CN105958512B - Multiple domain time-lag power system LOAD FREQUENCY control method containing high wind energy permeability - Google Patents

Abstract

The present invention relates to a multi-domain time-delay power system load frequency control method with high wind energy penetration rate, which is characterized in that it includes the following steps: S1, constructing a time-delay power system including multiple regions, and establishing the power generation system of each region Mathematical model; S2, according to the mathematical model of the generator, establish a state model containing uncertain items for each region; S3, design the integral sliding mode surface σ _i (t) according to the state model containing assembled uncertain items; S4, Design a sliding mode load frequency controller according to the integral sliding mode surface σ _i (t); S5, use the controller u _i (t) obtained in step S4 as a control instruction to optimize the load frequency deviation of the power system. Compared with the prior art, the wind power generator of the present invention participates in the system frequency adjustment, so that the wind power generation and the traditional thermal power generation can be closely matched, and the increment of the output power of each generator is reduced on average, ensuring the balance of power supply and demand in each region, effectively reducing the Frequency deviation for each region.

Description

Load frequency control method for multi-domain time-delay power system with high wind energy penetration

technical field

The invention relates to a load frequency control method of a power system, in particular to a load frequency control method of a multi-domain time-delay power system with high wind energy penetration rate.

Background technique

Frequency is one of the important indicators reflecting the safe and stable operation of the power system. In the normal operation of the power system, the frequency control is mainly completed by adjusting the active output of the generator. When a large disturbance occurs in the power system, that is, the power generation is seriously unbalanced, the frequency recovery of the power system needs to rely on load frequency control to keep the frequency within the range allowed by the power industry. At present, clean and renewable wind energy has attracted widespread attention, but the volatility of wind energy has led to unstable output power of wind turbines.

When large-scale wind turbines are connected to the traditional power generation system, the frequency deviation of the system is exacerbated. With the increase of wind energy penetration, it is expected that large wind farms can participate in frequency control. Generally, there are two ways to reduce wind energy volatility, one is coordinated control of energy storage batteries, and the other is to improve the control level of wind turbines.

At the same time, with the introduction of open communication network structure, fixed and random communication delays inevitably exist in traditional power system load frequency control. The introduction of time-delay will reduce the control effect of the control system and even cause the instability of the whole closed-loop system. Therefore, the time-delay effect becomes a key issue in the design of time-delay power system load frequency controller. Document "Yu, Xiaofeng, and K.Tomsovic. "Application of linear matrix inequalities for load frequency control with communication delays", IEEE Trans. Power Syst., vol.19, no.3, pp.1508-1515, 2004." Design of a robust load frequency controller based on linear matrix inequalities for uncertain delays in power system communication networks. Literature "Ama, Takashi Hiy, D. Zuo, and T. Funabashi."Multi-agent based automatic generation control of isolated standalone power system", Power System Technology, 2002. Proceedings. PowerCon2002. International Conference on IEEE, vol.1, pp .139-143, 2002." Aiming at the communication delay compensation method in the system, load frequency control is proposed. These literatures explicitly analyze the effect of signal delay on load frequency control.

Many literatures have extensively studied the application of different controllers in load frequency control. Traditional PID controllers have been widely used in system load frequency control. With the development of the power industry, the structure of the power system is becoming more and more complex, and the system is also affected by various load disturbances and fluctuating new energy sources, resulting in a large number of uncertain structures and parameters in the system. In order to solve the shortcomings of traditional frequency control, some advanced control theories are adopted, such as fuzzy control, neural network, predictive control and adaptive control, etc. These methods solve the influence of system uncertainty to a certain extent, but the algorithm is more complicated in practical application. The energy storage system can quickly provide active power compensation, so the literature "Kalyani, Sheetal, S. Nagalakshmi, and R. Marisha."Load frequency control using battery energy storage system in interconnected power system."Computing Communication & Networking Technologies (ICCCNT), 2012Third International Conference on.IEEE,2012" and "Aditya,S.K.,and D.Das."Application of battery energy storage system to load frequency control of an isolated power system."International journal of energy research,vol.23,no.3,pp.247 -258,1999" to improve the performance of system load frequency control. The document "Aditya, S.K., and D.Das."Application of battery energy storage system to load frequency control of an isolated power system."International journal of energy research, vol.23, no.3, pp.247-258,1999" will An incremental model for load frequency control is applied to an isolated power system with reheat thermal units and energy storage systems, improving system performance. With the increase of load disturbance, the storage capacity requirements of the energy storage system are also higher. Document "Jiang, L., et al. "Delay-dependent stability for load frequency control with constant and time-varying delays", IEEE Trans. Power Syst., vol.27, no.2, pp.932-941, 2012" For the load frequency control schemes of PID controllers in single-domain and multi-domain time-delay power systems, the relationship between PID controller delay profit and revenue is discussed. Although adjusting the gain of the PID controller can weaken the influence of time lag on the power system and keep the rated frequency within the range of deviation, the frequency deviation of each region always exists in each subsystem. Literature "Zhou Hui, Ya Fu, and RongCong."Fuzzy-based load frequency controller for interconnected power system with wind power integration", Electrical and Computer Engineering (CCECE), 2014IEEE 27th Canadian Conference on.IEEE,2014,pp.1-6 .” A fuzzy load frequency controller is designed for a two-area interconnected power system with wind power, but the model of the interconnected system has no communication delay.

As a typical nonlinear control, sliding mode control has the advantages of fast response and invariance to system parameter uncertainties and external disturbances. And the algorithm is simple and easy to realize in engineering, so it is widely used in the design of load frequency control of power system. literature" Tamara, and "Optimal sliding mode controller for power system's load-frequency control",UniversitiesPower Engineering Conference,2008.UPEC 2008.43rdInternational.IEEE,2008"The author used discrete-time sliding mode control in the load frequency control system to make the system stable and robust Therefore, it is necessary to obtain the optimal parameters of the discrete-time sliding mode model that determines the standard of the integral squared error. Literature "Al-Hamouz, ZM, and YLAbdel-Magid."Variable structure loadfrequency controllers for multiarea power systems", International Journal of Electrical Power&Energy Systems, vol.15, no.5, pp.293-300, 1993 "proposed variable structure controller is not sensitive to some parameter changes, and also considers the influence of power generation speed constraint and dead zone, but in large load disturbance The lower system may be unstable. Literature "Mi Yang, et al."Decentralized sliding mode load frequency control for multi-area power systems",IEEE Trans.Power Syst.,vol.28,no.4,pp.4301-4309,2013 "Aiming at the interconnected power system, a distributed sliding mode load frequency control is designed, which has dynamic stability under the constraints of a wide range of parameter changes and generation rates. It responds quickly to frequency and is insensitive to parameter changes and load disturbances, but The load frequency control design is not considered and there is no communication time delay problem in the new energy power system. Literature "Yang Mi, Yang Yang, Han Zhang, et al."Sliding mode based load frequency control for multi-area interconnected power system containing renewableenergy", Transportation Electrification Asia-Pacific (ITEC Asia-Pacific), 2014IEEE Conference and Expo.IEEE, 2014 "Decentralized sliding mode load frequency control solves the problem of wind The problem of load frequency control in electric multi-domain power system, but the problem of system communication delay is not raised.

Based on the above analysis, a novel sliding mode load frequency controller is designed and applied to the multi-domain time-delay hybrid power system.

Contents of the invention

The purpose of the present invention is to overcome the defects of the above-mentioned prior art and provide a multi-domain time-delay power system load frequency control method with high wind energy penetration rate that considers communication delay and effectively reduces the frequency deviation of each area of the redundant power system.

The purpose of the present invention can be achieved through the following technical solutions:

A multi-domain time-delay power system load frequency control method with high wind energy penetration is characterized in that it includes the following steps:

S1. Construct a time-delay power system including multiple regions, and establish a mathematical model of the power generation system in each region. Each region is connected by a tie line. Each region includes a thermal power generation system and a wind power generation system. The generator of the wind power generation system is wind power. turbine, so that the frequency deviation of the wind turbine participates in the system frequency adjustment as a coupling item in the system frequency deviation adjustment item;

S2. According to the mathematical model of the generator, a state model containing uncertain items is established for each area:

At the same time define the aggregation uncertainty term g _i (t):

Express the state model with aggregate uncertain terms as:

where the state variable is x _i (t):

x _i (t)＝[Δf _i (t) ΔP _mi (t) ΔP _vi (t) ΔE _i (t) Δδ _i (t) Δf _Ti (t) Δx _i1 (t) Δx _i2 (t) Δx _i3 ( t) Δx _i4 (t)] ^T In the formula, A′ _i is the system matrix, A′ _idi is the delay item coefficient matrix, B′ _i is the input matrix, E′ _ij is the interconnection item coefficient matrix, F′ _i is the disturbance Item coefficient matrix, ΔA _i , ΔA _idi , ΔE _ij , ΔB _i , ΔF _i are uncertain items of power system parameters corresponding to A′ _i , A′ _idi , E′ _ij , B′ _i , F′ _i respectively, The control variable u _i (t) is the sliding mode load frequency controller, Δf _i (t) is the system frequency deviation, ΔP _mi (t) is the generator output power increment, ΔP _vi (t) is the control valve position increment, ΔE _i (t) is frequency deviation integral controller increment, Δδ _i (t) is phase angle increment, Δf _Ti (t) is wind turbine frequency deviation, Δx _i1 (t), Δx _i2 (t), Δx _i3 (t), Δx _i4 (t) represent the state quantities in the i-th regional fan model;

S3, design the integral sliding mode surface σ _i (t) according to the state model containing aggregated uncertain items;

S4. Design the sliding mode load frequency controller u _i (t) according to the integral sliding mode surface σ _i (t):

u _i (t)＝-K _i x _i (t)-(G _i B _i ′) ^-1 ||G _i ||h _i -(G _i B _i ′) ^-1 (W _i +ε _i )sgn (σ _i (t)),

Among them, the aggregation uncertain term g _i (t) is bounded, and satisfies ||g _i (t)||≤h _i , where h _i is a bounded constant, h _i >0, ||*|| The Gyreed norm, the matrices G _i and K _i are the coefficient matrices of the integral sliding mode surface σ _i (t), sgn(*) is a symbolic function,

S5, optimize the load frequency deviation of the power system according to the controller u _i (t) obtained in step S4 as a control instruction.

In the step S1, among the adjustment items of the system frequency deviation Δf _i (t), the coupling item related to the wind turbine is where K _pi is the system gain, K _IGi is the fluid volume coupling coefficient, T _pi is the system time constant, and Δf _Ti (t) is the wind turbine frequency deviation.

The generator of the thermal power generation system is a non-reheating steam turbine or a reheating steam turbine.

In the step S1, the mathematical model of the thermal power generation system using a non-reheating steam turbine is:

The mathematical model of thermal power generation system using reheating steam turbine is:

In the formula, subscript i and subscript j indicate the number of the area, i=1,...,N, j=1,...,N, N is the number of areas, Δf _i (t) is the system Frequency deviation, Δf _Ti (t) is wind turbine frequency deviation, ΔP _mi (t) is generator output power increment, ΔP _vi (t) is regulator valve position increment, ΔE _i (t) is frequency deviation integral controller Increment, Δδ _i (t) is the phase angle increment, ΔP _di (t) is the system load disturbance, ΔP _GWi (t) is the wind turbine output power deviation in the i-th area, T _ij is the i-th area and the T _chi is the time constant of the steam turbine, T _rh is the time constant of the reheater, F _hp is the ratio of the power generated by the reheating steam turbine in this area to the total output power of all generators in this area ratio, T _pi is the system time constant, K _pi is the system gain, K _IGi is the coupling coefficient of liquid quantity, ΔP _Li (t) is the active power deviation of the area, ΔP _ri (t) is the state quantity output by the steam turbine governor, R _i is the speed adjustment of the governor, B _i is the area frequency offset coefficient, K _Ei is the integral control gain, d _i is the time lag constant, T _gi is the time constant of the governor;

The mathematical model of the wind power generation system is:

in

In the formula, Δx _i1 (t), Δx _i2 (t), Δx _i3 (t), Δx _i4 (t) represent the state variables in the i-th area fan model, T _wi is the time constant of the fan, K _PC1 is the propeller Pitch control feedback gain, K _P31 and T _P31 are pitch response data fitting control response coefficient and time constant respectively, K _P21 and T _P21 are hydraulic pitch actuator response coefficient and time constant respectively, K _PPi , K ₁₁ and K _P1i is the pitch control response coefficient, T _P1i is the pitch control response time constant, Δf _Ti (t) is the wind turbine frequency deviation, ΔP _mi (t) is the generator output power increment, K _IGi is the fluid coupling coefficient, K _TPi is the fan frequency feedback coefficient.

The step S3 is specifically: select the matrix G _i , make G _i B _i ' a non-singular matrix, and σ _i (t) satisfy the equation The matrix K _i satisfies λ(A _i ′-B _i ′K _i )<0, and λ(*) means to solve the eigenvalue.

Compared with the prior art, the present invention has the following advantages:

(1) Let the frequency deviation of the wind turbine participate in the system frequency adjustment as a coupling item in the system frequency deviation adjustment item, so that the wind power generation and the traditional thermal power generation can be closely matched, and the increment of the output power of each generator is reduced on average, ensuring that all regions Power supply and demand balance, effectively reducing the frequency deviation of each region.

(2) The frequency deviation increment Δf _i (t) (Hz) and the incremental change of generator output power ΔP _mi (t)( puMW), incremental change in governor valve position ΔP _vi (t)(puMW), zone control deviation integral control incremental change ΔE _i (t), angular frequency deviation Δδ _i (t), wind turbine speed deviation Δf _Ti (t) (Hz) and Δx _i1 (t), Δx _i2 (t), Δx _i3 (t), Δx _i4 (t) 10 power system states in the wind turbine model are optimized to realize tie-line Quick balancing of exchange power values and exchange power plan values.

(3) Establishing mathematical models for power generation systems using different types of (reheat and non-reheat) thermal generators can fully demonstrate that the frequency control method is applicable to areas with many different types of generators and has wide practicability.

(4) Compared with the traditional PID load frequency control and no energy storage system, the sliding mode control strategy reduces the frequency offset and tie line power fluctuations, fully ensuring the stability of the power system and faster response speed. In a certain range, when the load disturbance increases, the sliding mode load frequency control has better control performance than the PID control and energy storage system.

Description of drawings

Fig. 1 is the structural representation of the power system of the present embodiment;

Fig. 2 is the transfer function model of the power system area 1 of the present embodiment;

Fig. 3 is the transfer function model of the power system area 2 of the present embodiment;

Fig. 4 is a structural diagram of a fan and a generator in a certain area of the power system of the present embodiment;

Fig. 5 is the system frequency deviation response in area 1 when the power system d ₁ = 1.2s in this embodiment;

Fig. 6 is the power deviation response of the system tie line in area 1 when the power system d ₁ =1.2s in this embodiment;

Fig. 7 shows the system frequency deviation response in area 2 when the power system d ₁ = 1.2s in this embodiment;

Fig. 8 is the power deviation response of the system tie line in area 2 when the power system d ₁ = 1.2s in this embodiment;

Fig. 9 shows the system frequency deviation response in area 3 when the power system d ₁ = 1.2s in this embodiment;

Fig. 10 shows the power deviation response of the system tie line in area 3 when the power system d ₁ = 1.2s in this embodiment;

Fig. 11 shows the system frequency deviation response in area 1 when the power system d ₁ =3.0s in this embodiment;

Fig. 12 is the power deviation response of the system tie line in area 1 when the power system d ₁ =3.0s in this embodiment;

Fig. 13 shows the system frequency deviation response in area 2 when the power system d ₁ =3.0s in this embodiment;

Fig. 14 is the power deviation response of the system tie line in area 2 when the power system d ₁ =3.0s in this embodiment;

Fig. 15 shows the system frequency deviation response in area 3 when the power system d ₁ =3.0s in this embodiment;

Fig. 16 is the power deviation response of the system tie line in area 3 when the power system d ₁ =3.0s in this embodiment;

Fig. 17 is the random load disturbance response of the power system in this embodiment;

Fig. 18 shows the system frequency deviation response in area 1 when the power system d ₁ =1.494sin(t)+0.1 in this embodiment;

Fig. 19 shows the system frequency deviation response in area 2 when d ₂ =8sin(t)+0.3 of the power system in this embodiment;

Fig. 20 is the system frequency deviation response in area 3 when d ₂ =8sin(t)+0.3 of the power system in this embodiment;

Fig. 21 shows the system frequency deviation response in area 1 when the power system d ₁ = 1.2s in this embodiment;

Fig. 22 is the power deviation response of the system tie line in area 1 when the power system d ₁ = 1.2s in this embodiment;

Fig. 23 is the system frequency deviation response in area 2 when the power system d ₁ = 1.2s in this embodiment;

Fig. 24 is the power deviation response of the system tie line in area 2 when the power system d ₁ = 1.2s in this embodiment;

Fig. 25 shows the system frequency deviation response in area 3 when the power system d ₁ = 1.2s in this embodiment;

Fig. 26 is the power deviation response of the system tie line in area 3 when the power system d ₁ = 1.2s in this embodiment;

Fig. 27(a) is the system frequency deviation response in area 1 when the power system d ₁ =3.0s in this embodiment;

Figure 27(b) is an enlarged view of the 0-10s part in Figure 27(a);

Fig. 28 is the power deviation response of the system tie line in area 1 when the power system d ₁ =3.0s in this embodiment;

Fig. 29(a) is the system frequency deviation response in area 2 when the power system d ₁ =3.0s in this embodiment;

Figure 29(b) is an enlarged view of the 0-10s part in Figure 29(a);

Fig. 30 is the power deviation response of the system tie line in area 2 when the power system d ₁ =3.0s in this embodiment;

Fig. 31(a) is the system frequency deviation response in area 3 when the power system d ₁ =3.0s in this embodiment;

Figure 31(b) is an enlarged view of the 0-10s part in Figure 31(a);

Fig. 32 shows the power deviation response of the system tie line in area 3 when the power system d ₁ =3.0s in this embodiment.

Detailed ways

The present invention will be described in detail below in conjunction with the accompanying drawings and specific embodiments. This embodiment is carried out on the premise of the technical solution of the present invention, and detailed implementation and specific operation process are given, but the protection scope of the present invention is not limited to the following embodiments.

Example

For the multi-domain interconnected power system with high wind energy penetration rate as shown in Figure 1, in order to reduce the system frequency deviation caused by wind energy fluctuations, the distributed sliding mode controller is used to optimize the load frequency control to reduce the frequency deviation. The load frequency control method of the multi-domain time-delay power system with high wind energy penetration rate of the present invention comprises the following steps:

S1. Construct a time-delay power system including multiple regions, and establish a mathematical model of the power generation system in each region. Each region is connected by a tie line. Each region includes a thermal power generation system and a wind power generation system. Reheating steam turbine or reheating steam turbine, the generator of the wind power generation system is a wind turbine, and the mathematical model of the thermal power generation system using a non-reheating steam turbine is:

The mathematical model of thermal power generation system using reheating steam turbine is:

In the formula, subscript i and subscript j indicate the number of the area, i=1,...,N, j=1,...,N, N is the number of areas, Δf _i (t) is the system Frequency deviation (Hz), Δf _Ti (t) is wind turbine frequency deviation, ΔP _mi (t) is increment of generator output power (puMW), ΔP _vi (t) is increment of regulating valve position (puMW), ΔE _i (t) is frequency deviation integral controller increment, Δδ _i (t) is phase angle increment, ΔP _di (t) is system load disturbance (puMW), ΔP _GWi (t) is wind force in area i Turbine output power deviation (puMW), T _ij is the tie-line power synchronization factor between the i-th area and the j-th area, T _chi is the steam turbine time constant (s), T _rh is the reheater time constant (s) , F _hp is the proportion of reheat turbine power generation in the region to the total output power of all generators in the region, T _pi is the system time constant (s), K _pi is the system gain, K _IG is the fluid volume coupling coefficient, ΔP _Li (t) is the active power deviation of the i-th area, ΔP _ri (t) is the state quantity output by the steam turbine governor, R _i is the speed adjustment of the governor (Hz/ _puMW ), and Bi is the area frequency offset coefficient , K _Ei is the integral control gain, d _i is the time lag constant, T _gi is the governor time constant;

The mathematical model of the wind power generation system is:

in

In the formula, Δx _i1 (t), Δx _i2 (t), Δx _i3 (t), Δx _i4 (t) represent the state variables in the i-th area fan model, T _wi is the time constant of the fan, K _PC1 is the propeller Pitch control feedback gain, K _P31 and T _p31 are pitch response data fitting control response coefficient and time constant respectively, K _P21 and T _p21 are hydraulic pitch actuator response coefficient and time constant respectively, K _ppi , K ₁₁ and K _p1i is the pitch control response coefficient, T _p1i is the pitch control response time constant, Δf _Ti (t) is the wind turbine speed deviation, ΔP _mi (t) is the generator output power increment, K _IGi is the wind turbine fluid volume coupling coefficient , KT _p i is the fan frequency feedback coefficient.

S2. According to the mathematical model of the generator, a state model containing uncertain items is established for each area:

At the same time define the aggregation uncertainty term g _i (t):

Express the state model with aggregate uncertain terms as:

where the state variable is x _i (t):

x _i (t)＝[Δf _i (t) ΔP _mi (t) ΔP _vi (t) ΔE _i (t) Δδ _i (t) Δf _Ti (t) Δx _i1 (t) Δx _i2 (t) Δx _i3 ( t) Δx _i4 (t)] ^T

The control variable u _i (t) is the sliding mode load frequency controller, A′ _i is the system matrix, A′ _idi is the delay item coefficient matrix, B′ _i is the input matrix, E′ _ij is the interconnection item coefficient matrix, F′ _i is the coefficient matrix of the disturbance item, ΔA _i , ΔA _idi , ΔE _ij , ΔB _i , ΔF _i are the differences of power system parameters corresponding to A′ _i , A′ _idi , E′ _ij , B′ _i , F′ _i respectively Determined item;

S3, design the integral sliding mode surface σ _i (t) according to the state model containing aggregated uncertain items;

S4. Design the sliding mode load frequency controller u _i (t) according to the integral sliding mode surface σ _i (t):

u _i (t)＝-K _i x _i (t)-(G _i B _i ′) ^-1 ||G _i ||h _i -(G _i B _i ′) ^-1 (W _i +ε _i )sgn (σ _i (t)),

Among them, the aggregation uncertain term g _i (t) is bounded, and satisfies ||g _i (t)||≤h _i , where h _i is a bounded constant, h _i >0, ||*|| The Gyreed norm, the matrices G _i and K _i are the coefficient matrices of the integral sliding mode surface σ _i (t), sgn(*) is a symbolic function,

S5, optimize the load frequency deviation of the power system according to the controller u _i (t) obtained in step S4 as a control instruction.

The multi-domain interconnection power system of the present invention is embodied in that area 1, area 2, and area 3 in FIG. 1 are connected to each other through tie lines. Each regional power system includes thermal power generation system and wind power generation system. For thermal power generation systems, according to the type of generator, it can be divided into reheat thermal power generators and non-reheat thermal power generators. Figure 2 is the transfer function model of Zone 1, which uses a non-reheat thermal power generator. Figure 3 is the transfer function model of zone 2, using a reheat thermal generator. In addition to the different types of thermal power generators, these two transfer functions include auxiliary control links, communication delay links, and primary speed regulation links of thermal power generators, and a large-scale wind power generation system is connected. In the figure, G represents an equivalent generator, and WTG represents a wind power generator, that is, a wind turbine.

In the open-loop transfer function models shown in Figures 2 and 3, the first module is auxiliary control, also known as secondary control, through which the adaptation between power generation and load can be realized and the frequency deviation can be restored to zero. Proportional Integral Control, the role of integral is to ensure that the static frequency deviation is zero, and the role of proportion is to improve stability and increase response speed.

The second module is the communication delay link. With the increasing degree of interconnection among regional power systems and the development and application of power system information processing and network communication technology, the introduction of open communication network structure makes the load frequency control inevitably exist. Fixed and random communication delays. The introduction of time-delay will reduce the control effect of the control system and even cause the instability of the whole closed-loop system. Therefore, the time-delay effect becomes a key issue in the design of time-delay power system load frequency controller. This module is through the exponential function to achieve a certain delay.

The third module is the primary speed regulation link, which adopts the proportional control with static differential adjustment, and passes through the first-order inertial link accomplish.

The fourth module is the steam turbine. For non-reheat turbines, the non-reheat turbine exhibits a small time constant when the throttle valve position is changed due to the influence of the charging time of the vapor chamber and the duct leading to the HP cylinder. Since the time lag caused by the vapor chamber is relatively simple, a first-order inertial link is used express. For reheat turbines, the reheater and the transient steam flow downstream into the cylinder must be taken into account. The gas flow in the LP turbine section will vary with the pressure build-up process in the reheater volume.

The fifth module is the power system module, which adopts the first-order inertia link Synchronous power coefficients are introduced between interconnected regions.

In the present invention, the wind power generator participates in the adjustment of the system frequency. In the wind turbine transfer function model shown in Figure 4, the wind turbine module and the pitch angle controller are included. The frequency deviation Δf _i (t) is an input control signal, so that the wind turbine participates in system frequency regulation. The input of the frequency deviation Δf _i (t) changes the state quantity Δf _Ti (t) and the system matrix A _i ′ in the state equation, so that the design of the controller is different from the prior art. Wind generators participate in system frequency regulation, increasing the number of generators participating in frequency regulation in the entire power system. In the case of small frequency fluctuations, when multiple generators participate in frequency regulation, the incremental output power of each generator is reduced on average, so that load frequency regulation is easier and system frequency fluctuations are controlled within a smaller range . In the case of large frequency fluctuations, due to the increase in the number of generators participating in frequency regulation, each generator increases the generator capacity according to a certain ratio to achieve the purpose of frequency regulation. This reduces the involvement of the secondary frequency regulation of the thermal generator and makes frequency regulation easier. In severe cases, it also reduces the probability of system load shedding, thereby improving the reliability of power system operation.

The invention is effective in controlling the load frequency of the power system, and the designed sliding mode controller can control the load frequency within the allowable or even smaller range stipulated by the state. It can lay a certain foundation for future research on load frequency control.

(1) Mathematical model of time-delay interconnected power system with high wind power penetration rate

The multi-domain interconnected power system is decentralized and controlled, and each regional power system mainly includes a thermal power generation system and a wind power generation system. In order to design a decentralized sliding mode controller for a multi-domain time-delay interconnected hybrid power system including wind power generation and thermal power generation, the state model of each region is established to satisfy:

With the continuous change of power system load, the operation mode of the system must be adjusted. In different operating modes, the parameters of the system are different. Therefore, considering the uncertainty of the power system parameters, the power system is expressed as a model of uncertain items:

Among them, A _i 'is the system matrix, A' _idi is the delay term coefficient matrix, B _i 'is the input matrix, E' _ij is the interconnection term coefficient matrix, F _i is the disturbance term coefficient matrix, ΔA _i , ΔB _i , ΔA _idi , ΔF _i , ΔE _ij are uncertain items of power system parameters.

(2) The design principle of the load frequency control of the time-delay interconnected hybrid power system containing wind power in the present invention

In order to facilitate the design of the sliding mode controller, the power system with aggregate uncertain terms is expressed as

Before designing the controller, four assumptions are given first,

Assumption 1: (A _i ′, B _i ′) is controllable. Assumption 2: rank(B _i ′, g _i )≠rank(B _i ′). Hypothesis 3: The aggregation uncertain term g _i (t) is bounded and satisfies the following conditions: ||g _i (t)||≤h _i , where h _i >0 is a constant, i=1,... ., N. Hypothesis 4: The delay term of the system satisfies the following conditions

||x _p (td _i )||≤x _pmax , where x _pmax = max||x _p ||,p=i,j;

Design integral sliding mode surface satisfying equation Wherein, the matrix K _i satisfies λ(A _i ′-B _i ′K _i )<0, and an appropriate matrix G _i is selected to make the matrix G _i B _i ′ a non-singular matrix. The purpose of the present invention is to design a sliding mode load frequency controller for the time-delay hybrid power system in each area.

u _i (t)＝-K _i x _i -(G _i B _i ′) ^-1 ||G _i ||h _i -(G _i B _i ′) ^-1 (R _i +ε _i )sgn(σ _i (t)) to stabilize unmatched uncertain power systems. The stability of the sliding mode and the design of the controller can be realized by the following Theorem 1 and Theorem 2.

Theorem 1: When x∈B ^c (η), the sliding mode of the system is stable at any moment, where B ^c (η) is a closed spherical surface B(η) with x=0 as the center and η as the radius repair.

Proof: Construct the Lyapunov function as where P is the Lyapunov equation The solution of , Q _i is a given positive definite symmetric matrix.

Pick

Deriving v(t) gives:

From Assumption 3, we can get ||A _id1 ||≤α _i ,||E _ij ||≤γ _i ,

but

When the state trajectory of the system enters the closed spherical surface B ^c (η), λ(Q _i )>0, then the Lyapunov function established to ensure that the system is stable on the sliding surface.

Theorem 2: If the variable structure controller satisfies the following equation

u _i (t)＝-K _i x _i -(G _i B _i ′) ^-1 ||G _i ||h _i -(G _i B _i ′) ^-1 (R _i +ε _i )sgn(σ _i (t)), then the system satisfies the arrival condition.

in: represents a symbolic function.

Proof: construct Lyapunov function

When the system enters the sliding mode, it satisfies σ _i (t)=0 and but

but

From Assumption 3, we can get obvious The system state trajectory can reach the sliding surface in a finite time.

(3) Case analysis

In order to verify the effectiveness of the sliding mode load frequency controller designed in the large-scale wind power integrated time-delay hybrid power system, the following three simulation examples are used for simulation research, and compared with the optimized traditional proportional integral load frequency control, energy storage The proportional-integral load control of the system and the proportional-integral load control strategy of the non-energy storage system are compared. The system parameter values of the three calculation examples are listed in Table 1. In addition, there is a delay time of each system.

Table 1 Parameter values of each area of the power system

1) Calculation example 1 - the influence of wind power generation on the control of different load frequencies

In this calculation example, the interconnected system operates under rated conditions and has no uncertain parameters, and the system load disturbance is ΔP _di =0.02pu. With the increase of wind energy penetration, it is necessary to consider the participation of wind power in system load frequency regulation. Therefore, the high wind energy penetration interconnected time-delay power system will be studied in the following load frequency control scheme, (a) PID load frequency control with wind turbine active power control loop and wind turbine not participating in frequency modulation; (b) wind turbine Active power control loop, fan participates in PID load frequency control of frequency modulation; (c) includes wind turbine active power control loop, sliding mode load frequency control.

a. When the time delay of the interconnection system is d _i =1.2s (i=1,2,3), under the control of three load frequencies, the responses of the three areas are shown in Figure 5-10. Among them, the effect of scheme (c) is the most obvious, and the frequency deviation is approximately zero within 10s. At the same time, under the control of scheme (c), the power deviation of each regional tie line is approximately zero in a very short response time and the overshoot is small, while the frequency deviation Δf _i of schemes (a) and (b) ( t) and tie line power deviation ΔP _{tie i} (t) cannot be approximately zero for a long time. Comparing schemes (a) and (b) in Figure 5-10, it is obvious that when the fan participates in frequency regulation, the frequency deviation response fluctuates very little. The increment of wind power generation is determined by the system frequency increment and the weak inertia of wind power generation in the interconnected system can be improved.

b. When the time delay of the interconnection system is d _i =3.0s (i=1,2,3), under the control of three load frequencies, the responses of the three areas are shown in Figure 11-16. It is obvious that under the control of scheme (c), the response of load frequency deviation Δf _i (t) and tie line power deviation ΔP _{tie i} (t) in the three regions is fast and approximately zero with little overshoot.

2) Calculation example 2—the influence of changing time delay and load disturbance

In this calculation example, each zone operates under rated conditions and the fan participates in system frequency regulation. There are two schemes: sliding mode load frequency control and traditional PID load frequency control. Simulate the response of the controller under different time delays d ₁ =1.494sin(t)+0.1,d ₂ =8sin(t)+0.3,d ₃ =5sin(t)+0.1, the simulation results under random load disturbance are shown in the figure 18-20, the random load disturbance waveform of each region is shown in Figure 17. Compared with the load frequency deviation Δf _i (t) under PID control, sliding mode control has a slower influence on time-varying communication delay, and has smaller overshoot, faster response and less oscillation.

3) Calculation example 3—Energy storage battery participates in load frequency control of high wind energy penetration rate system

Since the energy storage battery can quickly provide active power compensation, it can be used to improve the performance of load frequency control. In order to extensively verify the effectiveness of sliding mode load frequency control, each area adopts sliding mode load frequency control, PID control with energy storage system and PID control without energy storage system, the system operates under rated conditions, and the fan participates in the system frequency regulation. The responses of the three regions are shown in Figures 21-32.

a. When the time delay in the interconnection system is d ₁ =d ₂ =d ₃ =1.2s, the response of load frequency deviation Δf _i (t) and tie line power deviation ΔP _{tie i} (t) in each area is shown in Figure 21- 26. After 10s, the PID control without energy storage system, Δf _i (t) and ΔP _{tie i} (t) oscillate with equal amplitude, while the PID control with energy storage system has a smaller overshoot than the PID control without energy storage system . Comparing the three control schemes, the overshoot of the sliding mode load frequency control is the smallest, the response speed is the fastest, and Δf _i (t) and ΔP _{tie i} (t) are approximately zero within 15s.

b. When the time delay in the interconnected system is d ₁ =d ₂ =d ₃ =3.0s, the response of load frequency deviation Δf _i (t) and tie line power deviation ΔP _{tie i} (t) in each area is shown in Figure 27( a) -32. With the increase of communication delay, under PID control without energy storage system, Δf _i (t) and ΔP _{tie i} (t) tend to disperse and oscillate greatly, and the frequency deviation is greater than 0.2Hz, so the system adopts energy storage system PID control is not feasible. Although the energy storage system can improve the variation of Δf _i (t) and ΔP _{tie i} (t) in PID load frequency control, it cannot be completely eliminated. However, by using the sliding mode load frequency control, the frequency deviation and tie line power deviation are relatively reduced, and the overshoot is reduced in a very short transition time.

Claims (4)

1. A multi-domain time-delay power system load frequency control method containing high wind energy penetration rate is characterized in that, comprising the following steps: S1. Construct a time-delay power system including multiple regions, and establish a mathematical model of the power generation system in each region. Each region is connected by a tie line. Each region includes a thermal power generation system and a wind power generation system. The generator of the wind power generation system is wind power. Turbine, let the frequency deviation of the wind turbine participate in the system frequency adjustment as a coupling item in the system frequency deviation adjustment item, and in the adjustment item of the system frequency deviation Δf _i (t), the coupling item related to the wind turbine is where K _pi is the system gain, K _IGi is the fluid volume coupling coefficient, T _pi is the system time constant, Δf _Ti (t) is the wind turbine frequency deviation; S2. According to the mathematical model of the generator, a state model containing uncertain items is established for each area: At the same time define the aggregation uncertainty term g _i (t): Express the state model with aggregate uncertain terms as: where the state variable is x _i (t): x _i (t)＝[Δf _i (t) ΔP _mi (t) ΔP _vi (t) ΔE _i (t) Δδ _i (t) Δf _Ti (t) Δx _i1 (t) Δx _i2 (t)Δx _i3 ( t) Δx _i4 (t)] ^T In the formula, t is the time variable, subscript i and subscript j represent the number of the region, x _j (t) is the state variable of the jth region, i=1,....,N, j=1,. ..., N, N is the number of regions, d _i is the delay constant, A' _i is the system matrix, A' _idi is the delay item coefficient matrix, B' _i is the input matrix, E' _ij is the interconnection item coefficient matrix, F′ _i is the disturbance term coefficient matrix, ΔA _i , ΔA _idi , ΔE _ij , ΔB _i , ΔF _i are the electric power corresponding to A′ _i , A′ _idi , E′ _ij , B′ _i , F′ _i respectively The uncertain items of system parameters, the control variable u _i (t) is the sliding mode load frequency controller, Δf _i (t) is the system frequency deviation, ΔP _di (t) is the system load disturbance, ΔP _mi (t) is the generator Output power increment, ΔP _vi (t) is the position increment of the regulating valve, ΔE _i (t) is the frequency deviation integral controller increment, Δδ _i (t) is the phase angle increment, Δf _Ti (t) is the wind turbine Frequency deviation, Δx _i1 (t), Δx _i2 (t), Δx _i3 (t), Δx _i4 (t) represent the state quantities in the i-th regional fan model; S3, design the integral sliding mode surface σ _i (t) according to the state model containing aggregated uncertain items; S4. Design the sliding mode load frequency controller according to the integral sliding mode surface σ _i (t): u _i (t)＝-K _i x _i (t)-(G _i B _i ′) ^-1 ‖G _i ‖h _i -(G _i B _i ′) ^-1 (W _i +ε _i )sgn(σ _i (t)), Among them, the aggregation uncertain term g _i (t) is bounded, and satisfies ||g _i (t)||≤h _i , where h _i is a bounded constant, h _i >0, ||*|| The Gyreed norm, the matrices G _i and K _i are the coefficient matrices of the integral sliding mode surface σ _i (t), x _pmax ＝max||x _p ||, p＝i, j, ε _i > 0, i＝1,...., N, sgn(*) is a sign function, S5, using the sliding mode load frequency controller u _i (t) obtained in step S4 as a control instruction to optimize the load frequency deviation of the power system. 2. The load frequency control method of a multi-domain time-delay electric power system containing high wind energy penetration according to claim 1, wherein the generator of the thermal power generation system is a non-reheating steam turbine or a reheating type steam turbine. 3. A kind of multi-domain time-delay power system load frequency control method containing high wind energy penetration rate according to claim 2 is characterized in that, in the described step S1, the thermal power generation system mathematics of non-reheating type steam turbine is adopted The model is: The mathematical model of thermal power generation system using reheating steam turbine is: In the formula, subscript i and subscript j indicate the number of the area, i=1,...,N, j=1,...,N, N is the number of areas, Δf _i (t) is the system Frequency deviation, Δf _Ti (t) is wind turbine frequency deviation, ΔP _mi (t) is generator output power increment, ΔP _vi (t) is regulator valve position increment, ΔE _i (t) is frequency deviation integral controller Δδ _i (t) is the phase angle increment, Δδ _j (t) is the phase angle increment of the j-th area, ΔP _di (t) is the system load disturbance, ΔP _GWi (t) is the i-th area T _ij is the tie line power synchronization factor between the i-th area and the j-th area, T _chi is the steam turbine time constant, T _rh is the reheater time constant, F _hp is the area The ratio of reheating steam turbine power generation to the total output power of all generators in the area, T _pi is the system time constant, K _pi is the system gain, K _IGi is the fluid coupling coefficient, ΔP _Li (t) is the active power deviation of the area, ΔP _ri (t) is the state quantity output by the steam turbine governor, R _i is the speed adjustment of the governor, _Bi is the area frequency offset coefficient, K _Ei is the integral control gain, d _i is the time delay constant, T _gi is governor time constant; The mathematical model of the wind power generation system is: in In the formula, Δx _i1 (t), Δx _i2 (t), Δx _i3 (t), Δx _i4 (t) represent the state variables in the i-th area fan model, T _wi is the time constant of the fan, K _PC1 is the propeller Pitch control feedback gain, K _P31 and T _P31 are pitch response data fitting control response coefficient and time constant respectively, K _P21 and T _P21 are hydraulic pitch actuator response coefficient and time constant respectively, K _PPi , K ₁₁ and K _P1i is the pitch control response coefficient, T _P1i is the pitch control response time constant, Δf _Ti (t) is the wind turbine frequency deviation, ΔP _mi (t) is the generator output power increment, K _IGi is the fluid coupling coefficient, K _TPi is the fan frequency feedback coefficient. 4. The load frequency control method of a multi-domain time-delay electric power system with high wind energy penetration rate according to claim 1, wherein the step S3 is specifically: select the matrix G _i such that G _i B _i ′is a non-singular matrix, σ _i (t) satisfies the equation The matrix K _i satisfies λ(A _i ′-B _i ′K _i )<0, and λ(*) means to solve the eigenvalue.