CN104778519A - VSC-MTDC power flow robust optimization method based on source and load uncertainty - Google Patents

Abstract

The invention discloses a VSC-MTDC power flow robust optimization method based on the source and load uncertainty. The VSC-MTDC power flow robust optimization method is designed for VSC-MTDC steady-state power flow, the uncertainty of the wind-turbine generating capacity and the uncertain of the load power consumption are comprehensively considered, and the robust optimization technology is used. By means of the VSC-MTDC power flow robust optimization method, a conventional power system power flow optimization method based on a wind power and load predicting and determining model is broken through, an uncertainty wind-turbine contribution model is built on the basis of a wind-speed and wind-turbine contribution relation, an uncertainty load model is built on the basis of load predicting, the flexible load peak valley leveling function is achieved, then a flexible direct-current power transmission network optimization objective function is built, solving of the optimization objective function is achieved in the mode that a robust optimization method is combined with an intelligent optimization algorithm under the condition that wind-turbine power generation and load power consumption are uncertain, reasonable power and voltage control reference values are provided for a VSC in a VSC-MTDC system, and the VSC-MTDC power flow robust optimization method strives to achieve the aim that the system safely, stably and economically runs under the uncertainty conditions.

Description

Based on source and lotus probabilistic VSC-MTDC robust Optimal Power Flow method

Technical field

The present invention relates to electric power system optimization running technology field, relate to voltage source converter-Multi-end flexible direct current transmission (VSC-MTDC) system specifically based on source and lotus probabilistic robust Optimal Power Flow research method.

Background technology

Since flexible DC power transmission development, support the access of large quantities of distributed power source particularly extensive centralized wind-powered electricity generation, but the natural quality of wind energy determines the strong undulatory property of wind-power electricity generation and strong intermittence, this uncertain factor brings new problem and challenge to the electric system tide optimization containing wind energy turbine set.In addition, whether electric load electricity consumption also has certain randomness, the flexible load in addition grown up under workload demand side management system, and energy active participate electrical network regulation mechanism, realize and mains side two-way interaction, this makes distribution optimizing research more complicated.

Tradition large-scale wind power grid-connected optimizing research be mostly be based upon determine wind-powered electricity generation and load prediction basis on carry out, but, comparatively big error is still there is in wind-powered electricity generation and load power prediction compared with actual, wind-powered electricity generation and load power predicted data is utilized to be optimized system, the impact that the uncertainty ignoring wind-powered electricity generation and load is brought, thus cannot the contingent extreme ruuning situation of answering system in time, cause the security of operation of system to be subject to serious threat.Therefore, when formulating the electric power system optimization strategy containing wind energy turbine set, the impact that wind-powered electricity generation uncertainty and negative rules bring to electrical network must be considered.

At present, the optimizing research considering the new forms of energy distributed generation system of wind-powered electricity generation and negative rules is very few, therefore on wind-powered electricity generation, load prediction basis, increase its uncertain factor electric power system optimization method and Optimized model are urgently set up, and have important research meaning based on wind-powered electricity generation and the research of flexible load probabilistic VSC-MTDC system robust Optimal Power Flow.

Summary of the invention

The technical matters that the present invention solves is: for existing electric power system optimization method on data prediction basis in the deficiency solving the Optimization Problems In Power Systems containing uncertain factor, herein to be optimized for example based on source and the probabilistic VSC-MTDC steady-state load flow of lotus, first on wind speed and load prediction basis, set up uncertain blower fan to exert oneself model and uncertain flexible load model, then flexible DC power transmission net busbar voltage expense is in violation of rules and regulations set up, the optimization object function that power transmission loss expense is minimum, robust optimisation technique is utilized to be converted into by the Solve problems of objective function containing the minimum conservative result under the worst case of uncertain factor, last combined with intelligent optimized algorithm solves Robust Optimization Model.

The technical scheme that technical solution problem of the present invention is taked is:

1, uncertain blower fan mathematical model is set up:

Forecasting wind speed data basis increases uncertain factor, and under making any instant, wind speed all changes between a uncertain neighborhood based on this point prediction wind speed, then uncertain wind speed mathematic model can be represented by the formula:

$V=\{v\∈R^{N_T},\underset{i\∈N_T}{\mathrm{\Σ}}\frac{|v_i-v_i^{\mathrm{fix}}|}{\widehat{v_i}}\≤\mathrm{\Δ},v_i\∈[v_i-\widehat{v_i},v_i+\widehat{v_i}],\∀i\∈N_T\}$

Wherein, V represents the indeterminacy section collection of wind speed v; N _trepresent that value is counted in time T; for t _imoment wind speed sample value is a determined value; The maximum uncertainty degree of wind speed passes through limited; Δ determines the restriction requirement of uncertain total departure.

Above uncertain wind speed mathematic model can be reduced to following formula:

$V=\{v\∈R^n:v_n^{\mathrm{fix}}{D}_n^{-}\≤v_n\≤v_n^{\mathrm{fix}}{D}_n^{+},\∀n\}$

From above formula, the scale of wind speed indeterminacy section collection is by parameter determine, when time, namely uncertainty is 0.

Funtcional relationship between the active-power P (v) that blower fan exports and wind speed v can approximate description be:

$P\left(v\right)=\left\{\begin{array}{cc}0& v<v_{\mathrm{ci}}\mathrm{orv}>v_{\mathrm{co}}\\ k_{1}v+k_2& v_{\mathrm{ci}}\&le;v<v_{\mathrm{rate}}\\ P_{\mathrm{rate}}& v_{\mathrm{rate}}\&le;v<v_{\mathrm{co}}\end{array}\right.$

In formula: k ₁=P _rate/ (v _rate-v _ci), k ₂=-k ₁v _ci, P _ratefor the rated power of aerogenerator; v _ci, v _rate, v _cobe respectively incision wind speed, wind rating and cut-out wind speed.And if only if v _ci≤ v < v _ratetime, blower fan output power is relevant with wind speed, and when namely wind speed is only in this is interval, uncertain just the exerting oneself on blower fan of wind speed produces uncertain impact.

Uncertain wind speed mathematic model is substituted into above formula determinacy blower fan model and can obtain uncertain blower fan mathematical model, be shown below:

$W=\{P\&Element;R^n:P_n\left(v_n^{\mathrm{fix}}{D}_n^{-}\right)\&le;P_n\&le;P_n\left(v_n^{\mathrm{fix}}{D}_n^{+}\right),\&ForAll;n\}$

2, uncertain flexible load mathematical model is set up:

Load prediction data basis increases uncertain factor, and under making any instant, load all changes between a uncertain neighborhood based on this point prediction load, then uncertain load mathematical model can be represented by the formula:

$D=\{L\&Element;R^n:L_n^{\mathrm{fix}}{D}_n^{-}\&le;L_n\&le;L_n^{\mathrm{fix}}{D}_n^{+},\&ForAll;n\}$

From above formula, the scale of load indeterminacy section collection is by parameter determine, when time, namely uncertainty is 0.

Because flexible load active response can realize peak load shifting effect under tou power price mechanism, namely uncertain flexible load initiatively increases load in low power consumption district, and recruitment is η (%), then uncertain load model interval limit is corrected as in peak of power consumption, district initiatively reduces load, and decrease is μ (%), then the interval upper limit of uncertain load model is corrected as this is uncertain flexible load mathematical model, and wherein η, μ are relevant with time, local electricity price etc.

3, optimization object function is set up:

For improving wind electricity digestion capability, blower fan exerts oneself a part for meeting the need for electricity of load nearby, remainder is then sent by Multi-end flexible direct current transmission Internet Transmission, and therefore in direct current transportation network, rational power division is conducive to network economy, safe operation.The optimization object function that the present invention sets up requires whole flexible DC power transmission net loss minimization, and the restriction of all DC bus-bar voltage is in the reasonable scope, and flexible DC power transmission net optimization object function is:

$\min F=\underset{t=1}{\overset{T}{\mathrm{\&Sigma;}}}{P}_{\mathrm{loss}}+\underset{t=1}{\overset{T}{\mathrm{\&Sigma;}}}\underset{j=1}{\overset{N_d}{\mathrm{\&Sigma;}}}(U_j-U_N)$

Wherein, P _lossrepresent system losses, comprise current conversion station loss and DC network loss.If current conversion station loss becomes κ % proportional relation with its transmission capacity, then n _drepresent DC bus node number, U _jrepresent DC bus-bar voltage, U _nrepresent DC bus rated voltage.

Flexible DC power transmission net optimization object function constraint condition can be divided into equality constraint and inequality constrain, and for the multiterminal element network of complexity, its equality constraint is:

$\begin{array}{cc}{s}_{\mathrm{pi}}{I}_{\mathrm{di}}-\underset{j\&Element;{\mathrm{\&Phi;}}_d}{\mathrm{\&Sigma;}}{g}_{\mathrm{dij}}{V}_{\mathrm{dj}}=0& i\&Element;{\mathrm{\&Phi;}}_d\end{array}$

Wherein, Φ _drepresent the connected node set of alternating current-direct current electrical network; When voltage source converter is rectifier, s _p=1; When voltage source converter is inverter, s _p=-1; g _dijit is the nodal-admittance matrix of DC network; I _diit is DC current vector; V _djit is DC voltage vector.

Inequality constrain condition is:

Blower fan is exerted oneself restriction:

P _min≤P _i≤P _maxi∈W

Direct current net node voltage, electric current, Power Limitation:

U _imin＜U _i＜U _imaxi∈Φ

I _imin＜I _i＜I _imaxi∈Φ

P _imin＜P _i＜P _imaxi∈Φ

Above Optimized model is mainly reflected in: can improve wind electricity digestion capability, improves direct current transportation net quality of voltage, realizes network security economical operation.

4, use based on source and lotus probabilistic VSC-MTDC robust optimisation technique:

This robust optimisation technique utilizes the thought of man and nature game, the final goal of game is, for the arbitrary uncertain factor interference in a certain limited set, design optimal strategy reaches minimum for target with the cost allowance making system and may suffer or operation risk, farthest to suppress the uncertain adverse effect to system, mathematically can be summarized as min-max-min and optimize thought.During min-max-min robust optimization thought is specifically applied to and studies based on source and lotus probabilistic VSC-MTDC robust Optimal Power Flow by the present invention.

Utilize robust Optimization Solution flexible DC power transmission net optimization object function, then Robust Optimization Model can be expressed as:

$\min F=\underset{t\&Element;T}{\min }\left(\underset{L\&Element;D}{\underset{P\&Element;W}{\max }}\underset{y}{\min }y(P,L)\right)$

When Robust Optimization Model solves, in all indefinite sets of variable, first try to achieve the conservative maximum solution meeting optimization object function one by one in conjunction with differential evolution intelligent optimization algorithm, object is the limiting case under making variable meet all uncertain factors; Then the minimum conservative solution under system all limit ruuning situation is asked under long time scale.

Compared with prior art advantage of the present invention is:

Compared with prior art, more gear to actual circumstances based on source and lotus probabilistic VSC-MTDC robust Optimal Power Flow technology, it considers randomness because of wind-force and intermittent and blower fan that is that cause is exerted oneself uncertainty, and flexible load itself is uncertain and stabilize wind electricity volatility function, game idea is utilized to drop to minimum by uncertain factor to the harm that system cloud gray model brings, the Optimization Solution of combined with intelligent optimized algorithm function to achieve the objective, for providing rational trend runtime value in VSC-MTDC system.

Accompanying drawing explanation

Fig. 1 is based on source and lotus probabilistic VSC-MTDC robust Optimal Power Flow simulation calculation flow process figure.

Embodiment

1, uncertain blower fan mathematical model is set up:

Lot of experimental data is indicated, an approximate Follow Weibull Distribution of regional wind speed change, therefore, by Weibull distribution random numbers generator v in MATLAB software _i=c (-lnx _i) ^1/kforecasting wind speed sample can be produced.Wherein, k is shape coefficient, and general span is 1.8-2.3; C is scale coefficient, the mean wind speed in the described area of reflection; x _ifor in [0,1] interval equally distributed random number.

Forecasting wind speed model basis increases uncertain factor, and under making any instant, wind speed all changes between a uncertain neighborhood based on this point prediction wind speed, then uncertain wind speed mathematic model can be represented by the formula:

$V=\{v\&Element;R^{N_T},\underset{i\&Element;N_T}{\mathrm{\&Sigma;}}\frac{|v_i-v_i^{\mathrm{fix}}|}{\widehat{v_i}}\&le;\mathrm{\&Delta;},v_i\&Element;[v_i-\widehat{v_i},v_i+\widehat{v_i}],\&ForAll;i\&Element;N_T\}$

Wherein, V represents the indeterminacy section collection of wind speed v; N _trepresent that value is counted in time T; for t _imoment wind speed sample value is a determined value; The maximum uncertainty degree of wind speed passes through limited; Δ determines the restriction requirement of uncertain total departure.

Above uncertain wind speed mathematic model can be reduced to following formula:

$V=\{v\&Element;R^n:v_n^{\mathrm{fix}}{D}_n^{-}\&le;v_n\&le;v_n^{\mathrm{fix}}{D}_n^{+},\&ForAll;n\}$

From above formula, the scale of wind speed indeterminacy section collection is by parameter determine, when time, namely uncertainty is 0; Generally, value between interval [0.8,1.2].

Funtcional relationship between the active-power P (v) that blower fan exports and wind speed v can approximate description be:

$P\left(v\right)=\left\{\begin{array}{cc}0& v<v_{\mathrm{ci}}\mathrm{orv}>v_{\mathrm{co}}\\ k_{1}v+k_2& v_{\mathrm{ci}}\&le;v<v_{\mathrm{rate}}\\ P_{\mathrm{rate}}& v_{\mathrm{rate}}\&le;v<v_{\mathrm{co}}\end{array}\right.$

In formula: k ₁=P _rate/ (v _rate-v _ci), k ₂=-k ₁v _ci, P _ratefor the rated power of aerogenerator; v _ci, v _rate, v _cobe respectively incision wind speed, wind rating and cut-out wind speed.And if only if v _ci≤ v < v _ratetime, blower fan output power is relevant with wind speed, and when namely wind speed is only in this is interval, uncertain just the exerting oneself on blower fan of wind speed produces uncertain impact.

Uncertain wind speed mathematic model is substituted into above formula determinacy blower fan model and can obtain uncertain blower fan mathematical model, be shown below:

$W=\{P\&Element;R^n:P_n\left(v_n^{\mathrm{fix}}{D}_n^{-}\right)\&le;P_n\&le;P_n\left(v_n^{\mathrm{fix}}{D}_n^{+}\right),\&ForAll;n\}$

2, uncertain flexible load mathematical model is set up:

Load prediction data basis increases uncertain factor, and under making any instant, load all changes between a uncertain neighborhood based on this point prediction load, then uncertain load mathematical model can be represented by the formula:

$D=\{L\&Element;R^n:L_n^{\mathrm{fix}}{D}_n^{-}\&le;L_n\&le;L_n^{\mathrm{fix}}{D}_n^{+},\&ForAll;n\}$

From above formula, the scale of load indeterminacy section collection is by parameter determine, when time, namely uncertainty is 0; Generally, value between interval [0.93,1.07].

Because flexible load active response can realize peak load shifting effect under tou power price mechanism, namely uncertain flexible load initiatively increases load in low power consumption district, and recruitment is η (%), then uncertain load model interval limit corrigendum in peak of power consumption, district initiatively reduces load, and decrease is μ (%), then the interval upper limit of uncertain load model is corrected as this is uncertain flexible load mathematical model, and wherein η, μ are relevant with time, local electricity price etc.

3, optimization object function is set up:

Optimization aim requires whole flexible DC power transmission net loss minimization, and the restriction of all DC bus-bar voltage is in the reasonable scope, and therefore flexible DC power transmission net optimization object function is:

$\min F=\underset{t=1}{\overset{T}{\mathrm{\&Sigma;}}}{P}_{\mathrm{loss}}+\underset{t=1}{\overset{T}{\mathrm{\&Sigma;}}}\underset{j=1}{\overset{N_d}{\mathrm{\&Sigma;}}}(U_j-U_N)$

Wherein, P _lossrepresent system losses, comprise current conversion station loss and DC network loss.If current conversion station loss becomes κ % proportional relation with its transmission capacity, then n _drepresent DC bus node number, U _jrepresent DC bus-bar voltage, U _nrepresent DC bus rated voltage.

Flexible DC power transmission net optimization object function constraint condition can be divided into equality constraint and inequality constrain, and for the multiterminal element network of complexity, its equality constraint is:

$\begin{array}{cc}{s}_{\mathrm{pi}}{I}_{\mathrm{di}}-\underset{j\&Element;{\mathrm{\&Phi;}}_d}{\mathrm{\&Sigma;}}{g}_{\mathrm{dij}}{V}_{\mathrm{dj}}=0& i\&Element;{\mathrm{\&Phi;}}_d\end{array}$

Wherein, Φ _drepresent the connected node set of alternating current-direct current electrical network; When voltage source converter is rectifier, s _p=1; When voltage source converter is inverter, s _p=-1; g _dijit is the nodal-admittance matrix of DC network; I _diit is DC current vector; V _djit is DC voltage vector.

Inequality constrain condition is:

Blower fan is exerted oneself restriction:

P _min≤P _i≤P _maxi∈W

Direct current net node voltage, electric current, Power Limitation:

U _imin＜U _i＜U _imaxi∈Φ

I _imin＜I _i＜I _imaxi∈Φ

P _imin＜P _i＜P _imaxi∈Φ

4, solve based on source and lotus probabilistic VSC-MTDC robust Optimal Power Flow:

Utilize the robust Optimization Solution flexible DC power transmission net optimization object function of all indefinite sets considering variable, then Robust Optimization Model can be expressed as:

$\min F=\underset{t\&Element;T}{\min }\left(\underset{L\&Element;D}{\underset{P\&Element;W}{\max }}\underset{y}{\min }y(P,L)\right)$

Above-mentioned Robust Optimization Model solution procedure is:

(1) in robust optimization interval, utilize intelligent optimization algorithm to carry out automatic search to variable, find out the optimum solution of flexible DC power transmission net optimization object function minF.

(2) under indefinite set, choose the conservative maximum solution meeting optimization object function in step (1).

(3) under long time scale, try to achieve the minimum conservative solution of system in step (2) under all limit ruuning situation.

Claims (6)

1. based on source and lotus probabilistic VSC-MTDC robust Optimal Power Flow method, it is characterized in that: first set up uncertain blower fan exert oneself model with can realize the uncertain flexible load model of peak load shifting, for improving wind electricity digestion capability, blower fan exerts oneself a part for meeting load need for electricity, remainder is then sent by Multi-end flexible direct current transmission Internet Transmission, by setting up flexible DC power transmission net voltage in violation of rules and regulations expense and the minimum optimization object function of power transmission loss expense, realize system with robust optimisation technique combined with intelligent optimized algorithm to generate electricity at blower fan, solving of optimization object function on all uncertain basis of load electricity consumption, for VSC in VSC-MTDC system provides rational voltage and power control reference value, guarantee system safety under condition of uncertainty, stable, economical operation.

2. to exert oneself model according to uncertain blower fan according to claim 1, it is characterized in that, uncertain Wind speed model is set up according to wind speed predicted data under long time scale, under realizing any instant, wind speed all changes in a certain scope, uncertain wind speed is substituted into blower fan and exerts oneself and can to obtain uncertain blower fan with wind speed relational expression and to exert oneself model.

3. according to uncertain flexible load model according to claim 1, it is characterized in that, uncertain load model is set up according to electro-load forecast data under long time scale, the load power consumption realized under any instant all changes in a certain scope, on this basis and then set up uncertain flexible load model, flexible load power levelling function is realized.

4. solve the VSC-MTDC optimization object function containing uncertain factor according to robust optimisation technique according to claim 1, it is characterized in that, first need to set up the indefinite set determining variable, then rational decision making uncertain variables, namely in uncertain parameter variation range, an optimum solution is sought, not only make constraint condition likely all be met under value in the institute of uncertain parameter, and make system improve economy under safe operation prerequisite.

5. solve the VSC-MTDC optimization object function containing uncertain factor according to robust optimisation technique according to claim 4, its concrete methods of realizing is: all conservative maximal value of first trying to achieve objective optimization function in all indefinite sets, and object is the limiting case under making optimum results meet all uncertainties; Then under long time scale, get the minimum conservative result of all conservative maximal values, can be mathematically the min-max-min optimization problem of a class belt restraining by this type of objective function min problem arises of asking with uncertain factor.

6. ask VSC-MTDC steady-state operation trend according to robust optimisation technique according to claim 5, it is characterized in that, be particularly suitable for solving the Optimization Problems In Power Systems with uncertain factor.