CN105205549B - Opportunity constraint planning-based optical storage system tracking day-ahead plan scheduling method - Google Patents
Opportunity constraint planning-based optical storage system tracking day-ahead plan scheduling method Download PDFInfo
- Publication number
- CN105205549B CN105205549B CN201510561197.0A CN201510561197A CN105205549B CN 105205549 B CN105205549 B CN 105205549B CN 201510561197 A CN201510561197 A CN 201510561197A CN 105205549 B CN105205549 B CN 105205549B
- Authority
- CN
- China
- Prior art keywords
- photovoltaic
- power
- energy storage
- storage system
- plan
- 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.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 33
- 238000003860 storage Methods 0.000 title claims abstract description 29
- 230000003287 optical effect Effects 0.000 title description 4
- 238000004146 energy storage Methods 0.000 claims abstract description 95
- 239000002245 particle Substances 0.000 claims abstract description 69
- 238000010248 power generation Methods 0.000 claims abstract description 18
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 16
- 238000005516 engineering process Methods 0.000 claims abstract description 12
- 230000003044 adaptive effect Effects 0.000 claims abstract description 7
- 238000013178 mathematical model Methods 0.000 claims abstract description 6
- 238000007599 discharging Methods 0.000 claims description 16
- 230000006870 function Effects 0.000 claims description 13
- 238000005457 optimization Methods 0.000 claims description 4
- 238000009826 distribution Methods 0.000 claims description 3
- 238000005315 distribution function Methods 0.000 claims description 3
- 238000012545 processing Methods 0.000 claims description 3
- 241000764238 Isis Species 0.000 claims description 2
- 150000001875 compounds Chemical class 0.000 claims description 2
- 238000009499 grossing Methods 0.000 claims description 2
- 238000005070 sampling Methods 0.000 claims description 2
- 230000000694 effects Effects 0.000 abstract description 12
- 238000004364 calculation method Methods 0.000 description 9
- 230000008859 change Effects 0.000 description 5
- 238000011217 control strategy Methods 0.000 description 5
- 238000004088 simulation Methods 0.000 description 5
- 238000000342 Monte Carlo simulation Methods 0.000 description 4
- 238000007781 pre-processing Methods 0.000 description 4
- 238000004458 analytical method Methods 0.000 description 3
- 238000011160 research Methods 0.000 description 3
- 230000009286 beneficial effect Effects 0.000 description 2
- 230000007613 environmental effect Effects 0.000 description 2
- 230000006872 improvement Effects 0.000 description 2
- 238000012797 qualification Methods 0.000 description 2
- 238000005096 rolling process Methods 0.000 description 2
- 230000002411 adverse Effects 0.000 description 1
- 230000002860 competitive effect Effects 0.000 description 1
- 230000007423 decrease Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000007812 deficiency Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000012983 electrochemical energy storage Methods 0.000 description 1
- 230000003203 everyday effect Effects 0.000 description 1
- 230000003631 expected effect Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 230000005855 radiation Effects 0.000 description 1
- 239000013589 supplement Substances 0.000 description 1
- 238000012795 verification Methods 0.000 description 1
Images
Landscapes
- Charge And Discharge Circuits For Batteries Or The Like (AREA)
- Supply And Distribution Of Alternating Current (AREA)
Abstract
The invention provides a light storage system tracking day-ahead plan scheduling method based on opportunity constraint planning, which comprises the following steps: (1) reading related data of the photovoltaic power station and the energy storage system; (2) simulating the actual photovoltaic power generation power in the day ahead by a Monte Carlo technology and making a photovoltaic planned output upper and lower limit range; (3) establishing an opportunity constraint planning mathematical model containing a control coefficient based on the short-term photovoltaic prediction power value and the random prediction deviation amount; (4) and determining the charge and discharge power of the energy storage system by adopting an improved adaptive particle swarm algorithm. According to the method, each forecast point is polled one day in advance, and the effect that the light storage tracking plan output is within the upper and lower limits of the plan is achieved by adopting an improved adaptive particle swarm algorithm; in addition, the tracking control is more flexible through adjusting the control coefficient in the objective function, the charge and discharge power and the charge state of the energy storage system are basically kept in the appropriate range, the charge and discharge capacity is improved, and the requirement on energy storage is reduced.
Description
Technical Field
The invention relates to a day-ahead plan scheduling method, in particular to a light storage system tracking day-ahead plan scheduling method based on opportunity constraint planning.
Background
Solar energy is recognized as one of the most competitive energy sources in the future, and has the characteristics of abundant resources, environmental protection and the like. According to the prediction of the International Energy Agency (IEA), the solar photovoltaic power generation accounts for 20% -25% of the global power generation amount by 2050 years and becomes one of basic energy. However, photovoltaic power generation is an intermittent energy source, is influenced by solar radiation intensity, environmental temperature and the like, has uncertainty in output power, generally causes adverse effects on power quality, power supply reliability and stability, power grid benefits and the like when grid connection is carried out, and through predicting output power of a photovoltaic power station, the photovoltaic power station power generation system contributes to coordinating and coordinating conventional energy sources and photovoltaic power generation by a power system dispatching department, timely adjusts a dispatching plan, reasonably arranges a power grid operation mode, effectively reduces the influence of photovoltaic access on a power system, and therefore improves stability and safety of power grid operation. However, at present, the photovoltaic prediction is influenced by natural conditions, so that the problems of overlarge prediction deviation and immature prediction technology still exist. In order to make up for the deficiency, the photovoltaic prediction error is reduced from the perspective of forming a light-storage combined system by using an energy storage technology to track planned output in time, and indirect improvement of prediction precision becomes a new research hotspot.
At present, a plurality of researches are carried out at home and abroad aiming at the photovoltaic prediction technology, wind storage conditions are mostly considered singly, analysis based on the light storage combined application is few, and especially, the research on the achievement of light storage tracking plan output aiming at the control of an energy storage system is less. In the prior art, a control strategy that the maximum approaching degree of wind and light storage capacity and fixed plan power is a target is provided, but the method ignores the upper and lower limit ranges of wind and light prediction, only one target plan can be fixedly formulated in each calculation, so that the flexibility of energy storage control is lost, and the requirements and the cost of an energy storage system are increased. The method can realize real-time tracking through a rolling change control coefficient, but only considers the ultra-short-term condition and the wind power range, and does not analyze the short-term condition and the photovoltaic tracking planned output.
Disclosure of Invention
In order to overcome the defects of the prior art, the invention provides a light storage system tracking day-ahead plan scheduling method based on opportunity constraint planning. The method aims at effectively controlling the energy storage system to limit the light storage capacity within a planned range, and flexibly controls the energy storage system to keep the charge-discharge power and the charge state within a proper range and reduce the requirements on the energy storage system to the maximum extent.
In order to achieve the purpose of the invention, the invention adopts the following technical scheme:
a light storage system tracking day-ahead plan scheduling method based on opportunity constraint planning comprises the following steps:
(1) reading related data of the photovoltaic power station and the energy storage system;
(2) simulating the actual photovoltaic power generation power in the day ahead by a Monte Carlo technology and making a photovoltaic planned output upper and lower limit range;
(3) establishing an opportunity constraint planning mathematical model containing a control coefficient based on the short-term photovoltaic prediction power value and the random prediction deviation amount;
(4) and determining the charge and discharge power of the energy storage system by adopting an improved adaptive particle swarm algorithm.
Preferably, in step (1), the relevant data includes: short-term photovoltaic prediction power value, energy storage system charge-discharge power upper and lower limit values and state of charge upper and lower limit values.
Preferably, the step (2) comprises the steps of:
step 2-1, simulating and processing the photovoltaic prediction deviation into a random variable ξ (t) by adopting a Monte Carlo technology;
step 2-2, taking the short-term photovoltaic prediction power value as a determination variable;
and 2-3, establishing an upper limit range and a lower limit range of photovoltaic planned output.
Preferably, in the step 2-1, the photovoltaic prediction deviation is a difference between a predicted power of photovoltaic power generation before the day and an actual photovoltaic power generation power of the day, and the actual photovoltaic power generation power Pact(t) is represented by the formula Pact(t)＝Ppre(t) + ξ (t) and Ppre(t) is the predicted power of photovoltaic power generation before the moment t day, the probability distribution function of the random variable ξ (t) adopts a method which satisfies that the mean value is 0 and the variance is sigma2(t) a normal probability distribution, σ (t) being represented by the formula σ (t) ═ 0.2Ppre(t)+0.02CapTo obtain CapIs the photovoltaic installed capacity.
Preferably, in step 2-3, the photovoltaic planned output upper and lower limits range is formulated according to ± 25% fluctuation based on the short-term photovoltaic predicted power value, and the photovoltaic planned output upper and lower limits are calculated by the following formula:
Plimit＝ξallow forCap
Pplan_up(t)＝Ppre(t)+Plimit
Pplan_dn(t)＝Ppre(t)-Plimit
In the formula, ξAllow forThe percentage value is an allowable percentage value of a prediction error in the day, and is 25% according to the specification; plimitIs ξAllow forDetermining a photovoltaic output fluctuation limit value; capIs the photovoltaic installed capacity; pplan_up(t) is the upper limit of the photovoltaic output, Pplan_dn(t) is a lower photovoltaic output limit value.
Preferably, the step (3) includes the following steps:
step 3-1, aiming at enabling the light storage output to be within the upper and lower limits of the planned output to the maximum extent, establishing an objective function as follows:
pplan_adj(t)＝u{Pplan_up(t)+c[Pplan_up(t)-Pplan_dn(t)]}
in the formula Pplan_adj(t) target control Power, Pplan_up(t) is the upper limit of the photovoltaic output, Pplan_dn(t) a lower limit value of a photovoltaic planned output force, f is a target function, u is a switching coefficient for controlling whether the stored energy works, u takes 1 to be in a working state of the energy storage system, u takes 0 to be in an idle state, and c is a target power control coefficient between 0 and 1; pbess(t) is a decision variable, namely the energy storage charge-discharge power at the moment t, Pbess(t) greater than zero indicates that the energy storage device is discharging and less than zero indicates charging;
step 3-2, establishing opportunity constraint planning conditions;
and 3-3, establishing constraint conditions of the energy storage system.
Preferably, in the step 3-2, the photovoltaic active power output is smoothed to ηeerIs less than or equal to the allowable range delta as an opportunity constraint planning condition, and the probability P of the establishment of the opportunity constraint planning conditionrNot less than the confidence level α, as follows:
ηeer＝[ppre(t)+pbess(t)+ξ(t)-pplan_adj(t)]/pplan_adj(t)
Pr{|ηeer|≤δ}≥α
in the formula, PrSmoothing out power η for photovoltaic active powereerA probability that is less than or equal to the allowable range δ.
Preferably, in step 3-3, the energy storage system constraint condition includes:
power constraints, i.e. charging, Pch.max≤Pbess(t) is less than or equal to 0, and P is less than or equal to 0 during dischargingbess(t)≤Pdis.maxIn the formula, Pch.maxA negative value is the maximum charging power of the energy storage system; pdis.maxPositive is the maximum discharge power of the energy storage system;
constraint condition of SOC, SOCmin≤SOC(t)≤SOCmax，
In the formula, SOCminThe minimum value of the charge state of the energy storage system; SOCmaxThe maximum value of the state of charge of the energy storage system; SOC (t) is the state of charge at time t; zeta is corresponding charge and discharge coefficient, zeta > 1 during discharging, show that there is certain loss in the course of discharging, zeta < 1 during charging, show that there is certain loss in the course of charging too; Δ t is the sampling time interval of the power; and C is the rated capacity of the energy storage system.
Preferably, in the power constraint, when P isact(t)＞Pplan_up(t) the energy storage system is in a charging state, i.e. Pch.max≤Pbess(t) is less than or equal to 0; when P is presentact(t)＜Pplan_dn(t) the energy storage system is in a discharge state, i.e. P is more than or equal to 0bess(t)≤Pdis.max。
Preferably, the step (4) includes the following steps:
step 4-1, setting parameters of a particle swarm optimization algorithm, comprising the following steps: the total number N of particle swarms, the number of iterations k, and the maximum number of iterations kmaxAn inertia constant omega and a learning factor c1And c2；
Step 4-2, initializing the position and speed of the particle swarm;
4-3, determining the fitness of each particle in the particle swarm according to the target function f;
4-4, comparing the fitness of each particle, and determining the individual optimal P of each particlebestOptimal P from all individualsbestTo determine global optimum Gbest(ii) a Updating the speed and the position of each particle according to the individual optimum and the global optimum;
step 4-5, calculating the fitness of each particle at the moment again according to the objective function f, and judging whether to update the individual optimal PbestAnd global optimum Gbest；
4-6, judging whether the search result reaches the iteration times, and if not, continuously updating the speed and the position of each particle; otherwise, stopping iteration and outputting the optimal solution.
Preferably, in step 4-4, the velocity and position of each particle are updated according to the following formula:
in the formula (I), the compound is shown in the specification,
respectively the speed and position of the ith particle from iteration to the kth generation;
respectively iterating to the speed and position of the ith particle of the (k + 1) th generation;
an individual extreme value for the ith particle for iteration to the kth generation;
is the global extremum of the particle group in the previous k generations; c. C1,c2The learning factor can accelerate convergence and avoid falling into local optimum; r is1,r2Is [0,1 ]]A random number in between; k is a constraint factor, and K is a constraint factor,
in the formula
Wherein k is the current particle iteration number; k is a radical ofmaxSetting the maximum iteration times for the particle swarm algorithm; omegamin、ωmaxMinimum and maximum inertial weights, respectively.
Preference is given toIn said step 4-5, said judging whether to update the individual optimum PbestAnd global optimum GbestThe method comprises the following steps: if true, then else, then
Compared with the prior art, the invention has the beneficial effects that:
the invention provides a light storage system tracking day-ahead plan scheduling method based on opportunity constraint planning, which realizes the effect that the output of a light storage tracking plan is in the upper and lower limit ranges of the plan by polling each forecast point one day in advance and adopting an improved adaptive particle swarm algorithm; in addition, the tracking control is more flexible through adjusting the control coefficient in the objective function, the charge and discharge power and the charge state of the energy storage system are basically kept in the appropriate range, the charge and discharge capacity is improved, and the requirement on energy storage is reduced.
Drawings
Fig. 1 is a flowchart of a method for tracking a planning and scheduling day ahead of a light storage system based on opportunity constraint planning provided by the invention.
FIG. 2 is a graph of photovoltaic predicted power, planned upper and lower limits, and simulated actual power at a certain day in an embodiment of the invention;
FIG. 3 is a graph showing the variation of the control coefficients c and u in a day under the condition of a fixed coefficient according to an embodiment of the present invention;
fig. 4 is a graph of the planned output effect of the optical storage tracking system when c is 0.5 under the condition of a fixed coefficient in the embodiment of the present invention;
fig. 5 is a graph of the planned output effect of the optical storage tracking system when c is 0.2 under the condition of the fixed coefficient in the embodiment of the present invention;
FIG. 6 is a graph showing the variation of the control coefficients c and u in one day under the condition of varying the coefficients in the embodiment of the present invention;
FIG. 7 is a graph of planned output effects of light storage tracking under varying coefficients in an embodiment of the present invention;
FIG. 8 is a graph of SOC variation under various conditions in an embodiment of the present invention;
FIG. 9 is a graph of the charging and discharging power of the energy storage system under the condition of the variation coefficient in the embodiment of the present invention;
FIG. 10 is a graph of algorithm convergence in an embodiment of the present invention;
Detailed Description
The present invention will be described in further detail with reference to the accompanying drawings.
In order to solve the problem that global consideration is lacked for charging and discharging of an energy storage system in the prior art, the embodiment of the invention provides a light storage system tracking day-ahead plan scheduling method based on opportunity constrained planning. The energy storage system mentioned in the method may be electromagnetic energy storage, mechanical energy storage, or electrochemical energy storage, and in this embodiment, a battery energy storage system is taken as an example for description, as shown in fig. 1, including the following steps:
the prediction time scale of the short-term photovoltaic power prediction value is 24h, 96 time intervals are counted, the prediction time resolution is 15min, namely: forecasting photovoltaic grid-connected power 24 hours in the future, wherein a forecasting point is one time every 15 minutes, and the forecasting is carried out in a rolling mode every day;
simulating the photovoltaic actual output at the day ahead by using a Monte Carlo technology and making a photovoltaic planned output upper and lower limit range;
for this reason, in this embodiment, the photovoltaic prediction deviation is processed into the random variable ξ (t) by adopting monte carlo simulation, the short-term photovoltaic prediction output is processed into the determined variable, and the photovoltaic actual output P is obtainedact(t) is represented by the formula Pact(t)＝PpreThe photovoltaic planned output upper and lower limit ranges are formulated according to the standard requirement and the +/-25% fluctuation according to the short-term photovoltaic power generation predicted value, and are calculated by the following formulas (1) to (3):
Plimit＝ξallow forCap(1)
Pplan_up(t)＝Ppre(t)+Plimit(2)
Pplan_dn(t)＝Ppre(t)-Plimit(3)
In the formula, ξAllow forThe percentage value is an allowable percentage value of a prediction error in the day, and is 25% according to the specification; plimitIs ξAllow forDetermining a photovoltaic output fluctuation limit value; capIs the photovoltaic installed capacity; ppreAnd (t) predicting the photovoltaic day-ahead power at the time t.
And (4) processing the photovoltaic prediction deviation into a random variable by adopting Monte Carlo simulation.
The probability distribution function of the random variable ξ (t) is taken to satisfy the conditions that the mean is 0 and the variance is sigma2(t) a normal probability distribution, σ (t) being represented by the formula σ (t) ═ 0.2Ppre(t)+0.02CapAnd (6) obtaining.
And 3, establishing an opportunity constraint planning mathematical model containing a control coefficient based on the short-term photovoltaic prediction power value and the random prediction deviation amount.
Aiming at enabling the light storage output to be within the upper and lower limits of the planned output to the maximum extent, establishing an objective function as follows:
pplan_adj(t)＝u{Pplan_up(t)+c[Pplan_up(t)-Pplan_dn(t)]} (4)
in the formula Pplan_adj(t) is target control power, f is a target function, u is a switching coefficient for controlling whether the energy storage works, 1 is taken as the energy storage system is in a working state, 0 is taken as an idle state, and c is a target power control coefficient between 0 and 1; pbess(t) is a decision variable, namely the energy storage charge-discharge power at the moment t, Pbess(t) is greater than zero, indicating that the energy storage device is discharging, less than zero.
Step 3.1, establishing opportunity constraint planning conditions;
taking the absolute value of the photovoltaic active power output smoothness rate not higher than the allowable range delta as an opportunity constraint planning condition, and enabling the probability of establishment not to be smaller than the confidence level α, as shown in formulas (6) to (7):
ηeer＝[ppre(t)+pbess(t)+ξ(t)-pplan_adj(t)]/pplan_adj(t) (6)
pr{|ηeer|≤δ}≥α (7)
step 3.2, establishing constraint conditions of the energy storage system
Power constraints, i.e. charging, Pch.max≤Pbess(t) is less than or equal to 0, and P is less than or equal to 0 during dischargingbess(t)≤Pdis.maxIn the formula, Pch.maxA negative value is the maximum charging power of the energy storage system; pdis.maxPositive is the maximum discharge power of the energy storage system;
constraint condition of SOC, SOCmin≤SOC(t)≤SOCmax，
In the formula, SOCminThe minimum value of the charge state of the energy storage system; SOCmaxThe maximum value of the state of charge of the energy storage system; SOCtThe state of charge at time t.
Step 3.3, improving the power constraint condition
When P is presentact(t)＞Pplan_up(t) the energy storage system is in a charging state, i.e. Pch.max≤Pbess(t)≤0；
When P is presentact(t)＜Pplan_dn(t) the energy storage system is in a discharge state, i.e. P is more than or equal to 0bess(t)≤Pdis.max。
Step 4, determining the charge and discharge power of the energy storage system by adopting an improved adaptive particle swarm algorithm
The specific calculation flow is as follows:
step 4.1, settingParameters of the particle swarm optimization algorithm comprise: the total number N of particle swarms, the number of iterations k, and the maximum number of iterations kmaxAn inertia constant omega and a learning factor c1And c2；
And 4.2, initializing the position and the speed of the particle swarm. Initializing particle speed, randomly selecting energy storage charging and discharging power corresponding to each time interval of each particle through energy storage system constraint conditions, verifying opportunity constraint planning conditions, repeating the step to initialize all particle positions if the conditions are met, and otherwise, continuously performing random selection until the verification is met.
And 4.3, determining the fitness of each particle through the objective function of the formula (5).
And 4.4, recording an extreme value. Comparing the fitness of each particle and determining the individual optimal P of each particlebestFrom all individual extrema PbestTo determine global optimum Gbest。
And 4.5, updating the speed and the position of each particle according to the individual extreme value and the global extreme value.
In the formulae (8) and (9),
respectively the speed and position of the ith particle from iteration to the kth generation;
respectively iterating to the speed and position of the ith particle of the (k + 1) th generation;
individual extremum for iterating to the ith particle of the k generation；
Is the global extremum of the particle group in the previous k generations; c. C1,c2The learning factor can accelerate convergence and avoid falling into local optimum; r is1,r2Is [0,1 ]]A random number in between; k is a constraint factor, and K is a constraint factor,
in the formula
Step 4.6, recalculating the fitness of each particle at the moment according to the objective function f, and judging whether to update the individual extreme value PbestAnd global extreme Gbest. If it is
Is established, then
If not, then,
then
Step 4.7, judging whether the search result reaches the iteration times, and if not, skipping to the step 45; otherwise, stopping iteration and outputting the optimal solution.
The embodiment of the invention also provides a day-ahead light-storage tracking planning system based on opportunity constraint planning, which comprises:
the data acquisition unit is used for reading related data of the photovoltaic power station and the energy storage system;
the data preprocessing unit is used for simulating the photovoltaic actual output at the day ahead and making a photovoltaic planned output upper and lower limit range by using a Monte Carlo technology;
the control module is used for establishing an opportunity constraint planning mathematical model containing a control coefficient based on the short-term photovoltaic prediction power value and the random prediction deviation amount;
and the calculation output module is used for determining the charging and discharging power of the energy storage system by adopting an improved self-adaptive particle swarm algorithm.
The data preprocessing unit further includes:
the first preprocessing unit is used for generating a prediction deviation random value and simulating the actual photovoltaic output in the day ahead;
the second preprocessing unit is used for calculating a fluctuation limit value according to the short-term photovoltaic prediction power and determining a planned upper limit range and a planned lower limit range;
the calculation output module further includes:
the setting module is used for setting parameters of the particle swarm optimization algorithm and comprises the following steps: the total number N of particle swarms, iteration number k, inertia constant omega and learning factor c1And c2；
The initialization module is used for initializing the position and the speed of the particle swarm;
the fitness calculation module is used for determining the fitness of each particle in the particle swarm;
an extreme value calculating module for comparing the fitness of each particle and determining the individual extreme value P of each particlebestAnd global extreme Gbest；
An updating module for updating the speed and position of each particle according to the individual extremum and the global extremum, recalculating the fitness of each particle at the moment, and judging whether to update the individual extremum PbestAnd global extreme Gbest；
The execution output module is used for judging whether the search result reaches the iteration times or not, and if not, continuously updating the speed and the position of each particle; otherwise, stopping iteration and outputting the optimal solution.
Example analysis
The method is characterized in that a wind-solar energy storage demonstration project is used as a background, short-term photovoltaic prediction data of a day in 7 months are selected as a case analysis object, the total installed capacity of photovoltaic power generation in the demonstration project is 40MW, the total installed capacity of energy storage is 20MW/70MW & h, the initial state of charge of an energy storage system is set to be 0.5, and the SOC is set to bemin＝0.3，SOCmax＝0.9。 PAnd (4) setting parameters in SO: population size 40, particle dimension 96, c1＝c21.4962, ω 0.7298, particle velocity range [ -3,3]The maximum number of iterations is 500. The number of Monte Carlo simulations was set to 1000.
FIG. 2 is a diagram of short-term photovoltaic prediction data of a certain day in 7 months of the exemplary project, according to actual physical conditions, the upper and lower limit output ranges of the solar photovoltaic scheduling plan are made according to +/-25% fluctuation of corresponding predicted output in the 'Specification', and actual photovoltaic output of 0-24 h on the previous day is simulated through a Monte Carlo simulation technology.
To verify the effectiveness and flexibility of the control strategy presented herein, the examples were compared with simulation calculations in the case of fixed coefficients and in the case of varying coefficients, respectively. The energy storage switch coefficient u is set to be 1 in the fixed coefficient condition, the energy storage system is kept in a working state all the time, the fixed target power control coefficient c is 0.5 and 0.2, simulation is respectively carried out as shown in fig. 3, the control strategy can effectively achieve the aim that the light storage combined output tracks the planned output, the actual output is basically limited in the upper and lower limit range of the plan under the supplement of the energy storage system, and the tracking effect is shown in fig. 4 and 5. In addition, although the tracking effect when c is fixed to 0.5 is obviously better than that when c is fixed to 0.2, in the SOC change curve, as shown in fig. 8, c is taken as 0.5, the energy storage system is in the working state in most of 96 periods, the discharge depth is deeper than that when c is 0.2, and finally the SOC lower limit in the calculation example is exceeded, which is not satisfactory. And when c is 0.2, the SOC variation trend of the energy storage system is basically reasonable, but the requirement on the energy storage system is high. In order to reduce the energy storage burden, on the basis of a fixed coefficient c being 0.2, according to the charging and discharging power constraint improvement condition, the energy storage system is only discharged in a time period when the actual photovoltaic output is lower than the planned lower limit, and is charged in a time period exceeding the planned upper limit, and the rest time periods are kept in an idle state to adjust a switching coefficient d as shown in fig. 6. As shown in fig. 7, the simulation results still achieve the expected effect.
As can be seen from fig. 8, the SOC variation curve of the energy storage system obtained by performing simulation after the variable coefficient is optimally adjusted when the fixed coefficient c is 0.2 is obviously optimal, the energy storage system operates only in a small part of time intervals after the variable coefficient, and the rest of the time intervals are in the idle state, which is very beneficial to prolonging the service life of the energy storage system. In addition, each charge and discharge is carried out within the upper and lower limits of the example SOC, a certain amount of discharge/charge is carried out before charge/discharge, and the charge and discharge capacity of the energy storage system is further improved.
In order to further verify that the simulation effect is better under the condition of the change coefficient, the probability that the prediction error is limited in the range under the conditions of the fixed coefficient (c is 0.2) after the stored energy is not added and the change coefficient is calculated specifically, and the prediction error is reduced and the degree of prediction accuracy is improved by comparing and analyzing various schemes. The results show that the energy storage requirement is reduced after the coefficient is changed, and the error can be reduced by 100% and limited within the qualified range, as shown in table 1.
TABLE 1 comparison of photovoltaic power error satisfaction requirement probabilities
| Condition of input energy storage | Without input of energy storage | Fixed coefficient case | Coefficient of variation situation |
| Probability within + -25% error band | 77.08％ | 98.96％ | 100％ |
Therefore, by comprehensively analyzing and tracking the planned output effect, the prediction accuracy degree and the working condition of the energy storage system are improved, the variable coefficient control strategy can be adopted as a reference scheme to control the energy storage of the day, and the specific charging and discharging power value of each time period is shown in fig. 9.
The population evolution process is shown in fig. 10, and it can be seen that the fitness value gradually decreases with the increase of the evolution algebra, which indicates that the light storage combined output curve and the target control power curve are closer and closer, and the fitness value reaches the optimum value and basically does not change when the iteration number reaches about 200 times, which indicates that the algorithm has good convergence.
Table 2 shows that when the confidence level is about 0.65, the error qualification rate is not much higher than that when no energy storage is added, which indicates that the tracking effect is not ideal, and the confidence level is above 0.75, and as the confidence level is increased, the qualification degree meeting the error requirement basically reaches above 90%, but the required total charge and discharge amount is gradually increased, the requirement on the energy storage system is more strict, and an appropriate confidence level can be selected for calculation according to the specific condition of the energy storage system in the actual tracking control.
TABLE 2 comparison of results of different confidence level calculations
According to the method for controlling the output of the day-ahead light storage combined tracking plan based on the opportunity constrained planning, the upper limit range and the lower limit range of the photovoltaic plan output are made according to the short-term predicted power, the randomness of prediction deviation is considered, the Monte Carlo technology is adopted to simulate the day-ahead actual power, the opportunity constrained planning mathematical model is established, the output of the energy storage system is solved by using the improved adaptive particle swarm algorithm, and the result shows that the output of the light storage combined tracking plan achieves a good effect. Meanwhile, feasibility of practical application of energy storage is considered, the tracking target power is adjusted at any time by setting a control coefficient, compared with a fixed coefficient control strategy, the method enables a day-ahead energy storage output control scheme to be more flexible, and requirements on an energy storage system are further reduced.
Finally, it should be noted that: the above embodiments are only for illustrating the technical solutions of the present invention and not for limiting the same, and although the present invention is described in detail with reference to the above embodiments, those of ordinary skill in the art should understand that: modifications and equivalents may be made to the embodiments of the invention without departing from the spirit and scope of the invention, which is to be covered by the claims.
Claims (7)
1. A light storage system tracking day-ahead plan scheduling method based on opportunity constraint planning is characterized in that,
the method comprises the following steps:
(1) reading related data of the photovoltaic power station and the energy storage system;
(2) simulating the actual photovoltaic power generation power in the day ahead by a Monte Carlo technology and making a photovoltaic planned output upper and lower limit range;
(3) establishing an opportunity constraint planning mathematical model containing a control coefficient based on the short-term photovoltaic prediction power value and the random prediction deviation amount;
(4) determining the charge and discharge power of the energy storage system by adopting an improved adaptive particle swarm algorithm;
the step (2) comprises the following steps:
step 2-1, simulating and processing the photovoltaic prediction deviation into a random variable ξ (t) by adopting a Monte Carlo technology;
step 2-2, taking the short-term photovoltaic prediction power value as a determination variable;
2-3, making an upper limit range and a lower limit range of the photovoltaic planned output;
in the step 2-1, the photovoltaic prediction deviation is a difference value between the predicted power of the photovoltaic power generation at the current day and the actual photovoltaic power generation power at the current day, and the actual photovoltaic power generation power Pact(t) is represented by the formula Pact(t)＝Ppre(t) + ξ (t) and Ppre(t) is the predicted power of photovoltaic power generation before the moment t day, the probability distribution function of the random variable ξ (t) adopts a method which satisfies that the mean value is 0 and the variance is sigma2(t) a normal probability distribution, σ (t) being represented by the formula σ (t) ═ 0.2Ppre(t)+0.02CapTo obtain CapIs the photovoltaic installed capacity;
in the step (3), the method comprises the following steps:
step 3-1, aiming at enabling the light storage output to be within the upper and lower limits of the planned output to the maximum extent, establishing an objective function as follows:
pplan_adj(t)＝u{Pplan_up(t)+c[Pplan_up(t)-Pplan_dn(t)]}
in the formula Pplan_adj(t) target control Power, Pplan_up(t) is the upper limit of the photovoltaic output, Pplan_dn(t) a lower limit value of a photovoltaic planned output force, f is a target function, u is a switching coefficient for controlling whether the stored energy works, the energy storage system is in a working state when u is 1, the energy storage system is in an idle state when u is 0, and c is a target power control coefficient between 0 and 1; pbess(t) is a decision variable, namely the energy storage charge-discharge power at the moment t, Pbess(t) greater than zero indicates that the energy storage device is discharging and less than zero indicates charging;
step 3-2, establishing opportunity constraint planning conditions;
3-3, establishing constraint conditions of the energy storage system, and in the step 3-2, outputting the photovoltaic active power with the smooth rate ηeerIs less than or equal to the allowable range delta as an opportunity constraint planning condition, and the probability P of the establishment of the opportunity constraint planning conditionrNot less than the confidence level α, as follows:
ηeer＝[ppre(t)+pbess(t)+ξ(t)-pplan_adj(t)]/pplan_adj(t)
Pr{|ηeer|≤δ}≥α
in the formula, PrSmoothing out power η for photovoltaic active powereerA probability that is established when the value is smaller than or equal to the allowable range δ; in step 3-3, the energy storage system constraint conditions include:
power constraints, i.e. charging, Pch.max≤Pbess(t) is less than or equal to 0, and P is less than or equal to 0 during dischargingbess(t)≤Pdis.maxIn the formula，Pch.maxA negative value is the maximum charging power of the energy storage system; pdis.maxPositive is the maximum discharge power of the energy storage system;
constraint condition of SOC, SOCmin≤SOC(t)≤SOCmax，
In the formula, SOCminThe minimum value of the charge state of the energy storage system; SOCmaxThe maximum value of the state of charge of the energy storage system; SOC (t) is the state of charge at time t; zeta is corresponding charge and discharge coefficient, zeta > 1 during discharging, show that there is certain loss in the course of discharging, zeta < 1 during charging, show that there is certain loss in the course of charging too; Δ t is the sampling time interval of the power; and C is the rated capacity of the energy storage system.
2. The scheduling method of claim 1, wherein in the step (1), the related data comprises: short-term photovoltaic prediction power value, energy storage system charge-discharge power upper and lower limit values and state of charge upper and lower limit values.
3. The dispatching method according to claim 1, wherein in steps 2-3, the planned upper and lower photovoltaic output limits are based on the short-term predicted photovoltaic power value+And (3) making 25% fluctuation, wherein the upper and lower photovoltaic planned output limits are calculated according to the following formula:
Plimit＝ξallow forCap
Pplan_up(t)＝Ppre(t)+Plimit
Pplan_dn(t)＝Ppre(t)-Plimit
In the formula, ξAllow forThe percentage value is an allowable percentage value of a prediction error in the day, and is 25% according to the specification; plimitIs ξAllow forDetermining a photovoltaic output fluctuation limit value; capIs the photovoltaic installed capacity; pplan_up(t) is the upper limit of the photovoltaic output, Pplan_dn(t) is a lower photovoltaic output limit value.
4. The scheduling method of claim 1 wherein the power constraint is when P isact(t)＞Pplan_up(t) the energy storage system is in a charging state, i.e. Pch.max≤Pbess(t) is less than or equal to 0; when P is presentact(t)＜Pplan_dn(t) the energy storage system is in a discharge state, i.e. P is more than or equal to 0bess(t)≤Pdis.max。
5. The scheduling method of claim 1, wherein the step (4) comprises the steps of:
step 4-1, setting parameters of a particle swarm optimization algorithm, comprising the following steps: the total number N of particle swarms, the number of iterations k, and the maximum number of iterations kmaxAn inertia constant omega and a learning factor c1And c2；
Step 4-2, initializing the position and speed of the particle swarm;
4-3, determining the fitness of each particle in the particle swarm according to the target function f;
4-4, comparing the fitness of each particle, and determining the individual optimal P of each particlebestOptimal P from all individualsbestTo determine global optimum Gbest(ii) a Updating the speed and the position of each particle according to the individual optimum and the global optimum;
step 4-5, calculating the fitness of each particle at the moment again according to the objective function f, and judging whether to update the individual optimal PbestAnd global optimum Gbest；
4-6, judging whether the search result reaches the iteration times, and if not, continuously updating the speed and the position of each particle; otherwise, stopping iteration and outputting the optimal solution.
6. The scheduling method according to claim 5, wherein in step 4-4, the velocity and position of each particle are updated according to the following formula:
in the formula (I), the compound is shown in the specification,
respectively the speed and position of the ith particle from iteration to the kth generation;
respectively iterating to the speed and position of the ith particle of the (k + 1) th generation;
an individual extreme value for the ith particle for iteration to the kth generation;
is the global extremum of the particle group in the previous k generations; c. C1,c2The convergence can be accelerated to avoid falling into local optimum for learning factors; r is1,r2Is [0,1 ]]A random number in between; k is a constraint factor, and K is a constraint factor,
in the formula
Wherein k is the current particle iteration number; k is a radical ofmaxSetting the maximum iteration times for the particle swarm algorithm; omegamin、ωmaxMinimum and maximum inertial weights, respectively.
7. The scheduling method of claim 5 wherein in step 4-5, said determining whether to update the individual optimal PbestAnd global optimum GbestThe method comprises the following steps: if it is
Is established, then
If not, then,
then
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510561197.0A CN105205549B (en) | 2015-09-07 | 2015-09-07 | Opportunity constraint planning-based optical storage system tracking day-ahead plan scheduling method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510561197.0A CN105205549B (en) | 2015-09-07 | 2015-09-07 | Opportunity constraint planning-based optical storage system tracking day-ahead plan scheduling method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN105205549A CN105205549A (en) | 2015-12-30 |
| CN105205549B true CN105205549B (en) | 2020-03-27 |
Family
ID=54953218
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510561197.0A Active CN105205549B (en) | 2015-09-07 | 2015-09-07 | Opportunity constraint planning-based optical storage system tracking day-ahead plan scheduling method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105205549B (en) |
Families Citing this family (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105680474B (en) * | 2016-02-22 | 2021-03-02 | 中国电力科学研究院 | Control method for restraining rapid power change of photovoltaic power station through energy storage |
| CN105809369B (en) * | 2016-03-31 | 2019-08-16 | 国电南瑞科技股份有限公司 | Consider the plan security check method a few days ago of new energy power Uncertainty distribution |
| CN106549378A (en) * | 2016-12-09 | 2017-03-29 | 国网江苏省电力公司金湖县供电公司 | It is a kind of to exert oneself probabilistic distribution coordinated dispatching method for distributed power source |
| CN106887858B (en) * | 2017-02-27 | 2021-09-10 | 中国电力科学研究院 | Energy storage system tracking planned output method and device for accessing new energy power generation |
| CN107465204B (en) * | 2017-08-31 | 2021-04-16 | 中国电力科学研究院 | Multi-battery pack power optimal distribution method and device in energy storage power station |
| CN109447372A (en) * | 2018-11-13 | 2019-03-08 | 广东电网有限责任公司 | One kind is avoided the peak hour load forecasting method and device |
| CN109995076B (en) * | 2018-12-12 | 2023-05-23 | 云南电网有限责任公司电力科学研究院 | Energy storage-based photovoltaic collection system power stable output cooperative control method |
| CN112994121A (en) * | 2020-12-07 | 2021-06-18 | 合肥阳光新能源科技有限公司 | New energy power generation power prediction deviation compensation method and system |
| CN114114909B (en) * | 2021-11-11 | 2024-03-22 | 海南师范大学 | Intermittent process 2D output feedback prediction control method based on particle swarm optimization |
| CN115207950B (en) * | 2022-07-27 | 2024-02-06 | 中国华能集团清洁能源技术研究院有限公司 | Random disturbance-based energy storage system control method and device |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7085660B2 (en) * | 2003-05-13 | 2006-08-01 | Siemens Power Transmission & Distribution, Inc. | Energy management system in a power and distribution system |
Family Cites Families (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2013086411A1 (en) * | 2011-12-09 | 2013-06-13 | The Aes Corporation | Frequency responsive charge sustaining control of electricity storage systems for ancillary services on an electrical power grid |
| US10079317B2 (en) * | 2013-07-15 | 2018-09-18 | Constantine Gonatas | Device for smoothing fluctuations in renewable energy power production cause by dynamic environmental conditions |
| CN103532143A (en) * | 2013-10-24 | 2014-01-22 | 东润环能(北京)科技有限公司 | New energy power generation system capable of making up power prediction accuracy |
| CN104779631B (en) * | 2014-12-31 | 2017-07-18 | 国家电网公司 | Energy storage tracking wind power output method of planning and its system based on the pre- power scale of wind-powered electricity generation |
-
2015
- 2015-09-07 CN CN201510561197.0A patent/CN105205549B/en active Active
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7085660B2 (en) * | 2003-05-13 | 2006-08-01 | Siemens Power Transmission & Distribution, Inc. | Energy management system in a power and distribution system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN105205549A (en) | 2015-12-30 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105205549B (en) | Opportunity constraint planning-based optical storage system tracking day-ahead plan scheduling method | |
| CN109462231B (en) | Load optimization scheduling method, system and storage medium for residential micro-grid | |
| CN104779631B (en) | Energy storage tracking wind power output method of planning and its system based on the pre- power scale of wind-powered electricity generation | |
| US11326579B2 (en) | Adaptive dynamic planning control method and system for energy storage station, and storage medium | |
| CN104993522B (en) | A kind of active distribution network Multiple Time Scales coordination optimization dispatching method based on MPC | |
| CN105162149B (en) | Generation schedule output method is tracked based on the light-preserved system that fuzzy self-adaption is adjusted | |
| CN106777487B (en) | A kind of photovoltaic plant containing energy-storage system is credible capacity calculation methods and system | |
| CN108875992B (en) | Virtual power plant day-ahead optimization scheduling method considering demand response | |
| CN109345019B (en) | Improved particle swarm algorithm-based micro-grid economic dispatching optimization strategy | |
| CN115693757A (en) | Photovoltaic energy optimization regulation and control method based on digital twinning technology | |
| CN105896575B (en) | Hundred megawatt energy storage power control method and system based on self-adaptive dynamic programming | |
| CN107069835B (en) | Real-time active distribution method and device for new energy power station | |
| CN110751365A (en) | Multi-target balanced scheduling method and system for cascade reservoir group | |
| CN111064192A (en) | Independent micro-grid capacity optimal configuration method considering source load uncertainty | |
| CN112836849A (en) | Virtual power plant scheduling method considering wind power uncertainty | |
| CN115423153A (en) | Photovoltaic energy storage system energy management method based on probability prediction | |
| CN105956708A (en) | Grey correlation time sequence based short-term wind speed forecasting method | |
| CN117674197B (en) | Frequency adjustment method, storage medium and equipment using virtual power plant active support | |
| CN109976155B (en) | Method and system for randomly optimizing and controlling interior of virtual power plant participating in gas-electricity market | |
| CN103440428A (en) | Method for determining self-adaption dynamic weight of combined prediction model for wind electricity power | |
| CN111080000A (en) | Ultra-short term bus load prediction method based on PSR-DBN | |
| CN111162551B (en) | Storage battery charging and discharging control method based on wind power ultra-short term prediction | |
| CN113872253A (en) | Pumped storage power station and new energy combined power generation optimal scheduling method and device | |
| CN114154790A (en) | Industrial park light storage capacity configuration method based on demand management and flexible load | |
| CN103679284A (en) | Accommodated wind power accessed fixed interval rolling scheduling method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |