Method, device, equipment, storage medium and system for determining power station dispatching strategy

Abstract

The application discloses a kind of method for determining power station scheduling strategy, device, equipment, storage medium and system.The method comprises: obtaining power transaction associated data, and inputting power transaction associated data into preset electricity price prediction model to obtain multiple initial predicted electricity prices, multiple initial predicted electricity prices are optimized using preset electricity price optimization model, to generate multiple target predicted electricity prices of power station, obtain the current state information of power station, and generate multiple target scheduling strategies according to multiple target predicted electricity prices and current state information using preset strategy optimization algorithm;Real electricity price in preset historical time period is used to determine the maximum profit target scheduling strategy from multiple target scheduling strategies, and the maximum profit target scheduling strategy is determined as optimal scheduling strategy.The method improves the income stability of power station, so that power station can more flexibly respond to market fluctuations, maximize power station income.

Description

Technical Field

[0001] This invention relates to the field of power plant dispatching technology, and in particular to a method, apparatus, equipment, storage medium, and system for determining power plant dispatching strategies. Background Technology

[0002] When participating in electricity market transactions, existing energy storage power stations rely on the personal experience of traders and manual operation by dispatchers, specifically including querying public data from the trading center. Subsequently, traders formulate dispatch strategies based on the retrieved data and their personal experience, and then dispatchers manually set dispatch curves to complete the actual dispatch tasks of the power station.

[0003] However, while this traditional process achieves peak-valley arbitrage for energy storage power stations to some extent, its level of automation and intelligence is low. Faced with ever-increasing market data volumes, manual operation is not only inefficient but also increases the risk of decision-making errors. The formulation of power station dispatch strategies relies on the personal experience and expertise of traders. As the amount of data continues to increase and the constraints become more complex, the limitations of manual processing become increasingly apparent, making it difficult to avoid potential decision-making errors that ultimately affect the overall profitability of the power station. Summary of the Invention

[0004] This invention provides a method, apparatus, equipment, storage medium, and system for determining power plant scheduling strategies to address the problem of suboptimal power plant scheduling strategies.

[0005] In a first aspect, the present invention provides a method for determining a power plant dispatching strategy, comprising:

[0006] Acquire electricity transaction related data and input the electricity transaction related data into a preset electricity price prediction model to obtain multiple initial predicted electricity prices, wherein the preset electricity price prediction model includes multiple electricity price prediction sub-models;

[0007] The multiple initial predicted electricity prices are optimized using a preset electricity price optimization model to generate multiple target predicted electricity prices for the power plant.

[0008] The current status information of the power plant is obtained, and multiple target scheduling strategies are generated based on the multiple target predicted electricity prices and the current status information using a preset strategy optimization algorithm.

[0009] Using the real electricity price within a preset historical time period, the maximum profit target scheduling strategy is determined from the multiple target scheduling strategies, and the maximum profit target scheduling strategy is determined as the optimal scheduling strategy.

[0010] Secondly, the present invention provides an apparatus for determining a power plant dispatching strategy, comprising:

[0011] An initial predicted electricity price determination module is used to acquire electricity transaction related data and input the electricity transaction related data into a preset electricity price prediction model to obtain multiple initial predicted electricity prices, wherein the preset electricity price prediction model includes multiple electricity price prediction sub-models;

[0012] The target predicted electricity price determination module is used to optimize the multiple initial predicted electricity prices using a preset electricity price optimization model to generate multiple target predicted electricity prices for the power plant.

[0013] The target scheduling strategy determination module is used to obtain the current status information of the power plant and generate multiple target scheduling strategies based on the multiple target predicted electricity prices and the current status information using a preset strategy optimization algorithm.

[0014] The optimal scheduling strategy determination module is used to determine the maximum profit target scheduling strategy from the multiple target scheduling strategies by using the real electricity price within a preset historical time period, and to determine the maximum profit target scheduling strategy as the optimal scheduling strategy.

[0015] Thirdly, the present invention provides an electronic device comprising:

[0016] At least one processor;

[0017] and memory that is communicatively connected to at least one processor;

[0018] The memory stores a computer program that can be executed by at least one processor, which enables the at least one processor to perform the method for determining the power plant scheduling strategy described in the first aspect above.

[0019] Fourthly, the present invention provides a computer-readable storage medium storing computer instructions for causing a processor to execute the method for determining a power plant scheduling strategy as described in the first aspect.

[0020] Fifthly, the present invention provides a power plant management system, which includes at least a computing platform, an MQTT server, a network gateway, and an energy management system within the power plant;

[0021] The MQTT server is used to enable information exchange between the energy management system on one side of the power plant's internal network gate and the computing platform on the other side of the network gate; the computing platform is capable of executing the method for determining the power plant scheduling strategy described in the first aspect above.

[0022] By adopting the above technical solution, future electricity prices were predicted using electricity trading correlation data. An electricity price optimization model was then used to optimize the predicted electricity price, resulting in a more profitable target predicted electricity price. Finally, a strategy optimization algorithm was used to generate the optimal power plant dispatch strategy based on the target predicted electricity price and the current state information of the power plant. This solution enables power plants to achieve a high degree of fully automated operation when participating in electricity trading, automatically generating accurate electricity price predictions and dispatch strategies. Compared with the traditional method of relying on manual determination of power plant dispatch strategies, this method improves efficiency, significantly reduces labor costs, greatly reduces the subjectivity and uncertainty of human decision-making, thereby improving the stability of power plant revenue and significantly enhancing the power plant's competitiveness in the electricity trading market. This allows power plants to respond more flexibly to market fluctuations, maximize power plant revenue, and promote the progress of intelligent electricity trading.

[0023] It should be understood that the description in this section is not intended to identify key or essential features of the invention, nor is it intended to limit the scope of the invention. Other features of the invention will become readily apparent from the following description. Attached Figure Description

[0024] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0025] Figure 1 This is a flowchart of a method for determining a power plant dispatching strategy according to Embodiment 1 of the present invention;

[0026] Figure 2 This is a flowchart of a method for determining a power plant dispatching strategy according to Embodiment 2 of the present invention;

[0027] Figure 3 This is a schematic diagram of an architecture for generating a power plant scheduling strategy according to Embodiment 2 of the present invention;

[0028] Figure 4 This is a power plant architecture diagram provided according to Embodiment 2 of the present invention;

[0029] Figure 5 This is a schematic diagram of the structure of a device for determining a power plant dispatching strategy according to Embodiment 3 of the present invention;

[0030] Figure 6 This is a schematic diagram of the structure of an electronic device according to Embodiment 4 of the present invention;

[0031] Figure 7This is a schematic diagram of a power plant management system according to Embodiment Six of the present invention. Detailed Implementation

[0032] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0033] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. In the description of this invention, unless otherwise stated, "a plurality of" means two or more. "And / or" describes the relationship between related objects, indicating that three relationships can exist; for example, A and / or B can represent: A alone, A and B simultaneously, and B alone. The character " / " generally indicates that the preceding and following related objects are in an "or" relationship. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or device that includes a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or devices.

[0034] Example 1

[0035] Figure 1 The flowchart of a method for determining a power plant scheduling strategy is provided in Embodiment 1 of the present invention. This embodiment is applicable to the situation of determining a power plant scheduling strategy. The method can be executed by a device for determining a power plant scheduling strategy. The device for determining a power plant scheduling strategy can be implemented in hardware and / or software. The device for determining a power plant scheduling strategy can be configured in an electronic device, which can be composed of two or more physical entities or a single physical entity.

[0036] like Figure 1 As shown in Embodiment 1 of the present invention, a method for determining a power plant dispatching strategy specifically includes the following steps:

[0037] S101. Obtain electricity transaction related data and input the electricity transaction related data into a preset electricity price prediction model to obtain multiple initial predicted electricity prices, wherein the preset electricity price prediction model includes multiple electricity price prediction sub-models.

[0038] In this embodiment, information related to electricity trading, such as weather information and electricity trading information, can be obtained in advance. After inputting the electricity trading-related data into each electricity price prediction sub-model in the preset electricity price prediction model, the corresponding (initial) predicted electricity price can be obtained. The electricity price prediction sub-model can be a machine learning model or similar type of model.

[0039] S102. Optimize the multiple initial predicted electricity prices using a preset electricity price optimization model to generate multiple target predicted electricity prices for the power plant.

[0040] In this embodiment, the preset electricity price optimization model can integrate multiple electricity prices through methods such as weighted averaging to output the optimal (target) predicted electricity price, such as the most economically optimal price. Multiple initial predicted electricity prices can be input into the preset electricity price optimization model, which can then output the target predicted electricity price.

[0041] S103. Obtain the current status information of the power station, and use a preset strategy optimization algorithm to generate multiple target scheduling strategies based on the multiple target predicted electricity prices and the current status information.

[0042] In this embodiment, the current state information of the power plant, such as its current operating status, can be obtained. The preset strategy optimization algorithm can process the obtained multiple target predicted electricity prices and current state information through deep reinforcement learning and other methods to generate a target scheduling strategy for the power plant.

[0043] S104. Using the real electricity price within a preset historical time period, determine the maximum profit target scheduling strategy from the multiple target scheduling strategies, and determine the maximum profit target scheduling strategy as the optimal scheduling strategy.

[0044] Specifically, the actual electricity price within a preset historical time period can be input into the target dispatch strategy, thereby obtaining the profit of the power plant executing each target dispatch strategy. The target dispatch strategy corresponding to the maximum profit can be determined as the optimal dispatch strategy.

[0045] This invention's technical solution utilizes electricity trading data to predict future electricity prices, optimizes these predictions using an electricity price optimization model to obtain a more profitable target predicted price, and finally generates an optimal power plant scheduling strategy based on the target predicted price and the power plant's current state information using a strategy optimization algorithm. This solution enables power plants to achieve a high degree of full automation when participating in electricity trading, automatically generating accurate electricity price predictions and scheduling strategies. Compared to traditional methods that rely on manual determination of power plant scheduling strategies, this improves efficiency, significantly reduces labor costs, greatly reduces the subjectivity and uncertainty of human decision-making, thereby improving the stability of power plant revenue and significantly enhancing the power plant's competitiveness in the electricity trading market. It allows power plants to respond more flexibly to market fluctuations, maximize power plant revenue, and promote the progress of intelligent electricity trading.

[0046] Optionally, the step of optimizing the multiple initial predicted electricity prices using a preset electricity price optimization model to generate multiple target predicted electricity prices for the power plant includes: obtaining a third-party predicted electricity price, and processing the third-party predicted electricity price and the multiple initial predicted electricity prices using the preset electricity price optimization model to obtain multiple target predicted electricity prices for the power plant.

[0047] Specifically, the electricity price predicted by a third party can be obtained in advance, such as the predicted electricity price at time A. A preset electricity price optimization model can be used to process the third-party predicted electricity price at time A and multiple initial predicted electricity prices at time A to obtain multiple target predicted electricity prices for the power plant at time A.

[0048] Optionally, the preset strategy optimization algorithm includes at least two of the following: branch and bound algorithm, genetic algorithm, mixed integer programming algorithm, and deep reinforcement learning algorithm; wherein, generating multiple target scheduling strategies based on the multiple target predicted electricity prices and the current state information using the preset strategy optimization algorithm includes: generating a target scheduling strategy corresponding to each algorithm in the preset strategy optimization algorithm based on the multiple target predicted electricity prices and the current state information.

[0049] Specifically, at least two of the following algorithms can be used: branch and bound algorithm, genetic algorithm, mixed integer programming algorithm, and deep reinforcement learning algorithm, to generate target scheduling strategies based on multiple target electricity price predictions and current state information.

[0050] Example 2

[0051] Figure 2 This is a flowchart of a method for determining a power plant scheduling strategy according to Embodiment 2 of the present invention. The technical solution of the present invention is further optimized based on the above optional technical solutions, and a specific method for determining the power plant scheduling strategy is given.

[0052] Optionally, the above method is applied to a computing platform within a power plant; wherein, obtaining the current status information of the power plant includes: obtaining the current status information through an MQTT server within the power plant; wherein, the MQTT server is used to realize information interaction between a target device on one side of the power grid gateway and the computing platform on the other side of the gateway, and the target device includes an energy management system. The advantage of this setup is that by deploying an MQTT server within the power plant, power plant-level data communication and scheduling strategy generation are achieved, enabling the power plant not only to handle complex external data but also to dynamically adjust forecast results according to changes in the electricity market, ensuring that the generated scheduling strategy can adapt to the real-time market environment.

[0053] Optionally, inputting the electricity transaction-related data into a preset electricity price prediction model to obtain multiple initial predicted electricity prices includes: inputting weather information, information released by the trading center, and electricity market information from the electricity transaction-related data into each electricity price prediction sub-model, so that each electricity price prediction sub-model outputs a corresponding initial predicted electricity price. The advantage of this setup is that it provides a comprehensive prediction of the electricity price on the prediction date through electricity transaction-related data from different dimensions.

[0054] Optionally, the step of processing the third-party predicted electricity price and the plurality of initial predicted electricity prices using a preset electricity price optimization model to obtain multiple target predicted electricity prices for the power plant includes: using a first preset electricity price optimization algorithm in the preset electricity price optimization model to determine a minimum electricity price difference based on the difference between the third-party predicted electricity price and the plurality of initial predicted electricity prices and the corresponding historical real electricity prices, and using the minimum electricity price difference to determine a first target predicted electricity price; using a second preset electricity price optimization algorithm in the preset electricity price optimization model to determine a minimum power plant operating profit difference based on the difference between the power plant operating profit of the third-party predicted electricity price and the power plant operating profit of the plurality of initial predicted electricity prices and the power plant operating profit of the corresponding historical real electricity prices, and using the minimum power plant operating profit difference to determine a second target predicted electricity price; and using a third preset electricity price optimization algorithm in the preset electricity price optimization model to determine a third target predicted electricity price based on the accuracy of the electricity price prediction sub-model and the plurality of initial predicted electricity prices and the third-party predicted electricity price. The advantage of this setup is that it enables multiple rounds of optimization of the predicted electricity price by utilizing the minimum electricity price difference, the minimum power plant operating profit difference, and the accuracy of the electricity price prediction sub-model.

[0055] like Figure 2 As shown in Embodiment 2 of the present invention, a method for determining a power plant dispatching strategy specifically includes the following steps:

[0056] S201. Obtain electricity transaction related data, and input the weather information, transaction center release information and electricity market information in the electricity transaction related data into each electricity price prediction sub-model, so that each electricity price prediction sub-model outputs the corresponding initial predicted electricity price.

[0057] Optionally, the electricity price prediction sub-model in the preset electricity price prediction model includes at least two of the following: gradient boosting tree model, extreme forest model, random forest model, linear regression model, and multilayer perceptron model.

[0058] Specifically, Figure 3 This is a schematic diagram of an architecture for generating power plant scheduling strategies, such as... Figure 3 As shown, weather information, information released by the trading center, and electricity market information from the electricity trading data can be input into gradient boosting tree model, extreme forest model, random forest model, linear regression model, and multilayer perceptron model, respectively. These models will output the initial predicted electricity price corresponding to the weather information, information released by the trading center, and electricity market information, respectively.

[0059] S202. Using the first preset electricity price optimization algorithm in the preset electricity price optimization model, determine the minimum electricity price difference based on the difference between the third-party predicted electricity price and the multiple initial predicted electricity prices and the corresponding historical real electricity prices, and use the minimum electricity price difference to determine the first target predicted electricity price.

[0060] Specifically, the third-party predicted electricity price and multiple initial predicted electricity prices can be weighted separately. The minimum absolute value of the difference between the weighted result and the corresponding historical actual electricity price is determined as the minimum electricity price difference. The associated initial predicted electricity price and third-party predicted electricity price are then weighted using the first electricity price weights α1, β1, γ1, ..., δ1 corresponding to the minimum electricity price difference. The result obtained is the first target predicted electricity price.

[0061] For example, an initial weight can be assigned to the initial predicted electricity price output by each electricity price prediction sub-model. A Bayesian search is then used to find the weight combination that minimizes the MAE (Money Amount Escalation) over the past N days; this weight combination is the first electricity price weight. The first electricity price weight can be determined in the following ways:

[0062]

[0063] in, This represents the predicted electricity price of the i-th type determined by the j-th method. For example, it could be the initial predicted electricity price corresponding to the first type of electricity transaction-related data output by the first electricity price prediction sub-model, or a predicted electricity price determined by a third party. The types of electricity transaction-related data include weather information, information released by the trading center, and electricity market information. α1, β1, γ1, ..., δ1 are the weights of the first electricity price, and n is the sample size over the past N days, where n = 96N. To disclose the true electricity price, The value within the parentheses represents the electricity price weight corresponding to the minimum value; this weight is the first electricity price weight.

[0064] S203. Using the second preset electricity price optimization algorithm in the preset electricity price optimization model, the minimum value of the power plant operating profit difference is determined based on the difference between the power plant operating profit of the third-party predicted electricity price and the power plant operating profit of the multiple initial predicted electricity prices and the power plant operating profit of the corresponding historical real electricity prices. The second target predicted electricity price is then determined using the minimum value of the power plant operating profit difference.

[0065] Specifically, the operating profit of the power plant based on the third-party predicted electricity price and multiple initial predicted electricity prices can be weighted separately. The minimum absolute value of the difference between the power plant operating profit corresponding to the weighted result and the power plant operating profit corresponding to the historical actual electricity price is determined as the minimum operating profit difference. Using the second electricity price weights α2, β2, γ2, ..., δ2 corresponding to the minimum operating profit difference, the associated initial predicted electricity price and the third-party predicted electricity price are weighted, and the result is the second target predicted electricity price.

[0066] For example, the daily operating profit of the power plant for each predicted electricity price (third-party predicted electricity price and initial predicted electricity price) over the past N days can be calculated, and the weighted combination that minimizes the difference between the predicted profit and the actual profit over the past N days (MAE) can be found. This weighted combination is the second electricity price weight. The second electricity price weight can be determined in the following ways:

[0067]

[0068] Where f() is the preset power plant operating profit function, its input is the daily electricity price of 96 points, and its output is the profit for that day. α2,β2,γ2,…,δ2 are the weights of the second electricity price.

[0069] S204. Using the third preset electricity price optimization algorithm in the preset electricity price optimization model, and based on the accuracy of the electricity price prediction sub-model, determine the third target predicted electricity price based on the multiple initial predicted electricity prices and the third-party predicted electricity price.

[0070] Specifically, the third preset electricity price optimization algorithm in the preset electricity price optimization model can be used to determine the third target predicted electricity price from the multiple initial predicted electricity prices and the third-party predicted electricity price based on the accuracy of the electricity price prediction sub-model.

[0071] For example, we can statistically analyze the high, low, and medium points of electricity prices over the past N days, and then predict the third target electricity price y at different time points. predict It can be represented as:

[0072]

[0073] Where, y predict You can take y k y m or y n .

[0074] h, m, and l represent the time points corresponding to the high, low, and median electricity prices obtained from statistics over the past N days, respectively. k ,y m and y n These are the initial predicted electricity prices output by different electricity price prediction sub-models.

[0075] y k ,y m and y n The initial predicted electricity price at time point h can be determined based on the accuracy of the electricity price prediction sub-model and the accuracy of the third-party prediction. For example, the initial predicted electricity price at time point h output by the electricity price prediction sub-model with the highest accuracy or the third-party prediction can be used as y. k The accuracy of both the electricity price prediction sub-model and the third-party prediction is determined based on the difference between historical predicted electricity prices and historical actual electricity prices. The smaller the difference between the historical actual electricity price and the initial predicted electricity price output by the electricity price prediction sub-model for the corresponding historical time, the higher the accuracy of the electricity price prediction sub-model.

[0076] For example, when a more extreme third target electricity price forecast is needed, y predict A larger predicted electricity price can be used, i.e., y. predict =y k Alternatively, a smaller predicted electricity price can be used, i.e., y. predict =y n .

[0077] S205. Obtain the current status information through the MQTT server in the power plant.

[0078] The MQTT server is used to enable information exchange between the target device on one side of the power plant's internal network gate and the computing platform on the other side of the network gate. The target device includes an energy management system.

[0079] Specifically, Figure 4 This is a diagram of a power plant architecture. (Example) Figure 4 As shown, the power station can be an energy storage power station, and can use the MQTT (Message Queuing Telemetry Transport) protocol for communication. The IoT system of the power station based on the MQTT protocol includes an Energy Management System (EMS), data acquisition equipment, a network gateway, Remote Terminal Units (RTUs), and an MQTT server. The MQTT system is mainly responsible for data acquisition within the station, scheduling control, and data exchange between Zone 1 and Zone 3. An algorithm platform is used to generate optimal scheduling strategies. The service cluster built using the MQTT protocol can support millions of concurrent connections and hundreds of millions of device accesses.

[0080] The specific communication process includes:

[0081] EMS is responsible for collecting various information from the power plant and transmitting this information to the data acquisition equipment via the IEC104 protocol. Upon receiving the information from EMS, the data acquisition equipment securely forwards the data to the RTU through a network gateway, ensuring data integrity and security during transmission. Data interaction between the RTU and the algorithm platform is implemented by the MQTT service. The MQTT service is responsible for transmitting data from the RTU to the algorithm platform and also for returning the processing results from the algorithm platform to the RTU, forming a closed-loop communication system. The algorithm platform adopts an advanced microservice architecture design to ensure system modularity and flexibility. Communication between microservices within the platform, as well as communication with external data sources, uses the HTTP protocol for data transmission. To ensure system security, communication between the platform's service backend and frontend uses encrypted transmission based on the HTTPS protocol, utilizing SSL certificates to protect data security during transmission and prevent data leakage and unauthorized access.

[0082] S206. Using each algorithm in the preset strategy optimization algorithm, a target scheduling strategy is generated according to the multiple target predicted electricity prices and the current state information. The preset strategy optimization algorithm includes at least two of the following: branch and bound algorithm, genetic algorithm, mixed integer programming algorithm, and deep reinforcement learning algorithm.

[0083] Specifically, such as Figure 3 As shown, the output of the preset electricity price optimization algorithm can be used as input for branch and bound algorithm, genetic algorithm, mixed integer programming algorithm and deep reinforcement learning algorithm, so as to output the target scheduling strategy accordingly.

[0084] For example, if there are 96 points per day, and each point has three states: charging, discharging, and resting, then there are a total of 3 charging / discharging combinations per day. 36After determining the charge / discharge combination, the target predicted electricity price is used to determine the charge / discharge combination that maximizes profit as the target scheduling policy. Since the exhaustive search method involves excessive computation, a branch-and-bound method can be used to add constraints from different dimensions, reducing the magnitude of policy generation and ultimately determining the initial scheduling policy. The constraint policies for branch-and-bound can include: SOC (State of Charge) constraints, start and end capacity constraints, price difference constraints, cost constraints, and time granularity constraints, etc.

[0085] The adaptive genetic algorithm uses chromosome encoding, initial population settings, and fitness function design to iteratively perform chromosome crossover and chromosome mutation, causing the algorithm to converge to the optimal state, obtain the highest fitness, and output the initial scheduling strategy.

[0086] Mixed-integer programming algorithms can be performed using a CBC solver to maximize spot market returns and obtain an initial scheduling strategy. The objective function aims to maximize spot market returns and can be expressed as:

[0087]

[0088] Constraints can be specifically expressed as follows:

[0089]

[0090] in:

[0091] P c,i and P d,i Let i be the charging power and discharging power at time i;

[0092] pr i Let i be the electricity price at time i;

[0093] t represents the time duration, and the electricity price at 96 points is t = 0.25;

[0094] P c_max and P d_max These are the maximum charging power and the maximum discharging power.

[0095] D i For charging and discharging operations, 0 indicates rest, and 1 indicates charging or discharging;

[0096] C effi and D effi For charging efficiency and discharging efficiency;

[0097] E0 and E -1 The initial and final SOCs for the running day;

[0098] Points time limit.

[0099] Deep reinforcement learning is a deep neural network with multiple inputs and a single output. The output action is a continuous value, including: rest, maximum chargeable power, and maximum dischargeable power. After the deep reinforcement learning algorithm or model generates an action, the environment changes accordingly, including information such as maximum chargeable / dischargeable power, state of charge (SOC), time point, and electricity price. The algorithm calculates the reward under the current state and action. The reward function maximizes the cumulative profit. The calculated reward function result and the next state are repeatedly fed into the deep reinforcement learning algorithm for iterative optimization, ultimately yielding the initial scheduling strategy.

[0100] S207. Using the real electricity price within a preset historical time period, determine the maximum profit target scheduling strategy from the multiple target scheduling strategies, and determine the maximum profit target scheduling strategy as the optimal scheduling strategy.

[0101] The method for determining power plant scheduling strategies provided in this invention, by deploying an MQTT server in the power plant, realizes power plant-level data communication and scheduling strategy generation. This enables the power plant to not only process complex external data but also dynamically adjust the forecast results according to changes in the electricity market, ensuring that the generated scheduling strategy can adapt to the real-time market environment. Furthermore, by using electricity transaction correlation data from different dimensions, a comprehensive forecast of the electricity price on the forecast day is performed. By utilizing the minimum electricity price difference, the minimum power plant operating profit difference, and the accuracy of the electricity price forecast sub-model, multiple rounds of optimization of the forecast electricity price are achieved. The optimal power plant scheduling strategy is generated using the optimized forecast electricity price.

[0102] Example 3

[0103] Figure 5 This is a schematic diagram of a device for determining a power plant dispatching strategy according to Embodiment 3 of the present invention. Figure 5 As shown, the device includes: an initial predicted electricity price determination module 301, a target predicted electricity price determination module 302, a target scheduling strategy determination module 303, and an optimal scheduling strategy determination module 304, wherein:

[0104] An initial predicted electricity price determination module is used to acquire electricity transaction related data and input the electricity transaction related data into a preset electricity price prediction model to obtain multiple initial predicted electricity prices, wherein the preset electricity price prediction model includes multiple electricity price prediction sub-models;

[0105] The target predicted electricity price determination module is used to optimize the multiple initial predicted electricity prices using a preset electricity price optimization model to generate multiple target predicted electricity prices for the power plant.

[0106] The target scheduling strategy determination module is used to obtain the current status information of the power plant and generate multiple target scheduling strategies based on the multiple target predicted electricity prices and the current status information using a preset strategy optimization algorithm.

[0107] The optimal scheduling strategy determination module is used to determine the maximum profit target scheduling strategy from the multiple target scheduling strategies by using the real electricity price within a preset historical time period, and to determine the maximum profit target scheduling strategy as the optimal scheduling strategy.

[0108] The apparatus for determining power plant dispatching strategies provided in this invention predicts future electricity prices using electricity trading correlation data, optimizes the predicted prices using an electricity price optimization model to obtain a target predicted electricity price with better returns, and finally generates the optimal power plant dispatching strategy by comprehensively considering the target predicted electricity price and the current state information of the power plant using a strategy optimization algorithm. This apparatus enables power plants to achieve highly automated operation when participating in electricity trading, automatically generating accurate electricity price predictions and dispatching strategies. Compared with the traditional method of relying on manual determination of power plant dispatching strategies, it improves efficiency, significantly reduces labor costs, greatly reduces the subjectivity and uncertainty of human decision-making, thereby improving the stability of power plant revenue and significantly enhancing the power plant's competitiveness in the electricity trading market. It allows power plants to respond more flexibly to market fluctuations, maximize power plant revenue, and promote the progress of intelligent electricity trading.

[0109] Optionally, the device is applied to a computing platform within a power plant.

[0110] Optionally, the target predicted electricity price determination module includes:

[0111] An information acquisition unit is used to acquire current status information through an MQTT server in the power plant; wherein, the MQTT server is used to realize information interaction between the target device on one side of the power plant's internal network gate and the computing platform on the other side of the network gate, and the target device includes an energy management system.

[0112] Optionally, the initial predicted electricity price determination module includes:

[0113] The initial predicted electricity price determination unit inputs weather information, information released by the trading center, and electricity market information from the electricity trading-related data into each electricity price prediction sub-model, so that each electricity price prediction sub-model outputs the corresponding initial predicted electricity price.

[0114] Furthermore, the electricity price prediction sub-model in the preset electricity price prediction model includes at least two of the following: gradient boosting tree model, extreme forest model, random forest model, linear regression model, and multilayer perceptron model.

[0115] Optionally, the target predicted electricity price determination module is specifically used to obtain a third-party predicted electricity price and process the third-party predicted electricity price and the multiple initial predicted electricity prices using a preset electricity price optimization model to obtain multiple target predicted electricity prices for the power plant.

[0116] Furthermore, the step of processing the third-party predicted electricity price and the multiple initial predicted electricity prices using a preset electricity price optimization model to obtain multiple target predicted electricity prices for the power plant includes: using a first preset electricity price optimization algorithm in the preset electricity price optimization model to determine the minimum electricity price difference based on the difference between the third-party predicted electricity price and the multiple initial predicted electricity prices and the corresponding historical real electricity prices, and using the minimum electricity price difference to determine the first target predicted electricity price; using a second preset electricity price optimization algorithm in the preset electricity price optimization model to determine the minimum power plant operating profit difference based on the difference between the power plant operating profit of the third-party predicted electricity price and the power plant operating profit of the multiple initial predicted electricity prices and the power plant operating profit of the corresponding historical real electricity prices, and using the minimum power plant operating profit difference to determine the second target predicted electricity price; and using a third preset electricity price optimization algorithm in the preset electricity price optimization model to determine the third target predicted electricity price based on the accuracy of the electricity price prediction sub-model and the multiple initial predicted electricity prices and the third-party predicted electricity price.

[0117] Optionally, the preset strategy optimization algorithm includes at least two of the following: branch and bound algorithm, genetic algorithm, mixed integer programming algorithm, and deep reinforcement learning algorithm.

[0118] Optionally, the preset strategy optimization algorithm includes at least two of the following: branch and bound algorithm, genetic algorithm, mixed integer programming algorithm, and deep reinforcement learning algorithm;

[0119] The target scheduling strategy determination module includes:

[0120] The target scheduling strategy determination unit is used to generate a target scheduling strategy based on each algorithm in the preset strategy optimization algorithm, according to the multiple target predicted electricity prices and the current state information.

[0121] The apparatus for determining power plant scheduling strategies provided in this embodiment of the invention can execute the method for determining power plant scheduling strategies provided in any embodiment of the invention, and has the corresponding functional modules and beneficial effects of the method.

[0122] Example 4

[0123] Figure 6A schematic diagram of an electronic device 40 that can be used to implement embodiments of the present invention is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital processors, cellular phones, smartphones, wearable devices (e.g., helmets, glasses, watches, etc.), and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the invention described and / or claimed herein.

[0124] like Figure 6 As shown, the electronic device 40 includes at least one processor 41 and a memory, such as a read-only memory (ROM) 42 or a random access memory (RAM) 43, communicatively connected to the at least one processor 41. The memory stores computer programs executable by the at least one processor. The processor 41 can perform various appropriate actions and processes based on the computer program stored in the ROM 42 or loaded into the RAM 43 from storage unit 48. The RAM 43 may also store various programs and data required for the operation of the electronic device 40. The processor 41, ROM 42, and RAM 43 are interconnected via a bus 44. An input / output (I / O) interface 45 is also connected to the bus 44.

[0125] Multiple components in electronic device 40 are connected to I / O interface 45, including: input unit 46, such as keyboard, mouse, etc.; output unit 47, such as various types of monitors, speakers, etc.; storage unit 48, such as disk, optical disk, etc.; and communication unit 49, such as network card, modem, wireless transceiver, etc. Communication unit 49 allows electronic device 40 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0126] Processor 41 can be a variety of general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of processor 41 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. Processor 41 performs the various methods and processes described above, such as methods for determining power plant scheduling strategies.

[0127] In some embodiments, the method for determining a power plant scheduling strategy may be implemented as a computer program tangibly contained in a computer-readable storage medium, such as storage unit 48. In some embodiments, part or all of the computer program may be loaded and / or installed on electronic device 40 via ROM 42 and / or communication unit 49. When the computer program is loaded into RAM 43 and executed by processor 41, one or more steps of the method for determining a power plant scheduling strategy described above may be performed. Alternatively, in other embodiments, processor 41 may be configured to perform the method for determining a power plant scheduling strategy by any other suitable means (e.g., by means of firmware).

[0128] Various embodiments of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), systems-on-a-chip (SoCs), payload-programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments may include implementations in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which may be a dedicated or general-purpose programmable processor, capable of receiving data and instructions from a storage system, at least one input device, and at least one output device, and transmitting data and instructions to the storage system, the at least one input device, and the at least one output device.

[0129] Computer programs used to implement the methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, such that when executed by the processor, the computer programs cause the functions / operations specified in the flowcharts and / or block diagrams to be performed. The computer programs may be executed entirely on a machine, partially on a machine, or as a standalone software package, partially on a machine and partially on a remote machine, or entirely on a remote machine or server.

[0130] The computer equipment provided above can be used to execute the method for determining power plant scheduling strategy provided in any of the above embodiments, and has corresponding functions and beneficial effects.

[0131] Example 5

[0132] In the context of this invention, the computer-readable storage medium may be a tangible medium, and the computer-executable instructions, when executed by a computer processor, are used to perform a method for determining a power plant dispatching strategy, the method comprising:

[0133] Acquire electricity transaction related data and input the electricity transaction related data into a preset electricity price prediction model to obtain multiple initial predicted electricity prices, wherein the preset electricity price prediction model includes multiple electricity price prediction sub-models;

[0134] The multiple initial predicted electricity prices are optimized using a preset electricity price optimization model to generate multiple target predicted electricity prices for the power plant.

[0135] The current status information of the power plant is obtained, and multiple target scheduling strategies are generated based on the multiple target predicted electricity prices and the current status information using a preset strategy optimization algorithm.

[0136] In the context of this invention, a computer-readable storage medium can be a tangible medium that may contain or store a computer program for use by, or in conjunction with, an instruction execution system, apparatus, or device. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination thereof. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof.

[0137] The computer equipment provided above can be used to execute the method for determining power plant scheduling strategy provided in any of the above embodiments, and has corresponding functions and beneficial effects.

[0138] It is worth noting that in the embodiments of the device for determining power plant dispatching strategies described above, the various units and modules included are only divided according to functional logic, but are not limited to the above division, as long as the corresponding functions can be achieved; in addition, the specific names of each functional unit are only for easy differentiation and are not used to limit the scope of protection of the present invention.

[0139] Note that the above description is merely a preferred embodiment of the present invention and the technical principles employed. Those skilled in the art will understand that the present invention is not limited to the specific embodiments described herein, and various obvious changes, readjustments, and substitutions can be made without departing from the scope of protection of the present invention. Therefore, although the present invention has been described in detail through the above embodiments, the present invention is not limited to the above embodiments, and may include many other equivalent embodiments without departing from the concept of the present invention, the scope of which is determined by the scope of the appended claims.

[0140] Example 6

[0141] Figure 7 A schematic diagram of a power plant management system that can be used to implement an embodiment of the present invention is shown.

[0142] like Figure 7 As shown, the power plant management system 70 includes at least a computing platform 71, an MQTT server 72, a network gateway 73, and an energy management system 74 within the power plant.

[0143] The MQTT server 72 is used to realize information interaction between the energy management system 74 on one side of the power plant intranet gate 73 and the computing platform 71 on the other side of the gate 73; the computing platform 71 can execute the method for determining the power plant scheduling strategy described in the above embodiments.

[0144] The power station can be an energy storage power station and can communicate using the MQTT (Message Queuing Telemetry Transport) protocol. For example... Figure 3 As shown, the power plant management system can include a computing platform, MQTT server, network gateway, and energy management system, as well as data acquisition equipment and remote terminal units (RTUs). The MQTT server is mainly responsible for data acquisition within the plant, scheduling control, and data exchange between Zone 1 and Zone 3. The algorithm platform is used to generate optimal scheduling strategies. The service cluster built using the MQTT protocol can support millions of concurrent connections and hundreds of millions of device accesses.

[0145] The computing platform provided above can be used to execute the method for determining power plant scheduling strategies provided in any of the above embodiments, and has corresponding functions and beneficial effects.

Claims

1. A method for determining a power plant dispatching strategy, characterized in that, include: Acquire electricity transaction related data and input the electricity transaction related data into a preset electricity price prediction model to obtain multiple initial predicted electricity prices, wherein the preset electricity price prediction model includes multiple electricity price prediction sub-models; The multiple initial predicted electricity prices are optimized using a preset electricity price optimization model to generate multiple target predicted electricity prices for the power plant. The current status information of the power plant is obtained, and multiple target scheduling strategies are generated based on the multiple target predicted electricity prices and the current status information using a preset strategy optimization algorithm. Using the real electricity price within a preset historical time period, the maximum profit target scheduling strategy is determined from the multiple target scheduling strategies, and the maximum profit target scheduling strategy is determined as the optimal scheduling strategy.

2. The method according to claim 1, characterized in that, A computing platform applied within a power plant; wherein, obtaining the current status information of the power plant includes: Obtain current status information through the MQTT server in the power plant; The MQTT server is used to enable information exchange between the target device on one side of the power plant's internal network gate and the computing platform on the other side of the network gate. The target device includes an energy management system.

3. The method according to claim 1, characterized in that, The step of inputting the electricity transaction-related data into a preset electricity price prediction model to obtain multiple initial predicted electricity prices includes: The weather information, information released by the trading center, and electricity market information in the electricity trading-related data are respectively input into each electricity price prediction sub-model so that each electricity price prediction sub-model outputs the corresponding initial predicted electricity price.

4. The method according to claim 3, characterized in that, The electricity price prediction sub-model in the preset electricity price prediction model includes at least two of the following: gradient boosting tree model, extreme forest model, random forest model, linear regression model, and multilayer perceptron model.

5. The method according to claim 1, characterized in that, The step of optimizing the multiple initial predicted electricity prices using a preset electricity price optimization model to generate multiple target predicted electricity prices for the power plant includes: Obtain the third-party predicted electricity price, and use a preset electricity price optimization model to process the third-party predicted electricity price and the multiple initial predicted electricity prices to obtain multiple target predicted electricity prices for the power plant.

6. The method according to claim 5, characterized in that, The process of using a preset electricity price optimization model to process the third-party predicted electricity price and the multiple initial predicted electricity prices to obtain multiple target predicted electricity prices for the power plant includes: Using the first preset electricity price optimization algorithm in the preset electricity price optimization model, the minimum electricity price difference is determined based on the difference between the third-party predicted electricity price and the multiple initial predicted electricity prices and the corresponding historical real electricity prices, and the first target predicted electricity price is determined using the minimum electricity price difference. Using the second preset electricity price optimization algorithm in the preset electricity price optimization model, the minimum value of the power plant operating profit difference is determined based on the difference between the power plant operating profit of the third-party predicted electricity price and the power plant operating profit of the multiple initial predicted electricity prices and the power plant operating profit of the corresponding historical real electricity prices. The second target predicted electricity price is then determined using the minimum value of the power plant operating profit difference. Using the third preset electricity price optimization algorithm in the preset electricity price optimization model, a third target predicted electricity price is determined based on the accuracy of the electricity price prediction sub-model, the multiple initial predicted electricity prices, and the third-party predicted electricity price.

7. The method according to any one of claims 1-6, characterized in that, The preset strategy optimization algorithm includes at least two of the following: branch and bound algorithm, genetic algorithm, mixed integer programming algorithm, and deep reinforcement learning algorithm; wherein, the generation of multiple target scheduling strategies based on the multiple target predicted electricity prices and the current state information using the preset strategy optimization algorithm includes: Each algorithm in the preset strategy optimization algorithm is used to generate a target scheduling strategy based on the multiple target predicted electricity prices and the current state information.

8. An apparatus for determining a power plant dispatching strategy, characterized in that, include: An initial predicted electricity price determination module is used to acquire electricity transaction related data and input the electricity transaction related data into a preset electricity price prediction model to obtain multiple initial predicted electricity prices, wherein the preset electricity price prediction model includes multiple electricity price prediction sub-models; The target predicted electricity price determination module is used to optimize the multiple initial predicted electricity prices using a preset electricity price optimization model to generate multiple target predicted electricity prices for the power plant. The target scheduling strategy determination module is used to obtain the current status information of the power plant and generate multiple target scheduling strategies based on the multiple target predicted electricity prices and the current status information using a preset strategy optimization algorithm. The optimal scheduling strategy determination module is used to determine the maximum profit target scheduling strategy from the multiple target scheduling strategies by using the real electricity price within a preset historical time period, and to determine the maximum profit target scheduling strategy as the optimal scheduling strategy.

9. The apparatus according to claim 8, characterized in that, The initial predicted electricity price determination module includes: The weather information, information released by the trading center, and electricity market information in the electricity trading-related data are respectively input into each electricity price prediction sub-model so that each electricity price prediction sub-model outputs the corresponding initial predicted electricity price.

10. The apparatus according to claim 8, characterized in that, The preset strategy optimization algorithm includes at least two of the following: branch and bound algorithm, genetic algorithm, mixed integer programming algorithm, and deep reinforcement learning algorithm; Specifically, the target scheduling strategy determination module is used to generate a target scheduling strategy based on each algorithm in the preset strategy optimization algorithm, according to the multiple target predicted electricity prices and the current state information.

11. An electronic device, characterized in that, The electronic device includes: At least one processor; and A memory communicatively connected to the at least one processor; wherein, The memory stores a computer program that can be executed by the at least one processor, the computer program being executed by the at least one processor to enable the at least one processor to perform the method for determining a power plant scheduling strategy as described in any one of claims 1-7.

12. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions that, when executed by a processor, implement the method for determining a power plant scheduling strategy as described in any one of claims 1-7.

13. A power plant management system, characterized in that, The system includes at least a computing platform, an MQTT server, a network gateway, and an energy management system within the power plant. The MQTT server is used to realize information interaction between the energy management system on one side of the power plant's internal network gate and the computing platform on the other side of the network gate; the computing platform is capable of executing the method for determining the power plant scheduling strategy as described in any one of claims 1-7.

Images