New energy power station power generation on-grid auxiliary decision method

Abstract

The application discloses a new energy power station power generation on-grid auxiliary decision method, the auxiliary decision method includes the following process: establishing the power load model, the new energy power generation model, the energy storage release model, using Monte Carlo simulation for prediction; according to the prediction result of the power load model, the new energy power generation model, the energy storage release model, the new energy power station power generation on-grid auxiliary decision is obtained; the new energy power station power generation on-grid auxiliary decision is through model prediction future set time's power load condition and new energy power generation condition, obtains the new energy power generation capacity strategy when the power load is in the peak stage;The new energy consumption is improved, but at the same time, the corresponding reserve is reserved, so as to provide an emergency plan when the set condition affects the new energy power generation capacity.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of coordinated peak shaving of centralized new energy power stations and energy storage power stations, and in particular to a guiding method for coordinated capacity based on data mining. BACKGROUND

[0002] In recent years, in order to achieve the "double carbon" goal, the composition of electric power energy has undergone major adjustments. With the rapid changes in the power grid pattern and power source structure, the State Grid is facing new challenges. The proportion of new energy access is increasing year by year, and new energy generation has its volatility and randomness. The access of new energy to the power grid has brought difficulties to active and reactive power adjustment. Currently, peak shaving of the power grid is almost entirely borne by conventional thermal power units, but as the carbon emission reduction task intensifies, traditional thermal power units will inevitably be gradually reduced.

[0003] New energy generation is usually greatly affected by geographical location and weather conditions. For example, wind power generation is affected by seasons, wind power and wind direction vary greatly, and power generation fluctuates greatly, and this fluctuation period is uncontrollable. If a large amount of electricity generated at a time when the power grid load is small is abandoned, it will cause great waste; and when the load is large, there may be no wind, and it cannot provide power generation. Therefore, if we want to improve the penetration rate of new energy, we must solve the problem of its large volatility.

[0004] Energy storage power stations are a relatively fast-growing form of power stations in recent years. The biggest difference between them and traditional power stations is that they do not really generate electricity, but convert electrical energy into other forms of energy (such as chemical energy, potential energy, etc.), and then convert these other forms of energy into electricity to supply the power grid when needed. The peak-to-valley regulation capability of energy storage power stations is its greatest feature.

[0005] The current dispatching mode of the State Grid is still mainly based on thermal power generation, and the peak shaving and valley regulation are also mainly based on thermal power units with peak shaving capability. Usually, the power generation of each station is dispatched according to the current load condition, and the traditional dispatching method is used, which lacks overall green energy consumption consideration supported by accurate prediction data, and is not conducive to improving the penetration rate of clean energy. To consider how to improve the consumption proportion of new energy and improve the cleanliness of power generation, so that the power grid is more secure and reliable, is the ultimate pursuit of the entire power system. SUMMARY

[0006] In order to solve the problems in the prior art, the present application aims to analyze historical electricity consumption and new energy generation data through data mining, establish a power load model and a new energy generation model, and provide data basis for auxiliary decision-making of new energy power station power generation, so as to improve the penetration rate of new energy and the ability to participate in peak shaving.

[0007] The present application achieves the above effects through the following technical solutions:

[0008] The new energy power station power generation on-grid auxiliary decision method comprises the following processes:

[0009] A power consumption load model, a new energy power generation model, and an energy storage release model are established, and Monte Carlo simulation is used for prediction;

[0010] According to the prediction results of the power consumption load model, the new energy power generation model, and the energy storage release model, a new energy power station power generation on-grid auxiliary decision is obtained;

[0011] The new energy power station power generation on-grid auxiliary decision predicts the power consumption load situation and the new energy power generation situation in the future set time through the model, obtains a new energy on-grid power amount strategy when the power consumption load is in the peak value stage, improves the consumption of new energy, and reserves a corresponding reserve at the same time to serve as an emergency plan when a set situation affects the new energy power amount.

[0012] Further, the new energy power generation amount strategy when the power consumption load is in the peak value stage is specifically:

[0013] The following formula is used to adjust the new energy on-grid proportion:

[0014]

[0015] Eupload refers to the on-grid power amount at the moment, Ev refers to the new energy power amount predicted through the new energy power generation model at the moment, Ef refers to the load predicted by using the power consumption load model at the moment, the denominator ∑En refers to the summary of the load in the working period of the day, and σ refers to the on-grid storage coefficient, which is adjusted according to the energy storage station storage ∑Ecan-release calculated by using the energy storage release model, is 1 when ∑Ecan-release is greater than ∑Eupload, is when ∑Ecan-release is less than ∑Eupload, and ∑Ecan-release / ∑Eupload.

[0016] Further, the specific process of establishing the power consumption load model, the new energy power generation model, and the energy storage release model is:

[0017] Historical data of power generation and power consumption of the new energy power station and the energy storage power station are obtained, and the historical data includes power consumption load data, new energy power generation data, new energy station power generation equipment information and real-time monitoring data, energy storage station energy storage equipment information and real-time monitoring data;

[0018] The historical data are analyzed to obtain a first constraint condition affecting the power consumption load number calculation, and then the power consumption load data are obtained according to the first constraint condition and the time series method and the discrete data sampling method to obtain the power consumption load model;

[0019] Segmenting and analyzing the historical data to obtain a second constraint condition affecting new energy power generation calculation, and then using time series method and discrete data sampling method to obtain the new energy power generation model according to the second constraint condition and the obtained new energy power generation data;

[0020] According to the characteristics of the energy storage types, clustering analysis is performed to determine the amount of electricity that can be received and stored by the energy storage station and the amount of electricity that can be released to the power grid at a set time, and an energy storage release model is established.

[0021] Further, the process of establishing the electricity load model is as follows:

[0022] The historical data is segmented into time slices in units of a set time length according to time sequence, and the segmented data is added with a set additional attribute to obtain processed historical data;

[0023] The processed historical data is extracted according to different attributes, and the distribution thereof is analyzed to obtain a first constraint condition affecting the electricity load, and then the electricity load model is obtained according to the first constraint condition and the obtained electricity load data:

[0024]

[0025] wherein X is a power generation data sequence retrieved from the database according to the additional attribute condition, fi is an electricity load weight value, a is a growth coefficient, s, t and w are additional attributes.

[0026] Further, the process of establishing the new energy power generation model is as follows:

[0027] The historical data is segmented into time slices in units of a set time length according to time sequence, and the segmented data is added with a set additional attribute to obtain processed historical data;

[0028] The processed historical data is extracted according to different attributes, and the distribution thereof is analyzed to obtain a second constraint condition affecting new energy power generation, and then the new energy power generation model is obtained according to the second constraint condition and the obtained new energy power generation data:

[0029]

[0030] wherein X is a power generation data sequence retrieved from the database according to the additional attribute condition, fi is an electricity load weight value, a is a growth coefficient, s, t and w are additional attributes.

[0031] Further, the process of establishing the energy storage release model is as follows:

[0032] The power generation process is divided into pumped storage, battery energy storage, hydrogen energy storage and compressed air energy storage according to the energy storage types;

[0033] According to the clustering analysis of the characteristics of the energy storage type, the energy storage station can receive the stored power and release the generated power to the power grid at a set time, and a storage release model is established.

[0034] As a preferred embodiment of the present application, the additional attributes are s: season, t: air temperature, and w: time period.

[0035] As a preferred embodiment of the present application, the additional attributes are l: latitude and longitude coordinates, s: season, t: air temperature, and w: time period.

[0036] As a preferred embodiment of the present application, the Monte Carlo simulation is used for prediction, specifically: using Monte Carlo simulation to analyze the influence of changes in the first constraint condition on the electricity load, and drawing a change trend curve; using Monte Carlo simulation to analyze the influence of changes in the second constraint condition on new energy power generation, and drawing a change trend curve.

[0037] Advantages

[0038] The advantage of the coordinated peak clipping and valley filling strategy of the new energy power station and the energy storage power station based on data mining is that, under the premise of ensuring the safe operation of the power grid, the green new energy penetration rate can be as high as possible, the new energy on-grid power strategy when the electricity load is in the peak stage can be obtained, the penetration ratio of new energy and the ability to participate in peak regulation can be improved, and carbon emissions can be reduced; and a new energy and energy storage peak regulation scheme is provided for the original single thermal power peak regulation mode. BRIEF DESCRIPTION OF DRAWINGS

[0039] The drawings accompanying the specification of the present application serve to provide a further understanding of the present application, and the illustrative embodiments of the present application and their descriptions serve to explain the present application, and do not constitute an improper limitation on the present application:

[0040] Figure 1 A flowchart of the model construction process of the present application is shown in the figure;

[0041] Figure 2 Influence of different temperatures and light intensities on photovoltaic power generation;

[0042] Figure 3 A sample day electricity load and new energy power generation comparison data analysis graph. DETAILED DESCRIPTION

[0043] The present application will be further described below in conjunction with the drawings:

[0044] Example 1

[0045] The present application provides a new energy power station power generation on-grid auxiliary decision-making method, which comprises the following processes:

[0046] Establishing a power consumption load model, a new energy generation model, and an energy storage release model, and using Monte Carlo simulation for prediction;

[0047] According to the prediction results of the power consumption load model, the new energy generation model, and the energy storage release model, obtaining a new energy power station generation on-grid auxiliary decision;

[0048] The new energy power station generation on-grid auxiliary decision predicts the power consumption load and the new energy generation in a future set time through a model, obtains a new energy on-grid power quantity strategy when the power consumption load is in a peak value stage, improves the consumption of new energy, and reserves a corresponding reserve to prepare for an emergency plan when a set condition affects the new energy power quantity.

[0049] Further, the new energy on-grid power quantity strategy when the power consumption load is in the peak value stage is specifically:

[0050] The following formula is used to adjust the new energy on-grid proportion:

[0051]

[0052] Eupload is the on-grid power quantity at the moment, Ev is the new energy power quantity predicted through the new energy generation model at the moment, Ef is the load predicted by using the power consumption load model at the moment, the denominator ΣEn is the summary of the daily working period load, and σ is the on-grid storage coefficient, which is adjusted according to the energy storage station storage quantity ΣEcan-release calculated by using the energy storage release model, is 1 when ΣEcan-release is greater than ΣEupload, is when ΣEcan-release is less than ΣEupload, and is ΣEcan-release / ΣEupload. By using the auxiliary decision method provided in the application, the new energy utilization is greatly improved, and the area of coal power is also greatly reduced under the condition of reaching the original load. The operation process of the technical solution is shown in the accompanying Figure 1 figure, and the specific steps are described as follows:

[0053] Power consumption load model

[0054] Basic idea: Collect power consumption load data in a certain geographical range, segment the data according to time sequence, combine and store the load data on a time slice with important factor data such as weather on the time slice, use discrete analysis to determine the normal distribution of the load data under the same conditions, determine the main points of attention, develop a value scheme, establish a load prediction model, the purpose of the model is to input specified influencing factors to obtain a load prediction value. The model is used to predict the load value under the same conditions in the future. It is used for auxiliary decision.

[0055] Example:

[0056] 1. For example, the electricity load data of a city in the past three years has been obtained (this data is constantly changing at intervals of 5 seconds);

[0057] 2. The data is divided into time slices of 5 seconds according to the time sequence (the time slice can only be greater than or equal to the minimum data sampling frequency, the smaller the time interval, the more convenient the data is to use), so that a time slice has its basic attributes (for example: date, season, time interval, etc.), some directly related attributes (for example: weather, temperature, pressure, light, wind speed, wind direction, etc.), some indirectly related attributes (for example: the number of users contained in the range, user capacity, user nature classification, etc.), and some additional attributes (for example: whether it is a working period, whether it is during the epidemic control period, etc.);

[0058] 3. Now there are 3*365*24*60*(60 / 5) time slices corresponding to the load data, and these time slices have their own attributes.

[0059] 4. Data extraction can be performed according to different attribute data, and the distribution of the data can be analyzed (for example: the data can be sampled and analyzed according to three dimensions of season, weather, and working period), and N discrete distributed load data can be extracted by filtering conditions, and the data distribution can be analyzed using mathematical methods such as central tendency, dispersion degree, and shape, and the use of which parameters to establish the electricity load model can be determined according to the actual distribution of the data. This model will have different types according to the data granularity and different conditions, in order to facilitate understanding, using the weighted mean method with the above three attribute inputs (s: season, t: temperature, w: working time), an abstract example is made, this model can be summarized as The form, where X is the power generation data sequence retrieved from the database according to the above search conditions, f is the power generation weight value, and a is the growth coefficient (such as indicating that the electricity consumption will increase year by year with the increase of CPI, a is usually greater than 1), and the previous analysis process is to determine the selection of the weighted value and analyze how to select the growth coefficient;

[0060] 5. When the model is established, input s, t, w to get f(s, t, w), which is the load prediction value under this condition; according to f(s, t, w), the future situation (such as weather forecast, etc.) can be used as input parameters to obtain the load prediction value at the future time point;

[0061] 6. By repeatedly calling this model, a load prediction curve for the next n hours can be generated, and by drawing the prediction curve, the trend of future load change can be analyzed, and the strategy can be adjusted in time.

[0062] New energy power generation model

[0063] Basic idea: Collect real-time power generation data of a new energy station, segment the data by time series, combine the load data on the time slice with the important factor data on this time slice, use the normal distribution of power generation data under the same conditions to determine the main focus, develop the value scheme, and establish the new energy power generation model. The purpose of this model is to input the specified influencing factors to get the predicted value of power generation. The model is used to predict the new energy power generation under the same conditions in the future. It is used to assist decision-making.

[0064] Example:

[0065] 1. For example, the basic data of all photovoltaic cell groups of a photovoltaic station has been obtained, including installed capacity, operation time, conversion rate, etc. Real-time power generation data of each unit of the photovoltaic station in the past three years (this data is constantly changing at intervals of 5 seconds);

[0066] 2. Segment it into time slices of 5 seconds according to the time series (the time slice can only be greater than or equal to the minimum data sampling frequency, the smaller the time interval, the easier the data to use), so a time slice has its basic attributes (such as date, season, time interval, etc.), some direct correlation attributes (such as weather, temperature, pressure, light, wind speed, wind direction, etc.), some indirect relationship attributes (such as station latitude and longitude coordinates, cell group failure rate, station power quality index, station high and low voltage ride-through capability, station anti-islanding capability, grid-connected inverter conversion rate, station grid-connected power, station control system power consumption, etc.), and some additional attributes (such as whether it is a working period or a maintenance period, etc.);

[0067] 3. Now there are 3*365*24*60*(60 / 5) time slices corresponding to real-time power generation, and these time slices have their own attributes.

[0068] 4. Data extraction can be performed according to different attribute data, and the distribution of the data can be analyzed (for example, data can be sampled and analyzed according to four dimensions of latitude, longitude, season, weather, and time period). Select the time slice with the coordinates of north latitude 31 degrees east longitude 121 degrees as the center point, winter, cloudy, and evening 16-17 hours for power generation analysis. Through filtering conditions, N discrete distributed power generation data is extracted, and data distribution analysis is performed using mathematical methods such as central tendency, dispersion degree, and shape. According to the actual distribution of the data, decide which parameters to use to establish the new energy power generation model. This model will have different types according to the data granularity and different conditions. In order to facilitate understanding, using the weighted mean method with the above four attribute inputs (l: latitude and longitude coordinates, s: season, t: temperature, w: time period), an abstract example is made. This model can be summarized as The form of f (X) = βf (X) + (1-β) f (X), where X is the power generation data sequence retrieved from the database according to the above search conditions, f is the power generation weight value, and β is the redundancy coefficient (usually less than 1). The previous analysis process is to determine the selection of the weighted value and analyze how to select the redundancy coefficient;

[0069] 5. When the model is established, input l, s, t, w to obtain f (l, s, t, w), which is the power generation prediction value under this condition; according to f (l, s, t, w), the future situation (such as weather forecast, etc.) can be taken as an input parameter to obtain the power generation prediction value at the future time point;

[0070] 6. By repeatedly calling this model, a power generation prediction curve for the next n hours can be generated, and by drawing the prediction curve, the trend of future power generation change can be analyzed, and the strategy can be adjusted in time.

[0071] Energy storage release model

[0072] Basic idea: Collect the energy storage index data of all operating energy storage stations, perform cluster analysis according to the characteristics of the energy storage type, determine the power that the energy storage station can receive and store and the power that the energy storage station can release to the grid at a certain time, and establish an energy storage release model. The purpose of this model is to input the station type and the current inventory to obtain the output of the storable power and the releasable power, which is used as an auxiliary decision for whether the new energy power generation is directly connected to the grid or stored for later use. In order to describe conveniently, this model does not consider the line loss of the transmission circuit from the new energy station to the energy storage station, and does not consider the energy loss caused by the storage of the energy storage station.

[0073] Example:

[0074] 1. For example, the basic data of all energy storage stations in a city have been obtained, such as the energy storage type, installed capacity, operation time, conversion rate, rated power, maximum output power, and the current storage data of each energy storage module of these stations (this data is constantly changing at intervals of 5 seconds) ;

[0075] 2. According to the energy storage type, it is classified into pumped storage, battery energy storage, hydrogen energy storage, compressed air energy storage, etc.;

[0076] 3. According to the commonness (such as installed capacity, rated power, maximum output power) and characteristics (such as pumped storage does not need inversion, battery energy storage has time-varying attenuation, different conversion rates of various types, etc.) of various types of energy storage, corresponding analysis is performed, and the storable power and the releasable power of the energy storage type are calculated according to the index of the energy storage type

[0077] where Erate is the rated capacity of the storage unit, P is the current power percentage of the storage unit, and Ez is the minimum reserved capacity of the storage unit​

[0078] 4. All the storable power is aggregated to obtain the city's storable power model

[0079]

[0080] 5. In the peak regulation auxiliary decision of the energy storage power station, the upper limit of the power that can be used for grid access and the power that can be stored can be calculated according to the real-time current storage data and new energy prediction data.

[0081] New energy grid access auxiliary decision

[0082] From the above analysis data, the characteristic curves of the power load, photovoltaic power generation and wind power generation of a certain feature day are extracted, as shown in the accompanying Figure 3

[0083] As can be seen from the figure, in the current state, due to the large fluctuation of new energy power generation, direct grid access has a greater impact on the stability of the power grid, so in order to ensure the stability of the power grid, a lower penetration rate is generally used, and when new energy problems occur, the defect of the power supply of new energy can be supplemented by thermal power. From this, a daily power consumption stacking diagram without energy storage participation under a fixed penetration rate is drawn, and the penetration rate is 27% analyzed from historical data.

[0084] In this embodiment, the coal power area ratio is large, and the peak value is very steep during the daytime peak period. It shows that the utilization rate of new energy is actually very low.

[0085] When the concept of energy storage power station regulation is introduced, because of the bottom-up of the energy storage power station, the penetration rate of new energy direct grid access can reach about 50%, and if the final grid access of energy storage is included, the penetration rate can reach a higher level. The following formula is used to adjust the new energy grid access ratio:

[0086]

[0087] E upload is the power at this moment, E v is the new energy output at this moment, E f is the predicted load at this moment, the denominator ∑E n is the aggregation of daily working period load, β is the grid storage coefficient, β increases with the storage ∑E can-release of the energy storage power station, and is greater than ∑E upload , 1, less than ∑E can-release , and ∑E upload , the utilization of new energy is greatly improved, and under the condition of reaching the original load, the area of coal power is also greatly reduced.

[0088] ​Currently, the dispatching strategy for new energy is to combine the short-term power prediction system provided by the new energy power station for the next 15 minutes to 4 hours with the real-time power grid load to dispatch the new energy power generation. The time resolution of the prediction value is 15 minutes. First, the granularity of the prediction value is too large, and due to the volatility of new energy, there may be a large change within 15 minutes. Moreover, this prediction can only reflect the prediction of new energy power generation, and does not comprehensively consider the change trend of power load and the status of energy storage. Once the accuracy of the prediction value changes, it will lead to a large fluctuation. Therefore, in order to ensure the safety of the power grid, the dispatching does not allow new energy to participate too much, resulting in a large amount of new energy being abandoned. By comprehensively using the power load model, the new energy generation model and the energy storage release model, the three key factors can be considered comprehensively. When the new energy power generation is large, the predicted output of the power load model and the output of the energy storage model can be considered. When there is enough energy storage, although the new energy generation has its uncertainty, the on-grid power of the new energy can be increased because of the energy storage. In addition, after the establishment of the three models, historical data or self-made data can be used for simulation and simulation. In special situations, the response can be simulated in advance to form an emergency plan. When a special situation occurs, it can also be calmly responded.

[0089] It should be noted that the actual model building process can be relatively simple, limited by the amount of data or the requirement for prediction accuracy is not high, it can also be very complex, in the data is very sufficient, considering many factors, the requirement for the accuracy of the prediction is very high, due to the limited space, the needs of each user are not the same, so here only do a simple description, only in the description of the idea and the step, the more historical data collected, the more accurate the model is established, the more accurate the prediction is, the more reliable the auxiliary decision-making is. The model building process is an iterative process. A preliminary model can be established through historical data analysis. In order to verify the accuracy of the model, machine learning techniques can be used to simulate a number of historical data, analyze the difference between the actual data and the summary value, mean value, and adjust the modeling parameters to make the final model more representative.

[0090] The above is only the preferred embodiment of the present application. It should be noted that for those skilled in the art, without departing from the principles of the present application, a number of improvements and adjustments can be made, which should also be considered as the protection scope of the present application.

Claims

1. A method for assisting decision-making in grid connection of new energy power plants, characterized in that, The decision support method The process includes the following: Establish electricity load models, new energy power generation models, and energy storage release models, and use Monte Carlo simulation for prediction; Based on the prediction results of the aforementioned electricity load model, new energy power generation model, and energy storage release model, auxiliary decision-making for grid connection of new energy power plants is obtained; The new energy power plant grid connection auxiliary decision-making uses a model to predict the electricity load and new energy power generation within a set time period in the future, and obtains a strategy for new energy power generation when the electricity load is at its peak; it also obtains a plan to improve the absorption of new energy, while reserving corresponding reserves to prepare for contingency plans in case the set situation affects the new energy power generation. The specific strategy for renewable energy generation during peak electricity load periods is as follows: Use the following formula to adjust the proportion of new energy connected to the grid: E upload This refers to the amount of electricity consumed while browsing the internet at this moment, E v E represents the amount of renewable energy generated at this moment. f To predict the current electricity load, σ is the grid-connected storage coefficient, which is adjusted according to the energy storage capacity ∑Ecan-release of the energy storage station, and ∑En is the comprehensive electricity load during the working hours of the day; The specific process for establishing the electricity load model, the new energy power generation model, and the energy storage release model is as follows: Acquire historical data on power generation and consumption of new energy power plants and energy storage power plants. The historical data includes power load data, new energy power generation data, information on power generation equipment and real-time monitoring data of new energy power plants, and information on energy storage equipment and real-time monitoring data of energy storage power plants. The historical data is segmented and analyzed to obtain the first constraint on the number of electricity loads. Then, based on the first constraint and the obtained electricity load data, the electricity load model is obtained using the time series method and the discrete data sampling method. The historical data is segmented and analyzed to obtain the second constraint on new energy power generation. Based on the second constraint and the acquired new energy power generation data, the new energy power generation model is obtained using the time series method and discrete data sampling method. Cluster analysis is performed based on the characteristics of energy storage types to determine the amount of electricity that an energy storage station can receive and store, as well as the amount of electricity that can be released to the grid at a given time, and an energy storage release model is established. The process of establishing the aforementioned electricity load model is as follows: The historical data is divided into time slices of a set duration according to the time series, and additional attributes are added to the segmented data to obtain the processed historical data. The processed historical data is extracted according to different attributes, and its distribution is analyzed to obtain the first constraint affecting the electricity load. Then, based on the first constraint and the obtained electricity load data, the electricity load model is obtained. Where X is the power generation data sequence retrieved from the database according to the additional attribute conditions, fi is the power load weight value, α is the growth coefficient, and s, t, and w are additional attributes.

2. The auxiliary decision-making method for grid connection of new energy power plants according to claim 1, characterized in that, The new energy power generation model is established as follows: The historical data is divided into time slices of a set duration according to the time series, and additional attributes are added to the segmented data to obtain the processed historical data. The processed historical data is extracted according to different attributes, and its distribution is analyzed to obtain the second constraint affecting new energy power generation. Then, based on the second constraint and the obtained new energy power generation data, the new energy power generation model is obtained. Where X is the power generation data sequence retrieved from the database according to the additional attribute conditions, fi is the power load weight value, β is the redundancy coefficient, and l, s, t, and w are additional attributes.

3. The auxiliary decision-making method for grid connection of new energy power plants according to claim 1, characterized in that, The energy storage and release model is established as follows: According to the type of energy storage, the power generation process is divided into pumped hydro storage, battery energy storage, hydrogen energy storage, and compressed air energy storage; Cluster analysis is performed based on the characteristics of energy storage types to determine the amount of electricity that an energy storage station can receive and store, as well as the amount of electricity that can be released to the grid at a given time, and an energy storage release model is established.

4. The auxiliary decision-making method for grid connection of new energy power plants according to claim 1, characterized in that, The additional attributes are s: season, t: temperature, and w: time period.

5. The auxiliary decision-making method for grid connection of new energy power plants according to claim 4, characterized in that, The additional attributes are: l: latitude and longitude coordinates, s: season, t: temperature, and w: time period.

6. The auxiliary decision-making method for grid connection of new energy power plants according to claim 1, characterized in that, The Monte Carlo simulation method for prediction involves: analyzing the impact of changes in the first constraint on electricity load using Monte Carlo simulation and plotting the trend curve; and analyzing the impact of changes in the second constraint on renewable energy generation using Monte Carlo simulation and plotting the trend curve.

Images