Energy storage optimization method based on accurate photovoltaic prediction

Abstract

The energy storage optimization method based on the accurate photovoltaic prediction solves the problems of the existing energy storage system optimization method that the energy storage capacity cannot be effectively utilized and the power smoothing effect is poor, and belongs to the field of new energy generation, energy storage and grid connection application.A kind of energy storage optimization method based on the accurate photovoltaic prediction, comprising: according to the photovoltaic prediction result, different time scales are distinguished, if it is a long time scale, based on the photovoltaic prediction result of the long time scale, the SOC of the energy storage system in the future prediction period is estimated, the current SOC is adjusted combined with the working condition limit, and the SOC out-of-limit phenomenon of the energy storage system in the future period is avoided, if it is a short time scale, based on the photovoltaic prediction result of the short time scale, the future short-term large photovoltaic power fluctuation is responded in advance, and the power smoothing effect is optimized under the premise of meeting the current power grid side power demand in the maximum limit.

Description

Technical Field

[0001] This invention relates to an energy storage optimization method based on accurate photovoltaic prediction, belonging to the field of new energy power generation and grid-connected energy storage applications. Background Technology

[0002] With more and more photovoltaic power generation systems being connected to the grid, especially with the development model of large-scale photovoltaic power sources being centrally connected to the grid, which is unique to my country, the traditional power system is developing towards a new type of power system characterized by a high proportion of new energy sources and a high degree of power electronics. The low damping and low inertia of the new power system itself, as well as the volatility and intermittency of photovoltaic power sources, place higher demands on power system dispatch and energy consumption.

[0003] Energy storage systems possess strong power storage and supply capabilities and fast response speeds. They can effectively mitigate power fluctuations caused by the intermittent nature of renewable energy generation and respond promptly to load changes, making them widely used in the renewable energy generation sector. Photovoltaic-energy storage integrated systems, which connect photovoltaic (PV) and energy storage systems via converters, play a crucial role in smoothing system power fluctuations and improving power supply and distribution stability. However, due to limitations imposed by their own State of Charge (SOC), the capacity of energy storage systems is often not fully utilized, affecting their power smoothing effect and reducing their economic efficiency. Furthermore, existing energy storage-based power smoothing control methods rely on real-time adjustments. Due to delays in information acquisition, energy storage systems often fail to respond promptly to large power fluctuations caused by uncontrollable PV output, resulting in significant peak-to-valley power differences in the power system. Simultaneously, large variations in output power can also affect the performance and lifespan of the energy storage devices themselves. Therefore, effective monitoring and management of energy storage systems and optimization of their operating status are among the key issues in the current application of renewable energy generation and grid-connected energy storage.

[0004] Currently, research on optimization methods for energy storage systems is only based on the assessment of the current SOC output capacity of the energy storage system, ignoring the impact of the SOC accumulation process and future photovoltaic power output, resulting in the ineffective utilization of energy storage capacity. In addition, some studies maintain SOC stability by reducing the number of charge and discharge cycles of the energy storage battery. This method avoids overcharging and over-discharging of the battery, but it cannot meet the power demand of the grid side, which is not conducive to the stable operation of the system. Summary of the Invention

[0005] To address the problems of ineffective utilization of energy storage capacity and poor power smoothing effect in existing energy storage system optimization methods, this invention provides an energy storage optimization method based on accurate photovoltaic prediction.

[0006] The present invention provides an energy storage optimization method based on accurate photovoltaic prediction, comprising:

[0007] S1. Based on the photovoltaic forecast results, distinguish different time scales. If it is a long time scale, proceed to S2; if it is a short time scale, proceed to S3.

[0008] S2. Dynamically adjust the SOC of the energy storage system based on long-term photovoltaic prediction results;

[0009] S3. Advanced response to power fluctuations based on short-timescale photovoltaic prediction results.

[0010] Preferably, S2 includes:

[0011] S21, the nth period T l This represents the current period, and the (n+1)th period T. l T represents a future forecast period; l For long-term photovoltaic forecasting periods;

[0012] S22. Calculate the SOC adjustment baseline value of the energy storage system using the optimal SOC planning model. The optimal SOC planning model is:

[0013]

[0014] Where t0 is the current time, t l To provide the time resolution for long-term photovoltaic prediction, SOC(t0) represents the measured SOC state value of the energy storage system at the current moment, and P... y-pv (t) represents the (n+1)th period T. l Internal photovoltaic power prediction results, P g (t) represents the power demand on the grid side, E b This refers to the rated capacity of the energy storage system. For the charging and discharging efficiency of the energy storage system; This indicates the lower limit of the normal operating range of the energy storage system's State of Charge (SOC). This indicates the upper limit of the normal operating range of the energy storage system's State of Charge (SOC). Indicates in SOC ref The (n+1)th period T calculated based on this... l The predicted SOC value of the internal energy storage system, of which Less than or equal to positive integers, ,therefore It is t l The result obtained by integration using the smallest unit of time One predicted value, T represents the (n+1)th period l Internal SOC Predictions The minimum value in, T represents the (n+1)th period l Inside SOC prediction value The maximum value in;

[0015] S23. Control the energy storage system to exchange energy with the grid, and adjust the SOC of the energy storage system to... Place.

[0016] Preferably, S21 includes:

[0017] S211. Based on the measured SOC state value SOC(t0) at the current moment, calculate the (n+1)th period T. l Internal energy storage system SOC prediction value ;

[0018] S212, with t l For the (n+1)th period T, the unit is... l Internal energy storage system SOC prediction value The system will determine and count the number of times the predicted State of Charge (SOC) value of the energy storage system exceeds the limit. Obtain the fluctuation state of the measured historical error value ΔSOC. According to the fluctuation state and the number of times the limit is exceeded Determine the smoothing factor ;

[0019]

[0020] α h-min α h-max These represent the minimum and maximum values ​​of the high-range smoothing factor, respectively; α l-min α l-max This represents the minimum and maximum values ​​of the low-range smoothing factor.

[0021] S213. Based on long-term photovoltaic prediction results, determine the nth period T. l Energy storage system SOC prediction error ΔSOC n The (n+1)th period T is obtained using the quadratic exponential smoothing method. l Estimate of SOC prediction error for internal energy storage system ;

[0022]

[0023] Among them, a n b n For smoothing coefficients;

[0024]

[0025] ΔSOC n For the nth period T lThe actual error in the SOC prediction of the energy storage system; , For the nth period and the (n-1)th period T respectively l The first-smoothed value of the SOC prediction error of the internal energy storage system; , For the nth period and the (n-1)th period T respectively l The second-order smoothed value of the SOC prediction error of the internal energy storage system.

[0026] As a preferred option, in S211, the (n+1)th period T l Internal energy storage system SOC prediction value y (t):

[0027] .

[0028] As a preferred option, in S212, the fluctuation state for:

[0029]

[0030] in, , T represents the past n periods l The actual error of the SOC prediction for the energy storage system is given by k, where k is a positive integer less than or equal to n.

[0031] As a preferred embodiment, S21 also includes:

[0032] S214, according to and the SOC adjustment benchmark value of energy storage system The (n+1)th period T is obtained l Predicted SOC value of the energy storage system at the end of the term :

[0033] .

[0034] Preferably, S3 includes:

[0035] S31. Determine the reasonable fluctuation range of grid-side power. Determine the prediction period T of two adjacent short timescales. s The range of power difference fluctuation in the energy storage system output is ±ΔP b ;

[0036] S32. Calculate the photovoltaic output power P at the current moment. pv (t0) and the predicted value of photovoltaic output The difference, if the difference is within the allowable fluctuation range ±ΔP b If it falls within the range, proceed to S33; otherwise, proceed to S34.

[0037] S33. The energy storage system normally compensates for grid demand. The expected output power of the energy storage system is: P b-ref =P g (t0) -P pv (t0);

[0038] S34. Using the optimal power planning model, calculate the reasonable and optimal expected output power P of the energy storage system. b-ref :

[0039]

[0040] Where t0 is the current time, This represents the power demand on the grid side at the current moment.

[0041] This invention also provides an energy storage control system based on accurate photovoltaic prediction, including a photovoltaic prediction unit, an information acquisition and power calculation unit, a coordinate transformation unit, an energy storage system SOC monitoring and management unit, a power reference value calculation unit, and a power and current control loop unit;

[0042] The photovoltaic prediction unit is connected to both the energy storage system SOC monitoring and management unit and the power reference value calculation unit. It is used to obtain photovoltaic prediction results, send the long-term photovoltaic prediction results to the energy storage SOC monitoring and management unit, and send the short-term photovoltaic prediction results to the power reference value calculation unit.

[0043] The signal acquisition and power calculation unit is connected to the energy storage system SOC monitoring and management unit, the power reference value calculation unit, and the power and current control loop unit. It is used to acquire the voltage and current of the grid side, photovoltaic power generation side, and AC side of the energy storage converter at the current moment, calculate the corresponding output power, send the output power of the grid side to the energy storage SOC monitoring and management unit, and send the output power of the grid side and photovoltaic power generation side to the power reference value calculation unit.

[0044] The coordinate transformation unit, connected to the power and current control loop unit, is used to obtain the voltage u at the grid connection point PCC via the phase-locked loop (PLL). pcc The phase θ is used to determine the output current i on the AC side of the energy storage converter. sh Perform the Park transform to obtain the d-axis and q-axis components i. shd and i shq And send it to the power and current control loop unit;

[0045] The energy storage SOC monitoring and management unit is used to obtain the SOC adjustment benchmark value of the energy storage system based on long-term photovoltaic forecast results and grid-side output power. refIt also controls the energy storage system to exchange energy with the grid, adjusting the current energy storage SOC to the desired SOC. ref Place;

[0046] The energy storage SOC monitoring and management unit calculates the SOC adjustment baseline value of the energy storage system through the optimal SOC planning model. The optimal SOC planning model is:

[0047]

[0048] Where t0 is the current time, t l To provide the time resolution for long-term photovoltaic prediction, SOC(t0) represents the measured SOC state value of the energy storage system at the current moment, and P... y-pv (t) represents the (n+1)th period T. l Internal photovoltaic power prediction results, P g (t) represents the power demand on the grid side, E b This refers to the rated capacity of the energy storage system. For the charging and discharging efficiency of the energy storage system; This indicates the lower limit of the normal operating range of the energy storage system's State of Charge (SOC). This indicates the upper limit of the normal operating range of the energy storage system's State of Charge (SOC). T represents the (n+1)th period l The predicted SOC value of the internal energy storage system, of which Less than or equal to positive integers, ,therefore It is t l The result obtained by integration using the smallest unit of time One predicted value, T represents the (n+1)th period l Internal SOC Predictions The minimum value in, T represents the (n+1)th period l Inside SOC prediction value The maximum value in;

[0049] The power reference value calculation unit, connected to the power and current control loop unit, is used to obtain the expected output power P of the energy storage system based on short-timescale photovoltaic forecast results and the output power of the grid side and the photovoltaic power generation side. b-ref Combined with the desired output power P b-ref The optimal power reference value is obtained and sent to the power and current control loop unit;

[0050] If the output power P of the photovoltaic power generation side at the current moment pv (t0) and the predicted value of photovoltaic output The difference is within the allowable fluctuation range ±ΔP b Within this range, the energy storage system is expected to output power P. b-ref =P g (t0) - P pv (t0), otherwise, using the optimal power planning model, calculate the reasonable and optimal expected output power P of the energy storage system. b-ref :

[0051]

[0052] Where t0 is the current time, Given the current power demand of the grid, the reasonable fluctuation range of the grid's output power is: Two adjacent short-timescale prediction periods T s The fluctuation range of the output power difference of the energy storage system is ±ΔP. b ;

[0053] The power and current control loop unit is used to control the power and current according to i shd i shq The phase θ and the optimal power reference value are used to control the energy storage converter for power compensation.

[0054] The beneficial effects of this invention are as follows: Based on accurate photovoltaic (PV) forecasting results, this invention fully combines the characteristics of PV forecasting over both long and short timescales to achieve precise monitoring and effective management of various aspects of the operating characteristics of the PV-storage integrated system. This invention utilizes the power exchange capability of the energy storage system to provide timely power compensation or absorb excess power from the system when PV output fluctuates, thereby smoothing power fluctuations and ensuring the safe and stable operation of the power system. Based on long-term PV forecasting results, this invention estimates the SOC of the energy storage system for the next forecast period and adjusts the current SOC in conjunction with operating condition constraints to avoid SOC exceeding limits in the next period. Compared with traditional methods, this invention can optimize the operating state of the energy storage system without sacrificing grid-side power demand, improve the utilization rate of energy storage capacity and power, and enhance the safety and economic benefits of system operation. This invention utilizes a quadratic exponential smoothing method to correct the accuracy of SOC budgeting. It also proposes a dynamic adjustment method for the smoothing factor, combining historical data fluctuations with the number of SOC exceedances in the next prediction period to dynamically adjust the error correction coefficient between the smoothing factor and the energy storage SOC, further improving the accuracy of monitoring and predicting energy storage system output. Based on short-term photovoltaic forecasting results, this invention proactively responds to significant short-term photovoltaic power fluctuations. While maximizing the satisfaction of current grid-side power demand, it optimizes power smoothing effects. Compared to traditional methods, this invention can smooth the output power of photovoltaic-energy storage combined systems, while avoiding the problems of excessively rapid changes in energy storage system output power and system response spikes caused by large power fluctuations, effectively protecting the performance of various aspects of the energy storage system. Attached Figure Description

[0055] Figure 1 Circuit connection diagram for connecting the photovoltaic-storage integrated system to the power grid;

[0056] Figure 2 The overall block diagram of the energy storage system optimization method based on accurate photovoltaic prediction at multiple time scales is shown.

[0057] Figure 3 This is a flowchart of an energy storage SOC management method based on accurate long-term photovoltaic prediction.

[0058] Figure 4 A block diagram of a forward response power smoothing method based on accurate short-timescale photovoltaic prediction;

[0059] Figure 5 This is a control block diagram for an energy storage system optimization method based on accurate photovoltaic prediction. Detailed Implementation

[0060] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0061] It should be noted that, unless otherwise specified, the embodiments and features described in the present invention can be combined with each other.

[0062] The present invention will be further described below with reference to the accompanying drawings and specific embodiments, but this is not intended to limit the scope of the invention.

[0063] The energy storage optimization method based on accurate photovoltaic prediction in this embodiment is applied to a photovoltaic-energy storage combined system based on a parallel energy storage converter. The circuit connection diagram of this system connected to the power grid is shown below. Figure 1 As shown, it includes: a three-phase bridge converter, a DC bus capacitor, an energy storage battery, a connecting reactor, and a three-phase isolation transformer.

[0064] exist Figure 1 In this circuit, the main structure of the energy storage converter is a three-phase bridge converter, with the DC side consisting of capacitor C. dc It is combined with an energy storage battery in parallel, and the DC side voltage is V. dc To indicate, I dc This is the DC-side output current. The energy storage converter is connected to reactor L. sh R is connected in parallel with the three-phase isolation transformer at the microgrid grid connection point. sh To account for the equivalent resistance of the connected reactor, the primary-to-secondary transformation ratio of each phase of the three-phase transformer is n:1. sh with i sh These represent the AC output voltage and current of the energy storage converter, respectively; u pv with i pv These represent the output voltage and current of the photovoltaic power generation system, respectively; u g with i g These represent the grid-side voltage and current, respectively. pcc R is the voltage at the grid connection point PCC. g and L g This is the equivalent impedance of the transmission line.

[0065] At the same time, according to the principle of power conservation, the following relationship can be obtained:

[0066]

[0067] Among them, P g and Q g These represent the active and reactive power absorbed by the grid side, respectively; P pv and Qpv These represent the active and reactive power outputs of the photovoltaic power generation system, respectively; P sh and Q sh These represent the active and reactive power outputs of the energy storage converter, respectively; P b The active power output of the energy storage system simultaneously satisfies P sh = P b The relationship is as follows: DC-side capacitors can compensate for and absorb reactive power, while energy storage systems mainly respond to the active power of the system.

[0068] This embodiment of the energy storage optimization method based on accurate photovoltaic prediction manages the state of charge (SOC) of the energy storage system by using photovoltaic power prediction data on both long and short time scales, while simultaneously smoothing the power output of the photovoltaic-storage combined system. The overall block diagram of this method is shown below. Figure 2 As shown, it includes:

[0069] Different time scales are distinguished based on the photovoltaic forecast results. If it is a long time scale, the SOC of the energy storage system is dynamically adjusted based on the photovoltaic forecast results of the long time scale. If it is a short time scale, the power fluctuation is responded to in advance based on the photovoltaic forecast results of the short time scale.

[0070] In this embodiment, long-term photovoltaic forecast results can be used to calculate the energy required by the energy storage system in the next forecast period, obtain the trend of the SOC of the energy storage system in that period, and adjust the current SOC by charging or discharging the energy storage system in advance according to the actual operating conditions. This ensures that the SOC remains within a reasonable operating range in the next forecast period, and maximizes the satisfaction of the power demand on the grid side.

[0071] In this embodiment, the power change within a future prediction period can be determined by using short-timescale photovoltaic forecast results. The energy storage system can then respond proactively to large photovoltaic power fluctuations, avoiding the impact of large power changes on the energy storage system and smoothing the power output of the photovoltaic-storage combined system.

[0072] It is necessary to comprehensively consider the characteristics of different forecast time scales and select an appropriate forecast period. For long time scales, the forecast period T... l The time range can be selected from 4 to 6 hours, with a time resolution of t. l Set to 15 minutes; for short timescales, the prediction period T s The time interval can be selected as 15 minutes, with a time resolution of t. s Set it to 15 minutes.

[0073] The method described in this embodiment for dynamically adjusting the SOC of the energy storage system based on long-term photovoltaic (PV) forecast results calculates the energy required by the energy storage system by combining the PV output over a future forecast period, obtains the trend of SOC change within that period, and determines the limit on SOC based on actual operating conditions. min ≤ SOC ≤ SOC max The specific idea behind this method is as follows: within a future prediction period, if the calculated energy storage SOC exceeds the upper limit, the energy storage system needs to be discharged in advance to reduce the current SOC; if the calculated energy storage SOC is lower than the lower limit, the energy storage system needs to be charged in advance to increase the current SOC, thereby leaving sufficient capacity and power margin to meet subsequent grid demand. Therefore, the core of this method is to use photovoltaic prediction results to obtain the optimal adjustment value of the current energy storage SOC, specifically including:

[0074] Step 21: Based on the long-term photovoltaic prediction results, determine the nth period T. l Energy storage system SOC prediction error ΔSOC n The (n+1)th period T is obtained. l Estimate of SOC prediction error for internal energy storage system That is, the error correction coefficient; the nth period T l This represents the current period, and the (n+1)th period T. l T represents a future forecast period; l For long-term photovoltaic forecasting periods;

[0075] Step 22: To ensure that in the next period T l Since the internal energy storage SOC remains within a reasonable operating range, the current energy storage SOC (t0) needs to be adjusted. Based on the above prediction results, the baseline value for SOC adjustment of the energy storage system is calculated using the optimal SOC planning model. The optimal SOC planning model is:

[0076]

[0077] Where t0 is the current time, t l To provide the time resolution for long-term photovoltaic prediction, SOC(t0) represents the measured SOC state value of the energy storage system at the current moment, and P... y-pv (t) represents the (n+1)th period T. l Internal photovoltaic power prediction results, P g (t) represents the power demand on the grid side, E b This refers to the rated capacity of the energy storage system. For the charging and discharging efficiency of the energy storage system; This indicates the lower limit of the normal operating range of the energy storage system's State of Charge (SOC). This indicates the upper limit of the normal operating range of the energy storage system's State of Charge (SOC). Indicates in SOC ref The (n+1)th period T calculated based on this... l The predicted SOC value of the internal energy storage system, of which Less than or equal to positive integers, ,therefore It is t l The result obtained by integration using the smallest unit of time One predicted value, T represents the (n+1)th period l Internal SOC Predictions The minimum value in, T represents the (n+1)th period l Inside SOC prediction value The maximum value in.

[0078] Step 23: Control the energy storage system to exchange energy with the grid, and adjust the SOC of the energy storage system to [normal value]. Place.

[0079] SOC ref To achieve a reasonable and optimal energy storage SOC adjustment benchmark, the energy storage system is controlled to exchange energy with the grid at time t0, thereby adjusting the energy storage SOC to the appropriate value. ref At this point, it can be guaranteed that in the next period T l When the internal energy storage system (SOC) operates stably within a safe range and can meet the power demand of the grid to the maximum extent, the energy exchange between the energy storage system and the grid is minimized, which improves the overall efficiency and capacity utilization to a certain extent and ensures the safe and stable operation of the system.

[0080] In a preferred embodiment, the limiting range of the energy storage SOC can be determined according to the actual operating conditions, i.e., the SOC. min ≤ SOC ≤ SOC max Based on photovoltaic prediction results, with a time resolution of t l Using the unit of calculation, for a future prediction period T l The system predicts and determines the SOC value of the stored energy, and counts the number of times the SOC exceeds the limit; the fluctuation status of the historical error value ΔSOC is calculated. Combined with fluctuation state With the next cycle T l Number of times the energy storage SOC exceeds the limit (m) l For smoothing factor α n Dynamic adjustments are made; a quadratic exponential smoothing mathematical model is used to calculate the future period T. lThe energy storage SOC prediction error within the system is used to obtain the error correction coefficient ΔSOC. y-n+1 The error correction coefficient ΔSOC y-n+1 Substituting these values ​​into the optimal SOC planning model, we obtain the energy storage SOC adjustment benchmark value SOC. ref And calculate the next period T l Predicted SOC value at the end of the storage period y-n+1 In this embodiment, step 21 includes:

[0081] Step 211: Based on the measured SOC state value SOC(t0) at the current moment, combined with the (n+1)th period T l Based on the internal photovoltaic output, calculate the energy consumed by the energy storage system, and obtain the (n+1)th period T. l Internal energy storage system SOC prediction value ;

[0082] The trend of energy storage SOC can be obtained through the following relationship:

[0083]

[0084] SOC y (t) represents the (n+1)th period T calculated based on the measured SOC state value SOC(t0) at the current moment. l The predicted SOC value of the internal energy storage system, of which Less than or equal to positive integers, Therefore, SOC y (t) is based on t l The result obtained by integration using the smallest unit of time One predicted value;

[0085] Step 212, with t l For the (n+1)th period T, the unit is... l Internal energy storage system SOC prediction value The system will determine and count the number of times the predicted State of Charge (SOC) value of the energy storage system exceeds the limit. Obtain the fluctuation state of the measured historical error value ΔSOC. According to the fluctuation state and the number of times the limit is exceeded Determine the smoothing factor ;

[0086] This implementation method is a dynamic adjustment method for the smoothing factor, which combines the fluctuation status of historical error value ΔSOC. With the next cycle T l Number of times the energy storage SOC exceeds the limit (m) l For smoothing factor αn To make adjustments, the fluctuation state of the measured historical error value ΔSOC is first obtained using the following formula. :

[0087]

[0088] Smoothing factor α n The value is typically taken in the range of 0.1-0.9. The magnitude of the value should take into account the fluctuations in historical data. When the historical data fluctuations are relatively mild, a smaller α value can be selected. n The numerical value is important, but when historical data fluctuates significantly, a larger α value needs to be selected. n Numerical values ​​are used to make the prediction model more sensitive and able to quickly keep up with changes in recent historical data. Therefore, the fluctuation status of historical error values ​​ΔSOC is incorporated. With the next cycle T l Number of times the energy storage SOC exceeds the limit (m) l The smoothing factor α is adjusted according to the following formula. n Adjustments will be made:

[0089]

[0090] α h-min α h-max These represent the minimum and maximum values ​​of the high-range smoothing factor, respectively; α l-min α l-max This represents the minimum and maximum values ​​of the low-range smoothing factor. In the formula above, For fluctuation threshold, when > When this occurs, it indicates that the historical data fluctuates significantly. In this case, let the smoothing factor α... n In [α] h-min α h-max If the value is within a higher range, then let the smoothing factor α be [value]. n In [α] l-min α l-max The value is taken within a lower range; considering that the charging and discharging performance of the energy storage system will change over time, therefore, when the SOC is predicted to be within a future period T... l When there are many instances of exceeding limits, recent historical data should be used to calculate ΔSOC more accurately. Therefore, it is necessary to combine the energy storage SOC to predict the number of limit exceedances, m. l Using the above formula, the smoothing factor α is adjusted in two different ranges. n Make dynamic adjustments.

[0091] Step 213: Based on the long-term photovoltaic prediction results, determine the nth period T. l Energy storage system SOC prediction error ΔSOC nThe (n+1)th period T is obtained using the quadratic exponential smoothing method. l Estimate of SOC prediction error for internal energy storage system :

[0092] To reduce the impact of changes in energy storage system parameters on the accuracy of SOC prediction, this implementation method uses a quadratic exponential smoothing method to reduce the SOC prediction error ΔSOC. n Calculations are performed, and the existing SOC predictions are corrected based on the calculation results. First, the prediction error ΔSOC is defined. n for:

[0093]

[0094] Wherein, ΔSOC n For the nth period T l The actual error in the SOC prediction of the energy storage system; SOC n For the nth period T l The measured SOC value of the energy storage at the end; SOC y-n For the nth period T l Predicted SOC value for energy storage at the end;

[0095] The quadratic exponential smoothing method calculation model is as follows:

[0096]

[0097] Among them, a n b n For smoothing coefficients;

[0098]

[0099] , For the nth period and the (n-1)th period T respectively l The first-smoothed value of the SOC prediction error of the internal energy storage system; , For the nth period and the (n-1)th period T respectively l The second-order smoothed value of the SOC prediction error of the internal energy storage system.

[0100] Step 214: Use the following formula to obtain the next period T. l The predicted SOC value at the end of the forecast period is used to participate in the calculation of the next forecast period, based on... and the SOC adjustment benchmark value of energy storage system The (n+1)th period T is obtained l Predicted SOC value of the energy storage system at the end of the term :

[0101] .

[0102] This embodiment uses a method based on the advance response to power fluctuations using short-timescale photovoltaic prediction results. The short-timescale photovoltaic prediction period is T. s The next forecast period T s The photovoltaic (PV) forecast results within the current period are compared with the PV output value at the current moment. If the difference is within the allowable range, power output is carried out normally according to the current grid demand. If the difference exceeds the allowable range, the next forecast period T is adjusted while ensuring that the current grid power is within a reasonable range. s To proactively respond to internal power demands, smooth power output, and avoid the impact of large power fluctuations on the energy storage system, specifically including:

[0103] First, determine the reasonable fluctuation range of the grid-side power. In other words, the output power of the photovoltaic-storage combined system must ensure that the grid-side power remains stable within the following range:

[0104]

[0105] Among them, P g (t0) represents the desired power on the grid side at the current moment, P g (t) represents the actual power on the grid side.

[0106] Furthermore, taking into account the impact of power fluctuation amplitude on the charging and discharging of the energy storage system and the response speed of the energy storage system control loop, the prediction period T between two adjacent prediction periods is determined. s The fluctuation range of the output power difference of the energy storage system is ±ΔP. b Calculate the current photovoltaic output power P pv (t0) and the predicted photovoltaic output value P y-pv (t0+T s The difference between the two values ​​satisfies the following relationship:

[0107]

[0108] The energy storage system is then allowed to charge and discharge normally to meet the grid demand. The expected output power of the energy storage system at this time is:

[0109]

[0110] If the difference between the two is not within the allowable range, then the reasonable and optimal expected output power of the energy storage system can be calculated according to the following optimal power planning model:

[0111]

[0112] In the above formula, it is assumed that the expected power on the grid side changes steadily, therefore the expected power value P at the current moment is used. g (t0) Calculate the expected output power for the next forecast period.

[0113] Advanced response power smoothing methods based on accurate short-timescale photovoltaic prediction, such as... Figure 4 As shown, the specific implementation steps of this method include:

[0114] Step 1: Determine the reasonable fluctuation range of grid-side power. Determine the prediction period T of two adjacent short timescales. s The range of power difference fluctuation in the energy storage system output is ±ΔP b ;

[0115] Step 2: Calculate the current photovoltaic output power P pv (t0) and the predicted photovoltaic output value P y-pv (t0+T s The difference between the two values ​​is within the allowable fluctuation range of ±ΔP. b If the result is within the acceptable range, proceed to step 3; otherwise, proceed to step 4.

[0116] Step 3: The energy storage system normally compensates for grid demand, with an expected output power of: P b-ref =P g (t0) - P pv (t0);

[0117] Step 4: Using the optimal power planning model, calculate the reasonable and optimal expected output power P of the energy storage system. b-ref .

[0118] This invention provides an optimization method for energy storage systems based on accurate photovoltaic prediction, the overall control block diagram of which is shown below. Figure 5 The system includes: a photovoltaic prediction unit, a signal acquisition and power calculation unit, a coordinate transformation unit, an energy storage SOC monitoring and management unit, a power reference value calculation unit, and a power and current control loop unit.

[0119] The signal acquisition and power calculation unit collects the voltage and current on the grid side, photovoltaic power generation side, and AC side of the energy storage converter at the current moment, and calculates the corresponding active and reactive power. The coordinate transformation unit obtains the voltage u at the grid connection point PCC through a phase-locked loop (PLL). pcc The phase θ is used to determine the output current i on the AC side of the energy storage converter. sh Perform the Park transformation to obtain its d-axis and q-axis components i. shd and i shq The energy storage SOC monitoring and management unit mainly utilizes... Figure 3 The process outlined below yields the optimal baseline SOC.ref It also controls the energy storage system to exchange energy with the grid, adjusting the current energy storage SOC to the desired SOC. ref The power reference value calculation unit utilizes... Figure 4 The process outlined below obtains the current optimal power reference value P. shref Q shref The data is then fed into the power and current control loop unit. The control flow of the power and current control loop unit is as follows: the difference between the power reference value and the actual value is calculated and sent to the PI regulator to obtain the d-axis and q-axis setpoints of the compensator output current. The difference between these values ​​is then calculated and sent to the PI regulator to obtain the modulation voltage signal. After the modulation voltage signal is modulated by the Park inverse transform and the PWM modulation module, the switching transistor control signal is obtained to control the energy storage converter for power compensation.

[0120] The present invention provides an energy storage control system based on accurate photovoltaic prediction, comprising a photovoltaic prediction unit, an information acquisition and power calculation unit, a coordinate transformation unit, an energy storage system SOC monitoring and management unit, a power reference value calculation unit, and a power and current control loop unit.

[0121] The photovoltaic prediction unit is connected to both the energy storage system SOC monitoring and management unit and the power reference value calculation unit. It is used to obtain photovoltaic prediction results, send the long-term photovoltaic prediction results to the energy storage SOC monitoring and management unit, and send the short-term photovoltaic prediction results to the power reference value calculation unit.

[0122] The signal acquisition and power calculation unit is connected to the energy storage system SOC monitoring and management unit, the power reference value calculation unit, and the power and current control loop unit. It is used to acquire the voltage and current of the grid side, photovoltaic power generation side, and AC side of the energy storage converter at the current moment, calculate the corresponding output power, send the output power of the grid side to the energy storage SOC monitoring and management unit, and send the output power of the grid side and photovoltaic power generation side to the power reference value calculation unit.

[0123] The coordinate transformation unit, connected to the power and current control loop unit, is used to obtain the voltage u at the grid connection point PCC via the phase-locked loop (PLL). pcc The phase θ is used to determine the output current i on the AC side of the energy storage converter. sh Perform the Park transform to obtain the d-axis and q-axis components i. shd and i shq And send it to the power and current control loop unit;

[0124] The energy storage SOC monitoring and management unit is used to obtain the SOC adjustment benchmark value of the energy storage system based on long-term photovoltaic forecast results and grid-side output power. refThe energy storage SOC monitoring and management unit mainly utilizes Figure 3 The process outlined below yields the optimal baseline SOC. ref It also controls the energy storage system to exchange energy with the grid, adjusting the current energy storage SOC to the desired SOC. ref The power reference value calculation unit, connected to the power and current control loop unit, is used to calculate the power reference value based on short-timescale photovoltaic forecast results and the output power of the grid side and the photovoltaic power generation side. Figure 4 The process outlined below obtains the current optimal power reference value P. shref Q shref And send it to the power and current control loop unit;

[0125] The power and current control loop unit is used to control the power and current according to i shd i shq The phase θ and optimal power reference value control the energy storage converter for power compensation. The control flow of the power and current control loop unit is as follows: the difference between the power reference value and the actual value is sent to the PI regulator to obtain the d-axis and q-axis setpoints of the compensator output current. The difference between these setpoints and the actual output current value is then sent to the PI regulator to obtain the modulation voltage signal. The modulation voltage signal is modulated by the Park inverse transform and the PWM modulation module to obtain the switching control signal to control the energy storage converter for power compensation.

[0126] While the invention has been described herein with reference to specific embodiments, it should be understood that these embodiments are merely examples of the principles and applications of the invention. Therefore, it should be understood that many modifications can be made to the exemplary embodiments, and other arrangements can be designed without departing from the spirit and scope of the invention as defined by the appended claims. It should be understood that different dependent claims and features described herein can be combined in ways different from those described in the original claims. It is also understood that features described in conjunction with individual embodiments can be used in other described embodiments.

Claims

1. An energy storage optimization method based on accurate photovoltaic prediction, characterized in that, The method includes: S1. Based on the photovoltaic forecast results, distinguish different time scales. If it is a long time scale, proceed to S2; if it is a short time scale, proceed to S3. S2. Dynamically adjust the SOC of the energy storage system based on long-term photovoltaic prediction results; S3. Anticipating power fluctuations based on short-timescale photovoltaic prediction results; S2 includes: S21, the nth period T l This represents the current period, and the (n+1)th period T. l T represents a future forecast period; l For long-term photovoltaic forecasting periods; S22. Calculate the SOC adjustment baseline value of the energy storage system using the optimal SOC planning model. The optimal SOC planning model is: ； in, For the current moment, For long-term photovoltaic forecasting, the temporal resolution is... This is the measured SOC (State of Charge) value of the energy storage system at the current moment. For the (n+1)th period Internal photovoltaic power prediction results For the power demand on the grid side, This refers to the rated capacity of the energy storage system. For the charging and discharging efficiency of the energy storage system; This indicates the lower limit of the normal operating range of the energy storage system's State of Charge (SOC). This indicates the upper limit of the normal operating range of the energy storage system's State of Charge (SOC). Indicates in The (n+1)th period calculated based on this The predicted SOC value of the internal energy storage system, of which Less than or equal to positive integers, ,therefore Therefore The result obtained by integration using the smallest unit of time One predicted value, T represents the (n+1)th period l Internal SOC Predictions The minimum value in, Indicates the (n+1)th period Inside SOC prediction value The maximum value in; S23. Control the energy storage system to exchange energy with the grid, and adjust the SOC of the energy storage system to... Place; S3 includes: S31. Determine the reasonable fluctuation range of grid-side power. Determine the prediction period of two adjacent short timescales. Range of output power difference in energy storage system ; S32. Calculate the photovoltaic output power at the current moment. Compared with the photovoltaic power output forecast The difference, if the difference is within the allowable fluctuation range If it falls within the range, proceed to S33; otherwise, proceed to S34. S33. The energy storage system normally compensates for grid demand. The expected output power of the energy storage system is: ; S34. Using the optimal power planning model, calculate the reasonable and optimal expected output power of the energy storage system. : ； Where t0 is the current time, This represents the power demand on the grid side at the current moment.

2. The energy storage optimization method based on accurate photovoltaic prediction according to claim 1, characterized in that, S21 includes: S211. Based on the measured SOC state value (t0) of the energy storage system at the current moment, calculate the (n+1)th period T. l Internal energy storage system SOC prediction value ; S212, with For the (n+1)th period, the unit is... Internal energy storage system SOC prediction value The system will determine and count the number of times the predicted State of Charge (SOC) value of the energy storage system exceeds the limit. Obtain the fluctuation state of the measured historical error value ΔSOC. According to the fluctuation state and the number of times the limit is exceeded Determine the smoothing factor ; ； These represent the minimum and maximum values ​​of the high-range smoothing factor, respectively. Represents the minimum and maximum values ​​of the low-range smoothing factor; The fluctuation threshold; S213. Based on long-term photovoltaic prediction results, determine the nth period T. l Energy storage system SOC prediction error ΔSOC n The (n+1)th period is obtained using the quadratic exponential smoothing method. Estimate of SOC prediction error for internal energy storage system ; ； in, For smoothing coefficients; ； ΔSOC n For the nth period T l The actual error in the SOC prediction of the energy storage system; , For the nth period and the (n-1)th period T respectively l The first-smoothed value of the SOC prediction error of the internal energy storage system; , For the nth period and the (n-1)th period T respectively l The second-order smoothed value of the SOC prediction error of the internal energy storage system.

3. The energy storage optimization method based on accurate photovoltaic prediction according to claim 2, characterized in that, In S211, the (n+1)th period T l Internal energy storage system SOC prediction value y (t): 。 4. The energy storage optimization method based on accurate photovoltaic prediction according to claim 2, characterized in that, In S212, the wave state σ is: ； in, , T represents the past n periods l The actual error of the SOC prediction for the energy storage system is given by k, where k is a positive integer less than or equal to n.

5. The energy storage optimization method based on accurate photovoltaic prediction according to claim 2, characterized in that, S21 further includes: S214, according to and the SOC adjustment benchmark value of energy storage system The (n+1)th period T is obtained l Predicted SOC value of the energy storage system at the end of the term : 。 6. A computer-readable storage device storing a computer program, characterized in that, When the computer program is executed, it implements the energy storage optimization method based on accurate photovoltaic prediction as described in any one of claims 1 to 5.

7. An energy storage optimization device based on accurate photovoltaic prediction, comprising a storage device, a processor, and a computer program stored in the storage device and executable on the processor, characterized in that, The processor executes the computer program to implement the energy storage optimization method based on accurate photovoltaic prediction as described in any one of claims 1 to 5.

8. An energy storage control system based on accurate photovoltaic prediction, comprising a photovoltaic prediction unit, an information acquisition and power calculation unit, a coordinate transformation unit, an energy storage system SOC monitoring and management unit, a power reference value calculation unit, and a power and current control loop unit, characterized in that, The photovoltaic prediction unit is connected to both the energy storage system SOC monitoring and management unit and the power reference value calculation unit. It is used to obtain photovoltaic prediction results, send the long-term photovoltaic prediction results to the energy storage SOC monitoring and management unit, and send the short-term photovoltaic prediction results to the power reference value calculation unit. The signal acquisition and power calculation unit is connected to the energy storage system SOC monitoring and management unit, the power reference value calculation unit, and the power and current control loop unit. It is used to acquire the voltage and current of the grid side, photovoltaic power generation side, and AC side of the energy storage converter at the current moment, calculate the corresponding output power, send the output power of the grid side to the energy storage SOC monitoring and management unit, and send the output power of the grid side and photovoltaic power generation side to the power reference value calculation unit. The coordinate transformation unit, connected to the power and current control loop unit, is used to obtain the voltage u at the grid connection point PCC via the phase-locked loop (PLL). pcc The phase θ is used to determine the AC output current i of the energy storage converter. sh Perform the Park transform to obtain the d-axis and q-axis components i. shd and i shq And send it to the power and current control loop unit; The energy storage SOC monitoring and management unit is used to obtain the SOC adjustment benchmark value of the energy storage system based on long-term photovoltaic forecast results and grid-side output power. ref It also controls the energy storage system to exchange energy with the grid, adjusting the current energy storage SOC to the desired SOC. ref Place; The energy storage SOC monitoring and management unit calculates the SOC adjustment baseline value of the energy storage system through the optimal SOC planning model. The optimal SOC planning model is: ； Where t0 is the current time, t l To provide the time resolution for long-term photovoltaic prediction, SOC(t0) represents the measured SOC state value of the energy storage system at the current moment, and P... y-pv (t) represents the (n+1)th period T. l Internal photovoltaic power prediction results, P g (t) represents the power demand on the grid side, E b η% represents the rated capacity of the energy storage system; η% represents the charge / discharge efficiency of the energy storage system. This indicates the lower limit of the normal operating range of the energy storage system's State of Charge (SOC). This indicates the upper limit of the normal operating range of the energy storage system's State of Charge (SOC). T represents the (n+1)th period l The predicted SOC value of the internal energy storage system, of which Less than or equal to positive integers, ,therefore It is t l The result obtained by integration using the smallest unit of time One predicted value, T represents the (n+1)th period l Internal SOC Predictions The minimum value in, T represents the (n+1)th period l Inside SOC prediction value The maximum value in; The power reference value calculation unit, connected to the power and current control loop unit, is used to obtain the expected output power P of the energy storage system based on short-timescale photovoltaic forecast results and the output power of the grid side and the photovoltaic power generation side. b-ref Combined with the desired output power P b-ref The optimal power reference value is obtained and sent to the power and current control loop unit; If the output power P of the photovoltaic power generation side at the current moment pv (t0) and the predicted value of photovoltaic output The difference is within the allowable fluctuation range ±ΔP b Within this range, the energy storage system is expected to output power P. b-ref =P g (t0) - P pv (t0), otherwise, using the optimal power planning model, calculate the reasonable and optimal expected output power P of the energy storage system. b-ref : ； Where t0 is the current time, Given the current power demand of the grid, the reasonable fluctuation range of the grid power output is ±w%. This applies to two adjacent short-timescale prediction periods T. s The fluctuation range of the output power difference of the energy storage system is ±ΔP. b ; The power and current control loop unit is used to control the power and current according to i shd i shq The phase θ and the optimal power reference value are used to control the energy storage converter for power compensation.

Images