CN106505635B

CN106505635B - Active scheduling model and scheduling system with minimum wind abandon

Info

Publication number: CN106505635B
Application number: CN201610835721.3A
Authority: CN
Inventors: 李焱; 杨金威; 邢海秋
Original assignee: Beijing E Techstar Co ltd
Current assignee: Beijing E Techstar Co ltd
Priority date: 2016-09-20
Filing date: 2016-09-20
Publication date: 2020-05-12
Anticipated expiration: 2036-09-20
Also published as: CN106505635A

Abstract

The invention provides an active scheduling model and a scheduling system with minimum wind curtailment, wherein the scheduling system comprises a unit role distribution module, a rolling scheduling module, a real-time scheduling module and an AGC control module; the active scheduling model with the minimum wind abandon is applied to the real-time scheduling module. Has the advantages that: according to the wind power real-time prediction result, the power generation requirement is corrected in real time, so that the output plan of each unit in the rest time period is corrected, the total output of the units is close to the actual power generation requirement step by step, the uncertainty of the day-ahead plan is reduced, and the output plan of each unit is more reasonable and more meaningful.

Description

Active scheduling model and scheduling system with minimum wind abandon

Technical Field

The invention belongs to the technical field of power dispatching, and particularly relates to an active dispatching model and an active dispatching system with minimum wind abandon.

Background

The economic dispatching of the power system refers to a dispatching method which achieves reasonable distribution of power generation loads among units at the most economic operation cost and ensures reliable power supply to users under the conditions of meeting network safety and power generation load balance. According to the difference of the optimization time interval, the optimization scheduling problem of the power system can be divided into two aspects: static optimized scheduling and dynamic optimized scheduling.

Static optimized scheduling refers to: the economic load of single operation discontinuous surface of the power system is optimized and distributed. The static optimization scheduling is mainly divided into two types in algorithm: the method comprises the following steps of classical economic dispatching based on the micro-increment rate of equal consumption and safe economic dispatching based on the optimal power flow.

Because the power system is a dynamic system in continuous operation, when large load demand change occurs in the system, the transitive capacity among all static scheduling results cannot be ensured due to the limitation of the adjusting capacity of the generator. Therefore, the continuous feasibility problem of the economic scheduling result, namely the dynamic economic scheduling problem needs to be researched.

The traditional open-loop dynamic scheduling mode optimizes the whole optimization period once at the initial optimization stage and completely transmits and executes the solution sequence, and the application effect of the scheduling mode in the traditional power system can basically meet the requirement due to high load prediction precision. However, after large-scale wind power is accessed, the wind power prediction precision is far lower than that of the traditional load, the wind power prediction error is remarkably increased along with the extension of optimization time, the difficult accurate prediction characteristic of wind power enables the deviation between the result of the traditional open-loop dynamic scheduling mode depending on the wind power prediction in the future and the actual power grid requirement to be large, and the scheduling mode needs to be improved urgently.

Disclosure of Invention

Aiming at the defects in the prior art, the invention provides a dry-state slag discharging device and method for a gasification furnace, which can effectively solve the problems.

The technical scheme adopted by the invention is as follows:

the invention provides an active scheduling model with minimum wind curtailment, which comprises the following optimization objective functions and optimization constraint conditions;

wherein:

r_ithe current adjustment cost of the generating unit of the conventional unit i is calculated;

△p_iadjusting the total output for the output of the ith conventional unit at the next moment to control the output;

w_jthe cost of abandoned wind power of a wind power plant is w for reducing abandoned wind_jIs far greater than the current adjustment cost r of the generating unit of the conventional unit in value_i；

The abandoned wind power of the wind power field j is equal to the predicted value of the wind power output predicted in the next period

And real-time scheduling plan value of next time period

A difference of (d);

N_G ^cagcthe units are dispatched in real time for the whole network, and the number of the wind turbine units is not included

N_G ^WindFor wind turbine generator systemCombining;

the current output value of the wind power plant j is obtained;

△ P is the total output adjustment amount of the real-time scheduling unit at the next moment:

wherein the content of the first and second substances,

is the increment of the predicted value of the ultra-short-term load,

Is the next time increment of the tie line plan,

△ P is the next moment increment of the planned unit outputⁿ _AGCThe last moment of AGC is unfinished quantity;

M_intrepresenting a set of network-wide lines and intranet safe power transmission profiles,

the upper limit of the power transmission of the cross section, _jTlower limit of power transmission of the cross section, T_jThe inequality constraint ensures that the transmission section is not overloaded, which is the current transmission power of the section;

S_ijadopting load balance sensitivity, wherein in order to achieve partition balance, partition load prediction information needs to be introduced into bus load factors;

S_kjis the load balance sensitivity, △ P_k ^wThe product of the abandoned wind electric quantity of the wind power plant k and the abandoned wind electric quantity of the wind power plant k represents the real-time influence of the abandoned wind electric quantity of the wind power plant on the section power;

△C_gjinfluence of planned adjustment quantity of a non-real-time scheduling unit on section power;

and solving the objective function to obtain a predicted value of the output of the future wind field and the single machine set.

The invention also provides a scheduling system applying the active scheduling model with the minimum wind curtailment, which comprises a unit role distribution module, a rolling scheduling module, a real-time scheduling module and an AGC control module;

the active scheduling model with the minimum wind abandon is applied to a real-time scheduling module;

the unit role allocation module divides the unit roles by adopting the following steps:

step 1, firstly, the probability of ACE falling in each control section is counted according to historical data, the probability of ACE falling in a dead zone is pro1, the probability of ACE falling in a normal zone is pro2, the probability of ACE falling in a cooperative zone is pro3, and the probability of ACE falling outside an emergency zone and an emergency zone is pro4, then:

step 2, the AGC units participating in adjusting ACE meet the requirement of total rotation standby at each time interval, so N AGC units participating in ACE control are totally arranged in N units in the system, N is more than or equal to 1 and less than or equal to N, and a set formed by the N units is marked as SetA; the lower limit of the spinning reserve is set to SRt according to the actual operation condition of the power grid, and the value of the spinning reserve is necessarily greater than ACEE;

the n traditional units belonging to the set SetA have 4 roles to choose from, namely: the control module corresponding to the unit role 1 is used for deviation adjustment; the control mode corresponding to the unit role 2 is a tracking real-time plan; the control mode corresponding to the unit role 3 is a tracking rolling plan; the control mode corresponding to the unit role 4 is a tracking day-ahead plan;

the variable RoleID represents the role of the unit, the values are 1,2,3 and 4, and the corresponding control modes are respectively as follows: deviation adjustment, tracking real-time plan, tracking rolling plan and tracking day-ahead plan; the corresponding unit roles are respectively as follows: an AGC unit, a real-time scheduling unit, a rolling planning unit and a day-ahead planning unit;

constructing 4 vectors Role (1), Role (2), Role (3) and Role (4) mapped with the RoleiD vector according to the RoleiD vector, wherein the 4 vectors are used for storing AGC unit subscripts with roles of 1,2,3 and 4 respectively;

the objective function thus constructed to the following optimization problem is:

wherein: p_it: the output value of the unit i at the moment t;

a_i: coefficients of quadratic terms of the nonlinear relationship;

b_i: first order coefficient of non-linear relation;

c_i: a constant term of the non-linear relationship;

d: the current output value is worth correcting the coefficient;

the objective function ensures that the expectation of the total ACE adjustment cost of all AGC units belonging to role (j) in one day is minimum;

after the role of the unit is allocated, it is necessary to ensure that the ACE has sufficient AGC adjustment margin when falling in each area, so that the following constraints are generated:

s_itthe rotation of the ith unit at the time t is reserved;

and solving the objective function under the constraint condition to obtain the finally determined AGC role.

The rolling scheduling module is configured to: rolling and correcting the generation planned output power of the day-ahead unit according to the load prediction result of the power grid extended short-term model and the extended short-term wind power prediction result on the basis of a day-ahead plan so that the total output power generated by the system gradually approaches the actual power generation requirement to obtain the economically optimal generation planned output of the unit, and acting the economically optimal generation planned output of the unit on the rolling plan unit;

the real-time scheduling module is used for: the method comprises the steps that the economically optimal unit power generation planned output is used as base point power, the unit output is adjusted according to a power grid ultra-short period load prediction result and an ultra-short period wind power prediction result, a real-time scheduling correction plan instruction for performing minimum unit output adjustment on a system state in a planning time period is generated, the real-time scheduling correction plan instruction acts on a real-time scheduling unit, and therefore the unbalance caused by power unbalance and random variation of wind power load in the economically optimal plan of the unit is eliminated;

the AGC control module is used for: and taking the real-time scheduling correction plan instruction given by the real-time scheduling module as a control base point, correcting the random prediction error generated in the advanced prediction link in real time, generating an AGC unit output adjustment instruction for controlling the AGC unit, and issuing the AGC unit output adjustment instruction to the AGC unit to realize the control of the AGC unit.

The active power dispatching model and the dispatching system with the minimum abandoned wind have the following advantages that:

according to the wind power real-time prediction result, the power generation requirement is corrected in real time, so that the output plan of each unit in the rest time period is corrected, the total output of the units is close to the actual power generation requirement step by step, the uncertainty of the day-ahead plan is reduced, and the output plan of each unit is more reasonable and more meaningful.

The dispatching model is essentially that a rolling plan dispatching and real-time dispatching plan stage is added between a day-ahead power generation plan and AGC power generation control, and a technical support link of intelligent decision and self-adaptive coordination control is established in the stage, so that the traditional manual regulation mode is replaced, the labor intensity of a dispatcher on duty is reduced, wind power is consumed to the maximum extent, and high-quality power supply is realized.

Drawings

Fig. 1 is a schematic structural diagram of a scheduling system 1 according to the present invention;

FIG. 2 is a schematic diagram of a scheduling system of the present invention in the 2 nd structure;

FIG. 3 is a schematic diagram of a scheduling system of the present invention in type 3;

FIG. 4 is a comparison graph of a specific actual load and planned load curve provided by the present invention;

FIG. 5 is a diagram illustrating the arrangement of the defined prediction concepts according to the prediction period;

fig. 6 is an ACE partitioning diagram.

Detailed Description

In order to make the technical problems, technical solutions and advantageous effects solved by the present invention more clearly apparent, the present invention is further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

For the convenience of understanding the present invention, related technical terms are first introduced:

(1) power dispatching:

the power dispatching is an effective management means which is adopted for ensuring safe and stable operation of a power grid, reliable external power supply and orderly operation of various power production works. The specific work content of power dispatching is that according to data information fed back by various information acquisition devices or information provided by monitoring personnel, the actual operation parameters of the power grid, such as voltage, current, frequency, load and the like, are combined, the development conditions of various production works are comprehensively considered, the safe and economic operation states of the power grid are judged, operation instructions are issued through a telephone or an automatic system, and field operators or automatic control systems are instructed to adjust, such as adjusting the output of a generator, adjusting load distribution, switching capacitors, reactors and the like, so that the continuous safe and stable operation of the power grid is ensured.

(2) And (3) load prediction:

the load prediction is to determine load data of a certain future moment according to various factors such as the operating characteristics, capacity increase decision, natural conditions, social influence and the like of a system under the condition of meeting a certain precision requirement, wherein the load refers to the power demand (power) or the power consumption; load prediction is an important content in economic dispatch of a power system and is an important module of an Energy Management System (EMS).

(3) And (3) power generation planning:

the power generation planning is the power generation power of each unit which is planned in advance by combining constraint conditions such as the upper and lower limits of the output force of each unit and the maximum climbing power of each unit and considering actual conditions such as the start-stop time and the minimum shutdown time of each unit on the premise of meeting the power balance according to load prediction.

(4)AGC：

Automatic Generation control (agc) is an important function in an Energy Management System (EMS), and controls the output of a frequency modulation unit to meet the changing power demand of a user and to keep the system in an economical operation state.

(5) And (3) rolling planning:

scrolling planning (also known as sliding planning) is a method of dynamically planning. It does not recompile the plan for the next epoch after the plan has been completely executed, as in the static analysis, but rather advances the plan one epoch forward in time, i.e., rolls forward once, each time the plan is compiled or adjusted.

Background of the invention

(1) The need for day-ahead planning

As time goes by, the influence of uncertainty of the power generation demand prediction will increase, and the accuracy of the power generation demand prediction will gradually decrease. Therefore, the rolling correction can be carried out on the power generation requirements of the rest time periods after each time period in one day in real time, so that the output plans of the units in the rest time periods are corrected in a rolling manner, and the total output of the units is closer to the actual power generation requirements. Therefore, the output plan of each unit can be more reasonable, and the rolling scheduling is to continuously correct the future plan.

(2) Need for real-time scheduling

The realization of active power balance and safety control is the ultimate goal of power grid dispatching, so that advanced intelligent real-time dispatching provides more reasonable scientific basis for dispatchers, and the correctness of power grid dispatching command is ensured.

(3) Need for load prediction

In order to improve the wind power acceptance of a power grid, wind power prediction and system load prediction are basic work, but the prediction error of a wind power prediction result is increased along with the increase of prediction time, so that a unit day-ahead plan made by the day-ahead wind power prediction and day-ahead system load prediction has larger deviation with the actual power generation requirement of the second day and the actual power generation capacity of wind power, and if the day-ahead plan result can be corrected in a rolling mode according to the more accurate day-ahead rolling wind power prediction result of the second day, the day-ahead wind power prediction deviation and the system load prediction deviation can be further eliminated, and the purpose of absorbing large-scale wind power is achieved.

The invention aims to:

a multi-dimensional power grid optimization scheduling decision technology adopting closed-loop feedback dynamic adjustment is based on the actual output of a unit, bus load, power grid running state and network topology structure, multi-time-interval multi-constraint safety check and blocking management are carried out, multi-objective rapid optimization decision of a scheduling plan is realized, the system calculation result not only ensures that the power and electric quantity balance and the line and section flow are not out of limit, but also realizes the functions of line N-1 scanning, fault set scanning and self-adaptive adjustment.

Finally, the aims of year-month power and electricity quantity plan balance, month-week scheduling plan decomposition management, day-ahead scheduling plan optimization compilation, intra-day scheduling plan rolling adjustment and real-time active power balance and coordination control are fulfilled; the construction of a water, fire, wind and nuclear combined optimization scheduling system is actively promoted, the influence of the thermal power peak regulation capacity is quantized, wind power is brought into the day-ahead and day-within power and electric quantity balance, and the probability characteristic of wind power prediction is fully considered, so that the target of the maximum wind power absorption capacity is met.

Key technology for optimizing scheduling system provided by the invention

(1) Optimized scheduling control mode with multi-time scale coordination

At present, the domestic scheduling mode mainly adopts two time scale scheduling modes of day-ahead optimized scheduling plan and AGC control, the time scale span is large, the scheduling mode is extensive, and the scheduling mode cannot adapt to the scheduling of a power grid after large-scale wind power is accessed.

According to the response capability of the unit and the current situation of scheduling production, the active scheduling strategy is divided into a daily level, a 1-hour level, a 15-minute level and a second level in the time dimension. According to the characteristics of load fluctuation and the control characteristics of the unit, the control can be divided into four stages: rolling scheduling of day-ahead plan and rolling plan, real-time scheduling plan and AGC control. Fig. 2 is a diagram of the overall control mode of the system, and fig. 1 is a schematic diagram of the 1 st structure of the scheduling system provided by the present invention; fig. 3 is a schematic diagram of a3 rd structure of a scheduling system provided by the present invention; the method is characterized by comprising the following steps:

1) the plan has enough time to carry out dynamic optimization calculation, and the control of the time level takes safety as constraint and economic as target, and can be called as optimal control.

2) The rolling schedule is based on short-term prediction, and the rolling schedule with 1 hour as a starting period fully utilizes the latest information to correct the schedule in the day ahead, so that the uncertainty is gradually reduced.

3) When the real-time scheduling plan is implemented, uncertain factors such as the fact that an operation point is close to the edge of a security domain, a unit does not effectively track the plan, and the capacity of an AGC unit is insufficient need to be faced. And selecting a unit with good performance and good execution plan in the system as a buffer unit, and adjusting output by taking 15 minutes as a period through ultra-short-term prediction to eliminate the uncertain factors. The buffer unit takes safety as a first target and economy as a second target, so that on one hand, the amount of power unbalance in the optimal control process is absorbed, the operation safety is improved, and the normal operation of the optimal control link is ensured; and on the other hand, an adjusting space is reserved for the second-level AGC unit, so that the normal operation of an AGC link is ensured.

4) AGC control, which handles the situation occurring at that time in real time, including corrective control and safety corrective control (congestion management). The correction control is to schedule a second-level AGC unit to enable the frequency and the tie line power to meet CPS (control performance Standard) assessment indexes; and safety correction control is used for immediately processing line section tidal current out-of-limit. The aim of AGC control is to quickly eliminate potential safety hazards and ensure the frequency quality of a system.

The scheduling mode is essentially characterized in that a rolling plan scheduling and real-time scheduling planning stage is added between a day-ahead power generation plan and AGC power generation control, and a technical support link of intelligent decision and self-adaptive coordination control is established in the rolling plan scheduling and real-time scheduling planning stage to replace a traditional manual adjustment mode, so that the labor intensity of a duty scheduler is reduced, wind power is consumed to the maximum extent, and high-quality power supply is realized.

In this control mode, all units are classified into four categories: day-ahead planning units, rolling scheduling units, coordinating units (real-time scheduling units) and manual fixing units. The day-ahead planning unit is executed strictly according to the day-ahead plan, the rolling scheduling unit is executed according to the plan after rolling correction, the coordination unit is responsible for the balance of load power change with large regularity and amplitude, and the control period is 15 min. Therefore, the AGC unit is kept with a larger adjustment margin by adjusting the output of the coordination unit, and the safety and the economy of system operation are improved; meanwhile, the wind power is consumed to the maximum extent through the adjustment of the real-time control unit, and the wind power utilization rate is improved.

(1) Extended short term load prediction

And in the rolling scheduling link, the execution condition of the power generation plan on the current day needs to be monitored, and under the condition that the original plan is seriously deviated from the actual load, the re-prediction of the load in the rest period of the day and the adjustment of the power generation plan are completed in time. Referring to fig. 4, a specific comparison graph of actual load and planned load curves is shown. The solid line in fig. 4 is the graph of the actual load operation of the system observed on a grid at 10:00 am on 11 months and 1. The projected load, shown in dashed lines, was predicted at 11:00 am the previous day. As shown in the figure, under the influence of load sensitive factors such as climate, the daily load deviates from the planned curve from the actual load operation curve of 9:00, and the deviation tends to be larger. In this case, a large load prediction error may be caused without correcting the daily load plan. And by using the newly obtained information, the load of the latter half day is predicted again, and the plan curve of the latter half day is adjusted, so that the deviation between the plan and the actual can be recovered to the greatest extent, and the load prediction error is reduced.

In order to meet the application requirements, an extended short-term load prediction concept is proposed: the load of unknown 1 hour to many hours after the current time of the day is predicted by using the latest information (including load information, weather information, electricity price and the like) which can be obtained at present.

From the prediction period, the expansion short-term load prediction is between the ultra-short-term and short-term load prediction. Fig. 5 shows a schematic diagram of the arrangement of the defined prediction concepts according to the prediction period.

The main application of extended short-term load forecasting and short-term load forecasting is to make daily load plans, the former is the extension of the latter on the forecasting cycle, and table 1 contrasts the main difference between the two (taking 96 sampling points per day as an example).

TABLE 1

The extended short-term load prediction uses not only historical information but also latest load, weather, fault, plan information and the like in the current day, so that the prediction accuracy can be improved. In summary, the goal of extended short-term load prediction is to make a reasonable and efficient prediction of the remaining load data of the day given the partial load data of the day. The load change situation can be predicted by tracking and detecting the change of the load-related factors. Under the condition that the actual load curve is predicted to be seriously deviated from the original planned curve, the method can be used for starting and expanding the short-term load prediction in advance, improving the accuracy of the load prediction and also being an important link for realizing the rolling scheduling of the power generation.

(2) Ultra short term load prediction

Ultra-short term load forecasting (from minutes to an hour) is a precondition for real-time active scheduling, and ensuring the precision is a key for realizing real-time scheduling. The ultra-short term load forecasting is characterized by short forecasting period and the key technology is precision. Statistical information that can be applied to the data as much as possible is required in the forecasting method. Compared with short-term and even medium-term load forecasting, the method has the following characteristics:

1) the forecasting period is short, so that the online operation is required, and the calculation time is higher;

2) in the concept of 'ultra-short term', the influence of weather change, holidays and double holidays is not obvious;

3) the load curve is not as stable as short-term, the higher harmonic component is relatively more, and the amplitude is large

The invention adopts an ultra-short-term load prediction method based on the similarity of load curve section shapes. And automatically matching similar time periods to predict according to the historical load output data. Different from the traditional method for selecting similar days according to the similarity of the values of the load curves, the method for dynamically selecting similar days according to the morphological similarity of the load curves can better improve the precision of ultra-short-term load prediction, particularly the prediction precision at the turning points.

(3) Rolling scheduling technique

In the day-ahead plan, the influence of uncertain factors of power generation demand prediction, particularly wind power output prediction, is increased along with the time, and the accuracy of the power generation demand and the wind power output prediction is gradually reduced, so that the rationality and the practicability of the planned output of the unit are influenced. Due to the fact that the wind power prediction error is increased along with the prediction time, the rolling scheduling of wind power is not suitable for adopting a too long time window, and the rolling wind power prediction result of the wind power plant is considered to give a predicted value of 4 hours in the future every 15 minutes, so that if rolling correction can be carried out on the power generation requirement of 4 hours in the future after each time period in one day in real time according to the wind power rolling prediction result, the output plans of all the units in the rest time period are corrected in a rolling mode, the total output of the units is close to the actual power generation requirement step by step, the uncertainty of the day-ahead plan can be reduced, the output plans of all the units are more reasonable, and the significance is higher. Therefore, we can consider rolling scheduling as a process of continuously revising and continuously refreshing the day-ahead plan.

The rolling scheduling is dynamic optimization from the current time period to the end time period, is a difficult problem in mathematics, and is complex and time-consuming in model. Therefore, how to obtain a practical optimization model suitable for the online application of the rolling scheduling link by decoupling and coordinating the time dimension and the space dimension of the dynamic optimization model needs to be researched. This puts requirements on the efficiency of the rolling scheduling algorithm; secondly, due to uncertainty caused by daily load fluctuation, the algorithm and the optimization model thereof also need to have good robustness.

When the rolling scheduling is established, not only the economic benefits of energy conservation and emission reduction need to be considered, but also the feasibility of the output of each unit in the rest time period is ensured, including meeting the unit climbing rate constraint, meeting the power generation-load power balance constraint, the network safety constraint and the like.

The time span of the mathematical model of rolling scheduling is [ t ]₀+1,T]The goal is to minimize the total cost of the system for some time in the future, which can be expressed as:

the online rolling optimization scheduling model comprises an optimization objective function and optimization constraint conditions;

the time span of the mathematical model of the online rolling plan is [ t ]₀+1,T]The goal is to minimize the total cost of the system over a future period of time, and in particular, may be

Wherein: i and j are respectively the serial numbers of the conventional unit and the wind generating set, and the value range is i belongs to [1, N ]],j∈[1,M](ii) a Wherein N is the total number of the conventional units; m is the total number of the wind turbine generators; t is a time interval number, and the value range is t epsilon [ t₀+1,T]；t₀To optimize the starting period; t is the length of the optimized time interval; p is a radical of_it,

Respectively the planned output values of the ith conventional unit and the jth wind turbine unit in the rolling plan in the t time period,

is at the t₀Predicting the predicted maximum power value of the jth wind turbine generator set in the tth time period in the time period; a is_i,b_i,c_iCoefficient of quadratic term, first term and constant term of generating cost of ith conventional unit, and lambda_jA wind curtailment cost factor for the wind turbine generator; when lambda is_jAnd b_iThe minimum wind abandon can be realized by the same order of magnitude and the positive value;

the optimization constraint conditions comprise:

(1) the unit output upper and lower bound constraint conditions are as follows:

wherein pa_it,pi_itRespectively an upper bound and a lower bound of the output of the ith conventional unit in the t-th time period; when a certain unit stops at a certain moment, the maximum output force and the minimum output force of the unit at the moment are set to be zero values;

(2) unit ramp rate constraint

p_i,t-1-△pd_it≤p_it≤p_i,t-1+△pu_it(3)

Wherein △ pd_it,△pu_itThe maximum value of the lowering force and the maximum value of the raising force allowed from the t-1 period to the t period for the ith conventional unit; p is a radical of_i,t-1A force value is planned for the ith conventional unit in the rolling plan in the t-1 th time period; the wind turbine generator is not limited by the climbing rate constraint;

(3) safety restraint of section tidal current

Wherein L and L respectively represent the number of the sections and the total number of the sections; g_li,

Sensitivity factors of the ith conventional unit and the jth wind turbine unit on the ith section can be obtained through admittance matrixes corresponding to direct current flow; _ltTL,

the minimum value and the maximum value of the section tidal current are obtained;

(4) load balancing constraints

Wherein D is_tForce values are plotted for the total;

because the constraint corresponding to the formula (2) and the formula (3) only contains the information of a single unit, the constraint is defined as a unit non-coupling constraint; the constraint of formula (4) and formula (5) includes information of a plurality of units, and thus the constraint is defined as a unit coupling constraint.

On-line rolling optimization scheduling model and plan output value p of conventional unit_itPlanned output value of wind turbine

For variables, equation (1) is the optimization objective function, and equations (2), (3), (4), (5) are the optimization constraints.

The rolling correction scheduling is an on-line application, and an algorithm of the rolling correction scheduling should have high calculation efficiency and strong robustness. In the invention, a Lagrange dual method is adopted for solving. Of course, in practical application, various methods can be flexibly adopted to solve the optimized objective function, which is not limited by the present invention.

(3) Real-time scheduling technique

1) Relation between real-time scheduling and other scheduling links

The load curve for a region can be decomposed into a fixed component, a trend component and a random component. The units are therefore classified into four categories for the three components, namely, a day-ahead planning unit, a rolling planning unit, a real-time unit and an AGC unit. The day-ahead and rolling planning unit is responsible for short-term scheduling plan execution, the real-time unit makes and executes a real-time plan according to ultra-short-term load prediction, and the AGC unit is responsible for correction control of fluctuating load.

In addition, the scheduling model provided by the invention also comprises a unit role distribution module, wherein the unit role distribution module is used for distributing the roles of the whole network unit in real time, and the unit roles comprise a day-ahead planning unit, a rolling planning unit, a real-time planning unit and an AGC unit;

the unit role distribution module adopts the following method to determine the roles of all units in the whole network at preset time intervals:

step 1, according to the deviation degree of the real-time frequency of the power grid and the normal set frequency, dividing the real-time frequency of the power grid into 4 control sections in sequence from light deviation degree to heavy deviation degree, wherein the control sections are respectively as follows: dead zone, normal zone, auxiliary zone and cooperation zone;

specifically, in order to ensure that AGC is smooth, stable, and effectively realizes real-time balance of power supply and demand, and avoid overshoot or undershoot in the process of reducing ACE, a Control Zone (Control Zone) needs to be divided. The control section is used to indicate the severity of ACE, and includes a Dead Zone (Dead Band Zone), a Normal Zone (Normal Zone) also called a Command Zone (Command Zone), an auxiliary Zone (Assist Zone) also called a permission Zone, and a Cooperation Zone (collaboration Zone) also called an Emergency Zone. Wherein the ACE partitioning diagram refers to fig. 6:

step 2, respectively determining the dead zone boundary value ACE through the following formula_DNormal zone boundary value ACE_NAuxiliary area boundary value ACE_AAnd a collaboration zone boundary value ACE_E：

Wherein: b is_i: a frequency deviation coefficient set in a control area is a positive sign in unit MW/0.1 HZ;

ε₁: the interconnected network controls the root mean square control target of the frequency average deviation of one minute all year round;

L₁₀: a control limit for the absolute value of the ten minute ACE average;

LOSS: a destabilizing power;

according to specific values of ACE, the coordination control strategy of each AGC unit is shown as the following table:

AGC set coordination control strategy table

In the table: "not controlled" means not making any adjustments; "bias adjustment" means to leave the baseline value and participate in ACE adjustment to promote ACE reduction. "base point close" means that the base point adjustment is performed directly, without considering the influence on ACE; "condition return" means that whether influence is caused on ACE or not when base point adjustment is performed, and if ACE is increased due to approaching to a base point value, the ACE is kept still; if approaching the base point value would cause ACE to decrease, a change is made.

The base point value is set in various modes, and the base point value adopted by the invention is a planned base point, namely the base point value is a continuous curve.

And 3, sequencing all the units in the whole network according to the unit performance, and recording the units as follows according to the sequence of the unit performance from high to low: the unit 1 and the unit 2 … obtain a sequencing list;

selecting the minimum m value meeting the following conditions in the sorting table to obtain the following serial numbers: m AGC units of unit 1 and unit 2 …; and the AGC unit is adjusted according to a control strategy of deviation adjustment:

wherein: cap_agc: AGC deviation adjustment capacity;

P_i,max: the maximum output value of the unit i;

P_i,min: the minimum output value of the unit i;

and 4, selecting the numbers as follows in sequence: the n units of the unit m +1 and the unit m +2 … unit n are used as real-time planning units, the real-time planning units are controlled according to a tracking real-time planning strategy, wherein n is the minimum value meeting the following constraints:

due to p_iThe output of (a) is a continuously varying value, and thus n is also a dynamically varying value;

and 5, selecting k units with the serial numbers of the unit n +1 and the unit n +2 … unit n + k as rolling planning units, and controlling the rolling planning units according to a tracking rolling planning strategy, wherein k is the minimum value meeting the following constraints:

and 6, the rest units in the ranking list are day-ahead planning units, and the day-ahead planning units are adjusted according to a control strategy for tracking the day-ahead plan.

The allocation of the roles of the unit can be automatically counted and selected through a program, in order to avoid frequently changing the roles, △ t can be classified once every a period of time, and for the unit adjusted according to deviation, the unbalanced power is proportionally allocated according to the requirement of energy-saving scheduling and the size of a coal consumption coefficient.

In addition, the unit role allocation module may also determine the unit role by using the following method:

step 1, firstly, the probability of the ACE falling in each control section is counted according to historical data, the probability of the ACE falling in a dead zone is pro1, the probability of the ACE falling in a normal zone is pro2, the probability of the ACE falling in a cooperative zone is pro3, and the probability of the ACE falling outside an emergency zone and an emergency zone is pro4, then:

step 2, the AGC units participating in adjusting ACE must meet the requirement of total rotation standby at each time interval, N AGC units participating in ACE control in N units in the system are not arranged (N is more than or equal to 1 and is less than or equal to N), and a set formed by the N AGC units is marked as SetA; the lower limit of the spinning reserve is set according to the actual operation condition of the power grid, and is not defined as SRt, and the value of the spinning reserve is necessarily greater than ACEE (namely 0.8 LOSS);

the n traditional units belonging to the set SetA have 4 roles to choose from:

unit role and control mode

The variable RoleID represents the role of the unit, the values are 1,2,3 and 4, and the corresponding control modes are respectively as follows: deviation adjustment, tracking real-time plan, tracking rolling plan and tracking day-ahead plan; the corresponding unit roles are respectively as follows: an AGC unit, a real-time unit, a rolling planning unit and a day-ahead planning unit;

4 vectors Role (1), Role (2), Role (3) and Role (4) mapped with the RoleiD vector can be constructed according to the RoleiD vector, and are used for storing AGC unit subscripts with roles of 1,2,3 and 4 respectively;

wherein: p_it: the output value of the unit i at the moment t;

a_i: coefficients of quadratic terms of the nonlinear relationship;

b_i: first order coefficient of non-linear relation;

c_i: a constant term of the non-linear relationship;

d: the current output value is worth correcting the coefficient;

the above objective function ensures that the expectation of the total cost of ACE adjustment in a day is minimized for all AGC groups belonging to role (j).

s_itthe rotation of the ith unit at the time t is reserved;

Real-time scheduling is advanced scheduling based on ultra-short term load prediction. Generally, t is t ═ t₁T is time pair₁And optimizing at the moment + T, and correcting the deviation between the scheduling plan and the prediction result.

The targets of real-time scheduling are coordinated rolling scheduling, coordinated AGC and coordinated network security.

2) Coordination of real-time scheduling with power generation plans

Real-time scheduling does not override the power generation schedule, but rather makes full use of the power generation schedule. The real-time scheduling may be based on the results of the day-ahead scheduling plan, or may be based on the results of the rolling scheduling plan, on which further checks and corrections are made. The coordination principle of real-time scheduling and power generation plan is 'dynamic and static connection and smooth transition'.

3) Coordination mode of real-time scheduling and AGC control

During the dispatching process of the power system, various accidents can continuously occur, such as fluctuation of wind power generation, load deviation, unplanned shutdown of the generator, line overload and the like. Real-time scheduling needs to be coordinated with not only a day-ahead plan but also AGC control, and plays a role in auxiliary adjustment of an AGC system. Real-time scheduling is responsible for the regular power distribution of loads with larger amplitudes, while AGC control is responsible for the fast random variation of loads with smaller amplitudes.

The coordination mode of real-time scheduling and AGC control is as follows:

first, the system should be operated at a safer operating point as much as possible.

AGC control is correction after occurrence of deviation, regardless of economy. The real-time scheduling advance control should be as accurate as possible, large deviation is not required, and large-amplitude adjustment of deviation correction control is not required, so that the method is economical on the whole. The accuracy of the power generation plan is ensured, and the accuracy of load prediction is improved firstly.

Second, enough adjustment space is reserved for AGC control in real-time scheduling.

The power system is always in an unordered dynamic variation. Various deviation correction control functions monitor the operating state of the power system in real time, and correct and control various deviations. Real-time scheduling should reserve a certain adjustment space for AGC control, and the requirement of deviation correction is met.

The control period of the real-time scheduling is one period (e.g., 15 minutes), and the control period of the AGC is around 10 s. The AGC unit is controlled by AGC software, is adjusted along with the change of ACE in the area and cannot be controlled in real-time scheduling. This consideration in this case preserves the problem of adjusting space for the AGC. Real-time scheduling can adjust the active balance of the system by controlling the SCHED mode unit, and reserves adjustable capacity for the AGC unit.

4) Real-time scheduling can adopt active scheduling model with minimum wind curtailment

The system should support real-time scheduling plan instruction adjustment of 1 hour and minute level, and realize a multi-time scale multi-level coordination scheduling mode which is comprehensive in economy, energy conservation and safety. Introducing a model predictive control theory MPC, researching a real-time scheduling model and an algorithm, mainly comprising:

A1) and establishing an active scheduling optimization model which meets the safety constraint and has the minimum wind abandon.

A2) Considering that in the future, due to wind power generation and the occurrence of steep rise and drop of loads, the system has main quick adjustable capacity to balance the loads and whether the network is congested or not.

A3) Although the real-time scheduling meets the network constraint in the security check, the situation that the power flow is out of limit still can occur in the operation of the power grid. Real-time scheduling requires on-line scanning of the operating conditions of all elements in the power grid, and emergency adjustment of power generation is performed when an out-of-limit condition occurs, namely blocking management is performed.

A4) An MPC model is introduced for real-time scheduling modeling of large-scale wind power, the self regulation and control capability of the wind power is fully exerted, and a load method that the wind power is simply equivalent to negative in the traditional scheduling is broken through.

And establishing an active scheduling optimization model meeting safety constraints and having the minimum wind abandon according to the ultra-short-term wind power predicted value by using the active real-time scheduling method having the minimum wind abandon.

Wherein, the real-time unit control cycle is 15 minutes (adjustable). On the premise of meeting the requirements of safety, economy, energy conservation and environmental protection, the tasks finished by the 15-minute real-time unit plan comprise: 1) make up for the bias of the load forecast before the day (expanded short term) and the load forecast in the 15 minutes ultra short term; 2) leaving enough spare capacity for AGC.

To this end, the following linear programming model may be constructed to describe:

in the model:

w_jthe cost of abandoned wind power of a wind power plant is reduced, and in order to reduce abandoned wind, the cost of abandoned wind power of the wind power plant is generally w_jIs far greater than the current adjustment cost r of the generating unit of the conventional unit in value_i；

And real-time scheduling plan value of next time period

A difference of (d);

N_G ^WindThe method comprises the steps of (1) collecting a wind turbine generator set;

the current output value of the wind power plant j is obtained;

wherein the content of the first and second substances,

is the increment of the predicted value of the ultra-short-term load,

Is the next time increment of the tie line plan,

△C_gjinfluence of planned adjustment quantity of a unit scheduled in non-real time on section power.

The objective of active scheduling is to accommodate as much wind power as possible. Therefore, the result of the real-time scheduling optimization model, i.e. the real-time planned value of the generated output of each wind farm, is generally the ultra-short-term forecast value. However, due to the constraints of power grid transmission capacity, power generation reserve capacity and the like, it cannot be ensured that the output of the wind power plant can always reach the predicted output, and at the moment, the real-time scheduling optimization result corresponding to the output of the wind power plant is the wind curtailment power required by the wind power plant. And sending the calculation result of real-time scheduling to each wind power plant as the planned output of the wind power plant in the next time period.

In addition, the following scheduling model can also be adopted for real-time scheduling:

f₁(p_it) Is a scheduling model objective function; t is the optimization time; t is t₀To optimize the starting period; t is_lOptimizing a period length for the base point tracking layer; n is the number of conventional units; a is_i、b_i、c_iThe coal consumption coefficient of a conventional unit i is obtained; p is a radical of_itPlanning the active power output of the conventional unit i in the t-th time period;

for rolling optimal planning layer plan of ith unit in t time period △ p_itTracking a planned adjustment quantity for the ith unit at a base point of the t-th time period to control the output quantity; g_windThe number of the wind power generator sets is; lambda [ alpha ]_jIs a cost factor of the abandoned wind;

expanding the short-term predicted output for the wind turbine;

and (4) planning the active power output of the wind turbine generator j in the t-th time period.

(5) Safety checking technology

The safety checking checks the system operation mode under the generator set output plan generated by the rolling scheduling module and the real-time scheduling module so as to ensure the safety and reliability of the system operation.

Firstly, a checking section intelligent generation function forms checking section power flow through alternating current power flow calculation; then, performing ground state power flow analysis on the check section, and judging the out-of-limit condition of the power grid under the ground state power flow; then, carrying out static safety analysis, and judging whether other branches are out of limit after the element is disconnected; and finally, according to the out-of-limit equipment, the heavy-load equipment and the stable section discovered by the static safety analysis, carrying out sensitivity analysis on the out-of-limit equipment, the heavy-load branch equipment and the out-of-limit and heavy-load stable section, carrying out sensitivity analysis on a voltage out-of-limit node, and providing a decision basis for subsequent auxiliary decision.

And analyzing and determining the static safety operation level of the check section through the static safety check function module, and providing a static safety check result for subsequent stable calculation check and auxiliary decision.

The system realizes management requirements such as energy-saving scheduling and economic scheduling by means of the project, realizes four-layer scheduling coordination control modes such as day-ahead planning, day-in rolling scheduling, day-in real-time scheduling and AGC control, and realizes closed-loop control from a scheduling side to a power plant side.

The optimized scheduling model provided by the invention has the general structure that:

1. integrated monitoring module

The comprehensive monitoring module collects information such as power grid real-time data, ultra-short-term prediction data, rolling optimization data, real-time scheduling data and the like in the day, organizes and analyzes the information according to business topics from a leader concern point and a dispatcher concern point through various advanced visual information display means, and covers aspects such as load, plan and power generation, sections, wind power, installed scale, electric quantity contract execution conditions and the like.

1) Load information monitoring submodule

And comparing and displaying the load prediction information, the expanded short-term load prediction information, the ultra-short-term load prediction information, the actual power grid load information, the heat supply load information and the like in a graphical mode.

2) Planning and power generation monitoring submodule

Comparing and displaying day-ahead plan information, rolling scheduling information, real-time scheduling plan information, actual power generation information and the like in a graphical mode; and displaying planned power generation conditions, actual power generation conditions, wind power absorption conditions, wind power limit conditions and hydropower peak regulation conditions of various power supplies.

3) Direct-adjusting section monitoring submodule

And displaying the active power and load conditions of the connecting line and the important section.

4) Wind power information monitoring submodule

And collecting all the wind power information in the day, wherein the wind power information comprises ultra-short-term wind power prediction, real-time output of a wind field, wind field reporting plan information, rolling optimization wind power conditions, real-time scheduling wind power conditions, wind power electricity limiting conditions and the like.

5) Installation condition analysis submodule

And the scale and proportion of installed hydropower, thermal power and wind power are displayed in a graphical mode, and the trend change condition is displayed.

6) Plan completion analysis submodule

And comparing and analyzing the electric quantity contract, and obtaining the total plan completion condition, the power plant information with higher plan completion rate, the power plant information with lower plan completion rate and the like.

2. Rolling scheduling module

The rolling scheduling module is used for carrying out re-prediction on the future 4-hour load and the power generation plan by fully utilizing the latest real-time information and the prediction information on the basis of the load information predicted by the extended short-term prediction module and taking 1 hour as a starting period, so that the day-ahead load and the day-ahead power generation plan predicted by the day-ahead plan module are corrected, and the uncertainty of the day-ahead plan is gradually reduced;

the rolling scheduling module comprises: the system comprises a connecting line plan management submodule, an ultra-short-period wind power prediction submodule, an extended short-period load prediction submodule, a constraint adjustment submodule, an online rolling optimization submodule and a plan instruction issuing submodule;

the tie line plan management submodule is used for providing functions of leading-in, copying, modifying and checking a tie line plan for 1-3 days in the future;

the ultra-short-term wind power prediction submodule is used for giving a wind power prediction value of 4 hours in the future every 15 minutes according to the actual value of the current grid wind regulation power; the system and the wind power prediction system establish an interface, and the ultra-short-period wind power prediction information of four hours in the future is regularly acquired every day.

The extended short term load prediction submodule is operable to: predicting a load value within unknown 1-many hours after the current time of the day according to the predicted wind power prediction value predicted by the ultra-short-term wind power prediction submodule; wherein the load value takes 15 minutes as a minimum unit, and further rolling correction is carried out on the day-ahead power generation plan;

the constraint adjusting submodule is used for providing various constraint conditions, and comprises: the method comprises the following steps of (1) unit output upper and lower boundary constraint, unit climbing rate constraint, starting mode constraint, section current constraint and load balance constraint;

the online rolling optimization submodule is used for: and (2) formulating an optimization strategy, taking the constraint conditions provided by the constraint adjusting submodule as constraints, taking the prediction structures of the ultra-short-term wind power prediction submodule and the extended short-term load prediction submodule as input, and performing online rolling optimization on the power generation output conditions of the units with different characteristics of hydropower, thermal power, cogeneration, wind power and pumped storage to obtain an online rolling optimization result, wherein the online rolling optimization result comprises the following steps: 4 hours of future wind field planning and rolling dispatching unit planning;

the algorithm input data includes: the method comprises the steps of real-time power grid model, load information, day-ahead plan execution condition, ultra-short-term wind power prediction information, extended short-term load prediction information, plan adjustment amount constraint, climbing rate constraint, heat supply load constraint, upper and lower output limit constraint, safety constraint and the like. In order to ensure the performability of the rolling optimization result, various constraint conditions are used as preconditions to carry out algorithm operation in the execution process of the rolling optimization program, and meanwhile, the plan result output by the algorithm is directly subjected to online static security check.

The rolling optimization strategy adopted by the rolling scheduling module is as follows: an economic optimal scheduling model based on the minimum abandoned wind is adopted, and the method is shown in the formula (1):

wherein f is₁(p_it) Is a scheduling model objective function; t is the optimization time; t is t₀To optimize the starting period; t is_hOptimizing the time interval length for the optimal planning layer; n is the number of conventional units; a is_i、b_i、c_iThe coal consumption coefficient of a conventional unit i is obtained; p is a radical of_itPlanning the active power output of the conventional unit i in the t-th time period; g_windThe number of the wind power generator sets is; lambda [ alpha ]_jIs a cost factor of the abandoned wind;

expanding the short-term predicted output for the wind turbine;

The plan instruction issuing sub-module is used for: and transmitting the online rolling optimization result obtained by the online rolling optimization submodule to a rolling planning unit.

(3) Real-time scheduling module

The real-time scheduling module is used for: based on ultra-short-term power generation demand prediction, wind power output prediction, a power grid model and real-time data, 15 minutes are used as a starting period, and an actual power generation plan of a real-time planning unit is formulated under the condition that the output limit, the climbing rate and the rolling power generation plan of the unit are considered, so that the load of the future 15 minutes is subjected to re-prediction and power generation plan adjustment, the deviation of a predicted value and a plan value is eliminated, the wind power access capacity of the power grid is improved, and the method is used as a link for coordinating a scheduling plan, AGC control and network safety and starting and stopping.

The real-time scheduling module comprises: an ultra-short-term load prediction submodule, an online real-time scheduling submodule, an online security check submodule and an automatic instruction issuing submodule;

the ultra-short-term load prediction submodule is used for automatically matching a similar time period according to historical load output data by adopting an ultra-short-term load prediction method based on the form similarity of a load curve section to obtain ultra-short-term load prediction data;

the online real-time scheduling submodule is used for performing unit combination operation by taking the ultra-short-term prediction data and the rolling scheduling unit plan of the rolling scheduling module as the basis and preset constraint conditions and optimization strategies as preconditions, and predicting the output of a wind field and a single unit in the future 15 minutes by taking 15 minutes as a period; specifically, the real-time scheduling module adopts a scheduling model of formula (2):

expanding the short-term predicted output for the wind turbine;

The online safety checking submodule is used for checking a real-time scheduling plan result of the online real-time scheduling submodule based on a current real-time power grid model and ultra-short-term prediction data, and simultaneously, considering the situation that a section is out of limit to obtain an online scheduling prediction result, and comprises the following steps: the output data of the wind field and the single machine set in the future 15 minutes;

and the automatic instruction issuing sub-module is used for issuing the real-time scheduling plan result of the remaining time period of the day to the real-time planning unit through the comprehensive data platform so as to realize the state control of the real-time planning unit.

(4) AGC control module

The AGC control module is used for carrying out real-time processing on the current condition of the unit in units of second level so as to control the output of the AGC unit; the device comprises a correction control submodule and a safety correction control submodule; the correction control submodule is used for scheduling a second-level AGC unit to enable the frequency and the tie line power to meet CPS (control performance Standard) evaluation indexes; and the safety correction control submodule is used for immediately processing line section tidal current out-of-limit.

In practical application, the safety check function provided by the system needs to call the generator set output arrangement data generated by the rolling scheduling module and the real-time scheduling module, and output the check result in a plan mode.

And based on the latest power grid model, the latest equipment state information and the latest prediction information in the day, performing static safety analysis on the plan, analyzing the power grid topology, calculating the power transfer distribution factor of each generator to the line load flow, counting the system blocking condition, and performing blocking management. The built-in functions comprise automatic generation of a check section, load flow analysis, static safety analysis, sensitivity analysis and the like, and the system provides 1-minute support for the running efficiency of a static safety check algorithm.

1) Automatic generation of check section

And the intelligent generation function of the check section combines system load prediction and bus load prediction according to the maintenance plan, the power generation plan, the short-term transaction plan, the temporary operation information and the equipment operation information, acquires reactive voltage information according to the user setting or similar solar current, intelligently integrates the data, and performs alternating current power flow calculation to form check section power flows aiming at different types of safety check requirements, wherein the check section power flows comprise an operation task check section, a maintenance plan check section, a generator plan check section and a short-term transaction plan check section.

2) Ground state power flow analysis

And analyzing and calculating the ground state power flow analysis according to the check section power flow formed by the check section intelligent generation function, comparing the power flow calculation result with the limit, and judging the power grid out-of-limit condition under the ground state power flow. Heavy equipment and corresponding load rates, out-of-limit equipment and corresponding out-of-limit percentages can be given. The objects of the out-of-limit inspection include line current, power transmitted by the section, capacity of the transformer branch and voltage of the bus.

3) Static security analysis

And (4) static safety analysis is performed on the checking section flow formed by the intelligent generating function of the checking section, and whether the N-1 fault and other elements are out of limit after the fault set specified by the user occur or not is checked.

The functional module supports various specified modes of cut-off elements, including cutting off the main equipment of the whole network one by one, cutting off the main equipment one by one according to the element types (generator type, transformer type and line type, wherein the line can be further divided into 500kV line and 220kV line), cutting off the main equipment one by one according to the voltage class, and cutting off the main equipment one by one according to the region. In addition, the user can customize the fault set according to needs, only carry out N-1 analysis on elements in the fault set and judge whether other elements are out of limit or not.

The functional module can simulate safety automatic devices such as automatic bus transfer equipment, a generator tripping machine and the like, and can automatically match a strategy table according to the operation mode of a power grid. The fault which causes heavy load and out-of-limit and the corresponding heavy load and out-of-limit equipment can be given, and the fault severity index is given.

4) Short circuit current analysis

And (4) intelligently generating a check section power flow according to the check section by short-circuit current analysis calculation, and judging whether the short-circuit capacity exceeds the standard in the check section or not by short-circuit current calculation. The method can not only scan the short-circuit fault of the whole network bus, but also scan the short-circuit fault according to the calculation range set by a user, and supports the short-circuit fault scanning according to the voltage grade and the partition selection calculation range.

The single-phase short-circuit fault scanning and the three-phase short-circuit fault scanning can be classified according to fault types. The calculation results of short-circuit faults with over-standard short-circuit current and close to over-standard short-circuit fault can be given, and the calculation results comprise short-circuit current of each bus and each line and corresponding faults.

5) Sensitivity analysis

And sensitivity analysis is performed on the out-of-limit and heavy-load branch circuits and the out-of-limit and heavy-load stable sections according to the out-of-limit and heavy-load equipment and the stable sections found by static safety analysis aiming at checking section tide, and the sensitivity analysis of the voltage out-of-limit nodes is performed. The function module supports the following functions:

1) and calculating the sensitivity between the active power of the branch or the stable section and the active power of the generator.

2) And calculating branch disconnection distribution factors, namely the change conditions of other lines or transformer power after the lines or the transformer branches are disconnected.

3) The sensitivity between the bus voltage and the node reactive injection (including the generator node and the capacitor reactor node), and the sensitivity between the bus voltage and the transformer transformation ratio are calculated.

4) Quasi-steady state sensitivity calculations are supported.

6) Plan issue

The plan release module provides functions including a plan data transparent transmission function and a plan data display function.

And (4) plan data transmission, namely, providing an interface by the system, transmitting the wind field plan, the water and electricity plan, the thermal power plan and the pumped storage plan calculated by the rolling scheduling module and the real-time scheduling module to a region III through a safety isolation device, interfacing with the OMS, carrying out approval circulation in the OMS, and transmitting the approval circulation to the power plant through the OMS. In order to meet the requirements of automatic issuing and execution of plan instructions, the system provides a function of issuing various kinds of plan data to the AGC system through a comprehensive data platform.

Planning data presentation

In order to facilitate leaders at all levels to directly check various kinds of plan data and latest wind power prediction information in a management area, the system provides a plan data checking client, and plan data, real-time data and contrastive analysis data are visually and vividly displayed by a visualization means.

(5) Effect evaluation module

Various benefits brought by power supply collaborative optimization with different characteristics are analyzed and displayed by adopting abundant visual display means; and analyzing and displaying the working effect brought by the introduction of a multi-period scheduling coordination mechanism in a day. The effect evaluation topic includes: clean energy utilization condition, energy conservation and emission reduction condition, generated energy, economy, starting mode and unit influence, wind power assessment and the like.

1) Power generation analysis

The display content comprises the power generation composition condition in the modes of day-ahead planning, rolling scheduling and real-time scheduling planning; comparing the generated energy of various power generation sources in different planning modes and changing trends; clean energy utilization rate, reduction of coal consumption, carbon emission and other macro index information.

2) Analysis of economics

The display content comprises the electricity purchasing cost constitution condition in the day-ahead plan, the rolling optimization plan and the real-time scheduling plan mode; comparing and changing the electricity purchasing cost of water, fire and wind power in different planning modes.

3) Boot method analysis

The display content comprises startup condition comparison, startup quantity change, startup capacity change, important unit attention influence analysis, system standby margin change, adjustable unit capacity analysis and the like of various units in a day-ahead plan, rolling optimization plan and real-time scheduling plan mode.

4) Wind power assessment

And assessing the reporting and executing conditions of the wind field plan. According to preset management rules, comparing and displaying the wind field reporting plan, the ultra-short-term wind power prediction information, the wind power rolling optimization information and the actual power generation output condition of the wind field with the same time scale and calculating indexes.

(6) System management module

The system management part provides functions including: basic data maintenance, user management, authority management and log management. Wherein:

basic data maintenance

The method provides the functions of batch import and maintenance of various basic data (such as static basic information of power plants, units and electric quantity combinations) required by rolling optimization and real-time scheduling, and the collected information of the thermal power plant comprises information of unit climbing rate, cost function, output limit, output influence of fire coal and heat supply and the like reported by the thermal power plant.

User management: and the maintenance and management of basic information and login information of a scheduling end user and a power plant end user are realized.

And (3) authority management: the functions of resource management, role management, empowerment and the like are provided, and system application safety and data access safety are achieved.

System logging: important operation in the system and operation conditions among system interfaces are tracked and recorded, and the follow-up memory is facilitated.

The system can establish a multi-period and multi-target coordinated optimal scheduling model by researching the operating characteristics and the influence mechanism on scheduling operation of power supplies with different characteristics such as wind power, cogeneration units, hydroelectric units, thermal power units and the like; by adopting a closed-loop feedback dynamic adjustment multi-dimensional power grid optimization scheduling decision-making technology, on the basis of the actual output of a unit, bus load, power grid running state and a network topological structure, through multi-period multi-constraint safety check and blocking management, the economic benefits of less wind abandon and energy conservation and emission reduction are considered, and multi-objective rapid optimization decision-making of scheduling plan rolling adjustment, real-time active balance and coordination control in the day is realized. The following effects are achieved:

1) the system carries out rolling online optimization on the plan from the current time period to the future 4-hour time period based on the extended short-term power generation demand prediction and the day-ahead plan information, not only can the economic benefits of less wind abandon and energy conservation and emission reduction be considered when rolling scheduling is made, but also the feasibility of the output of each unit in the rest time period can be ensured, including meeting the unit climbing rate constraint, meeting the power generation-load power balance constraint, the network safety constraint and the like.

2) The system can provide an advanced control strategy taking 15 minutes as a cycle, a power generation plan is compiled according to a scheduling period, the actual power generation plan of each unit is automatically arranged according to the principles of energy conservation, emission reduction and economic optimization on the basis of meeting the safe and stable operation of the system according to the power generation demand prediction and the wind power output prediction of the next scheduling period under the conditions of unit limit value, climbing rate and rolling power generation plan, the deviation of predicted values and plan values can be predicted and eliminated in advance, the wind power access capability of a power grid is improved, and the advanced control strategy serves as a link starting up for coordinating the scheduling plan, AGC control and network safety.

3) The system can be interfaced with the existing dispatching automation systems such as CC2000, D5000, a day-ahead planning system, a wind power forecasting system, an OMS and the like to obtain day-in forecasting information, planning information and real-time information, so that a data basis is provided for online expansion of short-term forecasting, ultra-short-term forecasting, rolling optimization and real-time dispatching control strategies, and closed-loop management of wind field control is formed.

4) The system can contrastively analyze the value and effect brought by online rolling optimization and real-time scheduling of the units with different multi-target characteristics, and forms comprehensive display of daily information by using visual technical means from the daily scheduling plan and control perspective and facing different user groups.

The foregoing is only a preferred embodiment of the present invention, and it should be noted that, for those skilled in the art, various modifications and improvements can be made without departing from the principle of the present invention, and such modifications and improvements should also be considered within the scope of the present invention.

Claims

1. A dispatching system of an active dispatching model with minimum wind abandon is characterized by comprising a unit role distribution module, a rolling dispatching module, a real-time dispatching module and an AGC control module;

the active scheduling model with the minimum wind curtailment comprises the following optimization objective functions and optimization constraint conditions;

wherein:

ΔP_iadjusting the total output for the output of the ith conventional unit at the next moment to control the output;

ΔP _ithe lower limit of the total output adjustment amount of the ith conventional unit;

the upper limit of the total output adjustment amount of the ith conventional unit is set;

w_jthe cost of abandoned wind power of a wind farm, w for reducing abandoned wind_jIs far greater than the current adjustment cost r of the generating unit of the conventional unit in value_i；

The abandoned wind electric quantity of the wind turbine generator j is equal to the predicted wind power output value predicted in the next period

And real-time scheduling plan value of next time period

A difference of (d);

N_G ^cagcthe method comprises the steps that the units are dispatched in real time for the whole network, and the number of the wind turbine units is not included;

P_j ⁰the current output value of the wind turbine generator j is obtained;

and delta P is the total output adjustment amount of the real-time scheduling unit at the next moment:

wherein the content of the first and second substances,

is the increment of the predicted value of the ultra-short-term load,

Is the next time increment of the tie line plan,

Is the next moment increment of the planned unit output in the day ahead; delta Pⁿ _AGCThe last moment of AGC is unfinished quantity;

S_ijthe load balance sensitivity of the ith conventional unit is that the partition load prediction information needs to be introduced into the bus load factor in order to achieve partition balance;

S_kjis the load balance sensitivity, Δ P, of the wind turbine k_k ^wThe product of the abandoned wind electric quantity of the wind turbine generator k and the abandoned wind electric quantity of the wind turbine generator k represents the real-time influence of the abandoned wind electric quantity of the wind turbine generator on the section power;

ΔC_gjinfluence of planned adjustment quantity of a non-real-time scheduling unit on section power;

solving the objective function to obtain a predicted value of the output of the future wind field and the single machine set;

step 2, the AGC units participating in adjusting ACE meet the requirement of total rotation standby at each time interval, so N AGC units participating in ACE control are totally arranged in N units in the system, N is more than or equal to 1 and less than or equal to N, and a set formed by the N units is marked as SetA; the lower limit of spinning reserve should be given by the actual operation of the grid, set to SRt, which must be greater than ACE_E；

wherein:

t: optimizing the time interval length;

a_i: coefficients of quadratic terms of the nonlinear relationship;

b_i: first order coefficient of non-linear relation;

c_i: a constant term of the non-linear relationship;

d: the current output value is worth correcting the coefficient;

s_itthe rotation of the ith unit at the time t is reserved;

solving the objective function under the constraint condition to obtain the finally determined AGC role;

the AGC control module is used for: taking a real-time scheduling correction plan instruction given by the real-time scheduling module as a control base point, correcting a random prediction error generated in an advanced prediction link in real time, generating an AGC unit output adjustment instruction for controlling an AGC unit, and issuing the AGC unit output adjustment instruction to the AGC unit to realize the control of the AGC unit;

the multi-dimensional power grid optimization scheduling decision making technology adopting closed-loop feedback dynamic adjustment is based on the actual output of a unit, bus load, power grid running state and network topology structure, multi-time-period multi-constraint safety check and blocking management are carried out, multi-objective rapid optimization decision making of a scheduling plan is realized, the system calculation result not only ensures that the power and electric quantity balance and the line and section flow are not out of limit, but also realizes the functions of line N-1 scanning, fault set scanning and self-adaptive adjustment; finally, the aims of year-month power and electricity quantity plan balance, month-week scheduling plan decomposition management, day-ahead scheduling plan optimization compilation, intra-day scheduling plan rolling adjustment and real-time active power balance and coordination control are fulfilled;

Wherein: i and j are respectively the serial numbers of the conventional unit and the wind generating set, and the value range is i belongs to [1, N ]],j∈[1,M](ii) a Wherein N is the number of conventional units; m is the total number of the wind turbine generators; t is the optimization time, and the value range is t epsilon [ t₀+1,T]；t₀To optimize the starting period; t is the length of the optimized time interval; p is a radical of_it,

the optimization constraint conditions comprise:

(1) the unit output upper and lower bound constraint conditions are as follows:

(2) unit ramp rate constraint

p_i,t-1-Δpd_it≤p_it≤p_i,t-1+Δpu_it(3)

Wherein: delta pd_it,Δpu_itThe maximum value of the lowering force and the maximum value of the raising force allowed from the t-1 period to the t period for the ith conventional unit; p is a radical of_i,t-1A force value is planned for the ith conventional unit in the rolling plan in the t-1 th time period; the wind turbine generator is not limited by the climbing rate constraint;

(3) safety restraint of section tidal current

(4) load balancing constraints

Wherein D is_tForce values are plotted for the total;

because the constraint corresponding to the formula (2) and the formula (3) only contains the information of a single unit, the constraint is defined as a unit non-coupling constraint; the constraint of the formula (4) and the constraint of the formula (5) contain the information of a plurality of units, so the constraint is defined as the coupling constraint of the units;

For variables, equation (1) is the optimization objective function, and equations (2), (3), (4), (5) are the optimization constraints;

specifically, in order to ensure that AGC is smooth, stable and effectively realize real-time balance of power supply and demand and avoid overshoot or undershoot in the process of reducing ACE, a control section needs to be divided, wherein the control section is used for indicating the severity of ACE and comprises a dead zone, a normal zone also called a command zone, an auxiliary zone also called an allowed zone and a cooperation zone also called an emergency zone;

L₁₀: a control limit for the absolute value of the ten minute ACE average;

LOSS: a destabilizing power;

AGC set coordination control strategy table

In the table: "not controlled" means not making any adjustments; "bias adjustment" means that the baseline value needs to be left and the ACE adjustment is participated to promote ACE reduction; "base point close" means that the base point adjustment is performed directly, without considering the influence on ACE; "condition return" means that whether influence is caused on ACE or not when base point adjustment is performed, and if ACE is increased due to approaching to a base point value, the ACE is kept still; if approaching to the base point value will promote ACE to reduce, then change;

the base point value is used as a planning base point, namely the base point value is a continuous curve;

wherein: cap_agc: AGC deviation adjustment capacity;

P_i,max: the maximum output value of the unit i;

P_i,min: the minimum output value of the unit i;

and 4, selecting the numbers as follows in sequence: the method comprises the following steps that n units of a unit m +1 and a unit m +2 … unit m + n are used as real-time planning units, the real-time planning units are controlled according to a tracking real-time planning strategy, wherein n is the minimum value meeting the following constraints:

step 6, the rest units in the ranking table are day-ahead planning units which are adjusted according to a control strategy for tracking the day-ahead plan;

the ultra-short-term wind power prediction submodule is used for giving a wind power prediction value of 4 hours in the future every 15 minutes according to the actual value of the current grid wind regulation power; the system and a wind power prediction system establish an interface, and ultra-short-term wind power prediction information of four hours in the future is regularly acquired every day;

the algorithm input data includes: the method comprises the steps of real-time power grid model, load information, day-ahead plan execution condition, ultra-short-term wind power prediction information, extended short-term load prediction information, plan adjustment amount constraint, climbing rate constraint, heat supply load constraint, upper and lower output limit constraint and safety constraint; in order to ensure the performability of the rolling optimization result, various constraint conditions are used as preconditions to carry out algorithm operation in the execution process of the rolling optimization program, and meanwhile, the plan result output by the algorithm is directly subjected to online static security check;

the rolling optimization strategy adopted by the rolling scheduling module is as follows: adopting an economic optimal scheduling model based on the minimum abandoned wind, see formula (6):

wherein f is₁(p_it) Is a scheduling model objective function; t is the optimization time; t is t₀To optimize the starting period; t is_hOptimizing the time interval length for the optimal planning layer; n is the number of conventional units; a is_i、b_i、c_iFor conventional units iA coal consumption coefficient; g_windThe number of the wind power generator sets is; lambda [ alpha ]_jIs a cost factor of the abandoned wind;

expanding the short-term predicted output for the wind turbine;

the plan instruction issuing sub-module is used for: transmitting the online rolling optimization result obtained by the online rolling optimization submodule to a rolling planning unit;

the real-time scheduling module is used for: based on ultra-short-term power generation demand prediction, wind power output prediction, a power grid model and real-time data, 15 minutes are used as a starting period, and an actual power generation plan of a real-time planning unit is formulated under the condition that the output limit, the climbing rate and the rolling power generation plan of the unit are considered, so that the load of the future 15 minutes is subjected to re-prediction and power generation plan adjustment, the deviation of a predicted value and a plan value is eliminated, the wind power access capacity of the power grid is improved, and the power generation plan is used as a link for coordinating a scheduling plan, AGC (automatic generation control) control and network safety and starting and stopping;

the online real-time scheduling submodule is used for performing unit combination operation on the basis of the ultra-short-term load prediction data and the rolling scheduling unit plan of the rolling scheduling module and taking a preset constraint condition and an optimization strategy as a precondition, and predicting the output of a wind field and a single unit in the future 15 minutes by taking 15 minutes as a period; specifically, the real-time scheduling module adopts a scheduling model of formula (7):

planning a rolling optimal planning layer of the ith unit in the t-th time period; Δ p_itTracking a planned adjustment quantity for the ith conventional unit at a base point of the t-th time period to control the output quantity; g_windThe number of the wind power generator sets is; lambda [ alpha ]_jIs a cost factor of the abandoned wind;

expanding the short-term predicted output for the wind turbine;

the online safety checking submodule is used for checking a real-time scheduling plan result of the online real-time scheduling submodule based on a current real-time power grid model and ultra-short-term load prediction data, and simultaneously considering the situation that a section is out of limit to obtain an online scheduling prediction result, and comprises the following steps: the output data of the wind field and the single machine set in the future 15 minutes;

the automatic instruction issuing submodule is used for issuing the real-time scheduling plan result of the remaining time period of the day to the real-time planning unit through the comprehensive data platform so as to realize the state control of the real-time planning unit;

the AGC control module is used for carrying out real-time processing on the current condition of the unit in units of second level so as to control the output of the AGC unit; the device comprises a correction control submodule and a safety correction control submodule; the correction control submodule is used for scheduling a second-level AGC unit to enable the frequency and the tie line power to meet CPS (control performance Standard) evaluation indexes; the safety correction control submodule is used for immediately processing line section tidal current out-of-limit;

in practical application, the safety check function provided by the system needs to call the generator set output arrangement data generated by the rolling scheduling module and the real-time scheduling module and output a check result in a plan mode;

based on the latest power grid model, equipment state information and prediction information in the day, performing static safety analysis on a plan, analyzing the power grid topology, calculating power transfer distribution factors of each generator to the line power flow, counting the system blocking condition, and performing blocking management; the built-in functions comprise automatic generation of a check section, load flow analysis, static safety analysis and sensitivity analysis, and the system provides the running efficiency support of a static safety check algorithm for 1 minute;

1) automatic generation of check section

The intelligent checking section generating function is used for acquiring reactive voltage information according to a user set or similar solar current by combining system load prediction and bus load prediction according to an overhaul plan, a power generation plan, a short-term transaction plan, temporary operation information and equipment operation information and intelligently integrating the data to perform alternating current power flow calculation to form checking section power flows aiming at different types of safety checking requirements, wherein the checking section power flows comprise an operation task checking section, an overhaul plan checking section, a generator plan checking section and a short-term transaction plan checking section;

2) ground state power flow analysis

The ground state power flow analysis carries out analysis and calculation according to the check section power flow formed by the check section intelligent generation function, the power flow calculation result is compared with the limit, and the power grid out-of-limit condition under the ground state power flow is judged; the heavy-load equipment, the corresponding load rate, the out-of-limit equipment and the corresponding out-of-limit percentage can be given; the objects of the out-of-limit inspection comprise line current, power transmitted by a section, capacity of a transformer branch and voltage of a bus;

3) static security analysis

Static safety analysis is performed on the checking section flow formed by the intelligent generating function of the checking section, and whether the N-1 fault and other elements are out of limit after a fault set specified by a user occur or not is checked;

the functional module supports various specified modes of disconnection elements, including disconnection of main equipment of the whole network one by one, and disconnection one by one according to element types, including generator types, transformer types and line types, wherein the line can be further divided into 500kV lines and 220kV lines, and is disconnected one by one according to voltage class and disconnected one by one according to regions; in addition, the user can customize the fault set according to needs, only carry out N-1 analysis on elements in the fault set and judge whether other elements are out of limit or not;

the functional module can simulate the automatic safety device of the automatic backup power switching machine and the automatic safety device of the generator tripping machine and can automatically match a strategy table according to the running mode of a power grid; faults causing heavy load and out-of-limit and corresponding heavy load and out-of-limit equipment can be given, and fault severity indexes are given;

4) short circuit current analysis

The short-circuit current analysis and calculation intelligently generates a check section power flow formed according to the check section, and whether the short-circuit capacity exceeds the standard in the check section is judged through the short-circuit current calculation; the method can not only scan the short-circuit fault of the whole network bus, but also scan the short-circuit fault according to the calculation range set by a user, and supports the short-circuit fault scanning according to the voltage grade and the partition selection calculation range;

the method can be divided into single-phase short-circuit fault scanning and three-phase short-circuit fault scanning according to fault types; the calculation results of short-circuit faults with over-standard short-circuit current and close to over-standard short-circuit fault can be given, wherein the calculation results comprise short-circuit current of each bus and each line and corresponding faults;

5) sensitivity analysis

The sensitivity analysis aims at checking section tide, and according to the out-of-limit equipment, the heavy-load equipment and the stable section found by static safety analysis, the sensitivity analysis of the out-of-limit branch, the heavy-load branch and the out-of-limit and heavy-load stable section is carried out, and the sensitivity analysis of a voltage out-of-limit node is carried out; the function module supports the following functions:

1) calculating the sensitivity between the active power of the branch or the stable section and the active power output of the generator;

2) calculating branch disconnection distribution factors, namely the change conditions of other lines or transformer power after the lines or transformer branches are disconnected;

3) calculating the sensitivity between the bus voltage and the node reactive power injection and the sensitivity between the bus voltage and the transformer transformation ratio; the node reactive power injection comprises a generator node and a capacitance reactor node;

4) supporting quasi-steady-state sensitivity calculation;

6) plan issue

The plan release module provides functions including a plan data transparent transmission function and a plan data display part;

the plan data is transmitted through a system providing interface, a wind field plan, a hydropower plan, a thermal power plan and a pumped storage plan which are calculated by a rolling scheduling module and a real-time scheduling module are transmitted to a region III through a safety isolation device, the safety isolation device is connected with an OMS (operation, maintenance and management) system to carry out approval circulation in the OMS system, and the approval circulation is transmitted to a power plant through the OMS system; in order to meet the requirements of automatic issuing and execution of plan instructions, the system provides a function of issuing various kinds of plan data to an AGC system through a comprehensive data platform;

planning data presentation

In order to facilitate leaders at all levels to directly check various kinds of plan data and latest wind power prediction information in a management area, the system provides a plan data checking client, and plan data, real-time data and contrastive analysis data are visually and vividly displayed by a visualization means;

the scheduling system further comprises an effect evaluation module; the effect evaluation module adopts abundant visual display means to analyze and display various benefits brought by power supply collaborative optimization with different characteristics; analyzing and displaying the working effect brought by the introduction of a multi-period scheduling coordination mechanism in a day; the effect evaluation topic includes: clean energy utilization condition, energy conservation and emission reduction condition, generated energy, economy, starting mode and unit influence and wind power assessment;

1) power generation analysis

The display content comprises the power generation composition condition in the modes of day-ahead planning, rolling scheduling and real-time scheduling planning; comparing the generated energy of various power generation sources in different planning modes and changing trends; clean energy utilization rate, coal consumption reduction and carbon emission macroscopic index information;

2) analysis of economics

The display content comprises the electricity purchasing cost constitution condition in the day-ahead plan, the rolling optimization plan and the real-time scheduling plan mode; comparing and changing the electricity purchasing cost of water, fire and wind power in different planning modes;

3) boot method analysis

The display content comprises startup condition comparison, startup quantity change, startup capacity change, important unit attention influence analysis, system standby margin change and adjustable unit capacity analysis of various units in a day-ahead plan, rolling optimization plan and real-time scheduling plan mode;

4) wind power assessment

Assessing the reporting and executing conditions of the wind field plan; according to preset management rules, comparing and displaying wind field reporting plans, ultra-short-term wind power prediction information, wind power rolling optimization information and actual power generation output conditions of the wind field with the same time scale and calculating indexes;

the scheduling system also comprises a system management module; the system management part provides functions including: basic data maintenance, user management, authority management and log management functions; wherein:

basic data maintenance

Providing various basic data required by rolling optimization and real-time scheduling, wherein the basic data comprises batch import and maintenance functions of static basic information of power plants, units and electric quantity contracts, and the collected information of the power plants comprises information of unit climbing rate, cost function, output limit, output influenced by fire coal and heat supply reported by the power plants;

user management: the maintenance and management of basic information and login information of a scheduling end user and a power plant end user are realized;

and (3) authority management: the functions of resource management, role management and empowerment are provided, and system application safety and data access safety are achieved;

system logging: important operation in the system and operation conditions among system interfaces are tracked and recorded, so that the follow-up memory is facilitated;

the system can establish a multi-period and multi-target coordinated optimal scheduling model by researching the operating characteristics and the influence mechanism on scheduling operation of power supplies with different characteristics of wind power generation units, cogeneration units, hydroelectric units and thermal power units; by adopting a closed-loop feedback dynamic adjustment multi-dimensional power grid optimization scheduling decision making technology, on the basis of the actual output of a unit, bus load, power grid running state and a network topological structure, through multi-period multi-constraint safety check and blocking management, the economic benefits of less wind abandonment and energy conservation and emission reduction are considered, and multi-objective rapid optimization decision making of scheduling plan rolling adjustment in the day and real-time active balance and coordination control is realized; the following effects are achieved:

1) the system performs rolling online optimization on a plan from a current time interval to a future 4-hour time interval based on extended short-term power generation demand prediction and day-ahead plan information, and not only can the economic benefits of less wind abandon and energy conservation and emission reduction be considered when rolling scheduling is formulated, but also the feasibility of output of each unit in the rest time interval can be ensured, including meeting unit climbing rate constraint, meeting power generation-load power balance constraint and network safety constraint;

2) the system can provide an advanced control strategy taking 15 minutes as a cycle, a power generation plan is compiled according to a scheduling period, the actual power generation plan of each unit is automatically arranged according to the principles of energy conservation, emission reduction and economic optimization on the basis of meeting the safe and stable operation of the system according to the power generation demand prediction and the wind power output prediction of the next scheduling period under the conditions of unit limit value, climbing rate and rolling power generation plan, the deviation of predicted values and plan values can be predicted and eliminated in advance, the wind power access capability of a power grid is improved, and the advanced control strategy is used as a link for coordinating the scheduling plan, AGC control and network safety and starting;

3) the system can interface with CC2000, D5000, a day-ahead planning system, a wind power prediction system and an OMS existing scheduling automation system to obtain day-in prediction information, planning information and real-time information, provide a data basis for online expansion of short-term prediction, ultra-short-term prediction, rolling optimization and real-time scheduling control strategies and form closed-loop management of wind field control;