Control method and device of optical storage power station, electronic equipment and storage medium

Abstract

The application provides a control method and device of a light storage power station, electronic equipment and a storage medium. The method comprises the following steps: correcting an initial daily photovoltaic power prediction sequence according to a decay correction coefficient and a pollution correction coefficient to obtain a target daily photovoltaic power prediction sequence; taking an energy storage scheduling reference as a constraint condition, generating an intra-day charging and discharging power instruction sequence of an energy storage system within a first time length in the future according to the target daily photovoltaic power prediction sequence; and dynamically distributing total power instructions in the intra-day charging and discharging power instruction sequence to each energy storage unit according to the energy storage scheduling reference and real-time operation state data of each energy storage unit to obtain real-time power set values of each energy storage unit within the first time length in the future, so as to control each energy storage unit in the target light storage power station. Through the application, long-term trends and short-term fluctuations can be considered, and the problems of overcharging and overdischarging of part of the energy storage units and accelerated decay of the service life can be avoided.

Description

Technical Field

[0001] This invention relates to the field of photovoltaic-storage synergistic control technology, and more specifically, to control methods, devices, electronic equipment, and storage media for photovoltaic-storage power plants. Background Technology

[0002] A photovoltaic (PV) power station is a distributed energy system that combines photovoltaic power generation units with electrochemical energy storage. Traditional control methods for PV power stations often employ single-time-scale models, making it difficult to account for both long-term trends and short-term fluctuations. Moreover, traditional control methods frequently use simple proportional allocation or threshold switching strategies, which can easily lead to overcharging and over-discharging of some energy storage units, resulting in accelerated lifespan degradation. Summary of the Invention

[0003] In view of this, the purpose of this application is to provide a control method, device, electronic equipment, and storage medium for a photovoltaic-storage power station, which can combine the day-ahead photovoltaic power prediction subsequence and the intraday photovoltaic power prediction sequence, taking into account both long-term trends and short-term fluctuations. Simultaneously, based on the energy storage scheduling benchmark and the real-time operating status data of each energy storage unit, dynamic power allocation is performed for each energy storage unit to avoid problems such as overcharging and over-discharging of some energy storage units and accelerated lifespan degradation.

[0004] In a first aspect, embodiments of this application provide a control method for a photovoltaic-storage power station, the method comprising: The method involves acquiring the initial intraday photovoltaic power prediction sequence, attenuation correction coefficient, and pollution correction coefficient of the photovoltaic array in the target photovoltaic-storage power station within the first future time period; as well as the energy storage scheduling benchmark of the energy storage system in the target photovoltaic-storage power station within the first future time period, and the real-time operating status data of each energy storage unit; the energy storage scheduling benchmark is determined based on the intraday photovoltaic power prediction sub-sequence of the photovoltaic array within the first future time period. The attenuation correction coefficient is used to characterize the degree of power output capacity attenuation caused by the increase in the service life of the photovoltaic array; the contamination correction coefficient is used to characterize the degree of power output capacity attenuation caused by surface contamination of the photovoltaic array. The initial intraday photovoltaic power prediction sequence is corrected based on the attenuation correction coefficient and the pollution correction coefficient to obtain the target intraday photovoltaic power prediction sequence. Using the energy storage scheduling benchmark as a constraint, and based on the target intraday photovoltaic power prediction sequence, the intraday charging and discharging power command sequence of the energy storage system within the first future time period is generated; Based on the energy storage scheduling benchmark and the real-time operating status data of each energy storage unit, the total power command in the intraday charging and discharging power command sequence is dynamically allocated to each energy storage unit to obtain the real-time power setting value of each energy storage unit operating in the first future time period, so as to control each energy storage unit in the target photovoltaic-energy storage power station.

[0005] In one possible implementation, the energy storage scheduling benchmark is obtained through the following steps: Obtain the day-ahead photovoltaic power prediction subsequence of the photovoltaic array within the first future time period; With the goal of minimizing the operating cost of the target photovoltaic-storage power station, an energy storage scheduling benchmark is generated based on the day-ahead photovoltaic power prediction subsequence.

[0006] In one possible implementation, before obtaining the day-ahead photovoltaic power prediction subsequence, the day-ahead photovoltaic power prediction subsequence is generated during the target day-ahead scheduling period according to the following steps: Obtain the target operating data of the photovoltaic array corresponding to the target day-ahead scheduling period; the target day-ahead scheduling period refers to the day-ahead scheduling period for predicting the power data of the photovoltaic array within the first future time period. The target operating data is input into the day-ahead prediction model to obtain the initial day-ahead photovoltaic power prediction full sequence corresponding to the photovoltaic array; According to a preset day-ahead correction period, the initial day-ahead photovoltaic power forecast sequence is corrected based on a meteorological correction factor to obtain the target day-ahead photovoltaic power forecast sequence; wherein, the meteorological correction factor is determined by the meteorological deviation vector between the actual meteorological data and the predicted meteorological data within the corresponding day-ahead correction period; the target day-ahead photovoltaic power forecast sequence includes the day-ahead photovoltaic power forecast subsequence.

[0007] In one possible implementation, the step of dynamically allocating the total power command in the intraday charge / discharge power command sequence to each energy storage unit based on the real-time operating status data of each energy storage unit according to the energy storage scheduling benchmark, to obtain the real-time power setpoint for each energy storage unit, includes: Based on the energy storage scheduling benchmark and the real-time operating status data of each energy storage unit, the target allocation weight corresponding to each energy storage unit is determined. According to the target allocation weight corresponding to each energy storage unit, the total power command in the intraday charging and discharging power command sequence is dynamically allocated to each energy storage unit to obtain the real-time power setting value of each energy storage unit.

[0008] In one possible implementation, determining the target allocation weight for each energy storage unit based on the energy storage scheduling benchmark and the real-time operating status data of each energy storage unit includes: For each energy storage unit, the initial allocation weight of the energy storage unit under each preset operating state parameter is determined based on the energy storage scheduling benchmark and the corresponding values ​​of each preset operating state parameter in the real-time operating state data of the energy storage unit. Based on the initial allocation weights of each energy storage unit under each preset operating state parameter, the target allocation weights corresponding to each energy storage unit are determined.

[0009] In one possible implementation, the initial allocation weight of any energy storage unit under various preset operating state parameters is determined according to the following steps: Based on the state of charge (SOC) reference value of the energy storage unit at the current moment in the energy storage scheduling benchmark and the real-time SOC in the real-time operating status data of the energy storage unit, the initial allocation weight of the energy storage unit under the SOC parameters is calculated. Based on the real-time health status value in the real-time operating status data, the initial allocation weight of the energy storage unit under the health status parameters is calculated.

[0010] In one possible implementation, determining the target allocation weight for each energy storage unit based on its initial allocation weight under various preset operating state parameters includes: Substituting the initial allocation weights of each energy storage unit under each preset operating state parameter into the following formula, we obtain the target allocation weights for each energy storage unit: ; in, Assign weights to the target corresponding to the i-th energy storage unit; Assign initial weights to the i-th energy storage unit under its state of charge parameters; Assign initial weights to the i-th energy storage unit under the health state parameters; The real-time maximum allowable charge and discharge power is the real-time operating status data of the i-th energy storage unit. The number of energy storage units; Assign initial weights to the j-th energy storage unit under its state of charge parameters; Assign initial weights to the j-th energy storage unit under the health state parameters; This represents the real-time maximum allowable charge / discharge power in the real-time operating status data of the j-th energy storage unit.

[0011] Secondly, embodiments of this application also provide a control device for a photovoltaic-storage power station, the device comprising: The acquisition module is used to acquire the initial intraday photovoltaic power prediction sequence, attenuation correction coefficient, and pollution correction coefficient of the photovoltaic array in the target photovoltaic-storage power station within the first future time period; as well as the energy storage scheduling benchmark of the energy storage system in the target photovoltaic-storage power station within the first future time period, and the real-time operating status data of each energy storage unit; the energy storage scheduling benchmark is determined based on the intraday photovoltaic power prediction sub-sequence of the photovoltaic array within the first future time period. The attenuation correction coefficient is used to characterize the degree of power output capacity attenuation caused by the increase in the service life of the photovoltaic array; the contamination correction coefficient is used to characterize the degree of power output capacity attenuation caused by surface contamination of the photovoltaic array. The correction module is used to correct the initial intraday photovoltaic power prediction sequence according to the attenuation correction coefficient and the pollution correction coefficient to obtain the target intraday photovoltaic power prediction sequence. The generation module is used to generate a daily charging and discharging power command sequence of the energy storage system within the first future time period, based on the target daily photovoltaic power prediction sequence and using the energy storage scheduling benchmark as a constraint. The allocation module is used to dynamically allocate the total power command in the intraday charge and discharge power command sequence to each energy storage unit according to the energy storage scheduling benchmark and the real-time operating status data of each energy storage unit, so as to obtain the real-time power setting value of each energy storage unit operating in the first future time period, so as to control each energy storage unit in the target photovoltaic-storage power station.

[0012] Thirdly, embodiments of this application also provide an electronic device, including: a processor, a storage medium, and a bus, wherein the storage medium stores machine-readable instructions executable by the processor, and when the electronic device is running, the processor communicates with the storage medium via the bus, and the processor executes the machine-readable instructions to perform the steps of the control method for a photovoltaic-storage power station as described in any of the first aspects.

[0013] Fourthly, embodiments of this application also provide a computer-readable storage medium storing a computer program, which, when executed by a processor, performs the steps of the control method for a photovoltaic-storage power station as described in any of the first aspects.

[0014] This application provides a control method, device, electronic equipment, and storage medium for a photovoltaic-storage power station. The method includes: correcting an initial intraday photovoltaic power prediction sequence based on attenuation and pollution correction coefficients to obtain a target intraday photovoltaic power prediction sequence; using an energy storage scheduling benchmark as a constraint, generating an intraday charging and discharging power command sequence for the energy storage system within a future first time period based on the target intraday photovoltaic power prediction sequence; and dynamically allocating the total power command in the intraday charging and discharging power command sequence to each energy storage unit based on the energy storage scheduling benchmark and the real-time operating status data of each energy storage unit, thereby obtaining a real-time power setpoint for each energy storage unit's operation within the future first time period, to control each energy storage unit in the target photovoltaic-storage power station. This application combines the day-ahead photovoltaic power prediction sub-sequence and the intraday photovoltaic power prediction sequence, taking into account both long-term trends and short-term fluctuations. Simultaneously, it dynamically allocates power to each energy storage unit based on the energy storage scheduling benchmark and the real-time operating status data of each energy storage unit, avoiding overcharging and over-discharging of some energy storage units and accelerated lifespan degradation. Attached Figure Description

[0015] To more clearly illustrate the technical solutions of the embodiments of this application, the accompanying drawings used in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of this application and should not be regarded as a limitation of the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0016] Figure 1 A flowchart of a control method for a photovoltaic energy storage power station provided in an embodiment of this application is shown; Figure 2 A flowchart illustrating the power allocation method of the energy storage unit provided in an embodiment of this application is shown. Figure 3 This paper shows a schematic diagram of the structure of a control device for a photovoltaic energy storage power station according to an embodiment of this application; Figure 4 A schematic diagram of the structure of an electronic device provided in an embodiment of this application is shown. Detailed Implementation

[0017] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. It should be understood that the accompanying drawings in this application are for illustrative and descriptive purposes only and are not intended to limit the scope of protection of this application. Furthermore, it should be understood that the schematic drawings are not drawn to scale. The flowcharts used in this application illustrate operations implemented according to some embodiments of this application. It should be understood that the operations in the flowcharts may not be implemented in sequence, and steps without logical contextual relationships may be reversed or implemented simultaneously. In addition, those skilled in the art, guided by the content of this application, may add one or more other operations to the flowcharts, or remove one or more operations from the flowcharts.

[0018] Furthermore, the described embodiments are merely some, not all, of the embodiments of this application. The components of the embodiments of this application described and illustrated herein can typically be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of this application provided in the accompanying drawings is not intended to limit the scope of the claimed application, but merely to illustrate selected embodiments of the application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without inventive effort are within the scope of protection of this application.

[0019] To enable those skilled in the art to utilize the content of this application, and in conjunction with the specific application scenario of "photovoltaic-storage synergistic regulation technology," the following embodiments are provided. For those skilled in the art, the general principles defined herein can be applied to other embodiments and application scenarios without departing from the spirit and scope of this application. Although this application primarily describes the field of "photovoltaic-storage synergistic regulation technology," it should be understood that this is merely an exemplary embodiment.

[0020] It should be noted that the term "comprising" will be used in the embodiments of this application to indicate the presence of the features declared thereafter, but does not exclude the addition of other features.

[0021] The following is a detailed description of a control method for a photovoltaic energy storage power station provided in the embodiments of this application.

[0022] Reference Figure 1 The diagram shown is a schematic flowchart of a control method for a photovoltaic-storage power station provided in an embodiment of this application. The exemplary steps of this embodiment are described below: S101. Obtain the initial intraday photovoltaic power prediction sequence, attenuation correction coefficient, and pollution correction coefficient of the photovoltaic array in the target photovoltaic-storage power station within the first time period in the future; and the energy storage scheduling benchmark of the energy storage system in the target photovoltaic-storage power station within the first time period in the future, and the real-time operating status data of each energy storage unit.

[0023] In this embodiment, the target photovoltaic-storage power station includes a photovoltaic array and an energy storage system. The photovoltaic array converts solar energy into electrical energy and transmits the surplus electricity to the energy storage system. The energy storage system includes energy storage units for storing electrical energy and reducing solar curtailment. During the operation of the target photovoltaic-storage power station, the power station is scheduled daily on a minute-by-minute basis.

[0024] Furthermore, the initial intraday photovoltaic power forecast sequence is the predicted photovoltaic power of the photovoltaic array within the first time period of the future (e.g., 15 minutes to 4 hours from the current moment), achieving short-term prediction of future photovoltaic power. Specifically, the initial intraday photovoltaic power forecast sequence is generated through the following steps: Step 1: Obtain the photovoltaic power data of the photovoltaic array within the current day's scheduling time and the predicted meteorological data for the first time in the future (such as irradiance, temperature, and photovoltaic array backsheet temperature).

[0025] In this embodiment of the application, taking a daily scheduling duration of 3 minutes as an example, if the current time is 3:05, then the photovoltaic power data includes the photovoltaic power of the photovoltaic array from 3:03 to 3:05.

[0026] Step 2: Input the photovoltaic power data within the current intraday scheduling period and the predicted meteorological data for the first future period into the pre-built intraday prediction model to obtain the initial intraday photovoltaic power prediction sequence.

[0027] In this embodiment of the application, the intraday prediction model is constructed using a hybrid model based on convolutional neural networks and long short-term memory networks. It is trained using first historical photovoltaic operation data (photovoltaic power data and predicted meteorological data) and is used to make short-term predictions of the power of the photovoltaic array in minutes.

[0028] In addition, the attenuation correction factor is used to characterize the degree of power output capacity degradation of a photovoltaic array due to its increasing service life. This attenuation correction factor is not fixed but dynamically calculated. The process is as follows: The system continuously collects historical meteorological data (such as actual output power and power generation) and historical photovoltaic power data of the photovoltaic array under the same historical meteorological data within an annual cycle (such as one or two years). By comparing the deviation changes between multiple historical photovoltaic power data under the same historical meteorological data within an annual cycle, regression analysis and other algorithms are used to quantify the percentage decrease in output power of the photovoltaic array within the annual cycle. The attenuation correction factor is determined based on the percentage decrease in output power of the photovoltaic array within the annual cycle.

[0029] In this embodiment, the corresponding attenuation correction coefficient can be retrieved from a pre-established attenuation correction coefficient table based on the percentage decrease in output power of the photovoltaic array over an annual period, and automatically updated according to a preset period (e.g., monthly / quarterly). Alternatively, a corresponding machine learning model can be trained to evaluate the attenuation correction coefficient; the method of determination is not limited.

[0030] The attenuation correction coefficient table corresponds to the service life of the storage photovoltaic array, the percentage decrease in the output power of the photovoltaic array within the annual cycle, and the attenuation correction coefficient.

[0031] In addition, the contamination correction coefficient is used to characterize the degree of power output degradation of the photovoltaic array due to surface contamination. It is determined in real time based on the amount of dust accumulation on the surface of the photovoltaic array (which can be obtained by establishing a relationship model between the amount of dust accumulation and the service life of the photovoltaic array; the model will vary depending on the regional climate) or the light transmittance (monitored using optical sensors). It can be implemented by looking up tables or by machine learning.

[0032] Furthermore, the energy storage dispatch benchmark is determined based on the day-ahead photovoltaic power prediction sub-sequence of the photovoltaic array within the next first time period, including the charge and discharge power benchmark values ​​of each energy storage unit at each time period within the next first time period and the state of charge benchmark trajectory (including the state of charge benchmark at each moment within the next first time period). Specifically, the energy storage dispatch benchmark is obtained through the following steps: Step 1: Obtain the day-ahead photovoltaic power prediction subsequence for the photovoltaic array in the first time period in the future.

[0033] In this embodiment, the day-ahead photovoltaic power prediction subsequence is the photovoltaic power time-series data of the photovoltaic array within a first future time period, predicted at the day-ahead scheduling level. Before obtaining this day-ahead photovoltaic power prediction subsequence, the subsequence needs to be generated within the target day-ahead scheduling period. The specific process is as follows: A. Obtain the target operation data of the photovoltaic array corresponding to the target day-ahead scheduling cycle; the target day-ahead scheduling cycle refers to the day-ahead scheduling cycle for predicting the power data of the photovoltaic array in the first hour of the future.

[0034] In this embodiment, the target operating data includes predicted meteorological data (such as irradiance, temperature, and photovoltaic array backsheet temperature) for the next day after the target scheduling cycle and photovoltaic power data of the photovoltaic array within the daytime scheduling duration before the target scheduling cycle. Taking a daytime scheduling duration of 18 hours as an example, if the scheduling time corresponding to the current target scheduling cycle is 21:00, then the photovoltaic power data refers to the photovoltaic power of the photovoltaic array from 3:00 to 21:00.

[0035] B. Input the target operating data into the day-ahead prediction model to obtain the initial day-ahead photovoltaic power prediction full sequence corresponding to the photovoltaic array.

[0036] In this embodiment, the day-ahead prediction model is constructed based on a time-series network with gated recurrent units and an attention mechanism, and is trained using second historical photovoltaic (PV) operation data. It is used for long-term prediction of the PV array's power on a daily basis. The initial day-ahead PV power prediction sequence includes the PV power for the next 24 hours following the scheduling time of the day-ahead scheduling cycle.

[0037] C. According to the preset day-ahead correction period (e.g., 1 hour), the initial day-ahead photovoltaic power prediction full sequence is corrected based on the meteorological correction factor to obtain the target day-ahead photovoltaic power prediction full sequence; the target day-ahead photovoltaic power prediction full sequence contains the day-ahead photovoltaic power prediction subsequence.

[0038] In this embodiment, the meteorological correction factor and the initial day-ahead photovoltaic power forecast full sequence are input into a pre-constructed correction model to obtain the target day-ahead photovoltaic power forecast full sequence. The correction model is pre-trained using historical day-ahead photovoltaic power forecast full sequences, corresponding meteorological correction factors, and actual photovoltaic power full sequences.

[0039] The meteorological correction factor is determined by the meteorological deviation vector between the actual and predicted meteorological data within the corresponding day-ahead correction period. A positive meteorological deviation vector with a larger absolute value indicates that the actual weather is better than the forecast, and a meteorological correction factor greater than 1 results in a larger upward adjustment to the entire initial day-ahead photovoltaic power forecast series. Conversely, a negative meteorological deviation vector with a larger absolute value indicates that the actual weather is worse than the forecast, and a meteorological correction factor less than 1 results in a larger downward adjustment to the entire initial day-ahead photovoltaic power forecast series. A pre-established mapping table between meteorological deviation vectors and meteorological correction factors can be used to determine the meteorological correction factor used to correct the entire initial day-ahead photovoltaic power forecast series.

[0040] Step 2: With the goal of minimizing the operating cost of the target photovoltaic-storage power station, generate an energy storage dispatch benchmark based on the day-ahead photovoltaic power prediction subsequence.

[0041] In this embodiment, the day-ahead optimization layer in the adaptive collaborative optimization model aims to minimize the daily operating cost of the target photovoltaic-storage power station. The mixed-integer linear programming algorithm is used in combination with the day-ahead photovoltaic power prediction subsequence to solve the problem and determine the energy storage scheduling benchmark. The solution constraints include upper and lower limits of the state of charge of each energy storage unit, upper and lower limits of charging and discharging power, mutual exclusion constraints of charging and discharging states, and power balance constraints.

[0042] Here, the embodiments of this application pre-establish an adaptive collaborative optimization model with dual time scales, including a day-ahead optimization layer and an intraday rolling optimization layer. The day-ahead optimization layer determines the day-ahead charge and discharge power reference values ​​and state-of-charge reference trajectories of each energy storage unit for each time period of the next day based on long-term prediction results (day-ahead photovoltaic power prediction subsequence).

[0043] S102. Based on the attenuation correction coefficient and the pollution correction coefficient, the initial intraday photovoltaic power prediction sequence is corrected to obtain the target intraday photovoltaic power prediction sequence.

[0044] In this embodiment, the initial photovoltaic power at each moment within the initial intraday photovoltaic power prediction sequence is multiplied by the attenuation correction coefficient and the contamination correction coefficient to calculate the target photovoltaic power at the corresponding moment within the target intraday photovoltaic power prediction sequence. The attenuation correction coefficient characterizes the power output capacity attenuation characteristic of the photovoltaic array as its operating years increase; the smaller the attenuation correction coefficient, the more significant the power output attenuation caused by aging of the photovoltaic array, and the greater the correction magnitude for the initial intraday photovoltaic power. The contamination correction coefficient characterizes the power output attenuation characteristic caused by dust and stains on the photovoltaic array panel; the smaller the contamination correction coefficient, the more significant the power output attenuation caused by surface contamination of the photovoltaic array, and the greater the correction magnitude for the initial intraday photovoltaic power.

[0045] S103. Using the energy storage dispatch benchmark as a constraint, generate the daily charging and discharging power command sequence of the energy storage system in the first time period in the future based on the target daily photovoltaic power prediction sequence.

[0046] In this embodiment, the intraday charging and discharging power command timing includes independent intraday charging and discharging power control commands for each energy storage unit; this step is solved by the intraday rolling optimization layer of the adaptive collaborative optimization model. The optimization objective of this layer is to output the optimal real-time scheduling strategy, which must simultaneously achieve the dual effects of smoothing photovoltaic power fluctuations and suppressing the degradation of energy storage equipment lifespan. The intraday rolling optimization layer is built based on a model predictive control architecture and constructs a multi-objective optimization function containing two optimization indicators: one is to minimize the system tracking error, and the other is to minimize the lifespan loss of each energy storage unit; the optimization solution process is strictly constrained by the energy storage scheduling benchmark, and a quadratic programming optimization problem is solved by rolling time domain, outputting the complete intraday charging and discharging power command timing in real time for each time period.

[0047] Minimizing system tracking error means that the actual charging and discharging power and real-time state of charge of the energy storage system must be synchronously aligned with the charging and discharging power benchmark value and state of charge benchmark trajectory defined by the energy storage scheduling benchmark. The deviation between the actual operating values ​​and the benchmark trajectory / benchmark value is used as the tracking error. During the optimization process, this deviation is continuously reduced so that the energy storage operating conditions accurately follow the preset scheduling benchmark, ensuring that the combined output of photovoltaic and energy storage is stable and controllable.

[0048] S104. Based on the energy storage scheduling benchmark and the real-time operating status data of each energy storage unit, the total power command in the daily charging and discharging power command sequence is dynamically allocated to each energy storage unit to obtain the real-time power setting value of each energy storage unit in the first time period in the future, so as to control each energy storage unit in the target photovoltaic-energy storage power station.

[0049] In the embodiments of this application, reference is made to Figure 2 The diagram shown is a power allocation flowchart of the energy storage unit provided in an embodiment of this application: S201. Based on the energy storage scheduling benchmark and the real-time operating status data of each energy storage unit, determine the target allocation weight corresponding to each energy storage unit.

[0050] In this application embodiment, the target allocation weight corresponding to each energy storage unit is determined through the following process: A. For each energy storage unit, determine the initial allocation weight of the energy storage unit under each preset operating state parameter based on the energy storage scheduling benchmark and the corresponding values ​​of each preset operating state parameter in the real-time operating state data of the energy storage unit.

[0051] In this embodiment, the preset operating state parameters include state of charge parameters and health state parameters. Specifically, the initial allocation weight of any energy storage unit under each preset operating state parameter is determined according to the following steps: (1) Calculate the initial allocation weight of the energy storage unit under the state of charge parameters based on the state of charge reference value of the energy storage unit at the current time in the energy storage scheduling benchmark and the real-time state of charge in the real-time operation status data of the energy storage unit.

[0052] In this application embodiment, the current state of charge reference value and the real-time state of charge in the real-time operating status data are substituted into the following formula to obtain the initial allocation weight of the energy storage unit under the state of charge parameters.

[0053] ; in, Let be the state-of-charge balancing factor (i.e., the initial weighting under the state-of-charge parameters) for the i-th energy storage unit, and exp(·) denotes an exponential function with the natural constant e as the base. Let be the state-of-charge deviation coefficient of the i-th energy storage unit. , This represents the real-time state of charge of the i-th energy storage unit. This represents the reference value of the state of charge (SOC) at the current moment in the SOC reference trajectory. This is the preset equilibrium strength coefficient.

[0054] (2) And calculate the initial allocation weight of the energy storage unit under the health status parameters based on the real-time health status value in the real-time operation status data.

[0055] In this embodiment of the application, the real-time health status value from the real-time operating status data is substituted into the following formula to obtain the initial allocation weight of the energy storage unit under the health status parameters: ; in, The health status weighting factor for the i-th energy storage unit (i.e., the initial weighting under the health status parameters). Let be the real-time health status value of the i-th energy storage unit. The first preset weighting coefficient, This is the second preset weighting coefficient; B. Determine the target allocation weight for each energy storage unit based on the initial allocation weight of each energy storage unit under each preset operating state parameter.

[0056] In this embodiment, the initial allocation weights of each energy storage unit under each preset operating state parameter are substituted into the following formula to obtain the target allocation weights corresponding to each energy storage unit: ; in, Assign weights to the target corresponding to the i-th energy storage unit; Assign initial weights to the i-th energy storage unit under its state of charge parameters; Assign initial weights to the i-th energy storage unit under the health state parameters; The real-time maximum allowable charge and discharge power is the real-time operating status data of the i-th energy storage unit. The number of energy storage units; Assign initial weights to the j-th energy storage unit under its state of charge parameters; Assign initial weights to the j-th energy storage unit under the health state parameters; This represents the real-time maximum allowable charge / discharge power in the real-time operating status data of the j-th energy storage unit.

[0057] S202. According to the target allocation weight corresponding to each energy storage unit, the total power command in the daily charging and discharging power command sequence is dynamically allocated to each energy storage unit to obtain the real-time power setting value of each energy storage unit.

[0058] In this embodiment, firstly, based on the total power demand in the daily charge / discharge power command sequence and combined with the target allocation weight corresponding to each energy storage unit, the initial power allocation value of each energy storage unit is calculated; then, boundary verification is performed on each initial power allocation value to determine whether it exceeds the physical constraints such as power and state of charge of each energy storage unit. If there is a power limit violation, the remaining power or power deficit of the energy storage unit that violates the limit is redistributed to other energy storage units with adjustment capabilities to obtain the actual charge / discharge power allocation value of each energy storage unit; finally, the actual charge / discharge power allocation value is sent to each energy storage unit as a real-time power setting value for execution.

[0059] Furthermore, feedback data after execution is simultaneously collected to drive the online update of the dual correction factor and dynamic weight allocation algorithm parameters. Specifically, the parameter update process includes: updating the dual correction factor online using a recursive least squares method based on the error sequence between actual photovoltaic power and predicted power to correct the power allocation benchmark; and adaptively adjusting the equilibrium intensity coefficient in the dynamic weight allocation algorithm according to the overall dispersion of the energy storage system's state of charge. α When the dispersion of the state of charge exceeds a preset threshold, increase α To enhance the balancing effect of the state of charge (SOC) of each energy storage unit, when the SOC dispersion is lower than a preset threshold, the [measurement / reduction] is achieved. α Prioritize meeting the system's response requirements to total power commands.

[0060] In summary, the control method for the photovoltaic-storage power station provided in this application has the following advantages: (1) In photovoltaic power prediction, existing technologies generally adopt single-timescale models (such as LSTM or ARIMA), and the model parameters are fixed, which cannot adapt to dynamic changes such as meteorological deviations, module degradation, and surface turbidity. The embodiments of this application construct a multi-timescale prediction architecture based on dual correction factors: the day-ahead prediction model uses gated recurrent units and attention mechanisms to capture daily trends, and the intraday prediction model uses convolutional neural networks and long short-term memory networks to capture minute-level fluctuations, and introduces meteorological correction factors (rolling correction based on real-time meteorological deviations), attenuation correction coefficients, and turbidity correction coefficients, while using recursive least squares to update parameters online. This enables the prediction model to respond simultaneously to long-term aging, medium-term meteorological deviations, and short-term cloud changes, and the prediction accuracy is significantly higher than that of existing single models.

[0061] (2) Regarding energy storage power allocation, existing technologies mostly employ proportional allocation, fixed priority allocation, or simple SOC proportional linear control, completely ignoring the differences in the health status of each energy storage unit. This leads to overcharging and over-discharging of some aging units, accelerating the overall system lifespan degradation. The embodiments of this application design a dynamic weight allocation algorithm, creatively multiplying and normalizing the exponential state-of-charge (SOC) balancing factor, the health status weighting factor, and the real-time maximum allowable charge / discharge power to obtain the power allocation weight for each energy storage unit. The SOC balancing factor uses an exponential function, causing the weight of units deviating from the baseline trajectory to adjust non-linearly and rapidly, avoiding the lag of linear balancing. The health status weighting factor automatically reduces the usage intensity of aging units, while healthy units undertake more power tasks. Furthermore, the algorithm also features boundary verification and remaining power redistribution mechanisms, and adaptively adjusts the balancing intensity coefficient based on the overall dispersion of the SOC of the energy storage units during system operation. This multi-factor collaborative allocation strategy achieves the dual objectives of SOC balancing and health status protection, fundamentally solving the lifespan degradation problem caused by unreasonable power allocation in existing technologies.

[0062] (3) Regarding the coordination between day-ahead scheduling and intraday control, existing technologies often treat the two separately: day-ahead scheduling independently generates the next day's plan with economic efficiency as the objective, while intraday control focuses only on real-time power tracking without referencing the day-ahead reference, resulting in a significant deviation from the economically optimal solution in actual operation. The embodiments of this application establish an adaptive collaborative optimization model with dual time scales. The day-ahead optimization layer not only outputs the charging and discharging power reference values ​​of each energy storage unit, but also outputs the complete state of charge reference trajectory. The intraday rolling optimization layer adopts a model predictive control framework with multiple objectives of minimizing system tracking error and minimizing energy storage unit lifetime loss. Soft constraints are introduced into the constraints—the intraday charging and discharging power and state of charge must follow the day-ahead reference value / reference trajectory, and a penalty term will be applied when there is a deviation. This "economic-oriented reference line + dynamic rolling correction" structure enables intraday control to respond quickly to real-time fluctuations in photovoltaics and loads without deviating from the economically optimal trajectory determined by the day-ahead. At the same time, when an energy storage unit fails and exits or reconnects, the day-ahead layer and the intraday layer automatically update the constraints and references, achieving seamless coordination under all operating conditions. Existing technologies cannot simultaneously meet the optimization requirements of long-term economic efficiency and short-term dynamic response, while this invention truly achieves the unity of the two through dual time scales and soft constraint mechanisms.

[0063] Based on the same inventive concept, this application also provides a control device for a photovoltaic power station corresponding to the control method of the photovoltaic power station. Since the principle of the device in this application is similar to the control method of the photovoltaic power station described above in this application, the implementation of the device can refer to the implementation of the method, and the repeated parts will not be described again.

[0064] Reference Figure 3The diagram shown is a schematic of a control device for a photovoltaic-storage power station provided in an embodiment of this application. The device includes: The acquisition module 301 is used to acquire the initial intraday photovoltaic power prediction sequence, attenuation correction coefficient, and pollution correction coefficient of the photovoltaic array in the target photovoltaic-storage power station within the first future time period; as well as the energy storage scheduling benchmark of the energy storage system in the target photovoltaic-storage power station within the first future time period, and the real-time operating status data of each energy storage unit; the energy storage scheduling benchmark is determined based on the intraday photovoltaic power prediction sub-sequence of the photovoltaic array within the first future time period; The attenuation correction coefficient is used to characterize the degree of power output capacity attenuation caused by the increase in the service life of the photovoltaic array; the contamination correction coefficient is used to characterize the degree of power output capacity attenuation caused by surface contamination of the photovoltaic array. The correction module 302 is used to correct the initial intraday photovoltaic power prediction sequence according to the attenuation correction coefficient and the pollution correction coefficient to obtain the target intraday photovoltaic power prediction sequence. The generation module 303 is used to generate the intraday charging and discharging power command sequence of the energy storage system within the first future time period based on the target intraday photovoltaic power prediction sequence, using the energy storage scheduling benchmark as a constraint. The allocation module 304 is used to dynamically allocate the total power command in the intraday charge and discharge power command sequence to each energy storage unit according to the energy storage scheduling benchmark and the real-time operating status data of each energy storage unit, so as to obtain the real-time power setting value of each energy storage unit operating in the first future time period, so as to control each energy storage unit in the target photovoltaic-storage power station.

[0065] In one possible implementation, the acquisition module 301 is specifically used to acquire the energy storage scheduling benchmark through the following steps: acquiring the day-ahead photovoltaic power prediction subsequence of the photovoltaic array within the future first time period; and generating the energy storage scheduling benchmark based on the day-ahead photovoltaic power prediction subsequence with the goal of minimizing the operating cost of the target photovoltaic-storage power station.

[0066] In one possible implementation, the acquisition module 301 is specifically configured to generate the day-ahead photovoltaic power prediction subsequence according to the following steps during the target day-ahead scheduling period before acquiring the day-ahead photovoltaic power prediction subsequence: acquiring the target operating data of the photovoltaic array corresponding to the target day-ahead scheduling period; the target day-ahead scheduling period refers to the day-ahead scheduling period for predicting the power data of the photovoltaic array within the first future time period; inputting the target operating data into the day-ahead prediction model to obtain the initial day-ahead photovoltaic power prediction full sequence corresponding to the photovoltaic array; correcting the initial day-ahead photovoltaic power prediction full sequence based on a meteorological correction factor according to a preset day-ahead correction period to obtain the target day-ahead photovoltaic power prediction full sequence; wherein, the meteorological correction factor is determined by the meteorological deviation vector between the actual meteorological data and the predicted meteorological data within the corresponding day-ahead correction period; the target day-ahead photovoltaic power prediction full sequence includes the day-ahead photovoltaic power prediction subsequence.

[0067] In one possible implementation, the allocation module 304 is specifically used to determine the target allocation weight corresponding to each energy storage unit based on the energy storage scheduling benchmark and the real-time operating status data of each energy storage unit; and to dynamically allocate the total power command in the intraday charging and discharging power command sequence to each energy storage unit according to the target allocation weight corresponding to each energy storage unit, so as to obtain the real-time power setting value of each energy storage unit.

[0068] In one possible implementation, the allocation module 304 is specifically used to determine the initial allocation weight of each energy storage unit under each preset operating state parameter based on the energy storage scheduling benchmark and the corresponding values ​​of each preset operating state parameter in the real-time operating state data of the energy storage unit; and to determine the target allocation weight of each energy storage unit based on the initial allocation weight of each energy storage unit under each preset operating state parameter.

[0069] In one possible implementation, the allocation module 304 is specifically configured to determine the initial allocation weight of any energy storage unit under various preset operating state parameters according to the following steps: calculating the initial allocation weight of the energy storage unit under the state of charge parameters based on the state of charge reference value of the energy storage unit at the current time in the energy storage scheduling benchmark and the real-time state of charge in the real-time operating state data of the energy storage unit; and calculating the initial allocation weight of the energy storage unit under the health state parameters based on the real-time health state value in the real-time operating state data.

[0070] In one possible implementation, the allocation module 304 is specifically used to substitute the initial allocation weights of each energy storage unit under each preset operating state parameter into the following formula to obtain the target allocation weights corresponding to each energy storage unit: ; in, Assign weights to the target corresponding to the i-th energy storage unit; Assign initial weights to the i-th energy storage unit under its state of charge parameters; Assign initial weights to the i-th energy storage unit under the health state parameters; The real-time maximum allowable charge and discharge power is the real-time operating status data of the i-th energy storage unit. The number of energy storage units; Assign initial weights to the j-th energy storage unit under its state of charge parameters; Assign initial weights to the j-th energy storage unit under the health state parameters; This represents the real-time maximum allowable charge / discharge power in the real-time operating status data of the j-th energy storage unit.

[0071] like Figure 4 As shown in the embodiment of this application, an electronic device 400 includes a processor 401, a memory 402, and a bus. The memory 402 stores machine-readable instructions that can be executed by the processor 401. When the electronic device is running, the processor 401 communicates with the memory 402 via the bus, and the processor 401 executes the machine-readable instructions to perform the steps of the control method of the photovoltaic-storage power station described above.

[0072] Specifically, the memory 402 and processor 401 mentioned above can be general-purpose memory and processor, without any specific limitations. When the processor 401 runs the computer program stored in the memory 402, it can execute the control method of the photovoltaic-storage power station mentioned above.

[0073] Corresponding to the control method of the photovoltaic-storage power station described above, this application embodiment also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, performs the steps of the control method of the photovoltaic-storage power station described above.

[0074] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems and devices described above can be referred to the corresponding processes in the method embodiments, and will not be repeated here. In the several embodiments provided in this application, it should be understood that the disclosed systems, devices, and methods can be implemented in other ways. The device embodiments described above are merely illustrative. For example, the division of modules is only a logical functional division, and in actual implementation, there may be other division methods. Furthermore, multiple modules or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the displayed or discussed mutual coupling or direct coupling or communication connection can be through some communication interfaces; the indirect coupling or communication connection of devices or modules can be electrical, mechanical, or other forms.

[0075] The modules described as separate components may or may not be physically separate. The components shown as modules may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0076] In addition, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0077] If the aforementioned functions are implemented as software functional units and sold or used as independent products, they can be stored in a processor-executable, non-volatile, computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, ROM, RAM, magnetic disks, or optical disks.

[0078] The above are merely specific embodiments of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A control method for a photovoltaic-storage power station, characterized in that, The method includes: The method involves acquiring the initial intraday photovoltaic power prediction sequence, attenuation correction coefficient, and pollution correction coefficient of the photovoltaic array in the target photovoltaic-storage power station within the first future time period; as well as the energy storage scheduling benchmark of the energy storage system in the target photovoltaic-storage power station within the first future time period, and the real-time operating status data of each energy storage unit; the energy storage scheduling benchmark is determined based on the intraday photovoltaic power prediction sub-sequence of the photovoltaic array within the first future time period. The attenuation correction coefficient is used to characterize the degree of power output capacity attenuation caused by the increase in the service life of the photovoltaic array; the contamination correction coefficient is used to characterize the degree of power output capacity attenuation caused by surface contamination of the photovoltaic array. The initial intraday photovoltaic power prediction sequence is corrected based on the attenuation correction coefficient and the pollution correction coefficient to obtain the target intraday photovoltaic power prediction sequence. Using the energy storage scheduling benchmark as a constraint, and based on the target intraday photovoltaic power prediction sequence, the intraday charging and discharging power command sequence of the energy storage system within the first future time period is generated; Based on the energy storage scheduling benchmark and the real-time operating status data of each energy storage unit, the total power command in the intraday charging and discharging power command sequence is dynamically allocated to each energy storage unit to obtain the real-time power setting value of each energy storage unit operating in the first future time period, so as to control each energy storage unit in the target photovoltaic-energy storage power station.

2. The control method for a photovoltaic-storage power station according to claim 1, characterized in that, The energy storage scheduling benchmark is obtained through the following steps: Obtain the day-ahead photovoltaic power prediction subsequence of the photovoltaic array within the first future time period; With the goal of minimizing the operating cost of the target photovoltaic-storage power station, an energy storage scheduling benchmark is generated based on the day-ahead photovoltaic power prediction subsequence.

3. The control method for a photovoltaic-storage power station according to claim 2, characterized in that, Before obtaining the day-ahead photovoltaic power prediction subsequence, the day-ahead photovoltaic power prediction subsequence is generated during the target day-ahead scheduling period according to the following steps: Obtain the target operating data of the photovoltaic array corresponding to the target day-ahead scheduling period; the target day-ahead scheduling period refers to the day-ahead scheduling period for predicting the power data of the photovoltaic array within the first future time period. The target operating data is input into the day-ahead prediction model to obtain the initial day-ahead photovoltaic power prediction full sequence corresponding to the photovoltaic array; According to a preset day-ahead correction period, the initial day-ahead photovoltaic power forecast sequence is corrected based on a meteorological correction factor to obtain the target day-ahead photovoltaic power forecast sequence; wherein, the meteorological correction factor is determined by the meteorological deviation vector between the actual meteorological data and the predicted meteorological data within the corresponding day-ahead correction period; the target day-ahead photovoltaic power forecast sequence includes the day-ahead photovoltaic power forecast subsequence.

4. The control method for a photovoltaic-storage power station according to claim 1, characterized in that, The step of dynamically allocating the total power command in the intraday charge / discharge power command sequence to each energy storage unit based on the real-time operating status data of each energy storage unit according to the energy storage scheduling benchmark, to obtain the real-time power setpoint for each energy storage unit, includes: Based on the energy storage scheduling benchmark and the real-time operating status data of each energy storage unit, the target allocation weight corresponding to each energy storage unit is determined. According to the target allocation weight corresponding to each energy storage unit, the total power command in the intraday charging and discharging power command sequence is dynamically allocated to each energy storage unit to obtain the real-time power setting value of each energy storage unit.

5. The control method for a photovoltaic-storage power station according to claim 4, characterized in that, The step of determining the target allocation weight for each energy storage unit based on the energy storage scheduling benchmark and the real-time operating status data of each energy storage unit includes: For each energy storage unit, the initial allocation weight of the energy storage unit under each preset operating state parameter is determined based on the energy storage scheduling benchmark and the corresponding values ​​of each preset operating state parameter in the real-time operating state data of the energy storage unit. Based on the initial allocation weights of each energy storage unit under each preset operating state parameter, the target allocation weights corresponding to each energy storage unit are determined.

6. The control method for a photovoltaic-storage power station according to claim 5, characterized in that, The initial allocation weight of any energy storage unit under various preset operating state parameters is determined according to the following steps: Based on the state of charge (SOC) reference value of the energy storage unit at the current moment in the energy storage scheduling benchmark and the real-time SOC in the real-time operating status data of the energy storage unit, the initial allocation weight of the energy storage unit under the SOC parameters is calculated. Based on the real-time health status value in the real-time operating status data, the initial allocation weight of the energy storage unit under the health status parameters is calculated.

7. The control method for a photovoltaic-storage power station according to claim 6, characterized in that, The step of determining the target allocation weight for each energy storage unit based on the initial allocation weight of each energy storage unit under various preset operating state parameters includes: Substituting the initial allocation weights of each energy storage unit under each preset operating state parameter into the following formula, we obtain the target allocation weights for each energy storage unit: ； in, Assign weights to the target corresponding to the i-th energy storage unit; Assign initial weights to the i-th energy storage unit under its state of charge parameters; Assign initial weights to the i-th energy storage unit under the health state parameters; The real-time maximum allowable charge and discharge power is the real-time operating status data of the i-th energy storage unit. The number of energy storage units; Assign initial weights to the j-th energy storage unit under its state of charge parameters; Assign initial weights to the j-th energy storage unit under the health state parameters; This represents the real-time maximum allowable charge / discharge power in the real-time operating status data of the j-th energy storage unit.

8. A control device for a photovoltaic-storage power station, characterized in that, The device includes: The acquisition module is used to acquire the initial intraday photovoltaic power prediction sequence, attenuation correction coefficient, and pollution correction coefficient of the photovoltaic array in the target photovoltaic-storage power station within the first future time period; as well as the energy storage scheduling benchmark of the energy storage system in the target photovoltaic-storage power station within the first future time period, and the real-time operating status data of each energy storage unit; the energy storage scheduling benchmark is determined based on the intraday photovoltaic power prediction sub-sequence of the photovoltaic array within the first future time period. The attenuation correction coefficient is used to characterize the degree of power output capacity attenuation caused by the increase in the service life of the photovoltaic array; the contamination correction coefficient is used to characterize the degree of power output capacity attenuation caused by surface contamination of the photovoltaic array. The correction module is used to correct the initial intraday photovoltaic power prediction sequence according to the attenuation correction coefficient and the pollution correction coefficient to obtain the target intraday photovoltaic power prediction sequence. The generation module is used to generate a daily charging and discharging power command sequence of the energy storage system within the first future time period, based on the target daily photovoltaic power prediction sequence and using the energy storage scheduling benchmark as a constraint. The allocation module is used to dynamically allocate the total power command in the intraday charge and discharge power command sequence to each energy storage unit according to the energy storage scheduling benchmark and the real-time operating status data of each energy storage unit, so as to obtain the real-time power setting value of each energy storage unit operating in the first future time period, so as to control each energy storage unit in the target photovoltaic-storage power station.

9. An electronic device, characterized in that, include: The device includes a processor, a storage medium, and a bus, wherein the storage medium stores machine-readable instructions executable by the processor, and when the electronic device is in operation, the processor communicates with the storage medium via the bus, and the processor executes the machine-readable instructions to perform the steps of the control method for a photovoltaic-storage power station as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program, which, when executed by a processor, performs the steps of the control method for a photovoltaic-storage power station as described in any one of claims 1 to 7.

Images