic cost, standard total coal consumption, pollutant discharge amount and electric load adjustment time are considered, and different weights are given according to the importance degree for calculation; the comprehensive economic cost objective function is:
Figure BDA0003277611270000031
wherein M is the comprehensive economic cost per hour; eta1、η2、η3Weight representing the degree of importance of expressions (4), (6) and (7), McThe standard coal unit price is Yuan/t, and the unit price of the coal in the invention is 900 Yuan/t; mnFor reducing treatment costs, Yuan/t, MadAnd adjusting time economic benefit coefficients for the load according to a power dispatching mechanism and unit benefit assessment values, yuan/min.
Furthermore, the constraint conditions of step 2 are specifically,
the generating power of the unit needs to be kept balanced:
Figure BDA0003277611270000032
medium and low pressure heating is also balanced, i.e.:
Figure BDA0003277611270000033
Figure BDA0003277611270000034
interval of generated power:
Pmin≤P≤Pmax (11)
difference between upper and lower limits of medium and low pressure heat supply
Figure BDA0003277611270000035
Wherein Dh1min、Dh1max、Dh2min、Dh2maxRespectively refers to the upper and lower limits of medium-pressure heat supply and the upper and lower limits of low-pressure heat supply.
Further, the step 3 is specifically to determine the constraints of the power generation load and the heating load of the cogeneration unit double-pump unit based on medium-pressure steam extraction and low-pressure steam extraction generated when the cogeneration unit double-pump unit generates power; and the standard coal consumption is reduced by adjusting the proportion of the thermoelectric load among a plurality of units.
Further, the particle swarm algorithm for annealing in the step 3 specifically comprises the following steps:
step 3.1: the method comprises the following steps that (1) the initialization temperature of a double-pump unit of a cogeneration unit and the current position and speed of each particle in a population are measured;
step 3.2: calculating the fitness of each particle, and storing the position and the fitness of each particle in pbest;
step 3.3: calculating the individual position with the best fitness in pbest and storing the fitness in gbest;
step 3.4: updating the speed and position of each particle by using a speed and position updating formula;
step 3.5: determining a global optimum with a roulette strategy;
step 3.6: calculating the current fitness of each particle and comparing the current fitness with the best position, and if the current fitness of the particle is better than the previous fitness, performing step 3.7; if the current fitness value of the particle is worse than the previous value, performing step 3.8;
step 3.7: recording this position as the best position;
step 3.8: accepting this value as the best position with a certain probability;
step 3.9: comparing all the calculated pbest with the gbest, if the value of the pbest is better than that of the gbest, updating the value of the gbest to be the current pbest, otherwise, keeping the value of the gbest unchanged;
step 3.10: based on step 3.9, if the condition is satisfied, stopping searching and outputting the result, otherwise, returning to step 3.3 to continue iteration until the condition is satisfied.
Further, the satisfaction of the condition in step 3.10 refers to reaching the number of iterations or reaching the desired accuracy.
The invention has the beneficial effects that:
the model established by the invention takes the coal consumption, the pollutant discharge amount and the load adjustment time as the targets, takes the generated power, the medium-pressure steam extraction flow and the low-pressure steam extraction flow as the constraint conditions, and reduces the comprehensive economic cost on the basis of the safe use of the cogeneration unit. In order to obtain a better optimization result, the results of the genetic algorithm and the GA _ PSO algorithm are compared in the research, the GA _ PSO algorithm is superior to the genetic algorithm in the aspects of total standard coal consumption, load adjustment time, comprehensive economic cost and the like, and the search precision can be improved and the GA _ PSO algorithm is prevented from falling into local optimization.
The invention achieves the purpose of lowest total standard coal consumption by reasonably distributing the power generation power, the medium-pressure steam extraction quantity and the low-pressure steam extraction quantity under the condition that the cogeneration unit is safely used and the heat supply load is met. Meanwhile, pollutant discharge amount and power grid load lifting assessment indexes are considered, and the optimal value is solved under multiple conditions, so that the power generation and heat supply of the unit are intelligentized.
Drawings
FIG. 1 is a flow chart of the method of the present invention.
FIG. 2 is a standard particle swarm algorithm fitness curve graph of the present invention.
FIG. 3 is a diagram of the operation result of the Matlab of the standard particle swarm algorithm of the invention.
FIG. 4 is a graph of fitness curve of simulated annealing particle swarm algorithm of the present invention.
FIG. 5 is a diagram of the operation result of the simulated annealing particle swarm algorithm matlab.
FIG. 6 is a graph of upper and lower limits of electrical load corresponding to low-pressure heating conditions of the unit.
FIG. 7 is a graph of the medium pressure steam versus electrical load of the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be described clearly and completely with reference to the accompanying drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
A multi-objective intelligent optimization-based load economic distribution method for a cogeneration unit comprises the following steps:
step 1: carrying out system analysis on the energy consumption cost influence decision of the cogeneration unit;
step 2: determining an objective function based on the system analysis of step 1;
and step 3: confirming the constraint conditions of the generating load and the heating load of the unit;
and 4, step 4: resolving the constraint conditions by using a simulated annealing particle swarm algorithm according to the constraint conditions in the step 3;
and 5: and performing load economic distribution of the cogeneration unit based on the calculation result of the step 4.
Further, the system analysis of the step 1 specifically includes researching the influence of the change of factors of main equipment, a system, operation, fuel and environment of the unit on the economy of the unit, and deeply analyzing the influence of each factor on the target value of the power supply coal consumption rate;
the influence factors specifically comprise design performance of the main equipment and the main auxiliary equipment, operation performance of the main equipment and the main auxiliary equipment, unit aging degree, fuel performance change, main operation parameters of the unit, start-stop process of the unit, ambient temperature and other operation and equipment factors;
various factors influencing the coal consumption and economic performance indexes of the unit are analyzed, energy consumption differences caused by the various factors are quantitatively analyzed, and the relation between the total energy consumption, the generating power and the heat supply capacity of the unit is determined.
Further, the objective function determined in step 2 is specifically that the double-extraction thermoelectric power generation unit is a common type of cogeneration unit at present, and can generate two types of pressure heat supply while generating power to meet different needs of users. The three loads can be independently adjusted on the premise of safe operation of the unit, and the characteristics of the unit are shown in tables 1-1 and 1-2.
Firstly, the optimization of the cogeneration unit must fulfill the aim of minimizing the total standard coal consumption by reasonably distributing the generated power, the medium-pressure heat supply and the low-pressure heat supply under the condition that the cogeneration unit is safely used and the heat supply load is satisfied. Secondly, because the environmental protection consciousness of China is greatly increased, the aim of only minimizing the total standard coal consumption is not enough, and the result that the reduction of the total standard coal consumption is reduced by increasing the pollutant discharge amount is probably caused. The aim of minimizing the emission of pollutants on the original target is to increase the emission of pollutants. In addition, the power grid load lifting task is an important assessment index for the power plant, so the task of adjusting the electric load is also important.
Firstly, determining standard total coal consumption, wherein the coal consumption of the cogeneration unit is mainly divided into heat supply coal consumption and power generation coal consumption, and the coal consumption b required when the unit generates powerP() As indicated by the general representation of the,
bPi(Pi)＝λ1Pi+λ2Dh1(i)+λ3Dh2(i)+λ4Pi 2+λ5Dh1(i)2+λ6Dh2(i)2+λ7 (1)
wherein λ isj J 1,2, 7 is a coefficient required for generating power by the unit, DhIs the total heat supply load, P is the power generation load;
coal consumption b required by unit for heat supplyD() As indicated by the general representation of the,
bDi(Dh)＝χ1Pi+χ2Dh1(i)+χ3Dh2(i)+χ4Pi 2+χ5Dh1(i)2+χ6Dh2(i)2+χ7 (2)
wherein, χjJ 1,2, 7 is a coefficient required by the unit for supplying heat;
total coal consumption B of unit for power generation and heat supplyi() As indicated by the general representation of the,
Bi(P,Dh)＝bp×Pi/103+bD×Dh(i)×H(i)/106 (3)
wherein, H () is the enthalpy value of the extraction steam, and P is the power generation load; specific values are shown in Table 1
In summary, the standard lowest overall coal consumption function is,
Figure BDA0003277611270000061
further, secondly, determining the pollutant discharge amount;
function f of emission of pollutants of computer seti() As indicated by the general representation of the,
fi(P,Dh)＝γ1Pi+γ2Dh1(i)+γ3Dh2(i)+γ4Pi 2+γ5Dh1(i)2+γ6Dh2(i)2+γ7 (5)
wherein, γjJ 1,2, 7 is a characteristic coefficient of the unit when the unit discharges pollutants;
the objective function for the lowest emission of pollutants is,
Figure BDA0003277611270000062
further, the electrical load adjustment time is the adjustment time after receiving the grid command, and the objective function is as follows:
T(Pi)＝maxti(Pi) (7)
wherein, ti(Pi) Means that the ith unit completes the charge PiRequired time, min, ti(Pi)＝|Pi-Pnow,i|/vi；Pnow,iIs the load, MW, currently borne by the ith unit; v. ofiThe rate of movement of the electrical load in the unit, MW/min;
when calculating the comprehensive economic cost, standard total coal consumption, pollutant discharge amount and electric load adjustment time are considered, and different weights are given according to the importance degree for calculation; the comprehensive economic cost objective function is:
Figure BDA0003277611270000063
wherein M is the comprehensive economic cost per hour; eta1、η2、η3Weight representing the degree of importance of the expressions (4), (6) and (7) (. eta.) in the present invention1,η2η3Set to 0.9, 0.05; mcThe standard coal unit price is Yuan/t, and the unit price of the coal in the invention is 900 Yuan/t; mnFor emission reduction treatment cost, yuan/t, the emission reduction treatment cost of the invention is 1260 yuan/t; madThe time economic benefit coefficient is adjusted for the load, and the cost is-4500 yuan/min according to the power dispatching mechanism and the unit benefit assessment value, yuan/min.
Furthermore, the constraint conditions of step 2 are specifically,
the generating power of the unit needs to be kept balanced:
Figure BDA0003277611270000071
medium and low pressure heating is also balanced, i.e.:
Figure BDA0003277611270000072
Figure BDA0003277611270000073
interval of generated power:
Pmin≤P≤Pmax (11)
medium-pressure heating and low-pressure heating:
Figure BDA0003277611270000074
wherein Dh1min、Dh1max、Dh2min、Dh2maxRespectively refers to the upper and lower limits of medium-pressure heat supply and the upper and lower limits of low-pressure heat supply. As shown in tables 4-1 and 4-2, as shown in fig. 6-7.
Specific values are shown in Table 3-2.
The standard Particle Swarm Optimization algorithm (PSO) was proposed in 1995 by Phd Eberhart and Phd Kennedy, and originated from studies on the predatory behavior of groups of birds. The central idea of the method is to share the information of the group by using independent individuals in the group, so that the whole group is subjected to an evolutionary process from disorder to order in the movement change, and a global optimal solution is found.
In the PSO algorithm, each solution to the problem is compared to a bird without mass and volume, and the bird is extended to n-dimensional space. A bird is a particle and the position in this n-dimensional space is considered as a vector, and the velocity at which the particles fly is also expressed as a vector.
The solution of the function of a problem to be optimized for all particles is the adaptive value (fitness), each particle has a speed that determines its flight direction and distance, and each particle knows clearly its current position and best position (pbest) of passing itself. As population intelligence, each particle also knows the best location (gbest) in the population where all particles have been found in the past. Briefly, gbest is the best value of pbest. Each particle will use its past experience and the best experience of other particles to determine how much distance it should fly in which direction.
The standard particle swarm algorithm comprises the following steps:
and (4) randomly initializing the current position and speed of each particle in the population.
Calculating the fitness of each particle, storing the position and the fitness of each particle in pbest, and storing the position and the fitness of the individual with the best fitness in pbest in gbest.
And updating the speed and position of each particle by using a speed and position updating formula.
And fourthly, calculating the current fitness of each particle to be compared with the best position, and if the current fitness is better than the previous fitness, recording the position as the best position.
Comparing all present pbest with gbest, if the value of pbest is better than that of gbest, updating gbest, otherwise not updating.
Stopping searching and outputting the result if the condition (reaching the iteration times or the expected precision) is met, and returning to the third step to continue the iteration.
The earliest idea of the Simulated Annealing (SA) algorithm was proposed by n.metropolis et al in 1953, the basic idea being derived from the process of solid cooling in the physical domain. The SA can receive a relatively poor solution with a certain probability in the process of continuously searching the optimal solution to replace the current solution, so that the local optimal solution is skipped to obtain the global optimal solution.
The simulated annealing particle swarm optimization is based on the standard particle swarm optimization, simulated annealing is introduced, the convergence rate of the particle swarm optimization and the local optimization capability of the simulated annealing algorithm are combined, and the actual application shows that the particle swarm optimization added with the simulated annealing algorithm is superior to the standard particle swarm optimization in both solution superiority and convergence rate.
Further, the step 3 is specifically to determine the constraints of the power generation load and the heating load of the cogeneration unit double-pump unit based on medium-pressure steam extraction and low-pressure steam extraction generated when the cogeneration unit double-pump unit generates power; and the standard coal consumption is reduced by adjusting the proportion of the thermoelectric load among a plurality of units.
Further, the particle swarm algorithm for annealing in the step 3 specifically comprises the following steps:
step 3.1: the method comprises the following steps that (1) the initialization temperature of a double-pump unit of a cogeneration unit and the current position and speed of each particle in a population are measured;
step 3.2: calculating the fitness of each particle, and storing the position and the fitness of each particle in pbest;
step 3.3: calculating the individual position with the best fitness in pbest and storing the fitness in gbest;
step 3.4: updating the speed and position of each particle by using a speed and position updating formula;
step 3.5: determining a global optimum with a roulette strategy;
step 3.6: calculating the current fitness of each particle and comparing the current fitness with the best position, and if the current fitness of the particle is better than the previous fitness, performing step 3.7; if the current fitness value of the particle is worse than the previous value, performing step 3.8;
step 3.7: recording this position as the best position;
step 3.8: accepting this value as the best position with a certain probability (10% -20%);
step 3.9: comparing all the calculated pbest with the gbest, if the value of the pbest is better than that of the gbest, updating the value of the gbest to be the current pbest, otherwise, keeping the value of the gbest unchanged;
step 3.10: based on step 3.9, if the condition is satisfied, stopping searching and outputting the result, otherwise, returning to step 3.3 to continue iteration until the condition is satisfied.
Further, the satisfaction of the condition in step 3.10 refers to reaching the number of iterations or reaching the desired accuracy.
In the experimental process, the particle swarm optimization is continuously improved, a large number of parameters are debugged, and other algorithms are added to optimize and improve the particle swarm optimization, so that the convergence rate of the particle swarm optimization is higher, and the optimization effect is better, so that the particle swarm optimization can be well practiced in daily production operation. In continuous attempts, the simulated annealing algorithm is added into the particle swarm algorithm, the convergence speed and the convergence effect are far higher than those of other algorithms, so that the simulated annealing algorithm and the particle swarm algorithm are selected to realize the optimization of the load distribution problem of the multi-target cogeneration unit.
The values shown in table 3-1 are the values of the unit characteristic coefficients obtained from the actual data analysis in the present application.
TABLE 3-1 characteristic coefficient values of unit
Figure BDA0003277611270000091
The values shown in table 3-2 are values of the upper and lower limits of medium and low pressure heat supply, the lower and upper limits of medium and low pressure heat supply, etc. in the above according to the actual production operation requirements.
TABLE 3-2 production run parameter values
Figure BDA0003277611270000101
In matlab simulation particle swarm optimization, the number of particles N is 50, C1 is C2 is 1.2, Vmin is-0.3, Vmax is 0.3, and the maximum number of iterations M is 500, and it is found through repeated experiments that the general particle swarm optimization converges after about 200 iterations, and the results are shown in fig. 2 and 3:
after the model is solved by using a standard particle swarm algorithm in Matlab, the standard particle swarm algorithm converges in about 200 generations to obtain a global optimal fitness value.
And (3) obtaining values of independent variables corresponding to the global optimal value obtained by utilizing a Matlab simulation standard particle swarm optimization (the independent variables are respectively the electric load quantity, the medium-voltage heat supply and the low-voltage heat supply of the unit 1 and the electric load quantity, the medium-voltage heat supply and the low-voltage heat supply of the unit 2), and corresponding values of standard heating coal consumption rate, heating coal consumption rate and pollution.
On the basis of matlab simulation particle swarm optimization, a simulated annealing algorithm is added, an annealing constant is introduced, the particle number N is 50, C1 is C2 is 1.2, Vmin is-0.3, Vmax is 0.3, the maximum iteration number M is 50, and the annealing constant lamda is 0.9, which are consistent with the standard particle swarm optimization. And (4) taking the average value of results of multiple experiments, adding the particle swarm algorithm subjected to simulated annealing, converging after iteration for 15 times approximately, and enabling the optimized fitness value to be superior to the value optimized by the standard particle swarm algorithm.
The results are shown in FIGS. 4 and 5:
the fitness curve obtained after the simulated annealing particle swarm algorithm is operated in Matlab can be obviously seen from the graph, the fitness curve presents a rapid convergence process, after 20 iterations are performed, the simulated annealing particle swarm algorithm finds a global optimum value, the convergence speed is higher compared with the standard particle swarm algorithm, and the value of the fitness is superior to that of the standard particle swarm algorithm.
In fig. 5, the data is the values of independent variables (respectively, the electric load, the medium-voltage heat supply, and the low-voltage heat supply of the unit 1 and the electric load, the medium-voltage heat supply, and the low-voltage heat supply of the unit 2) corresponding to the optimal solution after the simulated annealing particle swarm algorithm is run in Matlab to find the optimal solution in the mathematical model of the cogeneration unit, and the values of the heating coal consumption rate, the power generation coal consumption rate, and the pollution of the two corresponding units.
TABLE 3-3 comparison of results of Standard particle swarm Algorithm and simulated annealing particle swarm Algorithm
Figure BDA0003277611270000102
Figure BDA0003277611270000111
As shown in tables 3 to 3, it can be easily seen that, compared with the normally operating units, the cogeneration units optimized by the particle swarm optimization method only require 163.490 tons per hour for the total coal consumption of the whole plant, which is 1.53 tons per hour lower than 165.02 tons per hour in the standard operation, and the total comprehensive economic cost of the whole plant is 129627 yuan per hour, which is 1573 yuan per hour lower than 131200 yuan per hour in the standard operation.
Compared with a normally working unit, the combined heat and power generation unit optimized by the simulated annealing algorithm is added in the particle swarm optimization, the total coal consumption of the whole plant is only 162.502 tons per hour, 2.518 tons per hour is reduced compared with 165.02 tons per hour in standard working, and 0.988 tons per hour is reduced compared with 163.490 tons per hour of the combined heat and power generation unit optimized by the standard particle swarm; and the total comprehensive economic cost of the whole plant is 128826 yuan per hour, 2374 yuan per hour is reduced compared with 131200 yuan per hour in standard work, and 801 yuan per hour is reduced compared with 129627 yuan per hour in standard work. From the numerical point of view, the total standard coal consumption and the comprehensive economic cost of the whole plant of the combined heat and power generation unit optimized by the standard particle swarm optimization or the combined heat and power generation unit optimized by the simulated annealing algorithm added into the particle swarm optimization are reduced, and the production and operation cost of the whole plant is effectively reduced.
The invention comprehensively considers several important factors in the actual production, such as the power generation coal consumption, the heating coal consumption, the pollutant discharge amount and the load adjustment time of the double-extraction cogeneration unit, to establish a multi-target multi-constraint mathematical model to optimize by using an algorithm so as to reduce the comprehensive economic cost of the whole plant.
The method is based on energy consumption of double-extraction cogeneration units and comprehensive whole plant cost, comprises the steps of establishing a load optimization model for the double-cogeneration units of the plant by using a mathematical model, adding constraint conditions to adapt to actual production conditions, performing optimization calculation on the model by using a standard particle swarm optimization algorithm, introducing a simulated annealing algorithm to perform optimization calculation by using a particle swarm optimization algorithm, and not only combines the characteristics of local search capability of the simulated annealing algorithm and high convergence speed of the particle swarm optimization algorithm, but also ensures the accuracy of the algorithm and improves the convergence speed of the algorithm. Through comparison of actual operation, the effect of the particle swarm optimization algorithm after simulated annealing is obviously superior to that of a double-extraction cogeneration unit optimized by a common particle swarm optimization algorithm, the comprehensive economic cost of the whole plant is reduced, the effectiveness of the algorithm is verified, the algorithm can be quickly converged under the constraint condition, the accuracy of the algorithm is improved, and the optimized result is superior to that of the normally-working cogeneration unit which is not optimized. In summary, the particle swarm optimization algorithm combined with simulated annealing has excellent fitting effect and convergence rate in solving the load optimization distribution problem of the cogeneration unit, and has strong practical application value.
Figure BDA0003277611270000121
4-2 medium pressure steam corresponding to electric load relation
Figure BDA0003277611270000122

Claims (9)

1. The method for economically distributing the load of the cogeneration unit based on multi-objective intelligent optimization is characterized by comprising the following steps of:
step 1: carrying out system analysis on the energy consumption cost influence decision of the cogeneration unit;
step 2: an objective function is determined based on the system analysis of step 1.
And step 3: confirming the constraint conditions of the generating load and the heating load of the unit;
and 4, step 4: resolving the constraint conditions by using a simulated annealing particle swarm algorithm according to the constraint conditions in the step 3;
and 5: and performing load economic distribution of the cogeneration unit based on the calculation result of the step 4.
2. The multi-objective intelligent optimization-based load economic distribution method for the cogeneration unit according to claim 1, wherein the system analysis in the step 1 is specifically to study the influence of the changes of factors of main equipment, a system, operation, fuel and environment of the unit on the economy of the unit and deeply analyze the influence of each factor on the target value of the power supply coal consumption rate;
the influence factors specifically comprise design performance of the main equipment and the main auxiliary equipment, operation performance of the main equipment and the main auxiliary equipment, unit aging degree, fuel performance change, main operation parameters of the unit, start-stop process of the unit, ambient temperature and other operation and equipment factors;
various factors influencing the coal consumption and economic performance indexes of the unit are analyzed, energy consumption differences caused by the various factors are quantitatively analyzed, and the relation between the total energy consumption, the generating power and the heat supply capacity of the unit is determined.
3. The method for economically distributing the load of the cogeneration unit based on multi-objective intelligent optimization according to claim 1, wherein the objective function determined in the step 2 is specifically that standard total coal consumption is determined firstly, the coal consumption of the cogeneration unit is mainly divided into heating coal consumption and power generation coal consumption, and the coal consumption b required when the unit generates power is determinedP() As indicated by the general representation of the,
bPi(Pi)＝λ1Pi+λ2Dh1(i)+λ3Dh2(i)+λ4Pi 2+λ5Dh1(i)2+λ6Dh2(i)2+λ7 (1)
wherein λ isjJ 1,2, 7 is a coefficient required for generating power by the unit, DhFor total heating load, PiIs the load of power generation;
coal consumption b required by unit for heat supplyD() As indicated by the general representation of the,
bDi(Dh)＝χ1Pi+χ2Dh1(i)+χ3Dh2(i)+χ4Pi 2+χ5Dh1(i)2+χ6Dh2(i)2+χ7 (2)
wherein, χjJ 1,2, 7 is a coefficient required by the unit for supplying heat;
total coal consumption B of unit for power generation and heat supplyi() As indicated by the general representation of the,
Bi(P,Dh)＝bp×Pi/103+bD×Dh(i)×H(i)/106 (3)
wherein, H () is the extraction enthalpy value;
in summary, the standard lowest overall coal consumption function is,
Figure FDA0003277611260000021
4. the multi-objective intelligent optimization-based load economic distribution method for the cogeneration unit according to claim 3, wherein pollutant emission is determined secondly;
function f of emission of pollutants of computer seti() As indicated by the general representation of the,
fi(P,Dh)＝γ1Pi+γ2Dh1(i)+γ3Dh2(i)+γ4Pi 2+γ5Dh1(i)2+γ6Dh2(i)2+γ7 (5)
wherein, γjJ 1,2, 7 is a characteristic coefficient of the unit when the unit discharges pollutants;
the objective function for the lowest emission of pollutants is,
Figure FDA0003277611260000022
5. the economic distribution method of the load of the cogeneration unit based on multi-objective intelligent optimization according to claim 4, wherein the electric load adjustment time is the adjustment time after receiving the grid command, and the objective function is as follows:
T(Pi)＝maxti(Pi) (7)
wherein, ti(Pi) Means that the ith unit completes the charge PiRequired time, min, ti(Pi)＝|Pi-Pnow,i|/vi；Pnow,iIs the load, MW, currently borne by the ith unit; v. ofiThe rate of movement of the electrical load in the unit, MW/min;
when calculating the comprehensive economic cost, standard total coal consumption, pollutant discharge amount and electric load adjustment time are considered, and different weights are given according to the importance degree for calculation; the comprehensive economic cost objective function is:
Figure FDA0003277611260000023
wherein M is the comprehensive economic cost per hour; eta1、η2、η3Weight representing the degree of importance of expressions (4), (6) and (7), McThe standard coal unit price is Yuan/t, and the unit price of the coal in the invention is 900 Yuan/t; mnFor reducing treatment costs, Yuan/t, MadAnd adjusting time economic benefit coefficients for the load according to a power dispatching mechanism and unit benefit assessment values, yuan/min.
6. The method for economically distributing the load of the cogeneration unit based on multi-objective intelligent optimization according to claim 1, wherein the constraint conditions of the step 2 are specifically,
the generating power of the unit needs to be kept balanced:
Figure FDA0003277611260000024
medium and low pressure heating is also balanced, i.e.:
Figure FDA0003277611260000031
Figure FDA0003277611260000032
interval of generated power:
Pmin≤P≤Pmax (11)
medium-pressure heating and low-pressure heating:
Dh1min≤Dh1≤Dh1max
Dh2min≤Dh2≤Dh2max (12)
wherein Dh1min、Dh1max、Dh2min、Dh2maxRespectively refers to the upper and lower limits of medium-pressure heat supply and the upper and lower limits of low-pressure heat supply.
7. The multi-objective intelligent optimization-based load economic distribution method for the cogeneration unit according to claim 1, wherein the step 3 is specifically to determine the constraints of the power generation load and the heating load of the cogeneration unit double-pump unit based on medium-pressure steam extraction and low-pressure steam extraction generated during power generation of the cogeneration unit double-pump unit; and the standard coal consumption is reduced by adjusting the proportion of the thermoelectric load among a plurality of units.
8. The method for economically distributing the loads of the cogeneration units based on multi-objective intelligent optimization according to claim 7, wherein the pseudo-annealing particle swarm algorithm in the step 3 specifically comprises the following steps:
step 3.1: the method comprises the following steps that (1) the initialization temperature of a double-pump unit of a cogeneration unit and the current position and speed of each particle in a population are measured;
step 3.2: calculating the fitness of each particle, and storing the position and the fitness of each particle in pbest;
step 3.3: calculating the individual position with the best fitness in pbest and storing the fitness in gbest;
step 3.4: updating the speed and position of each particle by using a speed and position updating formula;
step 3.5: determining a global optimum with a roulette strategy;
step 3.6: calculating the current fitness of each particle and comparing the current fitness with the best position, and if the current fitness of the particle is better than the previous fitness, performing step 3.7; if the current fitness value of the particle is worse than the previous value, performing step 3.8;
step 3.7: recording this position as the best position;
step 3.8: accepting this value as the best position with a certain probability;
step 3.9: comparing all the calculated pbest with the gbest, if the value of the pbest is better than that of the gbest, updating the value of the gbest to be the current pbest, otherwise, keeping the value of the gbest unchanged;
step 3.10: based on step 3.9, if the condition is satisfied, stopping searching and outputting the result, otherwise, returning to step 3.3 to continue iteration until the condition is satisfied.
9. The method for economic distribution of load of cogeneration units based on multi-objective intelligent optimization according to claim 8, wherein said condition of satisfaction in step 3.10 refers to reaching the number of iterations or reaching the desired accuracy.
CN202111122814.9A 2021-09-24 2021-09-24 Multi-objective intelligent optimization-based load economic distribution method for cogeneration unit Pending CN113705935A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202111122814.9A CN113705935A (en) 2021-09-24 2021-09-24 Multi-objective intelligent optimization-based load economic distribution method for cogeneration unit

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202111122814.9A CN113705935A (en) 2021-09-24 2021-09-24 Multi-objective intelligent optimization-based load economic distribution method for cogeneration unit

Publications (1)

Publication Number Publication Date
CN113705935A true CN113705935A (en) 2021-11-26

Family

ID=78661789

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202111122814.9A Pending CN113705935A (en) 2021-09-24 2021-09-24 Multi-objective intelligent optimization-based load economic distribution method for cogeneration unit

Country Status (1)

Country Link
CN (1) CN113705935A (en)

Similar Documents

Publication Publication Date Title
CN109768573B (en) Power distribution network reactive power optimization method based on multi-target differential gray wolf algorithm
CN109345012B (en) Park energy Internet operation optimization method based on comprehensive evaluation indexes
CN105071433B (en) A kind of configuration scheme of distributed generation resource
CN106845626B (en) DG optimal configuration application method based on mixed frog-leaping particle swarm
CN107133840B (en) Multi-wind-field bidding method for promoting wind power consumption on site by electric heating
CN113783224A (en) Power distribution network double-layer optimization planning method considering operation of various distributed energy sources
CN111404204A (en) Optimized scheduling method for power system including wind power plant by using improved wolf pack algorithm
CN116307149A (en) Blast furnace performance optimization method based on attention LSTM and KBNSGA
Li et al. Active distribution network operational optimization problem: A multi-objective tuna swarm optimization model
CN108599142B (en) Mixed integer quadratic programming multi-target scheduling method based on weight
CN113255138A (en) Load distribution optimization method for power system
Zhou et al. Coal consumption prediction in thermal power units: A feature construction and selection method
CN113705935A (en) Multi-objective intelligent optimization-based load economic distribution method for cogeneration unit
CN113705934A (en) Supercritical wet-cold heat supply unit load economic distribution method based on multi-objective intelligent optimization
CN105552941A (en) Distributed power supply peak regulation capacity optimization method
CN113471989B (en) Intelligent micro-grid secondary frequency control method based on wolf optimization method
CN108629458B (en) Coal-electric unit optimized operation method based on big data
CN109506248B (en) Case reasoning boiler combustion optimization method based on online optimization
Kang et al. Brain storming optimization algorithm for heating dispatch scheduling of thermal power plant
CN117391402B (en) Intelligent factory multi-energy collaborative optimization method for carbon neutralization
CN113705911A (en) Thermoelectric load economical optimization distribution method based on wolf algorithm
CN113673851B (en) Provincial power grid day-ahead power generation planning method based on data driving
Kumari et al. Dynamic combined economic emission dispatch including wind generators by real coded genetic algorithm
Wu et al. Regional distributed photovoltaic short term power prediction method based on cluster analysis and stacking ensemble learning
Yuxin et al. Prediction of Boiler Control Parameters Based on LSTM Neural Network

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination