CN109995076B - A collaborative control method for power stable output of photovoltaic collection system based on energy storage - Google Patents

Abstract

The application provides a photovoltaic collection system power stable output cooperative control method based on energy storage, which comprises the following steps: identifying a photovoltaic power change mode of a current control period according to the photovoltaic output ultra-short-term prediction data, adaptively adjusting and optimizing objective functions according to different modes, acquiring an optimal filter coefficient through the objective function and a particle swarm algorithm, controlling energy storage output according to the optimal filter coefficient, smoothing fluctuation of photovoltaic power, and reducing fluctuation of light wave power; on the basis, the charge state of the energy storage battery is used as feedback to carry out secondary correction on the charge and discharge of the energy storage battery, so that the final light-storage combined output can be located in a predicted output interval, the photovoltaic predicted output is tracked, the accuracy of the photovoltaic prediction capability is improved, the problem that the photovoltaic output is stabilized to fluctuate in compensating the photovoltaic output prediction error is combined, and the two purposes of using the small-scale energy storage battery are achieved.

Description

A collaborative control method for power stable output of photovoltaic collection system based on energy storage

technical field

The present application relates to the technical field of photovoltaic systems, and in particular to a coordinated control method for stable power output of a photovoltaic collection system based on energy storage.

Background technique

In recent years, renewable energy represented by photovoltaics has developed rapidly. However, the utilization rate of photovoltaics in China is relatively low. There are two main reasons: one is the fluctuation of photovoltaic system output. Problems such as frequency fluctuations and excessive harmonics; photovoltaic output fluctuations can be divided into high-frequency fluctuations and low-frequency fluctuations. For low-frequency fluctuations, the power grid has enough response time to respond, so the impact of low-frequency fluctuations on the power grid can be ignored and reduced. Photovoltaic output fluctuations mainly refer to reducing the high-frequency fluctuations of photovoltaic output. Second, the accuracy of photovoltaic output prediction is low. The power grid predicts the photovoltaic power station output plan and system backup based on photovoltaic output. The prediction accuracy directly affects the photovoltaic grid connection space.

The photovoltaic power generation system based on energy storage is to add energy storage devices on the basis of the photovoltaic power generation system. Through the function of the energy storage system to quickly release and absorb electric energy, the high-frequency fluctuation of the output of the photovoltaic power generation system can be stabilized, and the smooth control of the output power of the system can be achieved. Purpose; however, in the prior art, the photovoltaic power generation system based on energy storage can smooth the photovoltaic power and suppress photovoltaic fluctuations. Although it can better use the historical data of photovoltaic output to predict photovoltaic output, it does not consider the impact of future photovoltaic output on the current moment. The impact of energy storage charging and discharging behavior; or directly inputting the predicted data into the simulation model instead of the actual output, resulting in a decrease in the accuracy of photovoltaic processing prediction.

Therefore, there is an urgent need for a power stable output system control method with dual goals of smoothing photovoltaic power and tracking photovoltaic predicted output.

Contents of the invention

The present application provides a coordinated control method for stable output of photovoltaic collection system power based on energy storage, which can improve the accuracy of predicted photovoltaic output by tracking and predicting output while smoothing photovoltaic power and stabilizing photovoltaic power output.

In order to solve the above technical problems, the embodiment of the present application discloses the following technical solutions:

The present application provides a method for collaborative control of power stable output of a photovoltaic collection system based on energy storage, the method comprising:

Ultra-short-term prediction of photovoltaic output to obtain predicted photovoltaic output power P _f ;

Determine the photovoltaic output trend mode according to the photovoltaic predicted output power _Pf , and the photovoltaic output trend mode includes a rising pattern, a falling pattern and a fluctuating pattern;

Acquire λ according to the photovoltaic output trend mode, and the λ is a weight coefficient representing the battery charge and discharge intensity item;

其中P_O为光储联合输出功率，P_b为储能电池输出功率；Establish objective function J, said objective function J is

Where P _O is the combined output power of optical storage, and P _b is the output power of the energy storage battery;

Obtaining the optimal filter coefficient α _opt by using particle swarm optimization algorithm according to the objective function J;

According to the optimal filter coefficient α _opt, the photovoltaic output power is smoothed based on the low-pass filter algorithm, and at the same time, the optimal filter coefficient α _opt is passed to the cooperative control module to obtain the preliminary joint output power P _{o, temp} and battery output value P _b,pri ;

According to the SOC of the energy storage battery, the preliminary optical storage joint output power P _o,temp is compensated for the prediction error to obtain the energy storage corrected power P _b,rec ;

The output power P _{b of the energy storage battery is calculated according to the battery output value P b,pri} and the energy storage correction power P _{b ,rec} .

Preferably, the determination of the photovoltaic output trend mode according to the photovoltaic forecast output power _Pf includes:

Define the function P＝[P _o (t-1), P _PV (t), P _f (t+1), P _f (t+2), P _f (t+3)] and the function ΔP _m ＝P _f (t+3)-P _o (t-1), where P _PV (t) is the photovoltaic output power at the current moment, P _o (t-1) is the combined output power of photovoltaics and storage at the previous moment, and P _f (t+1 ), P _f (t+2) and P _f (t+3) are respectively the predicted output power of photovoltaics at time t+1, time t+2 and time t+3 in the future;

Determine the monotonicity of the function P, when the function P is monotonically increasing, the photovoltaic output trend mode is an upward mode, and when the function P is monotonically decreasing, the photovoltaic output trend mode is a downward mode;

When the function P has a single pole and is a maximum value, if ΔP _m ≥ 0, the photovoltaic output trend mode is an upward mode, and if ΔP _m <-ε, the photovoltaic output trend mode is a downward mode, If -ε≤ΔP _m <0, the photovoltaic output trend mode is a fluctuation mode;

When the function P has a single pole and is a minimum value, if ΔP _m ≥ ε, the photovoltaic output trend mode is an upward mode, and if ΔP _m <0, the photovoltaic output trend mode is a downward mode, if 0≤ΔP _m <ε, then the photovoltaic output trend mode is a fluctuation mode;

When the function P has double poles, if ΔP _m >ε, the photovoltaic output trend mode is an upward mode, if ΔP _m <-ε, then the photovoltaic output trend mode is a downward mode, if -ε≤ΔP _m <ε0≤ΔP _m <ε, the photovoltaic output trend mode is a fluctuation mode.

Preferably, the obtaining λ according to the photovoltaic output trend mode includes:

When the photovoltaic output trend mode is rising mode, λ=SOC(t);

When the photovoltaic output trend mode is a descending mode, λ=100%-SOC(t);

When the photovoltaic output trend mode is a fluctuation mode, λ=2|50%-SOC(t)|;

Where SOC(t) is the state of charge value of the energy storage battery at the current moment.

Preferably, the transfer of the optimal filter coefficient α _opt to the cooperative control module to obtain the joint output power P _{o,temp of} optical storage and the output value P _b,pri of the battery includes:

_Obtain P _{o,temp according to P o,temp} =α _opt PPV(t)+(1-α _opt )Po(t-1);

P b, _pri is obtained according to P _b,pri =P _PV (t)−P _o,temp (t).

Preferably, the calculating the output power P _{b of the energy storage battery according to the battery output value P b,pri} and the energy storage correction power P _{b ,rec} includes:

The output power P _b of the energy storage battery is calculated according to P _b =P _b,pri +P _b,rec .

Compared with the prior art, the beneficial effects of the present application are:

(1) According to the ultra-short-term prediction data of photovoltaic output, this application identifies the photovoltaic power change mode in the current control period, adaptively adjusts and optimizes the objective function for different modes, and obtains the optimal filter coefficient through the objective function combined with the particle swarm algorithm. According to the optimal filter coefficient Control energy storage output, smooth fluctuations in photovoltaic power, stabilize light wave output, and improve photovoltaic utilization.

(2) The application transmits the optimal filter coefficient αo _pt to the cooperative control module to obtain the preliminary optical storage combined output power Po, _temp and battery output value P _{b, pri} ; Combined output power Po, _temp of optical storage to compensate prediction error to obtain energy storage correction power P _b,rec ; calculate energy storage battery output power P according to the battery output value P _b,pri and the energy storage correction power P _{b,rec b} , to achieve collaborative control, complete the secondary correction of the charge and discharge of the energy storage battery, so that the final joint output of photovoltaic storage can be within the predicted output range, track the predicted output of photovoltaics, improve the accuracy of photovoltaic prediction capabilities, and reduce the difference between photovoltaic output and The deviation of dispatching output improves the adjustment depth and space of joint output of solar storage system.

(3) This application combines the problem of stabilizing photovoltaic output fluctuations with compensating photovoltaic output prediction errors, and makes a secondary correction to the charge and discharge of the energy storage system, achieving the above-mentioned dual goals by using smaller-scale energy storage batteries .

Description of drawings

In order to illustrate the technical solution of the present application more clearly, the accompanying drawings that need to be used in the embodiments will be briefly introduced below. Obviously, for those of ordinary skill in the art, on the premise of not paying creative work, there are also Additional figures can be derived from these figures.

Fig. 1 is a schematic structural diagram of a photovoltaic collection system based on energy storage in the present application;

Fig. 2 is a schematic flow chart of a method for coordinated control of power stable output of a photovoltaic collection system based on energy storage provided by the present application;

Fig. 3 is the schematic flow chart of optimal filter coefficient solution in the present application;

Fig. 4 is a schematic diagram of a variation mode of photovoltaic output in an embodiment of the present invention.

Detailed ways

In order to enable those skilled in the art to better understand the technical solutions in the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described The embodiments are only some of the embodiments of the present application, not all of them. Based on the embodiments in this application, all other embodiments obtained by persons of ordinary skill in the art without creative efforts shall fall within the scope of protection of this application.

This application provides a coordinated control method for stable power output of photovoltaic collection systems based on energy storage, which can smooth photovoltaic power and reduce photovoltaic fluctuations while tracking and predicting output to improve the accuracy of photovoltaic forecast output and improve photovoltaic utilization.

The structure of the photovoltaic-storage combined system is shown in Figure 1. Figure 1 is a schematic structural diagram of the photovoltaic collection system based on energy storage in this application. It can be seen from Figure 1 that the system includes a large-scale photovoltaic electric field, energy storage batteries, DC/ DC converter, high ratio DC/DC converter, grid-connected DC/AC converter. The centralized energy storage battery is installed at the photovoltaic outlet to coordinate the output of renewable energy, and the controller provides the reference value of the charging and discharging power of the energy storage battery. Neglecting the energy loss during transmission and conversion, there are:

P _o =P _PV -P _b

Among them, P _O is the combined output power of photovoltaics and storage, PP _PV is the original output power of the photovoltaic power station, and P _b is the output power of the energy storage battery; P _b > 0 means that the battery is charging, otherwise it is discharging.

In order to achieve the dual goals of smoothing optical power fluctuations and compensating prediction errors, this application optimizes the filter coefficients for smooth fluctuations by using an adaptive optimization filter coefficient module, and then further adjusts the actual output of energy storage through a collaborative control module to complete the tracking prediction target of error. The coordinated control strategy for stable power output of photovoltaic storage system mainly includes three modules: ultra-short-term prediction of photovoltaic output by BPNN method, adaptive filter coefficient optimization, and coordinated control.

The traditional low-pass filter inputs the photovoltaic field output power sequence P _PV into the first-order inertial link whose time constant is T, and obtains a relatively smooth output P _o , and its transfer function is:

After discretizing the above formula, we get:

α∈[0,1]。Among them, Δt is the time interval, defining the filter coefficient

α∈[0,1].

表明滤波后输出功率平滑程度与滤波系数有关：α＝0时，P_o(t)＝P_o(t-1)，光储联合系统输出功率平稳，但光伏功率波动完全由储能电池补偿，电池单次充放电功率较大，易迅速饱和或放空。若要求储能电池持续参与光伏输出调节，则需要配置较大的电池容量，投资成本高；α＝1时P_o(t)＝P_PV(t)，电池闭锁，即不参与光伏电站波动平抑。Mode

It shows that the smoothness of the output power after filtering is related to the filter coefficient: when α=0, P _o (t)=P _o (t-1), the output power of the combined optical storage system is stable, but the photovoltaic power fluctuation is completely compensated by the energy storage battery, The single charge and discharge power of the battery is large, and it is easy to be saturated or empty quickly. If the energy storage battery is required to continue to participate in the regulation of photovoltaic output, it is necessary to configure a large battery capacity, and the investment cost is high; when α = 1, P _o (t) = P _PV (t), the battery is blocked, that is, it does not participate in the fluctuation stabilization of the photovoltaic power station .

Specifically, this application provides a coordinated control method for power stable output of a photovoltaic collection system based on energy storage. Specifically, refer to FIG. A schematic flow chart of a control method, the method comprising:

S01: Ultra-short-term prediction of photovoltaic output to obtain predicted photovoltaic output power P _f .

The ultra-short-term prediction of photovoltaic output is carried out to obtain the predicted photovoltaic output power P _f . Specifically, the sampling time interval of the original output data of the photovoltaic power station is T _s , and P _f is the predicted output power of the photovoltaic electric field. The BP neural network is used to predict the ultra-short-term output of the photovoltaic electric field , the forecast duration is T _l , the forecast data time interval is T _f , and T _f >T _s . The purpose of ultra-short-term forecasting of photovoltaic output is to optimize the filter coefficients using the forecast data on the T _f time scale. In this patent, the ultra-short-term prediction of photovoltaic output adopts the traditional BPNN method. The BPNN method is a commonly used technical means in the field, so it will not be repeated here.

S02: Determine a photovoltaic output trend mode according to the photovoltaic predicted output power _Pf , and the photovoltaic output trend mode includes a rising pattern, a falling pattern and a fluctuating pattern.

This application identifies the photovoltaic power change mode in the current control cycle based on the ultra-short-term forecast data of photovoltaic output, and therefore defines three photovoltaic output change modes: rising, falling and steady fluctuation mode. The power used for mode identification is the joint output power of photovoltaics and storage at the previous moment, the actual output power of photovoltaics at the current moment, and the predicted output power of photovoltaics in the future. Considering that the accuracy of photovoltaic prediction decreases with the prediction time, the prediction value within the next 30 minutes is selected to participate in the mode judgment.

Specifically, determining the photovoltaic output trend mode according to the photovoltaic forecast output power _Pf includes:

Define the function P＝[P _o (t-1), P _PV (t), P _f (t+1), P _f (t+2), P _f (t+3)] and the function ΔP _m ＝P _f (t+3)-P _o (t-1), where P _PV (t) is the photovoltaic output power at the current moment, P _o (t-1) is the combined output power of photovoltaics and storage at the previous moment, and P _f (t+1 ), P _f (t+2) and P _f (t+3) are respectively the predicted output power of photovoltaics at time t+1, time t+2 and time t+3 in the future;

Determine the monotonicity of the function P, when the function P is monotonically increasing, the photovoltaic output trend mode is an upward mode, and when the function P is monotonically decreasing, the photovoltaic output trend mode is a downward mode;

When the function P has a single pole and is a maximum value, if ΔP _m ≥ 0, the photovoltaic output trend mode is an upward mode, and if ΔP _m <-ε, the photovoltaic output trend mode is a downward mode, If -ε≤ΔP _m <0, the photovoltaic output trend mode is a fluctuation mode;

When the function P has a single pole and is a minimum value, if ΔP _m ≥ ε, the photovoltaic output trend mode is an upward mode, and if ΔP _m <0, the photovoltaic output trend mode is a downward mode, if 0≤ΔP _m <ε, then the photovoltaic output trend mode is a fluctuation mode;

When the function P has double poles, if ΔP _m >ε, the photovoltaic output trend mode is an upward mode, if ΔP _m <-ε, then the photovoltaic output trend mode is a downward mode, if -ε≤ΔP _m <ε0≤ΔP _m <ε, the photovoltaic output trend mode is a fluctuation mode.

The above content is expressed in Table 1 as:

Table 1 Judgment criteria for photovoltaic output trend mode

S03: Obtain λ according to the photovoltaic output trend mode, and the λ is a weight coefficient representing the battery charge and discharge intensity item. When the photovoltaic output trend mode is rising mode, λ=SOC(t);

When the photovoltaic output trend mode is a descending mode, λ=100%-SOC(t);

When the photovoltaic output trend mode is a fluctuation mode, λ=2|50%-SOC(t)|;

Where SOC(t) is the state of charge value of the energy storage battery at the current moment.

Please refer to FIG. 4 for scenarios corresponding to specific modes. FIG. 4 is a schematic diagram of a photovoltaic output variation mode in an embodiment of the present invention.

Under different modes of photovoltaic output, in order to meet the requirements of reducing power fluctuations, the battery will have different charging and discharging modes, and its state of charge has different restrictions on battery output. In the rising mode, the original output of photovoltaics continues to increase. In order to reduce the growth rate of the joint output of photovoltaics and storage within the control period, it is necessary to charge the energy storage battery in one direction. At this time, the working capacity of the energy storage battery is negatively correlated with the SOC, that is, as the SOC increases, the working space of the energy storage decreases. In order to reduce the power output of the battery at this time, it is necessary to appropriately increase the weight of the output power item of the energy storage battery. Therefore, in this mode, λ is set as the SOC value of the energy storage at the current moment, that is, λ=SOC(t); on the contrary, for the falling mode, the energy storage The battery needs to be discharged to slow down the decline of photovoltaic power. At this time, the working capacity of the energy storage battery is positively related to the SOC. In order to reduce the working intensity of the energy storage when the SOC is low, set λ=100%-SOC(t) in this mode , in order to increase the weight of the P _b item when the SOC is low; in the fluctuation mode, the energy storage battery charges and discharges in the control cycle at the same time, so to maintain the ideal state of SOC at 50% as the goal, set λ=2|50 %-SOC(t)|. The coefficient "2" is mainly to be consistent with the rising and falling modes, and to ensure that the range of λ is between [0,1].

其中P_O为光储联合输出功率，P_b为储能电池输出功率。S04: Establish an objective function J, the objective function J is

Where P _O is the combined output power of optical storage, and P _b is the output power of the energy storage battery.

In order to meet the smooth effect of photovoltaic output while taking into account the battery charge and discharge intensity, a multi-objective optimization problem model is constructed to solve the optimal filter coefficient. The objective function is as follows:

The two terms in the above formula represent the fluctuation constraint on the joint output power of optical storage and the charge and discharge power constraint of the energy storage battery respectively. 0, the optimization objective only considers the smoothing effect, and when λ increases, the influence of the P _b item on the optimization objective J will be enhanced, and the effect of smoothing fluctuations will be weakened.

多目标优化问题情境，本文提出一种自适应的权重系数确定方案，权重系数由当前及未来一段时间内光伏发电趋势及储能电池荷电状态SOC共同决定。When building a multi-objective optimization model, the existing methods mostly adopt the method of directly assigning values to the weight coefficients, which lacks configuration standards for the weight coefficients. In some cases, it is easy to cause the change of the objective function to mainly depend on a certain item, which cannot meet the goal of multi-objective optimization. Targeted

In the multi-objective optimization problem scenario, this paper proposes an adaptive weight coefficient determination scheme. The weight coefficient is jointly determined by the current and future photovoltaic power generation trends and the SOC of the energy storage battery.

S05: Obtain the optimal filter coefficient α _opt by using the particle swarm optimization algorithm according to the objective function J.

This application controls the energy storage output according to the optimal filter coefficient, smoothes the fluctuation of photovoltaic power, and stabilizes the light wave output. Therefore, the acquisition of the optimal filter coefficient α _opt is the key to this application. This application uses particle swarm optimization according to the objective function J The algorithm obtains the optimal filter coefficient _αopt ; the specific solution method refers to Fig. 3, and Fig. 3 is a schematic flow diagram of the optimal filter coefficient solution in this application; it can be seen from Fig. 3 that the solution process is as follows:

Set the control parameters of the particle swarm algorithm: the total number of particle swarms Γ, the inertia constant interval [W _min , W _max ], the learning factors c ₁ , c ₂ , and the number of iterations L;

Initialize the position and velocity of the particle swarm and the number of iterations k=0;

Calculate the particle fitness based on the objective function, and update the individual optimal and global optimal;

Update the number of iterations k=k+1, and determine whether the maximum number of iterations has been reached: if so, stop the iteration and record the current global optimal solution; otherwise, update the particle position and velocity, and continue to execute "calculate particle fitness based on the objective function, and update the individual optimal solution". optimal and global optimal".

S06: Smooth the photovoltaic output power based on the low-pass filter algorithm according to the optimal filter coefficient α _opt , and at the same time pass the optimal filter coefficient α _opt to the cooperative control module to obtain the preliminary joint output power P _o,temp and battery output Value P _b,pri .

In this application, the optimal filter coefficient α _opt in the current control period is obtained after the adaptive filter coefficient is optimized, and is passed to the cooperative control module. By inputting α _opt to pre-smooth the original output of the photovoltaic electric field, this application identifies the photovoltaic power change mode in the current control cycle based on the ultra-short-term prediction data of photovoltaic output, and adaptively adjusts and optimizes the objective function for different modes, and combines the objective function with the particle swarm algorithm Obtain the optimal filter coefficient, control the output of energy storage according to the optimal filter coefficient, smooth the fluctuation of photovoltaic power, stabilize the stability of light wave output, and improve the utilization rate of photovoltaic.

Specifically, the pre-smoothing method is based on the low-pass filtering algorithm, and the low-pass rate algorithm is a commonly used technical means in this field, so it will not be repeated here; the preliminary optical-storage combined output power P _o,temp and battery output value P _b,pri , compared with the original output of the photovoltaic power plant, the fluctuation rate of P _o,temp decreases. Afterwards, on the basis of P _o,temp , the charging and discharging action of the battery is further adjusted, the prediction error is compensated, and the joint output of optical storage is corrected. The SOC of the energy storage battery actively participates in the adjustment of the charging and discharging power of the battery as a closed-loop feedback quantity, and calculates the energy storage correction power P _b,rec . The reference output of the energy storage battery is jointly determined by P _b,pri and P _b, rec to achieve collaborative control.

Specifically, the transfer of the optimal filter coefficient α _opt to the cooperative control module to obtain the joint output power P _{o,temp of} optical storage and the output value P _b,pri of the battery includes:

_Obtain P _{o,temp according to P o,temp} =α _opt P _PV (t)+(1-α _opt )P _o (t-1);

P b, _pri is obtained according to P _b,pri =P _PV (t)−P _o,temp (t).

S07: Compensate the prediction error for the preliminary combined optical-storage output power P _o,temp according to the SOC of the energy storage battery to obtain a corrected energy storage power P _b,rec .

The energy storage battery charging and discharging power correction is based on different P _o,temp and SOC state intervals. To do this, first define the upper and lower forecast bounds:

ΔP _tol is the margin of error; based on the relationship between P _{o, temp} and the prediction interval: lower than the lower limit of prediction (P _{o, temp} < P _l ), within the allowable range of prediction (P _l ≤ P _{o, temp} ≤ P _h ) and higher than the predicted upper limit (P _o,temp >P _h ). At the same time, the state of charge of the battery is divided into three intervals: [0, SOC _low ], [SOC _low , SOC _high ], [SOC _high , 100%], SOC _high and SOC _low are the ideal working intervals of the energy storage battery The upper and lower boundary values of . See Table 2 for the secondary corrected power output of the energy storage battery under each interval combination.

Table 2 Energy storage correction power P _b,rec value corresponding to different SOC

Take the behavior of P _o,temp >P _h in the table as an example: in order to track and predict, the battery needs to be charged. If the battery SOC<SOC _low indicates that the battery has enough charging space, in order to improve the battery state of charge as soon as possible, we should make full use of P _o, The power space between _temp and P _l supplements the state of charge of the battery; on the contrary, if the battery SOC>SOC _high , the remaining charging space of the battery is insufficient. Correction; when the battery SOC is within the ideal working range, the secondary correction power of the energy storage battery is linearly related to the current SOC, and the lower the SOC, the greater the correction power. For the cases of P _l ≤ P _{o, temp} ≤ P _h and P _{o, temp} < P _l , the same idea is used to determine the secondary correction power of the energy storage battery.

S08: Calculate the output power P _{b of the energy storage battery according to the battery output value P b,pri} and the energy storage correction power P _{b ,rec} ; specifically, calculate the power according to P _b =P _b,pri +P _b,rec The output power P _b of the energy storage battery.

The final actual output power of the energy storage battery is jointly determined by the part used to smooth fluctuations and the secondary correction part of tracking prediction:

P _b =P _b,pri +P _b,rec

Since the joint output of photovoltaics and storage is adjusted after tracking the prediction interval, it has a certain impact on the smoothing effect of the joint output. On the premise of not adding a subsequent smoothing module, this paper requires that the photovoltaic forecast should be relatively accurate, and ΔP _tol should not be too large. Avoid |P _b,pri | >> |P _b,rec |.

This application realizes collaborative control and completes the secondary correction of the charge and discharge of the energy storage battery, so that the final joint output of photovoltaic storage can be located in the predicted output range, track the predicted output of photovoltaics, improve the accuracy of photovoltaic prediction ability, and reduce the difference between photovoltaic output and The deviation of dispatching output improves the adjustment depth and space of joint output of solar storage system.

At the same time, the application can adjust the state of charge of the energy storage battery in real time, realize the prevention of pre-charging of the battery, effectively stabilize the state of charge within a reasonable working range, and prolong the service life of the battery.

In this application, in order to quantitatively evaluate the effect of energy storage on smoothing photovoltaic power and compensating prediction errors, the following four evaluation indicators are used in this paper, and M in each indicator represents the number of data in the evaluation period.

Swing average:

Fluctuation max:

Forecasts track off-limits averages:

P _err (t)＝max(|P _o (t)-P _f (t)|-ΔP _tol ,0)

Predicted trace violation probability:

In the formula, the prediction tracking limit value P _err (t) indicates the distance between the output power and the adjacent prediction upper/lower limit when the output power is not within the prediction interval at time t.

Since the above implementation methods are described in conjunction with reference to other methods, different embodiments have the same parts, and the same and similar parts of the various embodiments in this specification can be referred to each other. No further elaboration here.

Other embodiments of the present application will be readily apparent to those skilled in the art from consideration of the specification and practice of the inventive disclosure herein. This application is intended to cover any modification, use or adaptation of the present invention, these modifications, uses or adaptations follow the general principles of the application and include common knowledge or conventional technical means in the technical field not disclosed in the application . The specification and examples are to be considered exemplary only, with the true scope and spirit of the application indicated by the contents of the appended claims.

The embodiments of the present application described above are not intended to limit the scope of protection of the present application.

Claims (5)

1. An energy storage-based photovoltaic collection system power stable output collaborative control method, characterized in that the method comprises: Ultra-short-term prediction of photovoltaic output to obtain predicted photovoltaic output power P _f ; Determine the photovoltaic output trend mode according to the photovoltaic predicted output power _Pf , and the photovoltaic output trend mode includes a rising pattern, a falling pattern and a fluctuating pattern; Acquire λ according to the photovoltaic output trend mode, and the λ is a weight coefficient representing the battery charge and discharge intensity item; Establish objective function J, said objective function J is

Where P _O is the combined output power of optical storage, and P _b is the output power of the energy storage battery; Obtaining the optimal filter coefficient α _opt by using particle swarm optimization algorithm according to the objective function J; According to the optimal filter coefficient α _opt, the photovoltaic output power is smoothed based on the low-pass filter algorithm, and at the same time, the optimal filter coefficient α _opt is passed to the cooperative control module to obtain the preliminary joint output power P _{o, temp} and battery output value P _b,pri ; According to the SOC of the energy storage battery, the preliminary optical storage joint output power P _o,temp is compensated for the prediction error to obtain the energy storage corrected power P _b,rec ; The output power P _{b of the energy storage battery is calculated according to the battery output value P b,pri} and the energy storage correction power P _{b ,rec} . 2. The method according to claim 1, wherein said determining the photovoltaic output trend mode according to the photovoltaic predicted output power _Pf comprises: Define the function P＝[P _o (t-1), P _PV (t), P _f (t+1), P _f (t+2), P _f (t+3)] and the function ΔP _m ＝P _f (t+3)-P _o (t-1), where P _PV (t) is the photovoltaic output power at the current moment, P _o (t-1) is the combined output power of photovoltaics and storage at the previous moment, and P _f (t+1 ), P _f (t+2) and P _f (t+3) are respectively the predicted output power of photovoltaics at time t+1, time t+2 and time t+3 in the future; Determine the monotonicity of the function P, when the function P is monotonically increasing, the photovoltaic output trend mode is an upward mode, and when the function P is monotonically decreasing, the photovoltaic output trend mode is a downward mode; When the function P has a single pole and is a maximum value, if ΔP _m ≥ 0, the photovoltaic output trend mode is an upward mode, and if ΔP _m <-ε, the photovoltaic output trend mode is a downward mode, If -ε≤ΔP _m <0, the photovoltaic output trend mode is a fluctuation mode; When the function P has a single pole and is a minimum value, if ΔP _m ≥ ε, the photovoltaic output trend mode is an upward mode, and if ΔP _m <0, the photovoltaic output trend mode is a downward mode, if 0≤ΔP _m <ε, then the photovoltaic output trend mode is a fluctuation mode; When the function P has double poles, if ΔP _m >ε, the photovoltaic output trend mode is an upward mode, if ΔP _m <-ε, then the photovoltaic output trend mode is a downward mode, if -ε≤ΔP _m <ε0≤ΔP _m <ε, the photovoltaic output trend mode is a fluctuation mode; Among them, ε is the fluctuation threshold. 3. The method according to claim 1, wherein said acquiring λ according to said photovoltaic output trend pattern comprises: When the photovoltaic output trend mode is rising mode, λ=SOC(t); When the photovoltaic output trend mode is a descending mode, λ=100%-SOC(t); When the photovoltaic output trend mode is a fluctuation mode, λ=2|50%-SOC(t)|; Where SOC(t) is the state of charge value of the energy storage battery at the current moment. 4. The method according to claim 1, characterized in that the transfer of the optimal filter coefficient _αopt to the cooperative control module to obtain the joint output power P _{o, temp of} optical storage and the battery output value P _{b, pri} includes : _Obtain P _{o,temp according to P o,temp} =α _opt P _PV (t)+(1-α _opt )P _o (t-1); P b, _pri is obtained according to P _b,pri =P _PV (t)−P _o,temp (t). 5. The method according to claim 1, wherein the calculating the output power Pb of the energy storage battery according to the battery output value Pb _,pri and the energy storage correction power _{Pb ,rec} comprises: The output power P _b of the energy storage battery is calculated according to P _b =P _b,pri +P _b,rec .

Images