Scheduling instruction optimization method and system for new energy power station

Abstract

The invention discloses a scheduling instruction optimization method and system for a new energy power generation station, and the method is characterized in that the method comprises the following steps: collecting historical operation data and historical scheduling instructions of the new energy power generation station, constructing a superior scheduling instruction prediction model of the new energy power generation station, based on the current operation state of the new energy station, a scheduling instruction set accepted by the station in the next 24 hours is extracted; collecting the standby power consumption of energy storage equipment in the new energy power generation station, constructing a superior scheduling instruction flexibility evaluation model, and calculating a flexibility index in the flexibility evaluation model through the standby power consumption of the energy storage equipment and the predicted power of the new energy power generation station; and substituting the scheduling instruction set and the flexibility index into a pre-constructed power prediction correction model to obtain generated power prediction data of the new energy station, and reporting the generated power prediction data as a scheduling instruction optimization suggestion to a new energy scheduling system.

Description

Technical Field

[0001] The present invention relates to the field of power systems, and more specifically, to a method and system for optimizing dispatching instructions for a new energy power generation station. Background Art

[0002] With the rapid development of new energy technologies, the proportion of renewable energy sources such as wind and solar power in the power system is gradually increasing. However, due to the volatility and uncertainty of these new energy sources, traditional power dispatch methods are unable to effectively cope with the complex and ever-changing changes in power demand, resulting in insufficient flexibility and stability in the dispatch system. Dispatching strategies for new energy power plants need to fully consider meteorological changes and their impact on power generation to improve energy efficiency and reduce operating costs.

[0003] Existing scheduling methods typically rely on static, single-timescale dispatch instructions, lacking comprehensive evaluation and dynamic adjustment mechanisms across different timescales, making them difficult to meet actual scheduling needs. Therefore, a method that can evaluate dispatch instructions across multiple timescales is urgently needed to optimize scheduling strategies and improve the operational efficiency and economic benefits of renewable energy power plants. Summary of the Invention

[0004] To address the deficiencies in the prior art, the present invention provides a method and system for optimizing dispatching instructions for renewable energy power generation stations. The method obtains the operating constraints of renewable energy stations by analyzing the feasible intervals for power dispatching at renewable energy stations and the flexible adjustment intervals for energy storage equipment, and provides dispatching recommendations on the station side using an economic optimal model.

[0005] The present invention adopts the following technical solutions.

[0006] The first aspect of the present invention relates to a method for optimizing dispatching instructions of a new energy power generation station, the method comprising the following steps: collecting historical operating data and historical dispatching instructions of the new energy power generation station, constructing a superior dispatching instruction prediction model for the new energy power generation station, and extracting a dispatching instruction set to be received by the station within the next twenty-four hours based on the current operating status of the new energy power generation station; collecting the standby power of the energy storage equipment in the new energy power generation station and constructing a superior dispatching instruction flexibility evaluation model, calculating the flexibility index in the flexibility evaluation model through the standby power of the energy storage equipment and the predicted power of the new energy power generation station; substituting the dispatching instruction set and the flexibility index into a pre-constructed power prediction correction model to obtain the power generation prediction data of the new energy station, and reporting it to the new energy dispatching system as a dispatching instruction optimization suggestion.

[0007] Preferably, historical operating data and historical dispatching instructions of the new energy power station are collected to construct a superior dispatching instruction prediction model for the new energy power station, including: using historical operating data and historical dispatching instructions to construct a joint probability distribution function between the operating status of the new energy power station and the dispatching instructions; using a Markov chain model to construct the state transition probability between different dispatching instructions to obtain the superior dispatching instruction prediction model of the new energy power station.

[0008] Preferably, historical operating data and historical dispatching instructions are used to construct a joint probability distribution function between the operating status of the new energy power station and the dispatching instructions, including: the operating status of the new energy power station at least includes the weather and load corresponding to the new energy power station; and the Bayesian method is used to solve the joint probability distribution between the operating status of the new energy power station and the dispatching instructions.

[0009] Preferably, the dispatch instruction set to be received by the station within the next twenty-four hours is extracted based on the current operating status of the new energy station, including: obtaining the predicted weather and predicted load, and the current operating status of the new energy station, and inputting them into the superior dispatch instruction prediction model of the new energy power generation station, so as to obtain the state transition implemented by the new energy power generation station based on the current operating status through a joint probability distribution function; the state transition is stored as a dispatch instruction set, and the dispatch instruction set includes one or more dispatch instructions whose transfer probability exceeds a preset threshold, and each dispatch instruction in the dispatch instruction set corresponds to a transfer probability starting from the current operating status.

[0010] Preferably, the backup power of the energy storage equipment in the new energy power generation station is collected and a flexibility evaluation model of the superior dispatch instruction is constructed, including: the flexibility evaluation model includes the energy storage SOC adjustment margin and the adjustable margin flexibility duration.

[0011] Preferably, the scheduling instruction set and flexibility index are substituted into a pre-built power prediction correction model to obtain the power generation prediction data of the new energy station, including: the power prediction correction model takes maximizing the comprehensive economic benefits as the objective function, and uses the value range of the flexibility index and the transfer probability of each scheduling instruction in the scheduling instruction set as constraints to implement the solution; the input of the power prediction correction model is the real-time power of the new energy power generation station and the real-time SOC of the energy storage device; the output of the power prediction correction model is the predicted power of the new energy power generation station and the SOC of the energy storage device.

[0012] Preferably, the solution is implemented with the value range of the flexibility index and the transition probability of each scheduling instruction in the scheduling instruction set as constraints, including: the transition probability of any scheduling instruction meets:

[0013] P(D t+1 )≥α

[0014] Where D t+1 is a scheduling instruction in the scheduling instruction set of the next moment predicted at time t to be scheduled, or any one of the multiple scheduling instructions,

[0015] P(D t+1 ) is the scheduling instruction D t+1 The probability of

[0016] α is the preset threshold of the transition probability;

[0017] The transition probability of any scheduling instruction under the adjustment of the value range of the flexibility index meets the following conditions:

[0018] P(flex)≥β

[0019] Where flex is the scheduling instruction D t+1 All dispatch instructions that can be covered by the flexibility of energy storage equipment;

[0020] And there are:

[0021] flex∈ψ([μ-σ,μ+σ])

[0022] Where ψ(·) is the scheduling instruction extraction function, which is the set of one or more scheduling instructions used when the predicted adjustment power falls into a known interval;

[0023] In the interval [μ-σ,μ+σ], μ is the scheduling instruction D t+1 The actual adjustment power, [-σ,σ] is the energy storage under the influence of flexibility for the dispatch instruction D t+1 The maximum offset range that can be provided.

[0024] The second aspect of the present invention relates to a dispatching instruction optimization system for a new energy power generation station; the system is implemented using a dispatching instruction optimization method for a new energy power generation station in the first aspect of the present invention; the system includes a construction module, a collection module and a reporting module; wherein the construction module is used to collect historical operating data and historical dispatching instructions of the new energy power generation station, construct a superior dispatching instruction prediction model for the new energy power generation station, and extract the dispatching instruction set received by the station in the next twenty-four hours based on the current operating status of the new energy station; the collection module is used to collect the standby power of the energy storage equipment in the new energy power generation station and construct a superior dispatching instruction flexibility evaluation model, and calculate the flexibility index in the flexibility evaluation model through the standby power of the energy storage equipment and the predicted power of the new energy power generation station; the reporting module is used to substitute the dispatching instruction set and the flexibility index into a pre-constructed power prediction correction model to obtain the power generation power prediction data of the new energy station, and report it to the new energy dispatching system as a dispatching instruction optimization suggestion.

[0025] A third aspect of the present invention relates to a terminal, comprising a processor and a storage medium; the storage medium is used to store instructions; and the processor is used to operate according to the instructions to execute the steps of the method described in the first aspect of the present invention.

[0026] A fourth aspect of the present invention relates to a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the method described in the first aspect of the present invention.

[0027] The beneficial effect of the present invention is that, compared with the prior art, the method and system for optimizing the dispatching instructions of a new energy power generation station in the present invention obtains the operating constraints of the new energy station by analyzing the feasible interval of power dispatching of the new energy station and the flexible adjustment interval of the energy storage equipment, and uses the economic optimal model to provide dispatching suggestions on the station side.

[0028] The beneficial effects of the present invention also include:

[0029] This method evaluates and provides feedback on instructions issued by the superior dispatch system in real time, ensuring the rationality and effectiveness of the dispatch instructions. By establishing an evaluation model and combining meteorological forecast data with the power plant's operating status, the method optimizes the dispatch strategy of energy storage equipment in real time, improving the overall economic benefits and operational efficiency of the power plant. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] Figure 1 The figure is a flow chart of a method for optimizing dispatching instructions of a new energy power station according to the present invention. DETAILED DESCRIPTION

[0031] In order to make the purpose, technical solutions and advantages of the present invention clearer and more accurate, the technical solutions of the present invention are described in detail below through multiple specific embodiments. The embodiments used in the present invention are only used to explain the present invention and are not intended to limit the content of the present invention.

[0032] During operation, renewable energy power plants often receive control instructions from their superior dispatching systems. These instructions aim to coordinate system-wide power generation and maintain power balance across different timescales. These dispatch instructions play a decisive role in the overall operation of power plants and have a profound impact on their economic performance within power market conditions. As the rate of renewable energy grid integration increases, renewable energy plants face increasing demands for dispatch accuracy during operation. Failure to do so will incur penalties, reducing their economic efficiency.

[0033] PV plants with energy storage have greater flexibility and regulatory resources than renewable energy plants without energy storage. This not only helps meet planned power generation assessment requirements but also improves overall operational economics through energy storage systems. Traditional approaches typically focus solely on the overall power response of energy plants after dispatch instructions are issued to meet the dispatch system's requirements for the PV plant's power generation plan. However, PV plants provide dispatching references to the dispatch system by reporting predicted power data. Power forecast information is provided by the plants themselves, and this forecast data can also be used as part of the plants' own flexibility and regulatory resources, a point that is often overlooked in existing research.

[0034] Figure 1 This is a flow chart of a method for optimizing dispatching instructions for a new energy power station according to the present invention. Figure 1 As shown, the first aspect of the present invention relates to a method for optimizing scheduling instructions of a new energy power generation station, and the method includes steps 1 to 3.

[0035] Step 1: Collect historical operating data and historical dispatching instructions of renewable energy power stations, build a superior dispatching instruction prediction model for renewable energy power stations, and extract the dispatching instruction set received by the stations in the next 24 hours based on the current operating status of the renewable energy stations.

[0036] Currently, renewable energy stations must fully respond to their flexible resources to meet dispatch requirements. Therefore, accurately evaluating dispatch requirements plays a crucial role in the economic benefits of renewable energy station power generation. This method specifically involves three phases: predicting dispatch requirements based on historical data, evaluating dispatch strategies based on current forecast information, and revising power forecasts to optimize the economic efficiency of renewable energy stations.

[0037] Because the decision-making basis for superior dispatch control instructions depends on the overall operating status of the system and the overall power generation status of different power plants, and the power generation and load power of new energy power plants and loads have certain periodicity, it is possible to combine a series of external data such as meteorological data and historical data with internal data to make certain predictions. To optimize their own operating economics, new energy manufacturers can combine the above data to provide site-side recommendations for dispatch instructions issued by the centralized forecasting and dispatching system of new energy sites. The present invention predicts dispatch instructions that may be received in the next 24 hours.

[0038] Preferably, historical operating data and historical dispatch instructions of a new energy power station are collected to construct a higher-level dispatch instruction prediction model for the new energy power station, including: using historical operating data and historical dispatch instructions to construct a joint probability distribution function between the operating state of the new energy power station and the dispatch instructions; using a Markov chain model to construct the state transition probability between different dispatch instructions to obtain the higher-level dispatch instruction prediction model for the new energy power station. The method of constructing a joint probability distribution function between the operating state of the new energy power station and the dispatch instructions using historical operating data and historical dispatch instructions includes: the operating state of the new energy power station at least includes the weather and load corresponding to the new energy power station; and using a Bayesian method to solve the joint probability distribution between the operating state of the new energy power station and the dispatch instructions.

[0039] In step 1, historical statistical analysis is performed and a histogram is used to analyze the correlation distribution between scheduling instructions and operating status. A joint probability distribution model of scheduling instructions and operating status is constructed, and the formula is as follows:

[0040] P(D t ∣S t )=Histogram(D t ,S t )

[0041] D t is the historical dispatching instruction of the new energy station at time t, S t For multiple variables, such as weather, load, and operating power.

[0042] This method can capture the distribution characteristics of scheduling instructions under different operating conditions, thus providing a basis for subsequent prediction models. t ,S t ) can be further optimized using Bayesian methods. Bayesian networks can be used to represent the conditional dependencies between dispatch instructions and multiple variables, such as weather and load. In one embodiment, the multiple variables may also include the operating power of renewable energy power stations.

[0043] Preferably, the dispatch instruction set to be received by the station within the next twenty-four hours is extracted based on the current operating status of the new energy station, including: obtaining the predicted weather and predicted load, and the current operating status of the new energy station, and inputting them into the superior dispatch instruction prediction model of the new energy power generation station, so as to obtain the state transition implemented by the new energy power generation station based on the current operating status through a joint probability distribution function; the state transition is stored as a dispatch instruction set, and the dispatch instruction set includes one or more dispatch instructions whose transfer probability exceeds a preset threshold, and each dispatch instruction in the dispatch instruction set corresponds to a transfer probability starting from the current operating status.

[0044] Using the Markov chain model, the state transition matrix of the scheduling instruction is constructed to predict the changes in the scheduling instruction in the next 24 hours. The state transition probability is defined as P(D t+1 ∣D t ), the formula is as follows:

[0045]

[0046] Where D t+1 is the scheduling instruction at the next moment, P(D t+1 ∣D t ) describes the state transition probability between two dispatch instructions under different charges at the station. Based on this method, the present invention can obtain the set of all possible dispatch methods at the next moment, that is, the set of dispatch instructions accepted by the station in the next 24 hours.

[0047] Step 2: Collect the reserve power of energy storage equipment in the new energy power generation station and construct a flexibility evaluation model for the superior dispatch instruction. Calculate the flexibility index in the flexibility evaluation model based on the reserve power of the energy storage equipment and the predicted power of the new energy power generation station.

[0048] The backup power of energy storage equipment in new energy power generation stations is collected and a flexibility evaluation model for superior dispatch instructions is constructed, including: the flexibility evaluation model includes the energy storage SOC adjustment margin and the adjustable margin flexibility duration.

[0049] When executing dispatch instructions from higher authorities, new energy stations meet the power generation requirements of these instructions by coordinating energy storage with the power of renewable energy generation. To characterize the regulatory capabilities within a station, this paper proposes establishing a flexibility index for new energy stations that include energy storage systems. This index comprehensively considers the uncertainty of renewable energy generation, the predicted value of dispatch instructions issued by the dispatch system, and the operating status and capacity of local adjustable resources to assess the impact of dispatch instructions on the operational flexibility of renewable energy power stations within the next 24 hours.

[0050] The energy storage SOC regulation margin indicator represents the expected value of the energy storage reserve power integrated over time in a random scenario over the next 24 hours. For the entire 24 hours, the energy storage reserve power integral value indicator is calculated as follows:

[0051]

[0052] Where, is the expectation, which is used to calculate the integral value of the energy storage SOC reserve power in each 24 hours. SOC(t) is the energy storage reserve power at time t.

[0053] The measurement method of time t is to add 1 to the time value for each hour, and the time within each hour is represented by a continuous real number in the interval [0,1].

[0054] If we consider the accumulated energy storage reserve power within 24 hours from time 0 to time T, we have:

[0055]

[0056] These indicators assess the energy storage device's ability to provide both energy and power support over the next 24 hours. A higher value indicates greater support capacity and greater ability to compensate for random disturbances, indicating greater flexibility.

[0057] The duration of adjustable margin flexibility represents the maximum operating time T that can meet the dispatch instruction without increasing the assessment cost, based on the current energy storage status and capacity conditions, within a 90% confidence interval, taking into account the probability of error in the power generation forecast of the new energy generation components. max The calculation formula is as follows:

[0058] T max =y(Ω t )

[0059] Where y(·) is a function related to the set Ω t It is related to the number of all elements in ;

[0060] Set Ω t How to obtain it:

[0061] Ω t ={t|P(|SOC(t)-SOC pred (t)|≤ΔSOC)≥0.90}

[0062] Where, SOC pred (t) is the energy storage reserve power predicted by the model in step 3,

[0063] ΔSOC is the allowable energy storage state deviation threshold,

[0064] If t is a discrete quantity, the above formula is used. If t is a continuous quantity, the maximum runtime can be obtained using piecewise integration, using a similar calculation method.

[0065] This indicator assesses the maximum operating time that can be achieved within a certain period of time after a dispatch order is issued without increasing assessment costs. A higher value indicates that the new energy station is more resistant to external disturbances and the randomness of dispatch orders, and is more flexible.

[0066] Through the above two indicators, the present invention proposes a method for comprehensively evaluating the flexibility of new energy stations, providing a theoretical basis for the scheduling optimization of new energy stations in complex operating environments.

[0067] Step 3: Substitute the dispatch instruction set and flexibility index into the pre-built power forecast correction model to obtain the power generation forecast data of the new energy station, and report it to the new energy dispatch system as a dispatch instruction optimization suggestion.

[0068] To optimize ultra-short-term power forecast data, this invention uses renewable energy sites as the primary resource and accurately predicts the actual power generated by renewable energy power generation components locally, covering both short-term and ultra-short-term power forecast data. Based on this data and dispatch instruction predictions, the overall operational performance of the power plant, including economic indicators and comprehensive lifespan indicators, is evaluated over multiple future timescales.

[0069] Preferably, the scheduling instruction set and flexibility index are substituted into a pre-built power prediction correction model to obtain the power generation prediction data of the new energy station, including: the power prediction correction model takes maximizing the comprehensive economic benefits as the objective function, and uses the value range of the flexibility index and the transfer probability of each scheduling instruction in the scheduling instruction set as constraints to implement the solution; the input of the power prediction correction model is the real-time power of the new energy power generation station and the real-time SOC of the energy storage device; the output of the power prediction correction model is the predicted power of the new energy power generation station and the SOC of the energy storage device.

[0070] By establishing a probabilistic model and combining it with the scheduling of the energy storage system, not only can the matching of power generation and weather forecast data be achieved, but the flexibility of the station's own operation can also be improved by modifying the weather forecast data, thereby achieving higher comprehensive economic benefits. Therefore, given the known actual photovoltaic power forecast data for a period of time in the future and the prediction results of the superior scheduling instructions, the present invention proposes to optimize the power forecast data of renewable energy power generation based on the regulation capacity of the local energy storage system, and to achieve an improvement in the comprehensive benefits of the station while comprehensively considering the assessment costs and the comprehensive benefits of the station. The key lies in optimizing the overall forecast data of the photovoltaic station, and therefore establishing a corresponding optimization model.

[0071] The objective function is defined as maximizing the comprehensive economic benefit E, taking into account power generation revenue, operation and maintenance costs, and penalty terms. The solution to the objective function is:

[0072]

[0073] The model uses discrete time t = 1 to 24 as the scheduling time. t =P t π t is the power generation revenue at the dispatch time, P t is the power generation of the new energy station, π t is the electricity price. oper (S t ) is the equipment operation and maintenance cost,

[0074] It represents the penalty term caused by the deviation between the actual power and the predicted power, and λ is the assessment penalty coefficient. The predicted power of the new energy station solved by the model, The energy storage SOC to be solved is related to the energy storage power at two adjacent moments. The difference between the predicted power and the real-time power of the new energy station is usually the dispatch instruction.

[0075] The constraints of the optimization model must comprehensively consider the predicted probability constraints and their confidence intervals, the energy storage system's operational constraints, and the site's ramp rate constraints. The transition probabilities and flexibility intervals obtained in steps 1 and 2 can be used to constrain the parameters to be solved, thereby finding the optimal solution to the objective function.

[0076] The power generation constraint is implemented as:

[0077] P t =P RE,t +P ESS,t

[0078] P RE,t The power generated by the new energy power generation unit is usually the same as the predicted power to be solved Equal or similar. ESS,t The power generated by the energy storage is usually The difference is related.

[0079] Preferably, the solution is implemented with the value range of the flexibility index and the transition probability of each scheduling instruction in the scheduling instruction set as constraints, including: the transition probability of any scheduling instruction meets:

[0080] P(D t+1 )≥α

[0081] Where D t+1 is a scheduling instruction in the next scheduling instruction set predicted at time t to be scheduled, or any one of the multiple scheduling instructions, P(D t+1 ) is the scheduling instruction D t+1 The probability of transition is α, and α is the preset threshold of the transition probability; the transition probability of any scheduling instruction under the adjustment of the value range of the flexibility index meets the following conditions:

[0082] P(flex)≥β

[0083] Where flex is the scheduling instruction D t+1 All dispatch instructions that can be covered by the flexibility of energy storage equipment; and there are:

[0084] flex∈ψ([μ-σ,μ+σ])

[0085] Where ψ(·) is the scheduling instruction extraction function, which is a set of one or more scheduling instructions used when the predicted adjustment power falls into a known interval; in the interval [μ-σ,μ+σ], μ is the scheduling instruction D t+1 The actual adjustment power, [-σ,σ] is the energy storage under the influence of flexibility for the dispatch instruction D t+1 The maximum offset range that can be provided.

[0086] As mentioned above, the probability constraint is used to ensure that the scheduling instructions meet the conditions of the probability distribution. In step 1, according to the Markov chain, the scheduling instruction D t+1 There are various possible power adjustment methods, and the transition probability of each adjustment method must be greater than or equal to α. The transition probability threshold can be obtained by solving the state transition probability in step 1. Some scheduling methods with higher probabilities are included in the scheduling instruction set, while those with lower probabilities are discarded. Confidence interval constraints can be used to set the confidence interval of scheduling instructions, ensuring that the instructions are within a certain range.

[0087] Energy storage system operation constraints are used to ensure the safe operation of the energy storage system based on the minimum and maximum operating states of the energy storage devices:

[0088] S min ≤S t ≤S max

[0089] The forecast data of power generation capacity of new energy power generation units is revised to:

[0090]

[0091] is the predicted amount of power generated by the generator set, is the correction amount.

[0092] To validate the method presented in this paper, a case study was conducted using a photovoltaic power station with energy storage. This station has a total installed capacity of 50MW and a 10MWh energy storage system. Located in an area with complex meteorological conditions, the power generation fluctuates significantly. The station needs to execute dispatch instructions issued by the superior dispatch system within the next 24 hours and improve overall operational economic efficiency by optimizing the dispatching strategy.

[0093] First, a new energy power generation prediction model is trained based on meteorological data, power generation data, and dispatch instruction data from the past year to obtain short-term and ultra-short-term power forecast data for the next 24 hours. At the same time, based on the current operating status and historical dispatch instructions, the dispatch instructions that may be issued by the superior dispatch system within the next 24 hours are predicted. Subsequently, the regulation capability of the energy storage system within the next 24 hours is evaluated in real time based on the predicted new energy power generation and dispatch instructions. The probability index of the energy storage SOC regulation margin and the duration index of the adjustable margin flexibility are evaluated to quantify the flexibility of the energy storage system in responding to random disturbances and changes in dispatch instructions. Finally, based on the actual power generation power of the monitoring station, the status of the energy storage system, and the execution of dispatch instructions. According to real-time data, the prediction model and dispatch strategy are dynamically adjusted to maximize the total economic benefits of the new energy station.

[0094] Through the above steps, the photovoltaic power station achieved significant optimized scheduling results over the next 24 hours. Specific calculation results show that the power station's comprehensive operating costs decreased by 15.2%, power generation revenue increased by 11.9%, and assessment costs decreased by 27.64%. The optimized scheduling strategy significantly improved the power station's economic efficiency and scheduling flexibility, effectively enhancing the overall economic benefits of the new energy station. Table 1 shows the benefits of the scheduling method of the present invention.

[0095]

[0096] Table 1 Income Statement

[0097] In summary, the present invention first predicts the dispatching instructions issued by the superior dispatching system by predicting the power generation of renewable energy and the current operating status; secondly, it combines the short-term and ultra-short-term power forecast data and the dispatching instruction forecast results to optimize the energy storage system and photovoltaic power regulation within the station to achieve the best overall operating economy; finally, it optimizes the forecast results of renewable energy power generation to improve the overall operational flexibility and economic benefits of renewable energy stations containing energy storage. The present invention can also establish a multi-time scale evaluation model, combine meteorological forecast data and power station operating status, and optimize the energy storage equipment dispatching strategy in real time to ensure the rationality and effectiveness of the dispatching instructions and improve the comprehensive economic benefits and operating efficiency of the new energy power station.

[0098] The second aspect of the present invention relates to a dispatching instruction optimization system for a new energy power generation station; the system is implemented using a dispatching instruction optimization method for a new energy power generation station in the first aspect of the present invention; the system includes a construction module, a collection module and a reporting module; wherein the construction module is used to collect historical operating data and historical dispatching instructions of the new energy power generation station, construct a superior dispatching instruction prediction model for the new energy power generation station, and extract the dispatching instruction set received by the station in the next twenty-four hours based on the current operating status of the new energy station; the collection module is used to collect the standby power of the energy storage equipment in the new energy power generation station and construct a superior dispatching instruction flexibility evaluation model, and calculate the flexibility index in the flexibility evaluation model through the standby power of the energy storage equipment and the predicted power of the new energy power generation station; the reporting module is used to substitute the dispatching instruction set and the flexibility index into a pre-constructed power prediction correction model to obtain the power generation power prediction data of the new energy station, and report it to the new energy dispatching system as a dispatching instruction optimization suggestion.

[0099] A third aspect of the present invention relates to a terminal, comprising a processor and a storage medium; the storage medium is used to store instructions; and the processor is used to operate according to the instructions to execute the steps of the method described in the first aspect of the present invention.

[0100] A fourth aspect of the present invention relates to a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the method described in the first aspect of the present invention.

[0101] Finally, it should be noted that the above embodiments are intended only to illustrate the technical solutions of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art will appreciate that the technical solutions of the present invention still include modifications or equivalent substitutions that may be made to the specific embodiments of the present invention. Any modifications or equivalent substitutions that do not depart from the spirit and scope of the present invention are intended to be covered by the claims of the present invention.

Claims

1. A method for optimizing dispatching instructions of a renewable energy power station, characterized in that: The method comprises the following steps: Collect historical operation data and historical dispatching instructions of renewable energy power stations, build a prediction model for superior dispatching instructions of renewable energy power stations, and extract the dispatching instruction set accepted by the stations within the next 24 hours based on the current operation status of renewable energy stations; Collect the reserve power of energy storage equipment in renewable energy power generation stations and construct a flexibility evaluation model for superior dispatching instructions. Calculate the flexibility index in the flexibility evaluation model through the reserve power of energy storage equipment and the predicted power of renewable energy power generation stations. The dispatch instruction set and flexibility indicators are substituted into the pre-built power forecast correction model to obtain the power generation forecast data of the new energy site, which is then reported to the new energy dispatch system as a dispatch instruction optimization suggestion.

2. The method for optimizing dispatching instructions of a new energy power generation station according to claim 1, characterized in that: The collecting of historical operation data and historical dispatching instructions of the renewable energy power generation station and building of a superior dispatching instruction prediction model of the renewable energy power generation station include: The joint probability distribution function between the operating status of the renewable energy power station and the dispatching instructions is constructed by using the historical operating data and historical dispatching instructions; The state transition probability between different dispatching instructions is constructed using the Markov chain model to obtain the upper-level dispatching instruction prediction model of the new energy power station.

3. The method for optimizing dispatching instructions of a new energy power generation station according to claim 2 is characterized in that: The method of constructing a joint probability distribution function between the operating status of a new energy power generation station and the dispatching instructions by using historical operating data and historical dispatching instructions includes: The operating status of the new energy power generation station at least includes the weather and load corresponding to the new energy power generation station; The Bayesian method is used to solve the joint probability distribution between the operating status of renewable energy power plants and dispatching instructions.

4. The method for optimizing dispatching instructions of a new energy power generation station according to claim 3 is characterized in that: The method of extracting the dispatching instruction set received by the station within the next 24 hours based on the current operating status of the new energy station includes: Obtain the forecast weather, forecast load and current operating status of the new energy power station, and input them into the superior dispatch instruction prediction model of the new energy power station, so as to obtain the state transition implemented by the new energy power station based on the current operating status through the joint probability distribution function; The state transition is stored as a scheduling instruction set, which includes one or more scheduling instructions whose transition probabilities exceed a preset threshold, and each scheduling instruction in the scheduling instruction set corresponds to a transition probability starting from the current running state.

5. The method for optimizing dispatching instructions of a new energy power generation station according to claim 4 is characterized in that: The method of collecting the reserve power of energy storage equipment in the new energy power generation station and constructing a flexibility evaluation model for the superior dispatching instruction includes: The flexibility evaluation model includes the energy storage SOC adjustment margin and the adjustable margin flexibility duration.

6. The method for optimizing dispatching instructions of a new energy power generation station according to claim 5 is characterized in that: Substituting the dispatch instruction set and the flexibility index into the pre-built power prediction correction model to obtain the power generation prediction data of the new energy station includes: The power prediction correction model takes the maximization of comprehensive economic benefits as the objective function, and takes the value range of the flexibility index and the transfer probability of each scheduling instruction in the scheduling instruction set as constraints to implement the solution; The input of the power prediction correction model is the real-time power of the new energy power generation station and the real-time SOC of the energy storage device; The output of the power prediction correction model is the predicted power of the new energy power generation station and the SOC of the energy storage device.

7. The method for optimizing dispatching instructions of a new energy power station according to claim 6 is characterized in that: The solving is performed with the value range of the flexibility index and the transition probability of each scheduling instruction in the scheduling instruction set as constraints, including: The transition probability of any of the scheduling instructions meets the following conditions: P(D t+1 )≥α Where D t+1 is a scheduling instruction in the scheduling instruction set of the next moment predicted at the time t to be scheduled, or any one of the multiple scheduling instructions, P(D t+1 ) is the scheduling instruction D t+1 The probability of α is the preset threshold of transition probability; The transfer probability of any scheduling instruction under the adjustment of the value range of the flexibility index meets the following conditions: P(flex)≥β Where flex is the scheduling instruction D t+1 All dispatch instructions that can be covered by the flexibility of energy storage equipment; And there are: flex∈ψ([μ-σ,μ+σ]) Where ψ(·) is the scheduling instruction extraction function, which is a set of one or more scheduling instructions used when the predicted adjustment power falls into a known interval; In the interval [μ-σ,μ+σ], μ is the scheduling instruction D t+1 The actual adjustment power, [-σ,σ] is the energy storage under the influence of flexibility for the dispatch instruction D t+1 The maximum offset range that can be provided.

8. A dispatching instruction optimization system for a new energy power station; characterized in that: The system is implemented by using a method for optimizing dispatching instructions of a new energy power station as described in any one of claims 1 to 7; The system includes a construction module, a collection module and a reporting module; wherein, The construction module is used to collect historical operation data and historical dispatching instructions of the new energy power generation station, build a superior dispatching instruction prediction model for the new energy power generation station, and extract the dispatching instruction set accepted by the station within the next 24 hours based on the current operation status of the new energy power generation station; The acquisition module is used to collect the reserve power of the energy storage equipment in the new energy power generation station and construct a flexibility evaluation model for the superior dispatching instruction, and calculate the flexibility index in the flexibility evaluation model through the reserve power of the energy storage equipment and the predicted power of the new energy power generation station; The reporting module is used to substitute the dispatching instruction set and flexibility index into the pre-built power prediction correction model to obtain the power generation prediction data of the new energy site, and report it to the new energy dispatching system as a dispatching instruction optimization suggestion.

9. A terminal comprising a processor and a storage medium; characterized in that: The storage medium is used to store instructions; The processor is configured to operate according to the instructions to execute the steps of the method according to any one of claims 1-7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

Images