CN102817335B - Method and system for optimal scheduling on joint flood control for cascade reservoir groups - Google Patents

Abstract

The invention discloses a method and a system for optimal scheduling on joint flood control for cascade reservoir groups. The method includes the steps of firstly, determining flood type according to real-time reservoir level and the process of forecasting runoff in case of flood; secondly, deciding a primary protection object according to the flood type and downstream control-point flood control standards; thirdly, automatically selecting flood control optimization objectives and corresponding optimization scheduling models in a system model base; and fourthly, solving and optimizing the optimization scheduling model by modified genetic algorithm. The method and the system dynamically judge the level of flood and finish hierarchical scheduling based on the level of flood. In addition, projection of downstream protection objects, dam safety and flood recycling are achieved by selecting proper optimization scheduling objectives according to comprehensive information of the reservoirs, such as regulation performance, scheduling period and reliability in forecast flood during hierarchical scheduling, efficiency in decisions for the reservoir groups is improved, and mass data statistics show that the decision efficiency is improved by about 30% by the method and the system.

Description

A kind of method and system of Cascade Reservoirs associating Flood Optimal Scheduling

Technical field

The present invention relates to the method and system of a kind of Cascade Reservoirs associating Flood Optimal Scheduling, belong to reservoir dispatching technical field.

Background technology

Along with the adjustment of national energy economy, the energy resource structure of China based on thermoelectricity will progressively develop to various energy resources, especially clean reproducible energy develops as preferential direction changes, and HYDROELECTRIC ENERGY will be the chief component of clean reproducible energy within following one period.

After general Hydropower Stations construction completes, the reform and development how adapting to electricity market and the requirement of separating the factory and network, namely built engineering how is made to give full play to economical, societal benefits, improve the enterprise new problem that competitiveness Shi Ge large watershed genco faces on electricity market to greatest extent, and how to carry out Dispatching Flood be exactly one of faced new problem.

Traditional Cascade Reservoirs not dynamic judges the rank of flood and realizes graded dispatching on this basis; Meanwhile, also cannot, according to integrated informations such as the reliabilities of the adjusting function of reservoir, residing schedule periods, forecast water in graded dispatching, realize the requirement of protection downstream protection object and dam safety and utilization of flood resources, and the efficiency of decision-making be low.

In addition, many storehouse series connection GROUP OF HYDROPOWER STATIONS associating Flood Control Dispatchs are very complicated, containing multiple linear and nonlinear restriction, use optimization routine Algorithm for Solving, usually there is the shortcomings such as amount of calculation is large, the low precision of solution, the improvement Dynamic Programming (POA) that developed recently gets up is commonly used to solve it, but the problem of " dimension " calamity will become in Practical Calculation the obstacle being difficult to overcome.

Summary of the invention

The object of the invention is to, provide a kind of Cascade Reservoirs to combine the method and system of Flood Optimal Scheduling, it dynamically can judge the rank of flood and realize graded dispatching on this basis; Meanwhile, the requirement protecting downstream protection object and dam safety and utilization of flood resources can be realized, improve the efficiency of decision-making.

For solving the problems of the technologies described above, the present invention adopts following technical scheme: a kind of method of Cascade Reservoirs associating Flood Optimal Scheduling, comprises the following steps:

S1, according to real-time database water level with face flood forecasting discharge process and judge flood classification;

S2, according to flood classification and the primary protection object of control point, downstream flood control standard decision-making;

S3, chooses Optimal Operation Model corresponding in flood control optimization aim and system model storehouse automatically;

S4, utilizes Revised genetic algorithum solve Optimal Operation Model and optimize.

Each Optimized Operation target all has corresponding scheduling model, can reach optimization aim corresponding to this model by using principle of optimality and numbered analog simulation to calculate.

Reservoir level one timing, it is larger then more dangerous that flood forecasting carrys out water inventory, when flood volume exceedes some thresholds, whole field flood will be caused all to let out under carrying out lower than downstream security discharge, and now should enter Yi Baoba is main Flood Control Dispatch.Equally, when forecasting that water inventory is certain, the lower then flood control capacity of reservoir level is stronger, can protect downstream flood control object better.Therefore, can Current Library water level, forecast peb process two factors directly determine reservoir and control the maximum stream flow at flood control control point, downstream below safety discharge.Need, on the basis gathering real time information and forecasting process information, to carry out anticipation by numbered analog simulation calculation.Flood classification is mainly divided into the little flood of below downstream level of protection and exceedes the great flood of downstream level of protection.

In the method for aforesaid Cascade Reservoirs associating Flood Optimal Scheduling, the judgement flood classification described in step S1 specifically comprises the following steps: downstream is controlled flood the safety discharge at control point as peak clipping flow by a.; B. peak clipping scheduling is carried out according to real-time database water level, forecast peb process and peak clipping flow; C. the reservoir level process that above-mentioned peak clipping is dispatched is detected, if the highest reservoir level exceedes the design flood level of dam, then this flood is the flood exceeding downstream flood control standard, if the highest reservoir level does not exceed the design flood level of dam, then this flood is the flood lower than downstream flood control standard.

For the little flood not exceeding downstream flood control standard, main optimization aim guarantees downstream protection object, more particularly: downstream flood control loss reduction and flood water resources utilization maximize; For the great flood exceeding downstream flood control standard, the main principle adopting graded dispatching, considers downstream flood control loss, downstream masses transfer time, dam safety and flood water resources utilization and maximizes several target.

In the method for aforesaid Cascade Reservoirs associating Flood Optimal Scheduling, specifically comprising according to flood classification and control point, the downstream primary protection object of flood control standard decision-making described in step S2: if a. exceedes the flood of downstream flood control standard, then primary protection object is the safety of dam self, when adjusting flood into reservoir flow in big vast process to be greater than flood control control point, downstream safety discharge, follow the tracks of under reservoir inflow carries out and let out; If b. lower than the flood of downstream flood control standard, then primary protection object is the flood control control point in downstream, and when adjusting flood into reservoir flow in big vast process to be greater than flood control control point, downstream safety discharge, reservoir storage outflow controls below downstream security discharge.

In the method for aforesaid Cascade Reservoirs associating Flood Optimal Scheduling, step S3, if storage capacity is limited or fixed, then using flood peak clipping flow to greatest extent as optimization aim, maximum peak clipping scheduling mode is adopted to be optimized scheduling, wherein, the scheduling model of described maximum peak clipping scheduling mode is:

Object function:

$\min {\∫}_{t_0}^{t_d}\{{[q_1\left(t\right)+Q_{\mathrm{1,2}}\left(t\right)]}^2+{[q_2\left(t\right)+Q_{\mathrm{2,3}}\left(t\right)]}^2+\mathrm{\Λ}+{[q_n\left(t\right)+Q_{n,n+1}\left(t\right)]}^2\}\mathrm{dt}$

That is:

$\min \underset{j=1}{\overset{M}{\mathrm{\Σ}}}\underset{i=1}{\overset{n}{\mathrm{\Σ}}}{(q_i^j+Q_{i,i+1}^j)}^2\mathrm{\Δt},$ $M=\frac{t_d-t_0}{\mathrm{\Δt}}$

Constraints:

Initial storage capacity constraint: $\underset{j=1}{\overset{M}{\mathrm{\Σ}}}(Q_i^j-q_i^j)\mathrm{\Δt}={\mathrm{\ΔV}}_i$

Maximum letdown flow constraint:

Period letdown flow constraint: $q_i^j\≤\min ({\stackrel{\‾}{Q}}_i^j,Q_i^j)$

Discharge capacity retrains: $q_i^j\≤q_i^j(Z_i^j,B_i^j)$

Water balance: $\frac{Q_i^j+Q_i^{j+1}}{2}-\frac{q_i^j+q_i^{j+1}}{2}=\frac{{\mathrm{\ΔV}}_i^j}{\mathrm{\Δt}}$

Muskingum advance of freshet: $Q_{i+1}^{j+1}=C_{0}q{Q}_i^{j+1}+C_{1}q{Q}_i^j+C_2{Q}_{i+1}^j$

s.t. $q{Q}_i^{j+1}=q_i^{j+1}+Q_{i,i+1}^{j+1}$

$q{Q}_i^j=q_i^j+Q_{i,i+1}^j$ i＝1,2,Λ,n；j＝1,2,Λ,M

Wherein:

be the flood into reservoir flow of i-th grade of reservoir in the jth period;

for solving the maximum peak clipping flow obtained according to object function;

be the storage outflow of i-th grade of reservoir in the jth period;

N is reservoir progression, from top to bottom, and i=1, Λ Λ n, integer;

M is calculation interval number, j=1, Λ Λ M, integer;

Δ t is the time segment length being divided into M period dispatching cycle;

Δ V _iit is the storage capacity of i-th grade of reservoir;

be i-th grade of reservoir and the i-th+1 grade reservoir local inflow in the jth period, and outbound place of reverse calculation to the i-th grade reservoir;

be i-th grade of reservoir the reservoir of jth period go out stream and interval flow and;

be the water level of i-th grade of reservoir in the jth period;

be the gatage of i-th grade of reservoir in the jth period;

be i-th grade of reservoir in the jth period, corresponding water level is aperture is time the maximum letdown flow that can provide.

In the method for aforesaid Cascade Reservoirs associating Flood Optimal Scheduling, in step S3, if in leading time total Incoming water quantity of flood deduct to participate between flood period generating set all the full value sending out the generating water supply volume consumed be a positive, there is maximum minus deviation in forecast, above-mentioned numerical value is still a positive, and when the current storage capacity being used for letting out in advance is greater than above-mentioned positive, to guarantee the safety of single reservoir and step reservoir, and make full use of flood resource, reducing abandoned water amount is as optimization aim to greatest extent, Optimized Operation mode is let out in employing in advance dispatches, make reservoir inflow exceed unit by sluicing position pre-before flood arrives and completely send out flow, be performed for more than retaining of part reservoir inflow, thus increment life insurance, wherein, the described scheduling model letting out Optimized Operation mode is in advance:

Object function:

$\min {\∫}_{t_0}^{t_d}\{{[q_1\left(t\right)+Q_{\mathrm{1,2}}\left(t\right)]}^2+{[q_2\left(t\right)+Q_{\mathrm{2,3}}\left(t\right)]}^2+\mathrm{\Λ}+{[q_n\left(t\right)+Q_{n,n+1}\left(t\right)]}^2\}\mathrm{dt}$

That is:

$\min \underset{j=1}{\overset{M}{\mathrm{\Σ}}}\underset{i=1}{\overset{n}{\mathrm{\Σ}}}{(q_i^j+Q_{i,i+1}^j)}^2\mathrm{\Δt},$ $M=\frac{t_d-t_0}{\mathrm{\Δt}}$

Constraints:

Initial storage capacity constraint: $\underset{j=1}{\overset{M}{\mathrm{\Σ}}}(Q_i^j-q_i^j)\mathrm{\Δt}={\mathrm{\ΔV}}_i$

Maximum letdown flow constraint:

Period letdown flow constraint: $q_i^j\≤\min ({\stackrel{\‾}{Q}}_i^j,Q_i^{j-{\mathrm{\τ}}_{i\backslash }})$

Discharge capacity retrains: $q_i^j\≤q_i^j(Z_i^j,B_i^j)$

Water balance: $\frac{Q_i^j+Q_i^{j+1}}{2}-\frac{q_i^j+q_i^{j+1}}{2}=\frac{{\mathrm{\ΔV}}_i^j}{\mathrm{\Δt}}$

Muskingum advance of freshet: $Q_{i+1}^{j+1}=C_{0}q{Q}_i^{j+1}+C_{1}q{Q}_i^j+C_2{Q}_{i+1}^j$

s.t. $q{Q}_i^{j+1}=q_i^{j+1}+Q_{i,i+1}^{j+1}$

$q{Q}_i^j=q_i^j+Q_{i,i+1}^j$ i＝1,2,Λ,n；j＝1,2,Λ,M

Wherein:

be the flood into reservoir flow of i-th grade of reservoir in the jth period;

for solving the maximum peak clipping flow obtained according to object function;

τ _iit is the leading time of i-th grade of reservoir;

be the storage outflow of i-th grade of reservoir in the jth period;

N is reservoir progression, from top to bottom, and i=1, Λ Λ n, integer;

M is calculation interval number, j=1, Λ Λ M, integer;

Δ t is the time segment length being divided into M period dispatching cycle;

Δ V _iit is the storage capacity of i-th grade of reservoir;

be i-th grade of reservoir and the i-th+1 grade reservoir local inflow in the jth period, and outbound place of reverse calculation to the i-th grade reservoir;

be i-th grade of reservoir the reservoir of jth period go out stream and interval flow and;

be the water level of i-th grade of reservoir in the jth period;

be the gatage of i-th grade of reservoir in the jth period;

be i-th grade of reservoir in the jth period, corresponding water level is aperture is time the maximum letdown flow that can provide.

In the method for aforesaid Cascade Reservoirs associating Flood Optimal Scheduling, utilize Revised genetic algorithum to solve Optimal Operation Model and optimize to comprise the following steps described in step S4:

S41, the generation of initial population, the i.e. formation of initial Flood Control Dispatch scheme individuality;

S42, calculates individual fitness, namely solves object function;

S43, the realization of genetic operator.

Realize the system of a kind of Cascade Reservoirs associating Flood Optimal Scheduling of preceding method, comprising: automatic Hydrological Telemetry System, communication system and waterpower scheduling automation, communication system is connected with automatic Hydrological Telemetry System and waterpower scheduling automation respectively.

In the system of aforesaid Cascade Reservoirs associating Flood Optimal Scheduling, described automatic Hydrological Telemetry System comprises: telemetry station, information transfer channel and console for centralized control, and information transfer channel is connected with telemetry station and console for centralized control respectively; Wherein, telemetry station is used for the real time data of automatically collecting rainfall, water level and other hydrologic parameters, by certain way, these data layouts is become pulse signal, be delivered to console for centralized control by information transfer channel under the control of console for centralized control; Telemetry station comprises: rainfall gauge, water-level gauge, encoder, data set, radio station and power-supply device; Information transfer channel connects the wave transmissions line between telemetry station and console for centralized control, comprises wired and wireless two classes; Console for centralized control is used for the hydrographic data concentrating each telemetry station in telemetry system, carries out calculating and arranges, make flood forecasting in time, and can open and close by regulating gate, carries out water project operation; Console for centralized control comprises communication station and electronic computer.

In the system of aforesaid a kind of Cascade Reservoirs associating Flood Optimal Scheduling, described communication system comprises optical fiber telecommunications system, microwave telecommunication system, production dispatch ring communication system and the power supply system for communications, the power supply system for communications respectively with optical fiber telecommunications system, microwave telecommunication system is connected with production dispatch ring communication system, optical fiber telecommunications system is used for bidirectional transfer of information between each branch center and centralized control center, microwave telecommunication system is used for data between telemetry-acquisition station and affiliated branch center to be transmitted, production dispatch ring communication system is used for centralized control center's internal information transmission, the power supply system for communications is used for providing reliable uninterrupted power source for communication equipment.

In the system of aforesaid a kind of Cascade Reservoirs associating Flood Optimal Scheduling, described waterpower scheduling automation comprises: information transmit-receive unit, Back ground Information memory cell, model storage unit and information process unit, Back ground Information memory cell is connected with information transmit-receive unit and information process unit respectively, and model storage unit is connected with information process unit.

Compared with prior art, the present invention, according to the current operating conditions of reservoir, the peb process faced and downstream security level of protection, dynamically judges the rank of flood and realizes graded dispatching on this basis; Simultaneously, according to integrated informations such as the reliabilities of the adjusting function of reservoir, residing schedule periods, forecast water in graded dispatching, by choosing appropriate Optimized Operation target, achieve the requirement of protection downstream protection object and dam safety and utilization of flood resources, improve the efficiency of decision-making of multi-reservoir simultaneously; In addition, from only to start to carry out iterative computation with an initial point in optimization routine algorithm different, simultaneously Revised genetic algorithum calculates from multiple initial point, finally try to achieve the optimal solution of problem, thus improve the efficiency of decision-making of Cascade Reservoirs, impel the economic and social benefit that final acquisition is maximum.Show according to mass data statistics, adopt the present invention to carry out Flood Dispatching Optimization, the efficiency of decision-making improves about 30%.

Accompanying drawing explanation

Fig. 1 is the structural representation of a kind of embodiment of the present invention;

Fig. 2 is the workflow diagram of a kind of embodiment of the present invention;

Fig. 3 is the flow chart scabbling an Optimized Operation;

Fig. 4 is the basic flow sheet that step hydropower station Flood Optimal Scheduling Floating-point Genetic Algorithm solves;

Fig. 5 is the flow chart letting out Optimized Operation in advance;

Fig. 6 is the adjustment concept map scabbling an Optimized Operation;

Fig. 7 is Reservoir Water Level process schematic when scabbling Optimized Operation;

Fig. 8 is the adjustment concept map letting out Optimized Operation in advance;

Fig. 9 is Reservoir Water Level process schematic when letting out Optimized Operation in advance;

Figure 10 is the schematic diagram of feasible individual.

Reference numeral: 1-automatic Hydrological Telemetry System, 2-communication system, 3-waterpower scheduling automation, 4-telemetry station, 5-information transfer channel, 6-console for centralized control, 7-optical fiber telecommunications system, 8-microwave telecommunication system, 10-production dispatch ring communication system, the 11-power supply system for communications, 12-information transmit-receive unit, 13-Back ground Information memory cell, 14-model storage unit, 15-information process unit.

Below in conjunction with the drawings and specific embodiments, the present invention is further illustrated.

Detailed description of the invention

Embodiments of the invention 1: apply the present invention in the Flood Optimal Scheduling of Cascade Stations on Wujiang River GROUP OF HYDROPOWER STATIONS.A system for Cascade Reservoirs associating Flood Optimal Scheduling, as shown in Figure 1, comprising: automatic Hydrological Telemetry System 1, communication system 2 and waterpower scheduling automation 3, communication system 2 is connected with automatic Hydrological Telemetry System 1 and waterpower scheduling automation 3 respectively.

Described automatic Hydrological Telemetry System 1 comprises: telemetry station 4, information transfer channel 5 and console for centralized control 6, and information transfer channel 5 is connected with telemetry station 4 and console for centralized control 6 respectively; Wherein, telemetry station 4, for automatically collecting the real time data of rainfall, water level and other hydrologic parameters, becomes pulse signal by certain way these data layouts, is delivered to console for centralized control 6 by information transfer channel 5 under the control of console for centralized control 6; Telemetry station 4 comprises: rainfall gauge, water-level gauge, encoder, data set, radio station and power-supply device; Information transfer channel 5 connects the wave transmissions line between telemetry station 4 and console for centralized control 6, comprises wired and wireless two classes; Console for centralized control 6, for the hydrographic data of telemetry station 4 each in concentrated telemetry system, carries out calculating and arranges, make flood forecasting in time, and can open and close by regulating gate, carries out water project operation; Console for centralized control 6 comprises communication station and electronic computer.

Described communication system 2 comprises optical fiber telecommunications system 7, microwave telecommunication system 8, production dispatch ring communication system 10 and the power supply system for communications 11.

Described waterpower scheduling automation 3 comprises: information transmit-receive unit 12, Back ground Information memory cell 13, model storage unit 14 and information process unit 15, Back ground Information memory cell 13 is connected with information transmit-receive unit 12 and information process unit 15 respectively, and model storage unit 14 is connected with information process unit 15.

The operating principle of above-described embodiment: (as shown in Figure 2)

Telemetry station 4 in automatic Hydrological Telemetry System 1 is as rainfall gauge, water-level gauge, encoders etc. collect rainfall automatically, the real time data of water level and other hydrologic parameters, the data collected above are transferred to console for centralized control 6(as communication station and electronic computer by information transfer channel 5 as wired or wireless by telemetry station 4), under the control of console for centralized control 6, by certain way, these data layouts are become pulse signal, console for centralized control 6 is according to real-time database water level and the classification facing flood forecasting discharge process test flood, and by test result information by communication system 2(as the optical fiber telecommunications system 7 of being powered by the power supply system for communications 11, microwave telecommunication system 8, production dispatch ring communication system 10) transfer to waterpower scheduling automation 3, the information transmit-receive unit 12 of waterpower scheduling automation 3 receives flood classification information and stored in Back ground Information memory cell 13, information process unit 15 also chooses according to flood classification and the primary protection object of control point, downstream flood control standard decision-making the Optimal Operation Model stored in flood control optimization aim and reading model memory cell 14 automatically, information process unit 15 utilizes Revised genetic algorithum solve Optimal Operation Model and optimize.

Wherein, judge that flood classification specifically comprises the following steps: downstream is controlled flood the safety discharge at control point as peak clipping flow by a.; B. peak clipping scheduling is carried out according to real-time database water level, forecast peb process and peak clipping flow; C. the reservoir level process that above-mentioned peak clipping is dispatched is detected, if the highest reservoir level exceedes the design flood level of dam, then this flood is the flood exceeding downstream flood control standard, if the highest reservoir level does not exceed the design flood level of dam, then this flood is the flood lower than downstream flood control standard.

Wherein, specifically comprise according to flood classification and control point, the downstream primary protection object of flood control standard decision-making: if a. exceedes the flood of downstream flood control standard, then primary protection object is the safety of dam self, when adjusting flood into reservoir flow in big vast process to be greater than flood control control point, downstream safety discharge, follow the tracks of under reservoir inflow carries out and let out; If b. lower than the flood of downstream flood control standard, then primary protection object is the flood control control point in downstream, and when adjusting flood into reservoir flow in big vast process to be greater than flood control control point, downstream safety discharge, reservoir storage outflow controls below downstream security discharge.

If storage capacity is limited or fixed, then as optimization aim, adopt maximum peak clipping scheduling mode to be optimized scheduling using flood peak clipping flow to greatest extent, wherein, the scheduling model of described maximum peak clipping scheduling mode is:

Object function:

$\min {\∫}_{t_0}^{t_d}\{{[q_1\left(t\right)+Q_{\mathrm{1,2}}\left(t\right)]}^2+{[q_2\left(t\right)+Q_{\mathrm{2,3}}\left(t\right)]}^2+\mathrm{\Λ}+{[q_n\left(t\right)+Q_{n,n+1}\left(t\right)]}^2\}\mathrm{dt}$

That is:

$\min \underset{j=1}{\overset{M}{\mathrm{\Σ}}}\underset{i=1}{\overset{n}{\mathrm{\Σ}}}{(q_i^j+Q_{i,i+1}^j)}^2\mathrm{\Δt},$ $M=\frac{t_d-t_0}{\mathrm{\Δt}}$

Constraints:

Initial storage capacity constraint: $\underset{j=1}{\overset{M}{\mathrm{\Σ}}}(Q_i^j-q_i^j)\mathrm{\Δt}={\mathrm{\ΔV}}_i$

Maximum letdown flow constraint:

Period letdown flow constraint: $q_i^j\≤\min ({\stackrel{\‾}{Q}}_i^j,Q_i^j)$

Discharge capacity retrains: $q_i^j\≤q_i^j(Z_i^j,B_i^j)$

Water balance: $\frac{Q_i^j+Q_i^{j+1}}{2}-\frac{q_i^j+q_i^{j+1}}{2}=\frac{{\mathrm{\ΔV}}_i^j}{\mathrm{\Δt}}$

Muskingum advance of freshet: $Q_{i+1}^{j+1}=C_{0}q{Q}_i^{j+1}+C_{1}q{Q}_i^j+C_2{Q}_{i+1}^j$

s.t. $q{Q}_i^{j+1}=q_i^{j+1}+Q_{i,i+1}^{j+1}$

$q{Q}_i^j=q_i^j+Q_{i,i+1}^j$ i＝1，2，Λ，n；j＝1，2，Λ，M

Wherein:

be the flood into reservoir flow of i-th grade of reservoir in the jth period;

for solving the maximum peak clipping flow obtained according to object function;

be the storage outflow of i-th grade of reservoir in the jth period;

N is reservoir progression, from top to bottom, and i=1, Λ Λ n, integer;

M is calculation interval number, i=1, Λ Λ M, integer;

Δ t is the time segment length being divided into M period dispatching cycle;

Δ V _iit is the storage capacity of i-th grade of reservoir;

be i-th grade of reservoir and the i-th+1 grade reservoir local inflow in the jth period, and outbound place of reverse calculation to the i-th grade reservoir;

be i-th grade of reservoir the reservoir of jth period go out stream and interval flow and;

be the water level of i-th grade of reservoir in the jth period;

be the gatage of i-th grade of reservoir in the jth period;

be i-th grade of reservoir in the jth period, corresponding water level is aperture is time the maximum letdown flow that can provide.

If in leading time total Incoming water quantity of flood deduct to participate between flood period generating set all the full value sending out the generating water supply volume consumed be a positive, there is maximum minus deviation in forecast, above-mentioned numerical value is still a positive, and when the current storage capacity being used for letting out in advance is greater than above-mentioned positive, to guarantee the safety of single reservoir and step reservoir, and make full use of flood resource, reducing abandoned water amount is as optimization aim to greatest extent, Optimized Operation mode is let out in employing in advance dispatches, make reservoir inflow exceed unit by sluicing position pre-before flood arrives and completely send out flow, be performed for more than retaining of part reservoir inflow, thus increment life insurance, wherein, the described scheduling model letting out Optimized Operation mode is in advance:

Object function:

$\min {\∫}_{t_0}^{t_d}\{{[q_1\left(t\right)+Q_{\mathrm{1,2}}\left(t\right)]}^2+{[q_2\left(t\right)+Q_{\mathrm{2,3}}\left(t\right)]}^2+\mathrm{\Λ}+{[q_n\left(t\right)+Q_{n,n+1}\left(t\right)]}^2\}\mathrm{dt}$

That is:

$\min \underset{j=1}{\overset{M}{\mathrm{\Σ}}}\underset{i=1}{\overset{n}{\mathrm{\Σ}}}{(q_i^j+Q_{i,i+1}^j)}^2\mathrm{\Δt},$ $M=\frac{t_d-t_0}{\mathrm{\Δt}}$

Constraints:

Initial storage capacity constraint: $\underset{j=1}{\overset{M}{\mathrm{\Σ}}}(Q_i^j-q_i^j)\mathrm{\Δt}={\mathrm{\ΔV}}_i$

Maximum letdown flow constraint:

Period letdown flow constraint: $q_i^j\≤\min ({\stackrel{\‾}{Q}}_i^j,Q_i^{j-{\mathrm{\τ}}_{i\backslash }})$

Discharge capacity retrains: $q_i^j\≤q_i^j(Z_i^j,B_i^j)$

Water balance: $\frac{Q_i^j+Q_i^{j+1}}{2}-\frac{q_i^j+q_i^{j+1}}{2}=\frac{{\mathrm{\ΔV}}_i^j}{\mathrm{\Δt}}$

Muskingum advance of freshet: $Q_{i+1}^{j+1}=C_{0}q{Q}_i^{j+1}+C_{1}q{Q}_i^j+C_2{Q}_{i+1}^j$

s.t. $q{Q}_i^{j+1}=q_i^{j+1}+Q_{i,i+1}^{j+1}$

$q{Q}_i^j=q_i^j+Q_{i,i+1}^j$ i＝1,2,Λ,n；j＝1,2,Λ,M

Wherein:

be the flood into reservoir flow of i-th grade of reservoir in the jth period;

for solving the maximum peak clipping flow obtained according to object function;

τ _iit is the leading time of i-th grade of reservoir;

be the storage outflow of i-th grade of reservoir in the jth period;

N is reservoir progression, from top to bottom, and i=1, Λ Λ n, integer;

M is calculation interval number, j=1, Λ Λ M, integer;

Δ t is the time segment length being divided into M period dispatching cycle;

Δ V _iit is the storage capacity of i-th grade of reservoir;

be i-th grade of reservoir and the i-th+1 grade reservoir local inflow in the jth period, and outbound place of reverse calculation to the i-th grade reservoir;

be i-th grade of reservoir the reservoir of jth period go out stream and interval flow and;

be the water level of i-th grade of reservoir in the jth period;

be the gatage of i-th grade of reservoir in the jth period;

be i-th grade of reservoir in the jth period, corresponding water level is aperture is time the maximum letdown flow that can provide.

To Cascade Stations on Wujiang River flood control operation, at present be adjusted to example with Goupitan to three cascades of husky a small bay in a river, when Goupitan meets with a flood, the magnitude (peak, amount, the frequency of occurrences) of this flood is known according to weather report, with current retaining situation for foundation, that drafts this flood retains the water yield, and the Flood Control of carrying out " scabbling head " formula by this calculates, namely carry out maximum peak clipping scheduling, finally realize the scheduling of flood in Goupitan.Because flood can occur develop and accompany by certain time lag between upstream and downstream reservoir, utilize Muskingun method to outbound flood routing, and obtain the flood into reservoir in subordinate storehouse with district flood superposition, equally subordinate storehouse is carried out to the Flood Control calculating of " scabbling head " formula, until most final stage, what complete this flood scabbles an Optimized Operation.

Mathematical Modeling

(I) least square and module (head module is scabbled in single storehouse)

The criterion that this module adopts is maximum letdown flow minimization principle, and its representation of concept is:

Min{Max[q(t)]} t∈[t ₀,t _d] （1.3）

Or Min{Max [q (t)+Q _district(t)] } t ∈ [t ₀, t _d] (1.4)

Q (t) represents letdown flow, Q _districtflow between (t) Representative Region.Front formula is that single storehouse is without district flood; Rear formula represents Dan Ku district flood.Its physical significance is through the lower flood discharge water (consider interval impact) of Optimized Operation, and the discharge process reaching flood control protection point reaches most uniform state.In Cascade Stations on Wujiang River Flood Dispatching Optimization, in order to realize the safety of step reservoir entirety, its subordinate's reservoir is all optimized scheduling as object of protection by each reservoir.Above-mentioned module criterion can realize by following object function, constraints.

Object function:

Single storehouse is without district flood: $\mathrm{Min}{\∫}_{t_0}^{t_d}{q}^2\left(t\right)\mathrm{dt}$

There is district flood in single storehouse:

Discrete form: $\mathrm{Min}\underset{j=1}{\overset{M}{\mathrm{\Σ}}}{\left[q_j\left(t\right)\right]}^2\mathrm{\Δt}$ j＝1,2,Λ,M； $M=\frac{t_d-t_0}{\mathrm{\Δt}};$

j＝1,2,Λ,M；

Constraints:

Storage capacity retrains

Without constraint q when letting out in advance _j(t)≤Q _j(t)

Gate discharge capacity constraint q _j(t)≤q (Z _j, B _j)

Reservoir water yield Constraints of Equilibrium $\frac{Q_j+Q_{j+1}}{2}-\frac{q_j+q_{j+1}}{2}=\frac{{\mathrm{\ΔV}}_j}{\mathrm{\Δt}}$

For Cascade Stations on Wujiang River multi-reservoir, the warehouse-in process of lower storage basin also needs to meet water flow advance constraint:

Q _i＝f(q _i-1)+Q _i′

(II) step reservoir least square and module (step scabbles head module)

Object function:

$\mathrm{Min}{\∫}_{t_0}^{t_d}\{{[q_1\left(t\right)+Q_{\mathrm{1,2}}\left(t\right)]}^2+{[q_2\left(t\right)+Q_{\mathrm{2,3}}\left(t\right)]}^2+\mathrm{\Λ}+{[q_n\left(t\right)+Q_{n,n+1}\left(t\right)]}^2\}\mathrm{dt}$

Discrete form: $\mathrm{Min}\underset{j=1}{\overset{M}{\mathrm{\Σ}}}\underset{i=1}{\overset{n}{\mathrm{\Σ}}}{(q_i^j+Q_{i,i+1}^j)}^2\mathrm{\Δt},$ $M=\frac{t_d-t_0}{\mathrm{\Δt}};$

I is step reservoir sequence number, arranges from top to bottom; N is step reservoir number;

Constraints: $\underset{j=1}{\overset{M}{\mathrm{\Σ}}}(Q_{j,i}-q_{j,i})\mathrm{\Δt}={\mathrm{\ΔV}}_i$ i＝1,2,Λ,n；

q _j,i≤Q _j,ii＝1,2,Λ,n；j＝1,2,Λ,M；

q _j,i≤q _j,i(Z _j,i,B _j,i)；

$\frac{Q_{j,i}+Q_{j+1,i}}{2}-\frac{q_{j,i}+q_{j+1,i}}{2}=\frac{{\mathrm{\ΔV}}_{j,i}}{\mathrm{\Δt}};$

Q _j+1,i+1＝c ₀Q′ _j+1,i+c ₁Q′ _j,i+c ₂Q _j,i+1；

Q′ _j+1,i＝q _j+1,i+ΔQ _j+1,i+1；

Q′ _j，i＝q _j，i+ΔQ _j,i+1；

Q _{j, i}be i-th grade of storehouse jth period average reservoir inflow;

Q _{j, i}be i-th grade of storehouse jth period average storage outflow;

Δ Q _{j, i+1}be the interval between i-th grade of storehouse and the i-th+1 grade storehouse, at the flow (and reverse calculation to the i-th grade outbound) of j period;

Q ' _j,iit is i-th grade of storehouse jth period average storage outflow (having contained at the corresponding levels to the interval flow of subordinate);

Z _j,ibe i-th grade of storehouse jth period mean water;

B _j,ibe i-th grade of storehouse jth period gatage;

Q _j,i(Z _j,i, B _j,i) be i-th grade of storehouse jth period be Z at water level _j,i, aperture is B _j,itime maximum under let out ability;

Q ' _ifor local inflow;

Q _ifor the reservoir inflow process in i level storehouse;

F (q _i-1) be the storage outflow process in i-1 level storehouse.

This model assumption flood into reservoir Q _t, district flood Q ' _t, storage capacity V _anti-and flood carrying capacity etc. is known.The object of Optimized Operation obtains the most uniform flash process of letdown flow exactly under above-mentioned various known conditions, i.e. storage capacity one timing, q _tquadratic sum is minimum, and to be equivalent to letdown flow the most even.Fig. 6 is with the schematic diagram of maximum peak clipping for criterion Optimized Operation.

As can be known from Fig. 7: t ₁to t ₂for the peak clipping period, be also that reservoir level goes up the stage, utilize flood control by reservoir regulation storage capacity water conservation during this period, t ₁time water level be still scheduling just water level, t ₂time water level be regulation goal water level; t ₁be the period before peak clipping, letdown flow is identical with reservoir inflow, and reservoir level remains unchanged before; t ₂be the period after peak clipping afterwards, letdown flow is identical with reservoir inflow, and it is constant that reservoir level maintains target water level.

Scabble the flow process of an Optimized Operation and maximum peak clipping scheduling as shown in Figure 3.

Wherein, utilize Revised genetic algorithum to solve (as shown in Figure 4) to Optimal Operation Model to comprise:

S41, the generation of initial population, the i.e. formation of initial Flood Control Dispatch scheme individuality;

S42, calculates individual fitness, namely solves object function;

S43, the realization of genetic operator.

(I) generation of initial population

Utilize Revised genetic algorithum and FPGA to solve Cascade Stations on Wujiang River power station Flood Optimal Scheduling, key is the selection of initial feasible solution.FPGA is directly with the real-valued conduct coding of decision variable, and the length of coding equals the number of decision variable.Decision variable when selecting the flood control storage outflow of step hydropower station to solve as FPGA, it is made up of the storage outflow vector of day part:

Q＝(q ₁,q ₂,Λ,q _i,Λ,q _M)

In formula: $q_1={(q_1^1,q_2^1,\mathrm{\Λ},q_n^1)}^T,$ ...， $q_i={(q_1^i,q_2^i,\mathrm{\Λ},q_n^i)}^T,$ ...， $q_M={(q_1^M,q_2^M,\mathrm{\Λ},q_n^M)}^T.$

Only to start to carry out iterative computation with an initial point different from optimization routine algorithm, and simultaneously genetic algorithm calculates from multiple initial point, finally tries to achieve the optimal solution of problem.Before execution genetic algorithm, first provide by P the individual set formed, i.e. initial population, they can be expressed as:

$Q_1=(q_1^1,q_2^1,\mathrm{\Λ},q_M^1)$

$Q_2=(q_1^2,q_2^2,\mathrm{\Λ}{,q}_M^2)$

$Q_P=(q_1^P,q_2^P,\mathrm{\Λ}{,q}_M^P)$

In schedule periods two Phase flow and local inflow known, the initial reservoir level of reservoir is known, scheduling end of term water level is determined, each power station can be put into operation under the known condition of the discharge capacity of unit number of units and earial drainage facility, in step associating Flood Optimal Scheduling, the flood control storage outflow process of each reservoir has certain constraint, so not all individuality is all feasible.Corresponding Optimized model, not only require that the flood control storage outflow of each reservoir day part meets flood discharge and conveyance capacity constraint, the initial storage capacity constraint of reservoir etc. in power station, the warehouse-in in subordinate storehouse need be superposed with district flood by the outbound in higher level storehouse and try to achieve after Muskingum advance of freshet simultaneously.Often produce body one by one for this reason, all must carry out feasibility checking to it.If a certain individuality is infeasible, then eliminate this individuality immediately, the schematic diagram of feasible individual as shown in Figure 10.

(II) fitness calculates

The genetic algorithm of floating-point encoding is adopted to eliminate the decode procedure adopting binary coding genetic algorithm, for solving of step hydropower station Flood Optimal Scheduling, its target function value always get on the occasion of, therefore, directly can set individual fitness and just equal corresponding target function value, getting the minimum solution of target function value is optimal solution.In algorithm, discharge capacity and artificial flood constrain in the coding of scheduling scheme and automatically meet, and other constraintss such as initial storage capacity constraint and water balance etc. are then paid attention to when designing genetic operator.Fitness function is:

$F=\min \underset{j=1}{\overset{M}{\mathrm{\Σ}}}\underset{i=1}{\overset{n}{\mathrm{\Σ}}}{(q_i^j+Q_{i,i+1}^j)}^2\mathrm{\Δt}$

(III) design of genetic operator

Selected initial population, determine fitness computational methods after, adopt determine formula sampling system of selection and Floating-point Genetic Algorithm the arithmetic crossover that is suitable for and uniformity variation means, the colony of new generation that generation outbound discharge process quadratic sum is less.For ensureing global convergence, after mutation operation, adopting best individual preservation strategy, in G generation, namely after variation, retaining the optimum individual in this colony and adaptive value thereof in generation.Circulation like this, until meet Optimality Criteria or tried to achieve satisfactory solution.The basic procedure that step hydropower station Flood Optimal Scheduling Floating-point Genetic Algorithm solves as shown in Figure 4.

Embodiments of the invention 2: the system that the present embodiment adopts is identical with embodiment 1.Adopt and let out Optimized Operation mode in advance.The general principle of letting out Optimized Operation is in advance instantaneous delivery balance, period water balance.Scheduling adheres to the basic principles as security doctrine and economic principle, i.e. not only this storehouse safety, also will create conditions safely for subordinate storehouse, abandon water less simultaneously as far as possible, under the condition of not abandoning water, make full use of step reservoir head benefit.

Mathematical Modeling

Let out the model of the substantially roughly the same maximum peak clipping scheduling mode of Mathematical Modeling of Optimized Operation mode in advance, but conceptually have fine difference, be explained as follows:

Take to abandon water and let out a part of storage capacity that soars in advance, have obvious benefit, one is that earial drainage process is more steady, from t ' ₀to t _ealmost always with q+Q _t(wherein q is for abandoning discharge, Q _tfor generating flow) under let out; Two is reduce large discharge flood discharge to the damage of mining under reservoir erosion control energy-dissipating installation; The fluctuation of generating tail water is comparatively mild, less on generated output impact; Maximum advantage is minimized further by letdown flow, and it is also less than the maximum letdown flow scabbling a scheduling mode.

Abandon water criterion:

W _flood-Δ V-Q _t(t _d-t ₀) (1.6)

When (1.6) formula is less than 0, this storehouse flood does not produce abandons water.And (1.6) formula is when being greater than 0, then abandon water, studying abe horizon (scabbling head) in fig. 8 makes area that abcja encloses equal area that befgb encloses, and under after this Flood Control of this storehouse actual, the process of letting out is abeh, and letdown flow is q+Q _t, under let out and last t _d-t ' ₀.Δ V is flood prevention storage capacity.

Fig. 9 is Reservoir Water Level process schematic when letting out Optimized Operation in advance.

Should be noted when changing theoretical procedure line into actual operating gate and letting out stage lowest water level value in advance, regulate in the less reservoir of storage capacity at some and likely can meet with level of dead water or corresponding limiting water level, the gatage lowest water level met in schedule periods need be adjusted and be not less than level of dead water or corresponding limiting water level.

Let out Optimized Operation flow process in advance as shown in Figure 5.

Claims (6)

1. a method for Cascade Reservoirs associating Flood Optimal Scheduling, is characterized in that, comprise the following steps:

S1, automatic Hydrological Telemetry System judges flood classification according to real-time database water level and the flood forecasting discharge process faced; Wherein, described judgement flood classification specifically comprises the following steps: downstream is controlled flood the safety discharge at control point as peak clipping flow by a.; B. peak clipping scheduling is carried out according to real-time database water level, forecast peb process and peak clipping flow; C. the reservoir level process that above-mentioned peak clipping is dispatched is detected, if the highest reservoir level exceedes the design flood level of dam, then this flood is the flood exceeding downstream flood control standard, if the highest reservoir level does not exceed the design flood level of dam, then this flood is the flood lower than downstream flood control standard;

S2, according to flood classification and the primary protection object of control point, downstream flood control standard decision-making, specifically comprise: if a. exceedes the flood of downstream flood control standard, then primary protection object is the safety of dam self, when adjusting flood into reservoir flow in big vast process to be greater than flood control control point, downstream safety discharge, follow the tracks of under reservoir inflow carries out and let out; If b. lower than the flood of downstream flood control standard, then primary protection object is the flood control control point in downstream, and when adjusting flood into reservoir flow in big vast process to be greater than flood control control point, downstream safety discharge, reservoir storage outflow controls below downstream security discharge;

S3, chooses Optimal Operation Model corresponding in flood control optimization aim and system model storehouse automatically;

S4, utilizes Revised genetic algorithum solve Optimal Operation Model and optimize.

2. the method for Cascade Reservoirs associating Flood Optimal Scheduling according to claim 1, it is characterized in that, in step S3, if storage capacity is limited or fixed, then using flood peak clipping flow to greatest extent as optimization aim, adopt maximum peak clipping scheduling mode to be optimized scheduling, wherein, the scheduling model of described maximum peak clipping scheduling mode is:

Object function:

$\min {\∫}_{t_0}^{t_d}\{{[q_1\left(t\right)+Q_{\mathrm{1,2}}\left(t\right)]}^2+{[q_2\left(t\right)+Q_{\mathrm{2,3}}\left(t\right)]}^2+\mathrm{\Λ}+{[q_n\left(t\right)+Q_{n,n+1}\left(t\right)]}^2\}\mathrm{dt}$

That is:

$\min \underset{j=1}{\overset{M}{\mathrm{\Σ}}}\underset{i=1}{\overset{n}{\mathrm{\Σ}}}{(q_i^j+Q_{i,i+1}^j)}^2\mathrm{\Δt},M=\frac{t_d-t_0}{\mathrm{\Δt}}$

Constraints:

Initial storage capacity constraint: $\underset{j=1}{\overset{M}{\mathrm{\Σ}}}(Q_i^j-q_i^j)\mathrm{\Δt}=\mathrm{\Δ}{V}_i$

Maximum letdown flow constraint:

Period letdown flow constraint: $q_i^j\≤\min ({\stackrel{\‾}{Q}}_i^j,Q_i^j)$

Discharge capacity retrains: $q_i^j\≤q_i^j(Z_i^j,B_i^j)$

Water balance: $\frac{Q_i^j+Q_i^{j+1}}{2}-\frac{q_i^j+q_i^{j+1}}{2}=\frac{\mathrm{\Δ}{V_i}^j}{\mathrm{\Δt}}$

Muskingum advance of freshet: $Q_{i+1}^{j+1}=C_{0}q{Q}_i^{j+1}+C_{1}q{Q}_i^j+C_2{Q}_{i+1}^j$

$\begin{array}{cc}s.t.& q{Q}_i^{j+1}=q_i^{j+1}+Q_{i,i+1}^{j+1}\end{array}$

$q{Q}_i^j=q_i^j+Q_{i,i+1}^{j}i=\mathrm{1,2},\mathrm{\Λ},n;j=\mathrm{1,2},\mathrm{\Λ},M$

Wherein:

be the flood into reservoir flow of i-th grade of reservoir in the jth period;

for solving the maximum peak clipping flow obtained according to object function;

be the storage outflow of i-th grade of reservoir in the jth period;

N is reservoir progression, from top to bottom, and i=1, Λ Λ n, integer;

M is calculation interval number, j=1, Λ Λ M, integer;

Δ t is the time segment length being divided into M period dispatching cycle;

Δ V _iit is the storage capacity of i-th grade of reservoir;

be i-th grade of reservoir and the i-th+1 grade reservoir local inflow in the jth period, and outbound place of reverse calculation to the i-th grade reservoir;

be i-th grade of reservoir the reservoir of jth period go out stream and interval flow and;

be the water level of i-th grade of reservoir in the jth period;

be the gatage of i-th grade of reservoir in the jth period;

be i-th grade of reservoir in the jth period, corresponding water level is aperture is time the maximum letdown flow that can provide.

3. the method for Cascade Reservoirs associating Flood Optimal Scheduling according to claim 1, it is characterized in that, in step S3, if in leading time total Incoming water quantity of flood deduct to participate between flood period generating set all the full value sending out the generating water supply volume consumed be a positive, there is maximum minus deviation in forecast, above-mentioned numerical value is still a positive, and when the current storage capacity being used for letting out in advance is greater than above-mentioned positive, to guarantee the safety of single reservoir and step reservoir, and make full use of flood resource, reducing abandoned water amount is as optimization aim to greatest extent, Optimized Operation mode is let out in employing in advance dispatches, make reservoir inflow exceed unit by sluicing position pre-before flood arrives and completely send out flow, be performed for more than retaining of part reservoir inflow, thus increment life insurance, wherein, the described scheduling model letting out Optimized Operation mode is in advance:

Object function:

$\min {\∫}_{t_0}^{t_d}\{{[q_1\left(t\right)+Q_{\mathrm{1,2}}\left(t\right)]}^2+{[q_2\left(t\right)+Q_{\mathrm{2,3}}\left(t\right)]}^2+\mathrm{\Λ}+{[q_n\left(t\right)+Q_{n,n+1}\left(t\right)]}^2\}\mathrm{dt}$

That is:

$\min \underset{j=1}{\overset{M}{\mathrm{\Σ}}}\underset{i=1}{\overset{n}{\mathrm{\Σ}}}{(q_i^j+Q_{i,i+1}^j)}^2\mathrm{\Δt},M=\frac{t_d-t_0}{\mathrm{\Δt}}$

Constraints:

Initial storage capacity constraint: $\underset{j=1}{\overset{M}{\mathrm{\Σ}}}(Q_i^j-q_i^j)\mathrm{\Δt}=\mathrm{\Δ}{V}_i$

Maximum letdown flow constraint:

Period letdown flow constraint: $q_i^j\≤\min ({\stackrel{\‾}{Q}}_i^j,Q_i^{j-{\mathrm{\τ}}_{i\backslash }})$

Discharge capacity retrains: $q_i^j\≤q_i^j(Z_i^j,B_i^j)$

Water balance: $\frac{Q_i^j+Q_i^{j+1}}{2}-\frac{q_i^j+q_i^{j+1}}{2}=\frac{\mathrm{\Δ}{V_i}^j}{\mathrm{\Δt}}$

Muskingum advance of freshet: $Q_{i+1}^{j+1}=C_{0}q{Q}_i^{j+1}+C_{1}q{Q}_i^j+C_2{Q}_{i+1}^j$

$\begin{array}{cc}s.t.& q{Q}_i^{j+1}=q_i^{j+1}+Q_{i,i+1}^{j+1}\end{array}$

$q{Q}_i^j=q_i^j+Q_{i,i+1}^{j}i=\mathrm{1,2},\mathrm{\Λ},n;j=\mathrm{1,2},\mathrm{\Λ},M$

Wherein:

be the flood into reservoir flow of i-th grade of reservoir in the jth period;

for solving the maximum peak clipping flow obtained according to object function;

τ _iit is the leading time of i-th grade of reservoir;

be the storage outflow of i-th grade of reservoir in the jth period;

N is reservoir progression, from top to bottom, and i=1, Λ Λ n, integer;

M is calculation interval number, j=1, Λ Λ M, integer;

Δ t is the time segment length being divided into M period dispatching cycle;

Δ V _iit is the storage capacity of i-th grade of reservoir;

be i-th grade of reservoir and the i-th+1 grade reservoir local inflow in the jth period, and outbound place of reverse calculation to the i-th grade reservoir;

be i-th grade of reservoir the reservoir of jth period go out stream and interval flow and;

be the water level of i-th grade of reservoir in the jth period;

be the gatage of i-th grade of reservoir in the jth period;

be i-th grade of reservoir in the jth period, corresponding water level is aperture is time the maximum letdown flow that can provide.

4. realize the system of a kind of Cascade Reservoirs associating Flood Optimal Scheduling of method described in claims 1 to 3, it is characterized in that, comprise: automatic Hydrological Telemetry System (1), communication system (2) and waterpower scheduling automation (3), communication system (2) is connected with automatic Hydrological Telemetry System (1) and waterpower scheduling automation (3) respectively; Wherein, described automatic Hydrological Telemetry System (1) comprising: telemetry station (4), information transfer channel (5) and console for centralized control (6), and information transfer channel (5) is connected with telemetry station (4) and console for centralized control (6) respectively; Wherein, telemetry station (4) is for automatically collecting the real time data of rainfall, water level and other hydrologic parameters, under the control of console for centralized control (6), by certain way, these data layouts are become pulse signal, be delivered to console for centralized control (6) by information transfer channel (5); Telemetry station (4) comprising: rainfall gauge, water-level gauge, encoder, data set, radio station and power-supply device; Information transfer channel (5) connects the wave transmissions line between telemetry station (4) and console for centralized control (6), comprises wired and wireless two classes; Console for centralized control (6), for the hydrographic data of telemetry station (4) each in concentrated telemetry system, carries out calculating and arranges, make flood forecasting in time, and can open and close by regulating gate, carries out water project operation; Console for centralized control (6) comprises communication station and electronic computer.

5. the system of Cascade Reservoirs associating Flood Optimal Scheduling according to claim 4, it is characterized in that, described communication system (2) comprises optical fiber telecommunications system (7), microwave telecommunication system (8), production dispatch ring communication system (10) and the power supply system for communications (11), the power supply system for communications (11) respectively with optical fiber telecommunications system (7), microwave telecommunication system (8) is connected with production dispatch ring communication system (10), optical fiber telecommunications system (7) is for bidirectional transfer of information between each branch center and centralized control center, microwave telecommunication system (8) transmits for data between telemetry-acquisition station and affiliated branch center, production dispatch ring communication system (10) transmits for centralized control center's internal information, the power supply system for communications (11) is for providing reliable uninterrupted power source for communication equipment.

6. the system of Cascade Reservoirs associating Flood Optimal Scheduling according to claim 4, it is characterized in that, described waterpower scheduling automation (3) comprising: information transmit-receive unit (12), Back ground Information memory cell (13), model storage unit (14) and information process unit (15), Back ground Information memory cell (13) is connected with information transmit-receive unit (12) and information process unit (15) respectively, and model storage unit (14) is connected with information process unit (15).