Optimization method and device for expected service electric quantity of new energy station and program product

Abstract

The invention provides a method and a device for optimizing expected service electric quantity of a new energy station and a program product. The optimization method comprises the following steps: obtaining new energy station basic data, new energy station environment data, preset supply and demand space data and initial expected power generation data of a new energy station for power business, wherein the initial expected power generation data for the power business comprises initial expected business electric quantity and initial expected power generation amount; inputting the new energy station basic data, the new energy station environment data, the preset supply and demand space data and the initial expected power generation data into an expected business electric quantity optimization model based on a self-adaptive genetic algorithm to obtain an optimized corrected expected business electric quantity, and based on a target cost function representing a target business parameter of the new energy station for the power business, training an expected business electric quantity optimization model based on an adaptive genetic algorithm, the target business parameter being the sum of an electricity price income of the new energy station for the power business and a variable parameter based on a preset constraint condition.

Description

Technical Field

[0001] The present disclosure relates to the fields of new energy and data processing, and more specifically, to a method, device and program product for optimizing expected business electricity volume of a new energy site. Background Art

[0002] In the field of new energy, industry goals related to "dual carbon" have led to rapid development of related technologies involved in the field of new energy. In the scenario where new energy is used as a power supply source, it is necessary to use the electricity generated by new energy as a power business to realize the supply of electricity. In this scenario, as the supply business of the power business becomes more and more perfect, the large-scale grid connection of new energy stations may have a relatively large impact on the power grid due to factors such as its volatility. In order to maintain the stable operation of the power grid, new energy stations are required to meet the following conditions: make a more accurate forecast of their power generation, rationalize the transaction volume, and reduce grid congestion.

[0003] Therefore, how to obtain the expected power generation of new energy sites that meet the above conditions so as to maintain the stability of the power grid when new energy is used as the power source has become an issue worthy of special attention. Summary of the invention

[0004] The embodiments of the present disclosure provide a method and device for optimizing the expected business electricity volume of a new energy station, establish an optimization model for the expected business electricity volume of the new energy station, and achieve the purpose of correcting the expected business electricity volume, thereby improving the target revenue of the new energy station for the business electricity volume while ensuring the stability of the power grid.

[0005] In a general aspect, a method for optimizing expected business electricity of a new energy station is provided, the method comprising: obtaining basic data of the new energy station, environmental data of the new energy station, preset supply and demand space data and initial expected power generation data of the new energy station for power business, the initial expected power generation data for power business comprising initial expected business electricity and initial expected power generation; inputting the basic data of the new energy station, the environmental data of the new energy station, the preset supply and demand space data and the initial expected power generation data into an expected business electricity optimization model based on an adaptive genetic algorithm to obtain an optimized revised expected business electricity, wherein the expected business electricity optimization model based on the adaptive genetic algorithm is trained based on a target cost function representing target business parameters of the new energy station for power business, the target business parameters being the sum of the electricity price revenue of the new energy station for power business and a variable parameter based on preset constraints.

[0006] Optionally, the new energy station basic data may include data related to the power generation capacity of the new energy station and data related to the location of the new energy station, and the preset supply and demand space data may include expected new energy power load data and expected new energy output data.

[0007] Optionally, the step of inputting the new energy station basic data, the new energy station environmental data, the preset supply and demand space data and the initial expected power generation data into an expected business electricity optimization model based on an adaptive genetic algorithm to obtain the optimized revised expected business electricity may include: inputting the new energy station environmental data and the preset supply and demand space data into a supply and demand space prediction model based on a preset machine learning algorithm to obtain first predicted supply and demand space data; inputting the new energy station basic data, the first predicted supply and demand space data and the initial expected power generation data into an expected business electricity optimization model based on an adaptive genetic algorithm to obtain the optimized revised expected business electricity.

[0008] Optionally, the expected business power optimization model based on the adaptive genetic algorithm can be trained in the following manner: in each iteration cycle, the following processing is performed: a plurality of initial population individuals corresponding to the sequence of the initial expected business power are obtained; the following processing is performed for every two initial population individuals: two initial population individuals are selected from the plurality of initial population individuals; the two selected initial population individuals are cross-processed based on a preset crossover probability to obtain two crossover population individuals after the crossover; the two crossover population individuals are mutated based on a preset mutation probability, a preset mutation intensity and a preset standard normal distribution coefficient to obtain two mutated population individuals after the mutation; the target cost function values ​​corresponding to the two mutated population individuals are compared with the target cost function values ​​corresponding to the corresponding population individuals in the two selected initial population individuals, and two population individuals with larger target cost function values ​​are selected to replace the corresponding selected initial population individuals as the corresponding initial population individuals for the next iteration cycle.

[0009] Optionally, the mutation process may include at least two stages of mutation process, and the values ​​of the preset mutation probability and the preset mutation strength may both decrease nonlinearly as the order of the stages of the mutation process increases.

[0010] Optionally, the mutation processing may include at least two stages of mutation processing, and for each stage of mutation processing, the current initial population individuals are mutated based on a preset mutation probability, a preset mutation strength and a preset standard normal distribution coefficient, and the step of obtaining the mutated mutant population individuals may include: when a preset parameter is less than a current preset mutation probability for the current stage, taking the sum of the product of the preset mutation strength for the current stage and the preset standard normal distribution coefficient and the current initial population individual as the mutated mutant population individual; when the preset parameter is greater than or equal to the current preset mutation probability for the current stage, taking the current initial population individual as the mutated mutant population individual.

[0011] Optionally, the supply and demand space prediction model based on the preset machine learning algorithm can be trained in the following manner: obtaining historical supply and demand space data and preset boundary conditions; performing correlation analysis on the historical supply and demand space data in combination with the preset boundary conditions to obtain supply and demand space characteristic data; inputting the supply and demand space characteristic data and the historical supply and demand space data into the supply and demand space prediction model based on the preset machine learning algorithm to obtain second predicted supply and demand space data; adjusting the model parameters of the supply and demand space prediction model based on the preset machine learning algorithm according to the historical supply and demand space data and the second predicted supply and demand space data.

[0012] Optionally, the variable parameters based on preset constraints may correspond to the following two items: the over-generation electricity that needs to be recovered at the new energy station corresponding to the preset transaction electricity gap for the power business, and the electricity recovery or compensation for the new energy station corresponding to the frequency and peak regulation power dispatching based on preset business conditions.

[0013] In another general aspect, there is provided an optimization device for expected business electricity of a new energy station, the optimization device comprising: a data acquisition module, configured to: acquire new energy station basic data, new energy station environmental data, preset supply and demand space data and initial expected power generation data of the new energy station for power business, the initial expected power generation data for power business comprising initial expected business electricity and initial expected power generation; a data correction module, configured to: input the new energy station basic data, the new energy station environmental data, the preset supply and demand space data and the initial expected power generation data into an expected business electricity optimization model based on an adaptive genetic algorithm to obtain an optimized corrected expected business electricity, wherein the expected business electricity optimization model based on the adaptive genetic algorithm is trained based on a target cost function representing a target business parameter of the new energy station for power business, the target business parameter being the sum of the price revenue of the new energy station for power business and a variable parameter based on preset constraints.

[0014] Optionally, the new energy station basic data may include data related to the power generation capacity of the new energy station and data related to the location of the new energy station, and the preset supply and demand space data may include expected new energy power load data and expected new energy output data.

[0015] Optionally, the data correction module inputs the new energy station basic data, the new energy station environmental data, the preset supply and demand space data and the initial expected power generation data into an expected business electricity optimization model based on an adaptive genetic algorithm to obtain the optimized revised expected business electricity. The operation may include: inputting the new energy station environmental data and the preset supply and demand space data into a supply and demand space prediction model based on a preset machine learning algorithm to obtain first predicted supply and demand space data; inputting the new energy station basic data, the first predicted supply and demand space data and the initial expected power generation data into the expected business electricity optimization model based on an adaptive genetic algorithm to obtain the optimized revised expected business electricity.

[0016] Optionally, the expected business power optimization model based on the adaptive genetic algorithm can be trained in the following manner: in each iteration cycle, the following processing is performed: a plurality of initial population individuals corresponding to the sequence of the initial expected business power are obtained; the following processing is performed for every two initial population individuals: two initial population individuals are selected from the plurality of initial population individuals; the two selected initial population individuals are cross-processed based on a preset crossover probability to obtain two crossover population individuals after the crossover; the two crossover population individuals are mutated based on a preset mutation probability, a preset mutation intensity and a preset standard normal distribution coefficient to obtain two mutated population individuals after the mutation; the target cost function values ​​corresponding to the two mutated population individuals are compared with the target cost function values ​​corresponding to the corresponding population individuals in the two selected initial population individuals, and two population individuals with larger target cost function values ​​are selected to replace the corresponding selected initial population individuals as the corresponding initial population individuals for the next iteration cycle.

[0017] Optionally, the mutation process may include at least two stages of mutation process, and the values ​​of the preset mutation probability and the preset mutation strength may both decrease nonlinearly as the order of the stages of the mutation process increases.

[0018] Optionally, the mutation processing may include at least two stages of mutation processing, and for each stage of mutation processing, the operation of performing mutation processing on the current initial population individuals based on a preset mutation probability, a preset mutation strength and a preset standard normal distribution coefficient to obtain the mutated mutated population individuals may include: when a preset parameter is less than a current preset mutation probability for the current stage, taking the sum of the product of the preset mutation strength for the current stage and the preset standard normal distribution coefficient and the current initial population individual as the mutated mutated population individual; when the preset parameter is greater than or equal to the current preset mutation probability for the current stage, taking the current initial population individual as the mutated mutated population individual.

[0019] Optionally, the supply and demand space prediction model based on the preset machine learning algorithm can be trained in the following manner: obtaining historical supply and demand space data and preset boundary conditions; performing correlation analysis on the historical supply and demand space data in combination with the preset boundary conditions to obtain supply and demand space characteristic data; inputting the supply and demand space characteristic data and the historical supply and demand space data into the supply and demand space prediction model based on the preset machine learning algorithm to obtain second predicted supply and demand space data; adjusting the model parameters of the supply and demand space prediction model based on the preset machine learning algorithm according to the historical supply and demand space data and the second predicted supply and demand space data.

[0020] Optionally, the variable parameters based on preset constraints may correspond to the following two items: the over-generation electricity that needs to be recovered at the new energy station corresponding to the preset transaction electricity gap for the power business, and the electricity recovery or compensation for the new energy station corresponding to the frequency and peak regulation power dispatching based on preset business conditions.

[0021] In another general aspect, a computer program product is provided, comprising computer instructions, which, when executed by at least one processor, cause the at least one processor to execute the method for optimizing the expected business electricity volume of the new energy station as described above.

[0022] In another general aspect, a computer-readable storage medium is provided, which, when instructions in the computer-readable storage medium are executed by at least one processor, causes the at least one processor to execute the method for optimizing the expected business electricity volume of the new energy station as described above.

[0023] In another general aspect, a computer device is provided, comprising: at least one processor; and at least one memory storing computer executable instructions, wherein the computer executable instructions, when executed by the at least one processor, cause the at least one processor to execute the method for optimizing the expected business power of the new energy station as described above.

[0024] According to the optimization method and device of the expected business electricity of the new energy station of the embodiment of the present disclosure, by considering the information related to the new energy station and the business demand related information, combined with the preset constraints, an optimization model for the expected business electricity of the new energy station is established, and the purpose of correcting the expected business electricity is achieved, thereby improving the target revenue of the new energy station for the business electricity while ensuring the stability of the power grid. In addition, by setting the adaptive crossover probability, mutation probability and mutation intensity, the search ability, convergence speed and accuracy of the optimization model algorithm for the expected business electricity of the new energy station are effectively improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] Figure 1 It is a flowchart showing a method for optimizing expected business electricity volume of a new energy station according to an embodiment of the present disclosure.

[0026] Figure 2 is a flowchart illustrating a method for training a supply and demand space prediction model based on a preset machine learning algorithm according to an embodiment of the present disclosure.

[0027] Figure 3 is a flowchart illustrating a training method of an expected business power optimization model based on an adaptive genetic algorithm according to an embodiment of the present disclosure.

[0028] Figure 4 It is an execution logic block diagram showing an example of a method for optimizing expected business electricity volume of a new energy station according to an embodiment of the present disclosure.

[0029] Figure 5 It is a logical block diagram showing an optimization system for expected business electricity volume of a new energy station according to an embodiment of the present disclosure.

[0030] Figure 6 is a block diagram showing an optimization device for expected business electricity volume of a new energy station according to an embodiment of the present disclosure.

[0031] Figure 7 is a block diagram illustrating a computer device according to an embodiment of the present disclosure. DETAILED DESCRIPTION

[0032] The following specific embodiments are provided to help the reader obtain a comprehensive understanding of the methods, devices and / or systems described herein. However, after understanding the disclosure of the present application, various changes, modifications and equivalents of the methods, devices and / or systems described herein will be clear. For example, the order of operations described herein is only an example and is not limited to those orders set forth herein, but can be changed as will be clear after understanding the disclosure of the present application, except for operations that must occur in a specific order. In addition, for greater clarity and simplicity, the description of features known in the art may be omitted.

[0033] The features described herein can be implemented in different forms and should not be construed as being limited to the examples described herein. Rather, the examples described herein have been provided to illustrate only some of the many possible ways to implement the methods, devices, and / or systems described herein, which will be clear after understanding the disclosure of the present application.

[0034] As used herein, the term "and / or" includes any one of the associated listed items and any combination of any two or more.

[0035] The terms used herein are only used to describe various examples and are not intended to limit the disclosure. Unless the context clearly indicates otherwise, the singular form is also intended to include the plural form. The terms "comprise", "include" and "have" indicate the presence of the described features, quantities, operations, components, elements and / or combinations thereof, but do not exclude the presence or addition of one or more other features, quantities, operations, components, elements and / or combinations thereof.

[0036] Unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as those generally understood by a person of ordinary skill in the art to which the present disclosure belongs after understanding the present disclosure. Unless explicitly defined as such herein, terms (such as those defined in a general dictionary) should be interpreted as having a meaning consistent with their meaning in the context of the relevant art and the present disclosure, and should not be interpreted in an idealized or overly formal manner.

[0037] Furthermore, in the description of examples, when it is considered that a detailed description of a well-known related structure or function would cause vague interpretation of the present disclosure, such a detailed description will be omitted.

[0038] Reference will now be made in detail to the embodiments of the present disclosure, examples of which are illustrated in the accompanying drawings, wherein like reference numerals refer to like parts throughout. The embodiments will be described below with reference to the drawings in order to explain the present disclosure.

[0039] In the relevant technology, as mentioned above, in the scenario where new energy is used as a power supply source, the large-scale grid connection of new energy stations may have a greater impact on the power grid due to its volatility and other factors. In order to ensure the stability of the power grid and fully consider the competitive relationship between different power supply entities in the power business, the present disclosure fully considers the following two rule constraints: in the scenario of the scheduled power business, the over-generation of the new energy station corresponding to the preset transaction power gap needs to be recovered, and the power recovery and compensation corresponding to the frequency and peak regulation of the power dispatching of the new energy station according to the preset business requirements. For example, in the scenario of the scheduled power business for the preset power generation area, the above two rule constraints can be specifically reflected as follows: the over-generation of the new energy station corresponding to the power gap of the predetermined proportion of the transaction power needs to be recovered, and according to the preset business requirements, the power recovery and compensation corresponding to the frequency and peak regulation of the power dispatching (for example, the relevant power dispatching realized by other power supply entities) of the new energy station.

[0040] Here, from the perspective of new energy stations, it is necessary to balance the relationship between the expected business electricity and the expected power generation of new energy stations, so as to balance the relationship between the revenue of new energy stations and the stability of the power grid, so as to achieve a win-win situation. In the process of correcting the expected business electricity of new energy stations, many factors need to be considered, such as weather conditions, business supply and demand space, user-side electricity consumption information, etc., but the weather conditions and business supply and demand space themselves are uncertain, which has a great impact on the correction of expected business electricity. In addition, when considering the competitive relationship between various supply entities, the predicted optimal correction data of the expected business electricity cannot meet the actual business requirements in actual applications, so it is necessary to make reasonable adjustments based on the conditions of the new energy stations themselves.

[0041] To this end, the present disclosure proposes a method for optimizing the expected business electricity volume of a new energy station, which can successfully correct the expected business electricity volume and achieve a balance between the revenue of the new energy station and the stability of the power grid to the greatest extent possible.

[0042] Below, refer to Figures 1 to 7 A method and device for optimizing the expected business power of a new energy station according to an embodiment of the present disclosure are described in detail.

[0043] Figure 1 It is a flowchart showing a method for optimizing expected business electricity volume of a new energy station according to an embodiment of the present disclosure.

[0044] Reference Figure 1 In step S101, basic data of new energy stations, environmental data of new energy stations, preset supply and demand space data, and initial expected power generation data of new energy stations for power business are obtained.

[0045] Here, the initial expected power generation data for the power business includes the initial expected business power and the initial expected power generation.

[0046] For example, the basic data of the new energy station may include the power generation capacity related data of the new energy station and the location related data of the new energy station. Here, the power generation capacity related data of the new energy station may include the capacity and power generation capacity curve of the new energy station, and the location related data of the new energy station may represent the physical conditions of the new energy station, and may include the geographical location coordinates of the new energy station, the location information of each power generation equipment in the new energy station, etc.

[0047] In addition, for example, the new energy station environmental data may include climate data of the area where the new energy station is located.

[0048] As an example, the preset supply and demand space data may include expected new energy power load data and expected new energy output data. Here, the expected new energy power load data may include power consumption plan data on the user side, and the expected new energy output data may include output plan data of the new energy station for the scheduled power business.

[0049] By providing preset supply and demand space data for training and solving the optimization model according to the present disclosure, the adverse effects of data fluctuations on the correction results at the user side can be effectively avoided.

[0050] In step S102, the new energy station basic data, the new energy station environmental data, the preset supply and demand space data and the initial expected power generation data are input into the expected business power optimization model based on the adaptive genetic algorithm to obtain the optimized revised expected business power.

[0051] According to an embodiment of the present disclosure, an expected business electricity quantity optimization model based on an adaptive genetic algorithm may be trained based on a target cost function representing target business parameters for power business of a new energy station.

[0052] Here, the target business parameter is the sum of the electricity price revenue of the new energy station for the power business and the variable parameters based on the preset constraints.

[0053] By taking the maximization of the sum of the electricity price revenue of the new energy station for the power business and the variable parameters based on the preset constraints as the goal, considering the physical conditions, supply and demand space, capacity, power generation capacity curve and other constraints of the new energy station, taking the expected business electricity sequence of the new energy station at each moment as the decision variable, and taking the transaction electricity and reported power generation capacity as constraints, an optimization model of the expected business electricity of the new energy station for the power business based on rule constraints is established.

[0054] As an example, the variable parameters based on preset constraints may correspond to the following two items: the over-generation electricity that needs to be recovered at the new energy station corresponding to the preset transaction electricity gap for the power business, and the electricity recovery or compensation for the new energy station corresponding to the frequency and peak regulation power dispatching based on the preset business conditions.

[0055] By considering the above-mentioned variable parameters based on preset constraints in the process of solving the optimization model using the adaptive genetic algorithm mentioned later, it is possible to ultimately obtain the corrected data of the expected business volume of the new energy station for the scheduled power business based on rule constraints, thereby improving the total revenue of the new energy station in the scheduled power business while ensuring the stable operation of the power grid.

[0056] In addition, by considering variable parameters based on preset constraints for constraining new energy sites in terms of power business, the impact of new energy volatility on the power grid can be effectively reduced.

[0057] In addition, as an example, step S102 may further include the following steps S1021 and S1022:

[0058] In step S1021, the new energy station environment data and the preset supply and demand space data are input into a supply and demand space prediction model based on a preset machine learning algorithm to obtain the first predicted supply and demand space data.

[0059] As an example, one can use Figure 2 The method shown is used to train the above-mentioned supply and demand space prediction model based on the preset machine learning algorithm. Figure 2 is a flowchart illustrating a training method 200 for a supply-demand space prediction model based on a preset machine learning algorithm according to an embodiment of the present disclosure.

[0060] Reference Figure 2 , in step S201, historical supply and demand space data and preset boundary conditions are obtained.

[0061] Here, the historical supply and demand space data may also be referred to as the historical actual supply and demand space, and include historical new energy power load data and historical new energy output data, and these data may be standardized before being used, but the present disclosure is not limited thereto.

[0062] In addition, the preset boundary conditions can represent the conditions corresponding to some extreme situations in the supply and demand space forecasting process, and can help to better fit the forecast curve of extreme situations. For example, during the scheduled holiday, the user-side electricity consumption for the scheduled group of people is significantly reduced, so this situation can be used as an extreme situation to preset the boundary conditions.

[0063] In step S202, the historical supply and demand space data is subjected to correlation analysis by combining preset boundary conditions to obtain supply and demand space characteristic data. Here, the supply and demand space characteristic data represents the relevant characteristics used to predict the supply and demand space.

[0064] In step S203, the supply and demand space characteristic data and the historical supply and demand space data are input into a supply and demand space prediction model based on a preset machine learning algorithm to obtain second predicted supply and demand space data.

[0065] In step S204, model parameters of a supply and demand space prediction model based on a preset machine learning algorithm are adjusted according to the historical supply and demand space data and the second predicted supply and demand space data.

[0066] As an example, the preset machine learning algorithm may be a gradient boosting tree (also referred to as CatBoostRegressor) machine learning algorithm, but the present disclosure is not limited thereto.

[0067] As a specific example, when using a trained supply and demand space prediction model based on a preset machine learning algorithm to perform step S1021, the input of the supply and demand space prediction model includes the disclosed supply and demand space data for the first predetermined number of days in the future and other data including climate data, and the output of the supply and demand space prediction model includes the predicted supply and demand space data for the second predetermined number of days in the future. Here, the second predetermined number of days can be changed according to business needs.

[0068] For example, public data such as climate, load, and output of the area to be predicted can be integrated, and the time span for predicting the supply and demand spatial data can be set according to business needs. If the supply and demand data for the next three days is set to be disclosed according to business needs, and the business requires the time span to be extended to the next seven days, the first predetermined number of days can be three days, and the predicted supply and demand spatial data can be the supply and demand spatial data for the next 4-7 days.

[0069] By utilizing the predicted supply and demand space data generated by the above supply and demand space prediction model, more effective supply and demand space data can be provided for the expected business power optimization model based on the adaptive genetic algorithm, so as to help obtain more accurate revised expected business power.

[0070] In step S1022, the new energy station basic data, the first predicted supply and demand space data and the initial expected power generation data are input into the expected business electricity optimization model based on the adaptive genetic algorithm to obtain the optimized revised expected business electricity.

[0071] As an example, one can use Figure 3 The method shown is used to train the above-mentioned expected business power optimization model based on the adaptive genetic algorithm. Figure 3is a flowchart showing a training method 300 of an expected service power optimization model based on an adaptive genetic algorithm according to an embodiment of the present disclosure.

[0072] Reference Figure 3 , in each iteration cycle, the following steps S301 to S306 are performed:

[0073] First, in step S301, a plurality of initial population individuals corresponding to a sequence of initial expected business electricity are obtained.

[0074] As an example, the above step S301 may represent an operation of initializing the population, for example, To represent, where i = 1, 2, ..., N p , N p is the number of individuals in the population, D is the expected transaction quantity sequence Q to be solved l,t As an example, we can set N p =50, but not limited to this.

[0075] Next, the following steps S302 to S305 are performed for every two initial population individuals:

[0076] In step S302, two initial population individuals are selected from multiple initial population individuals. and Here, g represents the generation to which the individual belongs, that is, the g-th generation individual.

[0077] Here, when selecting individuals from the population, for example, a proportional selection operator method may be used to select several individuals for crossover according to the fitness values ​​of the individuals from the population, and individuals with larger fitness values ​​have higher probabilities of being selected.

[0078] In step S303, the two selected initial population individuals are crossover processed based on the preset crossover probability p to obtain two crossover population individuals after the crossover. and The specific expressions are shown in the following formula (1) and formula (2):

[0079]

[0080] in, Represent the individuals of the gth generation The value after the crossover operation of the dth dimension.

[0081] In addition, the expression of the crossover probability is shown in the following formula (3):

[0082]

[0083] in,

[0084] In addition, p max 、p min are the maximum and minimum crossover probabilities, respectively. For example, in the example p max 、p min The values ​​can be 0.7 and 0.3 respectively.

[0085] According to the embodiments of the present disclosure, the fitness value proportions of the two crossover individuals are neutralized by taking the average of the fitness value proportions of the above two probabilities, thereby effectively enhancing the search capability.

[0086] also, All represent fitness functions. Taking as an example, its expression is the following formula (4):

[0087]

[0088] Among them, R E is the electricity fee income of the new energy station for the scheduled power business (for example, for the scheduled electric energy business), and its expression is the following formula (5):

[0089]

[0090] Among them, Q r,t is the expected power generation of the new energy station in period t, P r,t is the real-time node electricity price of the new energy station in period t, Q l,t is the expected transaction volume of the new energy station in period t, P l,t is the expected average transaction price of the new energy station in period t, P a,t is the unified settlement reference point electricity price for the entire network. Here, P r,t and P a,t The interval value of can be obtained using the above-mentioned supply and demand space prediction model.

[0091] In addition, R L It is the cost of recovering the shortfall of electricity volume of the new energy station for the scheduled electricity business (at this time, the power generation status of the new energy station is over-generation, and the value is negative), and its expression is the following formula (6):

[0092]

[0093] Among them, λ 1 To calculate the position ratio coefficient for the transaction power shortage recovery fee, for example, it can be 85% for the preset business; R P is the weighted average price of the nodes of the new energy station for the scheduled power business; L The weighted average price for a power generation area over a preset period (e.g., medium to long term).

[0094] In addition, R X The recovery fee R for the new energy station for the scheduled power business based on the preset business conditions B and compensation cost R H , and its expression is the following formula (7):

[0095] R X =R B +R H (7)

[0096] Among them, when hour, R H =0.

[0097] when hour, R B =0.

[0098] When both of the above conditions are not met, R X =0.

[0099] Here, λ 2 is the compensation coefficient, λ 3 is the recovery factor.

[0100] For example, for the first business, the compensation factor may be 80% and the recovery factor may be 115%, and for the second business, the compensation factor may be 70% and the recovery factor may be 120%.

[0101] That is to say, taking the first business as an example, if the actual electricity charge is less than 80% of the predetermined electricity charge, it corresponds to being too low compared to the average transaction price, and compensation fees will be incurred; if the actual electricity charge is greater than 115% of the predetermined electricity charge, it corresponds to being too high compared to the average transaction price, and recovery fees will be incurred.

[0102] In the present disclosure, the above-mentioned recovery fee and compensation fee can be expressed as follows: when a new energy station participates in a scheduled power business (for example, power business), on the one hand, if the new energy station trades too much electricity, the electricity fee income will be negative, and the basic operation of the station cannot be guaranteed at this time; on the other hand, if the new energy station trades too little electricity, the electricity fee income will be too high, which is not conducive to the good order of business; for the above two aspects of the measurement factors, the power grid side will recover or compensate for the electricity fee that exceeds or is lower than the average transaction price within a certain range. Therefore, by adding the measurement factor of the recovery fee or compensation fee, the sustainable development of the new energy station can be guaranteed while maintaining the good order of business.

[0103] In step S304, the two crossover population individuals after the crossover are mutated based on the preset mutation probability, the preset mutation strength and the preset standard normal distribution coefficient to obtain two mutated population individuals after the mutation.

[0104] As an example, step S304 may further include the following steps S3041 and S3042:

[0105] In step S3041, when the preset parameter is less than the current preset mutation probability for the current stage, the product of the preset mutation intensity for the current stage and the preset standard normal distribution coefficient and the sum of the current initial population individuals are used as the mutated population individuals after mutation.

[0106] In step S3042, when the preset parameter is greater than or equal to the current preset mutation probability for the current stage, the current initial population individual is used as the mutated population individual after mutation.

[0107] According to an embodiment of the present disclosure, the specific variation formulas involved in step S304 are shown in the following formulas (8) and (9):

[0108]

[0109] Among them, r is a random number in the range [0,1), Z is a random variable of the standard normal distribution N(0,1), and p m represents the mutation probability, and σ represents the mutation intensity.

[0110] Here, according to an embodiment of the present disclosure, the above-mentioned mutation processing may include at least two stages of mutation processing, and the values ​​of the preset mutation probability and the preset mutation strength may both decrease nonlinearly as the order of the stages of the mutation processing increases.

[0111] As a specific example, in the early stage of iteration, the mutation probability and mutation strength values ​​can be set to be larger, which can enhance the algorithm's exploration ability, achieve algorithm performance balance, reduce space complexity, and avoid falling into local optimality; in the middle and late stages of iteration, the mutation probability and mutation strength values ​​can be set to be smaller, which can improve the algorithm's convergence speed and accuracy.

[0112] For example, as an example, for the relevant requirements of the preset power business, the mutation probability and mutation intensity can be set as follows: when the number of iterations g<500, p m =0.07, σ = 0.6; when the number of iterations is 500≤g<800, p m =0.035, σ =0.38; when the number of iterations g ≥ 800, p m =0.02, σ=0.15. Experiments have shown that for the preset power business, under the above settings, the model is more efficient in obtaining the optimal profit point. However, the values ​​of the above parameters are only examples, and other examples of the present disclosure are not limited thereto.

[0113] By adopting multi-stage mutation processing, compared with a single mutation probability and mutation intensity, a balance of algorithm performance can be achieved, avoiding the problem that a single numerical value is set too large, resulting in deviation in the optimization result, and a single numerical value is set too small, resulting in the inability to find the optimal solution within a fixed number of iterations.

[0114] Reference Figure 3 In step S305, the target cost function values ​​corresponding to the two mutated population individuals are compared with the target cost function values ​​corresponding to the corresponding population individuals in the two selected initial population individuals, and the two population individuals with larger target cost function values ​​are selected to replace the corresponding selected initial population individuals as the corresponding initial population individuals of the next iteration cycle. That is, the population individuals with larger fitness values ​​are selected to join the original population as the initial population individuals of the next generation cycle.

[0115] Here, in the iterative solution process to maximize the target cost function (for example, the above fitness function) (the decision variable is the expected transaction volume Q in each period l,t ), the following constraints are set, as shown in formula (10):

[0116]

[0117] Among them, Q R is the capacity of the new energy station, Q s Report the power generation capacity for the new energy site (for example, report the power generation capacity for the current month).

[0118] Next, in step S306, it is determined whether all the initial population individuals have been processed. If yes, this iteration ends, and if no, the next group of initial population individuals is processed.

[0119] Finally, when a preset termination condition is met, for example, a preset number of iterations is reached, the training method 300 of the expected service power optimization model based on the adaptive genetic algorithm ends, otherwise, the selection, crossover, and mutation cycle operations continue.

[0120] By adaptively and dynamically adjusting the crossover probability, mutation probability and mutation intensity, the search capability, convergence speed and accuracy of the population are improved, so as to obtain the optimal configuration strategy for the transaction volume of the new energy station for the scheduled power business in the preset period.

[0121] Below, refer to Figure 4 and Figure 5 To illustrate the execution logic of the optimization method according to the present disclosure and the system logic block diagram of the corresponding optimization system. Figure 4 is an execution logic block diagram showing an example of a method for optimizing the expected business power of a new energy station according to an embodiment of the present disclosure, and Figure 5It is a logical block diagram showing an optimization system for expected business electricity volume of a new energy station according to an embodiment of the present disclosure.

[0122] Reference Figure 4 and Figure 5 The server side can be used to provide input data of the optimization model to the control side (for example, including but not limited to new energy station capacity information, station information, weather information, supply and demand space information, expected power generation information, etc.). The input data can be received by the data collection module on the control side and further provided to the supply and demand space prediction module on the control side.

[0123] In addition, the control end can be used to execute the optimization method of the expected business power of the new energy station of the embodiment of the present disclosure, and as Figure 5 Each module of the control end shown can execute the corresponding steps in the optimization method 100 as described above, which will not be repeated here.

[0124] In addition, the user end may include a new energy station visualization platform system, an expected transaction power correction data reporting dispatch and business center. For example, the user end completes the expected transaction power correction operation of the new energy station in the predetermined business system by receiving the expected transaction power correction data from the control end.

[0125] As a specific example, according to business arrangements, the expected transaction volume correction data for a predefined number of days in the future (for example, 3-7 days) can be generated based on the latest public data at a fixed time every day, and the correction operation of the expected transaction volume can be completed when the correction channel is opened.

[0126] By timely and effectively correcting the expected transaction volume of new energy stations for scheduled power business, it is possible to achieve a balance between obtaining maximum total revenue from scheduled power business and maintaining grid stability for new energy stations.

[0127] Figure 6 is a block diagram showing an optimization device 600 for expected business electricity quantity of a new energy station according to an embodiment of the present disclosure.

[0128] Reference Figure 6 The device 600 for optimizing the expected business electricity of a new energy station includes: a data acquisition module 610 and a data correction module 620 .

[0129] According to an embodiment of the present disclosure, the data acquisition module 610 can perform the following operations: obtain new energy station basic data, new energy station environmental data, preset supply and demand space data and the initial expected power generation data of the new energy station for the power business, and the initial expected power generation data for the power business includes the initial expected business electricity and the initial expected power generation.

[0130] According to an embodiment of the present disclosure, the data correction module 620 can perform the following operations: input the new energy station basic data, the new energy station environmental data, the preset supply and demand space data and the initial expected power generation data into the expected business power optimization model based on the adaptive genetic algorithm to obtain the optimized corrected expected business power.

[0131] In the present disclosure, an expected business electricity optimization model based on an adaptive genetic algorithm can be trained based on a target cost function representing a target business parameter for a new energy station for a power business. Here, the target business parameter is the sum of the electricity price revenue for the new energy station for the power business and a variable parameter based on a preset constraint condition.

[0132] For example, the basic data of the new energy station may include data related to the power generation capacity of the new energy station and data related to the location of the new energy station, and the preset supply and demand space data may include expected new energy power load data and expected new energy output data.

[0133] In addition, according to an embodiment of the present disclosure, variable parameters based on preset constraints may correspond to the following two items: the over-generation power that needs to be recovered at the new energy station corresponding to the preset transaction power gap for the power business and the power recovery or compensation corresponding to the frequency and peak regulation power dispatch of the new energy station based on the preset business conditions.

[0134] As an example, the data correction module inputs the new energy station basic data, the new energy station environment data, the preset supply and demand space data and the initial expected power generation data into the expected business power optimization model based on the adaptive genetic algorithm, and the operation of obtaining the optimized corrected expected business power may include the following operations 1) and 2):

[0135] In operation 1), the new energy station environment data and the preset supply and demand space data are input into a supply and demand space prediction model based on a preset machine learning algorithm to obtain the first predicted supply and demand space data.

[0136] Here, as an example, the supply and demand space prediction model based on the preset machine learning algorithm can be trained through the following operations 11) to 14):

[0137] In operation 11), historical supply and demand space data and preset boundary conditions are obtained.

[0138] In operation 12), the historical supply and demand space data is subjected to correlation analysis (ie, data classification) by combining preset boundary conditions to obtain supply and demand space characteristic data.

[0139] In operation 13), the supply and demand space characteristic data and the historical supply and demand space data are input into a supply and demand space prediction model based on a preset machine learning algorithm to obtain second predicted supply and demand space data.

[0140] In operation 14), according to the historical supply and demand space data and the second predicted supply and demand space data, the model parameters of the supply and demand space prediction model based on the preset machine learning algorithm are adjusted.

[0141] In operation 2), the new energy station basic data, the first predicted supply and demand space data and the initial expected power generation data are input into the expected business electricity optimization model based on the adaptive genetic algorithm to obtain the optimized revised expected business electricity.

[0142] As an example, the expected service power optimization model based on the adaptive genetic algorithm can be trained in the following manner: In each iteration cycle, the following processes 1) and 2) are performed:

[0143] In process 1), a plurality of initial population individuals corresponding to the sequence of initial expected business electricity are obtained.

[0144] In process 2), the following processes 21) to 24) are performed for every two initial population individuals:

[0145] In process 21), two initial population individuals are selected from a plurality of initial population individuals.

[0146] In process 22), the two selected initial population individuals are crossover processed based on a preset crossover probability to obtain two crossover population individuals after the crossover.

[0147] In process 23), the two crossover population individuals after the crossover are mutated based on the preset mutation probability, the preset mutation intensity and the preset standard normal distribution coefficient to obtain two mutated population individuals after the mutation.

[0148] Here, as an example, the mutation process may include at least two stages of mutation process, and the values ​​of the preset mutation probability and the preset mutation strength may both decrease nonlinearly as the order of the stages of the mutation process increases.

[0149] In addition, as an example, for the mutation processing of each stage, the specific operation of processing 23) may include: when the preset parameter is less than the current preset mutation probability for the current stage, the product of the preset mutation intensity for the current stage and the preset standard normal distribution coefficient and the sum of the current initial population individuals are used as the mutated population individuals after mutation; when the preset parameter is greater than or equal to the current preset mutation probability for the current stage, the current initial population individuals are used as the mutated population individuals after mutation.

[0150] In processing 24), the target cost function values ​​corresponding to the two mutated population individuals after mutation are compared with the target cost function values ​​corresponding to the corresponding population individuals in the two selected initial population individuals, and the two population individuals with larger target cost function values ​​are selected to replace the corresponding selected initial population individuals as the corresponding initial population individuals for the next iteration cycle.

[0151] It should be noted that the operations performed on the above-mentioned structural frames can be referred to in Figure 1 The related contents described are similar and will not be repeated here.

[0152] Figure 7 is a block diagram illustrating a computer device 700 according to an embodiment of the present disclosure.

[0153] Reference Figure 7 According to an embodiment of the present disclosure, a computer device 700 may include a processor 710 and a memory 720. The processor 710 may include (but not limited to) a central processing unit (CPU), a digital signal processor (DSP), a microcomputer, a field programmable gate array (FPGA), a system on a chip (SoC), a microprocessor, an application-specific integrated circuit (ASIC), etc. The memory 720 may store computer executable instructions to be executed by the processor 710. The memory 720 includes a high-speed random access memory and / or a non-volatile computer-readable storage medium. When the processor 710 executes the computer executable instructions stored in the memory 720, the optimization method of the expected business power of the new energy station as described above can be implemented.

[0154] In addition, the computer device 700 may also include a video display (such as a liquid crystal display) and a user interaction interface (such as a keyboard, a mouse, a touch input device, etc.) All components of the computer device 700 may be connected to each other via a bus and / or a network.

[0155] According to the optimization method of the expected business power of the new energy station according to the embodiment of the present disclosure, it can be written as a computer program / instruction to form a computer program product and stored on a computer-readable storage medium. When the computer program / instruction is executed by the processor, the optimization method as described above can be implemented. When the instructions in the computer-readable storage medium are executed by the processor of the electronic device / server, the electronic device / server is enabled to execute the optimization method as described above. Examples of computer-readable storage media include: read-only memory (ROM), random access programmable read-only memory (PROM), electrically erasable programmable read-only memory (EEPROM), random access memory (RAM), dynamic random access memory (DRAM), static random access memory (SRAM), flash memory, non-volatile memory, CD-ROM, CD-R, CD+R, CD-RW, CD+RW, DVD-ROM, DVD-R, DVD+R, DVD-RW, DVD+RW, DVD-RAM, BD-ROM, BD-R, BD-R LTH, BD-RE, Blu-ray or optical disk storage, hard disk drive (HDD), solid state drive (SSD), card storage (such as, multimedia card, secure digital (SD) card or extreme digital (XD) card), magnetic tape, floppy disk, magneto-optical data storage device, optical data storage device, hard disk, solid state disk and any other device, any other device is configured to store computer program and any associated data, data files and data structures in a non-transitory manner and provide the computer program and any associated data, data files and data structures to a processor or computer so that the processor or computer can execute the computer program. In one example, the computer program and any associated data, data files and data structures are distributed on a networked computer system so that the computer program and any associated data, data files and data structures are stored, accessed and executed in a distributed manner by one or more processors or computers.

[0156] According to the optimization method and device for the expected business electricity volume of the new energy station according to the embodiments of the present disclosure, an optimization model for the expected business electricity volume of the new energy station is established by considering the information related to the new energy station and the information related to the business demand and combining with the preset constraints, thereby achieving the purpose of correcting the expected business electricity volume, thereby improving the target profit of the new energy station for the business electricity volume while ensuring the stability of the power grid.

[0157] On the other hand, by setting adaptive crossover probability, mutation probability and mutation intensity, the search capability, convergence speed and accuracy of the optimization model algorithm for the expected business power of new energy stations are effectively improved.

[0158] Although some embodiments of the present disclosure have been disclosed and described, it will be appreciated by those skilled in the art that modifications and variations may be made to the embodiments without departing from the concept and spirit of the present disclosure, the scope of which is defined by the claims and their equivalents.

Claims

1. A method for optimizing the expected business electricity of a new energy station, characterized in that: The optimization method comprises: Obtaining new energy station basic data, new energy station environmental data, preset supply and demand space data, and initial expected power generation data for power business of the new energy station, wherein the initial expected power generation data for power business includes initial expected business electricity and initial expected power generation; Input the new energy station basic data, the new energy station environmental data, the preset supply and demand space data and the initial expected power generation data into the expected business power optimization model based on the adaptive genetic algorithm to obtain the optimized revised expected business power. Among them, based on the target cost function representing the target business parameters of the new energy station for the power business, the expected business electricity optimization model based on the adaptive genetic algorithm is trained, and the target business parameters are the sum of the electricity price revenue of the new energy station for the power business and the variable parameters based on preset constraints.

2. The optimization method according to claim 1, characterized in that: The new energy station basic data includes data related to the power generation capacity of the new energy station and data related to the location of the new energy station, and the preset supply and demand space data includes expected new energy power load data and expected new energy output data.

3. The optimization method according to claim 1, characterized in that: The step of inputting the new energy station basic data, the new energy station environmental data, the preset supply and demand space data and the initial expected power generation data into the expected business power optimization model based on the adaptive genetic algorithm to obtain the optimized revised expected business power includes: Inputting the new energy station environment data and the preset supply and demand space data into a supply and demand space prediction model based on a preset machine learning algorithm to obtain first predicted supply and demand space data; The new energy station basic data, the first predicted supply and demand space data and the initial expected power generation data are input into an expected business electricity quantity optimization model based on an adaptive genetic algorithm to obtain an optimized revised expected business electricity quantity.

4. The optimization method according to claim 1, characterized in that: The expected business power optimization model based on the adaptive genetic algorithm is trained in the following manner: In each iteration cycle, the following processing is performed: Acquire a plurality of initial population individuals corresponding to the sequence of the initial expected business electricity; The following processing is performed for every two initial population individuals: Selecting two initial population individuals from the multiple initial population individuals; Performing crossover processing on the two selected initial population individuals based on a preset crossover probability to obtain two crossover population individuals after the crossover; Based on a preset mutation probability, a preset mutation strength and a preset standard normal distribution coefficient, the two crossover population individuals after the crossover are mutated to obtain two mutated population individuals after the mutation; The target cost function values ​​corresponding to the two mutated population individuals after the mutation are compared with the target cost function values ​​corresponding to the corresponding population individuals in the two selected initial population individuals, and two population individuals with larger target cost function values ​​are selected to replace the corresponding selected initial population individuals as the corresponding initial population individuals of the next iteration cycle.

5. The optimization method according to claim 4, characterized in that: The mutation process includes at least two stages of mutation process, and the values ​​of the preset mutation probability and the preset mutation strength both decrease nonlinearly as the order of the stages of the mutation process increases.

6. The optimization method according to claim 4, characterized in that: The mutation processing includes at least two stages of mutation processing, and for each stage of mutation processing, the steps of performing mutation processing on the current initial population individuals based on a preset mutation probability, a preset mutation intensity and a preset standard normal distribution coefficient to obtain the mutated mutated population individuals include: When the preset parameter is less than the current preset mutation probability for the current stage, the sum of the product of the preset mutation intensity for the current stage and the preset standard normal distribution coefficient and the current initial population individual is used as the mutant population individual after the mutation; In the case where the preset parameter is greater than or equal to the current preset mutation probability for the current stage, the current initial population individual is used as the mutated population individual after the mutation.

7. The optimization method according to claim 3, characterized in that: The supply and demand space prediction model based on the preset machine learning algorithm is trained in the following manner: Obtain historical supply and demand spatial data and preset boundary conditions; By combining preset boundary conditions, the historical supply and demand space data is subjected to correlation analysis to obtain supply and demand space characteristic data; Inputting the supply and demand space characteristic data and the historical supply and demand space data into the supply and demand space prediction model based on the preset machine learning algorithm to obtain second predicted supply and demand space data; According to the historical supply and demand space data and the second predicted supply and demand space data, the model parameters of the supply and demand space prediction model based on the preset machine learning algorithm are adjusted.

8. The optimization method according to claim 1, characterized in that: The variable parameters based on preset constraints correspond to the following two items: the over-generation power that needs to be recovered at the new energy station corresponding to the preset transaction power gap for the power business, and the power recovery or compensation corresponding to the frequency and peak regulation power dispatch of the new energy station based on the preset business conditions.

9. An optimization device for expected business electricity of a new energy station, characterized in that: The optimization device comprises: The data acquisition module is configured to: acquire new energy station basic data, new energy station environmental data, preset supply and demand space data, and initial expected power generation data for power business of the new energy station, wherein the initial expected power generation data for power business includes initial expected business power and initial expected power generation; The data correction module is configured to: input the new energy station basic data, the new energy station environmental data, the preset supply and demand space data and the initial expected power generation data into the expected business power optimization model based on the adaptive genetic algorithm to obtain the optimized corrected expected business power; Among them, based on the target cost function representing the target business parameters of the new energy station for the power business, the expected business electricity optimization model based on the adaptive genetic algorithm is trained, and the target business parameters are the sum of the electricity price revenue of the new energy station for the power business and the variable parameters based on preset constraints.

10. A computer program product comprising computer instructions, characterized in that: When the computer instructions are executed by at least one processor, the at least one processor is prompted to execute the method for optimizing the expected business power of the new energy station as described in any one of claims 1 to 8.

Images