A method for improving the accuracy of photovoltaic power plant output prediction using energy storage

Abstract

This invention belongs to the field of photovoltaic power generation technology, specifically relating to a method for improving the output prediction accuracy of photovoltaic power plants using energy storage. Addressing the shortcomings of existing energy storage photovoltaic power plants, such as high energy storage capacity and high charging / discharging frequencies, the invention adopts the following technical solution: A method for improving the output prediction accuracy of photovoltaic power plants using energy storage, comprising: acquiring an output prediction curve and the current total power of the photovoltaic power plant, and comparing them to obtain the prediction accuracy; comparing the prediction accuracy with a preset accuracy threshold to determine whether the prediction accuracy meets the standard; when the prediction accuracy meets the standard, the energy storage system is in standby mode; when the prediction accuracy does not meet the standard, the energy storage system is controlled to perform charging / discharging operations. The beneficial effect of this invention is that when the prediction accuracy meets the standard, the energy storage system is in standby mode, and only when the prediction accuracy does not meet the standard does the energy storage system perform charging / discharging operations, thereby reducing the charging / discharging frequency of the energy storage system while meeting the prediction accuracy requirements.

Description

Technical Field

[0001] This invention belongs to the field of photovoltaic power generation technology, specifically relating to a method for improving the accuracy of photovoltaic power plant output prediction by utilizing energy storage. Background Technology

[0002] The power-side energy storage mode, which is jointly deployed with photovoltaic power plants, can make full use of the advantages of energy storage, such as strong peak-shaving and frequency regulation capabilities and fast response speed, to improve the overall grid friendliness of photovoltaic power plants, enhance the peak-shaving and frequency regulation capabilities of photovoltaic power plants, and actively support the healthy and stable operation of the power grid.

[0003] According to the "Implementation Rules for the Management of Ancillary Services of Grid-Connected Power Plants in East China" and the "Implementation Rules for the Management of Grid-Connected Operation of Power Plants in East China," photovoltaic (PV) power plants are required to submit their photovoltaic power forecasts to the power dispatching and trading agency. PV power forecasts are conducted in two ways: day-ahead forecasts and ultra-short-term forecasts. When the actual power generation of a PV power plant deviates significantly from the forecasted power, the PV power plant must undergo an assessment and compensate for the assessment costs. However, due to the volatility and intermittency of PV power output, the accuracy of its forecasts is relatively low, making it difficult to meet the assessment targets of the dispatching agency.

[0004] Current research on the application of energy storage technology in photovoltaic power plants mainly focuses on scheduling methods that use energy storage to track planned output and smooth photovoltaic output. However, to achieve good tracking and smoothing effects of planned output, a high capacity of energy storage is required, and frequent large-scale adjustments to energy storage charging and discharging have a significant impact on the lifespan of energy storage, which will greatly increase the overall operating cost of the entire plant.

[0005] Therefore, it is necessary to propose a scheme to improve the output prediction accuracy of photovoltaic power plants based on factors such as energy storage characteristics and lifespan, so as to improve the power prediction accuracy of photovoltaic power plants, reduce the amount of electricity required for photovoltaic power plants, and at the same time reduce the charging and discharging frequency of energy storage. Summary of the Invention

[0006] This invention addresses the shortcomings of existing energy storage photovoltaic power plants, such as high energy storage capacity and high energy storage charging and discharging frequency. It provides a method to improve the output accuracy of photovoltaic power plants by utilizing energy storage, thereby improving the power prediction accuracy of photovoltaic power plants, reducing the amount of electricity required for photovoltaic power plant assessment, and lowering the overall operating cost of the entire plant.

[0007] To achieve the above objectives, the present invention adopts the following technical solution: a method for improving the output prediction accuracy of photovoltaic power plants using energy storage, the method comprising the following steps:

[0008] Obtain the power output prediction curve and the current total power of the photovoltaic power station, and compare them to obtain the prediction accuracy;

[0009] The prediction accuracy is compared with the preset accuracy threshold to determine whether the prediction accuracy meets the standard.

[0010] When the prediction accuracy meets the standard, the energy storage system is in standby mode; when the prediction accuracy does not meet the standard, the energy storage system is controlled to perform charging and discharging operations.

[0011] This invention discloses a method for improving the output prediction accuracy of photovoltaic power plants using energy storage. The method controls the energy storage system based on the prediction accuracy. When the prediction accuracy meets the target, the energy storage system is in standby mode. Charging and discharging operations are only performed when the prediction accuracy fails to meet the target. This reduces the charging and discharging frequency of the energy storage system, extends its lifespan, and lowers the overall operating cost of the entire power plant. The accuracy threshold can be used as a performance evaluation standard.

[0012] As an improvement, the power output forecast curves include the day-ahead 24-hour power output forecast curve and the ultra-short-term 4-hour power output forecast curve.

[0013] As an improvement, the following steps are included:

[0014] Step S1: Obtain the AC power P of the photovoltaic array inverter. s Total power of photovoltaic power station P N 24-hour output forecast curve of photovoltaic power station P short , Photovoltaic power plant ultra-short-term 4-hour output forecast curve P ultra-short Energy storage PCS (Power Conversion System) AC side power P E ;

[0015] Step S2: Set the current time T i Total power of photovoltaic power station P Ni Compared with the 24-hour output forecast curve of the photovoltaic power plant P short_i By comparing, we can obtain the current time T. i Real-time photovoltaic power plant day-ahead average forecast accuracy μ short_i :

[0016]

[0017] Step S3: Set the current time T i Total power of photovoltaic power station P Ni Compared with the ultra-short-term 4-hour output prediction curve of photovoltaic power plants P ultra-short By comparing, we can obtain the current time T. i Real-time photovoltaic power plant ultra-short-term daily average forecast accuracy μ ultra-short_i :

[0018]

[0019] Step S4: Set the current time T to... i The daily average forecast accuracy μ short_i Ultra-short-term daily average forecast accuracy μultra-short_i Compared with the assessment criteria, the system enters four different operating conditions: PR0, PR1, PR2, and PR3, depending on the prediction accuracy.

[0020] As an improvement, in step S4, the assessment standard for the daily average forecast accuracy is 0.8, and the assessment standard for the ultra-short-term daily average forecast accuracy is 0.85.

[0021] As an improvement, step S4 includes:

[0022] Step S401: When μ short_i >0.8, μ ultra-short_i When the accuracy is greater than 0.85, the photovoltaic power station is in the condition where the prediction accuracy meets the standard. The photovoltaic power station generates electricity freely, the energy storage system is in standby mode, and the coordination control is in the PR0 condition.

[0023] Step S402: When μ short_i <0.8, μ ultra-short_i When the value is greater than 0.85, the day-ahead forecast accuracy of the photovoltaic power station is unqualified, the ultra-short-term forecast accuracy is qualified, the energy storage system is in operation, and the integrated photovoltaic power station and energy storage system enter the PR1 operating condition.

[0024] Step S403: When μ short_i >0.8, μ ultra-short_i When the accuracy is less than 0.85, the day-ahead forecast accuracy of the photovoltaic power station is qualified, the ultra-short-term forecast accuracy is unqualified, the energy storage system is in operation, and the integrated photovoltaic power station and energy storage system enter the PR2 operating condition.

[0025] Step S404: When μ short_i <0.8, μ ultra-short_i When the accuracy is less than 0.85, the day-ahead forecast accuracy and ultra-short-term forecast accuracy of the photovoltaic power station are both unqualified, the energy storage system is in operation, and the photovoltaic power station and energy storage system enter the PR3 operating condition.

[0026] As an improvement, under PR1 operating conditions, the coordinated control system controls the energy storage system to perform charging and discharging operations. Specifically, let the charging and discharging power of the energy storage system at the next moment be P. Ei+1 The accuracy of the day-ahead forecast for the entire station at the next moment is μ. short_i+1 The overall cost of the entire site at the next moment is M. ai+1 The assessment amount M is based on the accuracy of photovoltaic power plant forecasts. p Energy storage operating life depreciation cost M E The overall site assessment cost M was obtained. ai+1 :

[0027] M ai+1 =M p +M E =λ p (0.8-μshort_i+1 )P n tαC 全站 +λ E M Ei+1

[0028] Where, λ p λ is the predicted cost coefficient for photovoltaic power plants. E P is the cost factor for energy storage operation life depreciation. n Let t be the total installed capacity of the photovoltaic power station, α be the assessment hours, and C be the assessment management coefficient. 全站 This is the highest approved on-grid tariff for the entire photovoltaic power station.

[0029] The optimal P is obtained through a minimization optimization algorithm. Ei+1_best To reduce the overall site assessment cost M ai+1 lowest.

[0030] As an improvement, under PR2 operating conditions, the coordinated control system controls the energy storage system to perform charging and discharging operations. Specifically, let the charging and discharging power of the energy storage system at the next moment be P. Ei+1 The accuracy of the day-ahead forecast for the entire station at the next moment is μ. short_i+1 The overall cost of the entire site at the next moment is M. ai+1 The assessment amount M is based on the accuracy of photovoltaic power plant forecasts. p Energy storage operating life depreciation cost M E The overall site assessment cost M was obtained. ai+1 :

[0031] M ai+1 =M p +M E =λ p (0.85-μ ultra-short_i+1 )P n tαC 全站 +λ E M Ei+1

[0032] The optimal P is obtained through a minimization optimization algorithm. Ei+1_best To reduce the overall site assessment cost M ai+1 lowest.

[0033] As an improvement, under PR3 operating conditions, the coordinated control system controls the energy storage system to perform charging and discharging operations. Specifically, let the charging and discharging power of the energy storage system at the next moment be P. Ei+1 The accuracy of the day-ahead forecast for the entire station at the next moment is μ. short_i+1 The accuracy of ultra-short-term prediction is μ ultra-short_i+1 The next moment, the overall site cost M ai+1 as follows:

[0034]

[0035] The optimal P is obtained through a minimization optimization algorithm. Ei+1_best To reduce the overall site assessment cost M ai+1 lowest.

[0036] As an improvement, when the day-ahead forecast accuracy and ultra-short-term forecast accuracy of the photovoltaic power station change from unqualified to qualified, the energy storage management system is put into standby mode. The energy storage unit and the photovoltaic array of the photovoltaic power station share a box-type step-up transformer. The photovoltaic power generation unit is connected to the low-voltage side of the box-type step-up transformer through a DC / AC inverter, and the energy storage battery unit is connected to the low-voltage side of the box-type step-up transformer through a PCS, together forming a photovoltaic-energy storage unit.

[0037] A photovoltaic power station employs the aforementioned method for improving the accuracy of photovoltaic power station output prediction by utilizing energy storage.

[0038] The beneficial effects of the method for improving the output prediction accuracy of photovoltaic power plants by utilizing energy storage in this invention are as follows: the energy storage system is controlled according to the prediction accuracy. When the prediction accuracy meets the standard, the energy storage system is in standby mode. Only when the prediction accuracy does not meet the standard will the energy storage system perform charging and discharging operations, thereby reducing the charging and discharging frequency of the energy storage system while meeting the prediction accuracy requirements. Attached Figure Description

[0039] Figure 1 This is a schematic diagram of the photovoltaic and energy storage hybrid configuration mode of a centralized photovoltaic power station according to Embodiment 1 of the present invention.

[0040] Figure 2 This is a communication architecture diagram of a centralized photovoltaic power station according to Embodiment 1 of the present invention.

[0041] Figure 3 This is a flowchart of the steps in Embodiment 1 of the present invention. Detailed Implementation

[0042] The technical solutions of the embodiments of the present invention will be explained and described below with reference to the accompanying drawings. However, the following embodiments are only preferred embodiments of the present invention and not all embodiments. Other embodiments obtained by those skilled in the art based on the embodiments in the implementation methods without creative effort are all within the protection scope of the present invention.

[0043] See Figures 1 to 3 The present invention discloses a method for improving the output prediction accuracy of photovoltaic power plants using energy storage, the method comprising the following steps:

[0044] Obtain the power output prediction curve and the current total power of the photovoltaic power station, and compare them to obtain the prediction accuracy;

[0045] The prediction accuracy is compared with the preset accuracy threshold to determine whether the prediction accuracy meets the standard.

[0046] When the prediction accuracy meets the standard, the energy storage system is in standby mode; when the prediction accuracy does not meet the standard, it enters the photovoltaic-storage coordinated operation mode, controlling the energy storage system to coordinate with the photovoltaic array to perform charging and discharging operations.

[0047] The present invention provides a method for improving the output prediction accuracy of photovoltaic power plants using energy storage. The method controls the energy storage system based on the prediction accuracy. When the prediction accuracy meets the standard, the energy storage system is in standby mode; it only performs charging and discharging operations when the prediction accuracy fails to meet the standard, thereby reducing the charging and discharging frequency of the energy storage system. The accuracy threshold can be used as an evaluation standard.

[0048] Example 1

[0049] See Figure 1 Taking a centralized photovoltaic power station in Zhejiang Province as an example, a hybrid photovoltaic-energy storage configuration mode is adopted to achieve the goal of adding a power-side energy storage system. This configuration mode forms the basis for improving the overall forecasting accuracy of the station. This mode uses a shared box-type step-up transformer between the energy storage units and the photovoltaic array, with the energy storage system and the existing photovoltaic array being mixed. The existing photovoltaic power generation units are connected to the low-voltage side of the box-type step-up transformer via DC / DC and DC / AC inverters. All 55 box-type step-up transformers in the station are connected to the 35kV bus via seven 35kV collector lines. The newly added energy storage battery units are mixed and configured within several photovoltaic power generation units, connected to the 315V bus via a PCS, and then connected to the low-voltage side of the box-type step-up transformer, forming several photovoltaic-energy storage units. These units, along with the existing photovoltaic power generation units, are connected to the 35kV bus via seven 35kV collector lines.

[0050] See Figure 2 The data interaction of the photovoltaic energy storage hybrid configuration system is divided into a local equipment layer and a field control layer, with corresponding architectures. The local equipment layer includes a containerized energy storage BMS system, an energy storage PCS converter, metering meters, and photovoltaic output information sampling devices. The field control layer mainly includes a process control platform, workstations, and a photovoltaic power plant power prediction server. The process control platform is responsible for real-time data interaction with local equipment and issuing coordinated control strategies. Data acquisition includes: energy storage battery BMS operation information, energy storage PCS operation information, photovoltaic power plant operation information, and photovoltaic power plant power prediction information.

[0051] See Figure 3 The specific implementation steps are as follows:

[0052] Step S1: The photovoltaic-storage coordinated control system communicates with the photovoltaic centralized control system and the photovoltaic power prediction system via the IEC104 protocol (communication) to collect the AC side power P of the photovoltaic array inverter. sTotal power of photovoltaic power station P N 24-hour output forecast curve of photovoltaic power station P short , Photovoltaic power plant ultra-short-term 4-hour output forecast curve P ultra-short It communicates with the energy storage management system via the MOBUS TCP protocol to collect the AC side power P of the energy storage PCS. E And related information such as voltage and current.

[0053] Step S2: The photovoltaic-storage coordinated control system is configured with two control modes: "independent energy storage operation" and "photovoltaic-storage coordinated operation". The power station can select the appropriate operating mode according to its actual operating status to ensure stable and safe operation. When "photovoltaic-storage coordinated operation" is activated, the function of improving the output prediction accuracy of the photovoltaic power station can be realized by utilizing energy storage.

[0054] Step S3: Based on the current time T i Total power of photovoltaic power station P Ni Compared with the 24-hour output forecast curve of the photovoltaic power plant P short_i Get the current time T i Real-time photovoltaic power plant day-ahead average forecast accuracy μ short_i .

[0055] The real-time daily average forecast accuracy for photovoltaic power plants is the total power P of the photovoltaic power plants from 00:00 to the current time. N Compared with the 24-hour power output forecast P short The accuracy of the squared mean of the deviation. Specific evaluation of T. i The method is as follows:

[0056] Step S301: Starting from 00:00 during the day, sample and record the current T every 15 minutes. i Total power of photovoltaic power station P Ni and the current T i The predicted power output of the photovoltaic power plant at that time (P) short_i .

[0057] Step S302: Calculate the current T i The predicted power output of the photovoltaic power plant at that time (P) short_i With the total power P of the photovoltaic power station Ni Deviation rate δ Ni And record and store.

[0058]

[0059] Among them, P Ni When the value is 0, the default deviation rate δ Ni It is 0. δ Ni When δ > 1, Ni The value is 1.

[0060] Step S303: Calculate the time from 00:00 during the day to the current time T. i The total deviation of the photovoltaic power plant power prediction at time n is accumulated, and the average of the squared deviation rates of the n sampling points throughout the day is calculated to obtain the current T value of the photovoltaic power plant. i Daily average prediction accuracy μ before the time date short_i .

[0061]

[0062] Step S4: Based on the current time T i Total power of photovoltaic power station P Ni Compared with the ultra-short-term 4-hour output prediction curve of photovoltaic power plants P ultra-short Get the current time T i Real-time photovoltaic power plant ultra-short-term daily average forecast accuracy μ ultra-short_i .

[0063] The real-time photovoltaic power station ultra-short-term daily average forecast accuracy is from 00:00 within the day to the current time T. i During this period, the total power of the photovoltaic power station P N Compared with ultra-short-term 4-hour power output forecast P ultra-short The daily average accuracy of the deviation. The specific evaluation method is as follows:

[0064] Step S401: Since the ultra-short-term predicted power is updated every 15 minutes, starting from 00:00 within the day, the current T is sampled, recorded, and stored every 15 minutes. i Total power of photovoltaic power station P Ni And the next moment T i+1 Predicted power output P of photovoltaic power station in the ultra-short term (4 hours) ultra-short_i+1 .

[0065] Step S402: Use the previous time step T i-1 Real-time recorded ultra-short-term predicted power P of photovoltaic power plants ultra-short_i With the current time T i Total power of photovoltaic power station P Ni Deviation rate δ Ni And record and store.

[0066]

[0067] Among them, P Ni When the value is 0, the default deviation rate δ Ni It is 0. δ Ni When δ > 1, Ni The value is 1.

[0068] Step S403: Calculate the time from 00:00 during the day to the current time T.i The total deviation of the ultra-short-term power prediction of the photovoltaic power station at time n is accumulated, and the average of the squared deviation rates of the n sampling points throughout the day is calculated to obtain the current T value of the photovoltaic power station. i Time-based ultra-short-term daily average forecast accuracy μ ultra-short_i .

[0069]

[0070] Step S5: Set the current time T i Daily average prediction accuracy μ short_i Compare with the assessment standards. At this point, two scenarios exist:

[0071] (1) If μ short_i <0.8, day-to-day forecast accuracy μ short_i If the assessment is unsuccessful, proceed to step S601.

[0072] (2) If μ short_i ≥0.8, day-to-day forecast accuracy μ short_i If the assessment is passed, proceed to step S602.

[0073] Step S6: Set the current time T i Ultra-short-term daily average forecast accuracy and μ ultra-short_i The assessment standards were compared.

[0074] Step S601: The day-ahead forecast accuracy assessment fails, and:

[0075] (3) If μ ultra-short_i <0.85, ultra-short-term forecast accuracy μ ultra-short_i If the assessment is unsuccessful, proceed to step S701.

[0076] (4) If μ ultra-short_i ≥0.85, ultra-short-term forecast accuracy μ ultra-short_i If the assessment is passed, proceed to step S702.

[0077] Step S602: The day-ahead forecast accuracy assessment is qualified, and:

[0078] (5) If μ ultra-short_i <0.85, ultra-short-term forecast accuracy μ ultra-short_i If the assessment is unsuccessful, proceed to step S703.

[0079] (6) If μ ultra-short_i ≥0.85, ultra-short-term forecast accuracy μ ultra-short_i If the assessment is passed, proceed to step S704.

[0080] Step S7: Set the current time T i Daily average prediction accuracy μ short_iUltra-short-term daily average forecast accuracy and μ ultra-short_i Compared with the assessment standards, after the "photovoltaic-storage coordinated operation" is put into operation, the energy storage system is used to improve the prediction accuracy.

[0081] Step S701: If the day-ahead forecast accuracy and ultra-short-term forecast accuracy of the photovoltaic power station are both unqualified, put the energy storage management system into operation and coordinate the control system to control the energy storage system to carry out charging and discharging operations.

[0082] The charging and discharging power of the energy storage system at the next moment is P. Ei+1 The total power output of the photovoltaic power station is P. Ni+1 as follows:

[0083] P Ni+1 =P si+1 +P Ei+1

[0084] The accuracy of the total station-wide day-ahead forecast for the next moment is μ. short_i+1 The accuracy of ultra-short-term prediction is μ ultra-short_i+1 The next moment, the overall site cost M ai+1 The following is a breakdown of the overall assessment cost, which includes the assessment amount M for the accuracy of photovoltaic power plant forecasts. p Energy storage operating life depreciation cost M E :

[0085]

[0086] λ p λ is the predicted cost coefficient for photovoltaic power plants. E P is the cost factor for energy storage operation life depreciation. n Let t be the total installed capacity of the photovoltaic power station, α be the assessment hours, and C be the assessment management coefficient. 全站 This is the highest approved on-grid electricity price for the entire photovoltaic power station.

[0087] The optimal P is obtained through a minimization optimization algorithm. Ei+1_best To reduce the overall site assessment cost M ai+1 lowest.

[0088] At the next moment T i+1 The command is issued to control the energy storage system to output power P. Ei+1_best .

[0089] Step S702: If the day-ahead forecast accuracy of the photovoltaic power station is unqualified, but the ultra-short-term forecast accuracy is qualified, put the energy storage management system into operation and coordinate the control system to control the energy storage system to perform charging and discharging operations.

[0090] The charging and discharging power of the energy storage system at the next moment is P. Ei+1 The total power output of the photovoltaic power station is P. Ni+1 as follows:

[0091] P Ni+1 =P si+1 +P Ei+1

[0092] The accuracy of the total station-wide day-ahead forecast for the next moment is μ. short_i+1 The accuracy of ultra-short-term prediction is μ ultra-short_i+1 The next moment, the overall site cost M ai+1 The comprehensive assessment cost includes the assessment amount M for the accuracy of photovoltaic power plant forecasts, as follows. p Energy storage operating life depreciation cost M E :

[0093] M ai+1 =λ p (0.8-μ short_i+1 )P n tαC 全站 +λ E M Ei+1

[0094] Where, λ p λ is the predicted cost coefficient for photovoltaic power plants. E P is the cost factor for energy storage operation life depreciation. n Let t be the total installed capacity of the photovoltaic power station, α be the assessment hours, and C be the assessment management coefficient. 全站 This is the highest approved on-grid electricity price for the entire photovoltaic power station.

[0095] The optimal P is obtained through a minimization optimization algorithm. Ei+1_best To reduce the overall site assessment cost M ai+1 lowest.

[0096] At the next moment T i+1 The command is issued to control the energy storage system to output power P. Ei+1_best .

[0097] Step S703: If the day-ahead forecast accuracy of the photovoltaic power station is qualified, but the ultra-short-term forecast accuracy is unqualified, the energy storage management system is put into operation, and the coordination control system controls the energy storage system to carry out charging and discharging operations.

[0098] The charging and discharging power of the energy storage system at the next moment is P. Ei+1 The total power output of the photovoltaic power station is P. Ni+1 as follows:

[0099] P Ni+1 =P si+1 +P Ei+1

[0100] The accuracy of the total station-wide day-ahead forecast for the next moment is μ. short_i+1 The accuracy of ultra-short-term prediction is μ ultra-short_i+1The next moment, the overall site cost M ai+1 as follows:

[0101] M ai+1 =λ p (0.85-μ ultra-short_i+1 )P n tαC 全站 +λ E M Ei+1

[0102] Where, λ p λ is the predicted cost coefficient for photovoltaic power plants. E P is the cost factor for energy storage operation life depreciation. n Let t be the total installed capacity of the photovoltaic power station, α be the assessment hours, and C be the assessment management coefficient. 全站 This is the highest approved on-grid electricity price for the entire photovoltaic power station.

[0103] The optimal P is obtained through a minimization optimization algorithm. Ei+1_best To reduce the overall site assessment cost M ai+1 lowest.

[0104] At the next moment T i+1 The command is issued to control the energy storage system to output power P. Ei+1_best .

[0105] Step S704: When the day-ahead forecast accuracy and ultra-short-term forecast accuracy of the photovoltaic power station are both qualified, switch the energy storage management system to standby mode.

[0106] The beneficial effects of the method for improving the output prediction accuracy of photovoltaic power plants by utilizing energy storage in Embodiment 1 of the present invention are as follows: the energy storage system is controlled according to the prediction accuracy. When the prediction accuracy meets the standard, the energy storage system is in standby mode. Only when the prediction accuracy does not meet the standard will the energy storage system perform charging and discharging operations, which can reduce the charging and discharging frequency of the energy storage system while ensuring the prediction accuracy. Relevant data is collected in real time, and the working status of the energy storage system changes in a timely manner according to whether the prediction accuracy is qualified.

[0107] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Those skilled in the art should understand that the present invention includes, but is not limited to, the content described in the above specific embodiments. Any modifications that do not depart from the functional and structural principles of the present invention will be included within the scope of the claims.

Claims

1. A method for improving the prediction accuracy of a photovoltaic power station output by using energy storage, characterized in that: The method for improving the prediction accuracy of photovoltaic power station output by using energy storage comprises the following steps: Obtaining the output prediction curve and the current total power of the photovoltaic power station, and comparing to obtain the prediction accuracy; Comparing the prediction accuracy with the preset accuracy threshold to determine whether the prediction accuracy meets the standard; When the prediction accuracy meets the standard, the energy storage system is in standby state; when the prediction accuracy does not meet the standard, the energy storage system is controlled to perform charging and discharging operation; Step S4: Set the current time... Daily average forecast accuracy Ultra-short-term daily average forecast accuracy Compared with the evaluation criteria, the system enters different states depending on the prediction accuracy. PR0, PR 1. PR 2 , PR 3. Four different working conditions; PR0 Under normal working condition, photovoltaic power station generates electricity freely, and energy storage system is in standby state. Under the working condition, the coordinated control system controls the energy storage system to perform charging and discharging operation, specifically, the next time energy storage system charging and discharging power is set as , the next time the whole station day-ahead prediction accuracy is , the next time the whole station comprehensive assessment cost is , according to the photovoltaic power station prediction accuracy assessment amount , energy storage operation life loss cost , the whole station comprehensive assessment cost is obtained : Wherein, is the cost coefficient of the photovoltaic power station prediction, is the cost coefficient of the energy storage operation life loss, is the total installed capacity of the photovoltaic power station, t is the number of hours, is the examination management coefficient, is the highest approved on-grid electricity price of the photovoltaic power station. The optimal solution is obtained by minimizing the optimization algorithm So that the overall cost of the station comprehensive examination The lowest Under the working condition, the coordinated control system controls the energy storage system to perform charging and discharging operation, specifically, the next time energy storage system charging and discharging power is set as , the next time the whole station day-ahead prediction accuracy is , the next time the whole station comprehensive assessment cost is , according to the photovoltaic power station prediction accuracy assessment amount , energy storage operation life loss cost , the whole station comprehensive assessment cost is obtained : The optimal solution is obtained by minimizing the optimization algorithm So that the overall cost of the station comprehensive examination The lowest Under the working condition, the coordinated control system controls the energy storage system to perform charging and discharging operation, specifically, the next time energy storage system charging and discharging power is set as , the next time the whole station day-ahead prediction accuracy is , the ultra-short-term prediction accuracy is , and the next time the whole station comprehensive assessment cost is as follows: The optimal solution is obtained by minimizing the optimization algorithm , so that the overall cost of the station comprehensive examination is the lowest.

2. The method for improving the prediction accuracy of a photovoltaic power station output by using energy storage according to claim 1, characterized in that: The output prediction curve comprises a 24-hour-ahead output prediction curve and a 4-hour-ahead output prediction curve.

3. The method for improving the prediction accuracy of a photovoltaic power station output by using energy storage according to claim 2, characterized in that: The method comprises the following steps: Step S1, acquiring AC side power of photovoltaic array inverter , total power of photovoltaic power station , day-ahead 24h output prediction curve of photovoltaic power station , ultra-short-term 4h output prediction curve of photovoltaic power station , AC side power of energy storage PCS ; Step S2: compare the current time Total power of the photovoltaic power station With the photovoltaic power station day-ahead 24h output prediction curve The current time Real-time photovoltaic power station day-ahead daily prediction accuracy : ； Step S3: comparing the current time Total power of the photovoltaic power plant with the photovoltaic power plant ultra-short-term 4h output prediction curve to obtain the current time Real-time photovoltaic power plant ultra-short-term daily average prediction accuracy : 。 4. The method for improving the prediction accuracy of a photovoltaic power station output by using energy storage according to claim 3, characterized in that: In step S4, the assessment standard of the 24-hour-ahead daily average prediction accuracy is 0.8, and the assessment standard of the 4-hour-ahead daily average prediction accuracy is 0.

85.

5. The method for improving the prediction accuracy of a photovoltaic power station output by using energy storage according to claim 4, characterized in that: Step S4 It comprises: Step S401: when , the photovoltaic power station is in a predicted accuracy up-to-standard working condition, the photovoltaic power station carries out free power generation, the energy storage system is in a standby state, and the coordinated control is in a working condition; Step S402: When , the day-ahead prediction accuracy of the photovoltaic power station is unqualified, the ultra-short-term prediction accuracy is qualified, the energy storage system is in an operating state, and the integrated photovoltaic power station and energy storage system enter operation mode; Step S403: When , the photovoltaic power station day-ahead prediction accuracy is qualified, the ultra-short-term prediction accuracy is unqualified, the energy storage system is in the running state, and the integrated photovoltaic power station and energy storage system enters the working condition; Step S404: When , the day-ahead prediction accuracy and the ultra-short-term prediction accuracy of the photovoltaic power station are both unqualified, the energy storage system is in an operating state, and the integrated photovoltaic power station and energy storage system enters operation mode.

6. The method for improving the prediction accuracy of a photovoltaic power station output by using energy storage according to claim 5, characterized in that: When the 24-hour-ahead prediction accuracy and the 4-hour-ahead prediction accuracy of the photovoltaic power station are both qualified, the energy storage energy management system is put into standby state; the energy storage unit of the photovoltaic power station shares a box-type step-up transformer with the photovoltaic array, the photovoltaic power generation unit is connected to the low-voltage side of the box-type step-up transformer through a DC / AC inverter, the energy storage battery unit is connected to the low-voltage side of the box-type step-up transformer through a PCS, and the photovoltaic power generation unit and the energy storage battery unit jointly form a photovoltaic energy storage unit.

7. A photovoltaic power plant, characterized by: The method for improving the prediction accuracy of photovoltaic power station output by using energy storage comprises the following steps:

Images