A power supply and demand balance determination method and system, a storage medium and a program product according to power supply

Abstract

The application provides a power supply and demand balance determination method and system based on power supply, a storage medium and a program product, and relates to the technical field of data processing specially applied to business forecasting. Based on the current power transaction scheme and the corresponding credibility, the power supply range sequence is generated to convert each scheme combined with its credibility into the expected supply situation on the time sequence. Then, the power supply range sequence of all power transaction schemes is accumulated to obtain the total power supply range sequence, and the historical power load time sequence data is input into the power load prediction model to generate the corresponding demand range sequence, so that the supply and demand can be compared in the same time sequence dimension. The pre-supply and demand situation can be extracted, and the subsequent power supply failure and power cut phenomenon caused by the mismatch between supply and demand can be avoided, thereby ensuring the stability of power supply.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of data processing specially applicable to business forecasting, and in particular to a power supply and demand balance determination method and system based on power supply, a storage medium and a program product. BACKGROUND

[0002] In the current environment, each participant faces technical difficulties in comparing power supply capacity and future demand capacity in the process of implementing power supply according to contracts. Related technical means are difficult to calculate future power supply capacity and future demand capacity, which can easily lead to subsequent power supply failure and even power cuts. SUMMARY

[0003] The present application provides a power supply and demand balance determination method and system based on power supply, a storage medium and a program product, which can reduce subsequent supply risks and avoid economic losses while maintaining costs.

[0004] In a first aspect, the present application provides a power supply and demand balance determination method based on power supply, comprising: obtaining all power trading schemes of a current region; the power trading scheme defines a set of one or more power purchase and sale information, including contract power and execution period; generating a power supply range sequence based on the current power trading scheme and the corresponding credibility; the power supply range sequence is a time sequence of expected power supply range on a discrete time unit in a future preset period; the current power trading scheme is any power trading scheme; accumulating the power supply range sequences of all power trading schemes to obtain a total power supply range sequence; inputting historical power load time sequence data of the current region into a power load prediction model to generate a power predicted demand range sequence corresponding to the total power supply range sequence at the discrete time unit; the power predicted demand range sequence is a time sequence of expected power demand range on a discrete time unit in a future preset period; extracting the current expected power supply range in the total power supply range sequence at the current discrete time unit and the current expected power demand range in the power predicted demand range sequence at the current discrete time unit, the current discrete time unit being any discrete time unit; when the minimum value of the current expected power supply range falls within the current expected power demand range, determining a first range of the maximum value of the current expected power demand range and the minimum value of the current expected power supply range; calculating a first proportion of the first range in the current expected power demand range; if the first proportion is less than an importance threshold of the current region; determining that the current discrete time unit is in a sufficient supply state, and regarding the next discrete time unit under the current discrete time unit as the current discrete time unit; if the first proportion is not less than the importance threshold of the current region; determining that the current discrete time unit is in a insufficient supply state, and regarding the next discrete time unit under the current discrete time unit as the current discrete time unit.

[0005] By adopting the technical solution, based on the current power transaction scheme and the corresponding credibility, the power supply range sequence is generated to combine each scheme with its credibility into the expected supply situation on the time sequence. Then, the power supply range sequence of all power transaction schemes is accumulated to obtain the total power supply range sequence, and the historical power load time sequence data is input into the power load prediction model to generate the corresponding demand range sequence, so that the supply and demand can be compared in the same time sequence dimension. The pre-supply and demand situation can be extracted, and the subsequent power supply and the power cut phenomenon caused by the mismatch between supply and demand can be avoided, the stability of power supply is ensured, and the orderly development of various power activities in the region is ensured.

[0006] In some embodiments in combination with the first aspect, based on the current power transaction scheme and the corresponding credibility, the power supply range sequence is generated; the power supply range sequence is a time sequence of expected power supply range on discrete time units in a future preset period; the step of the current power transaction scheme is any power transaction scheme, specifically including: in a plurality of historical discrete time units, the contract power and the corresponding actual delivery power of the current power transaction scheme are obtained; for each historical discrete time unit, the ratio of the actual delivery power to the contract power is calculated to obtain a set of performance ratios; the performance ratio is determined as the credibility.

[0007] By adopting the technical solution, the contract power and the actual delivery power are obtained in a plurality of historical discrete time units, and the ratio thereof is calculated to obtain the performance ratio, which is determined as the credibility. The actual execution situation is integrated into consideration, which means that when generating the power supply range sequence subsequently, it is not simply based on the contract agreement, but the actual performance factor is integrated. The problem of relying only on contract estimation and deviating from reality is avoided, and the entire transaction decision assistance process based thereon is more reliable, the probability of supply risk caused by inaccurate supply estimation is reduced, and the smooth development of power transaction is ensured.

[0008] In some embodiments of the first aspect, in some embodiments, for each historical discrete time unit, a ratio of the actual delivery electricity quantity to the contract electricity quantity is calculated to obtain a set of performance ratios; after the step of determining the performance ratios as the credibility, the method further comprises: normalizing the performance ratios to generate normalized performance ratios; selecting a plurality of representative point values from a value range of the normalized performance ratios according to a preset interval; mapping each value in the normalized performance ratios to the nearest representative point value according to a minimum distance criterion; counting frequencies of the normalized performance ratio values of the representative point values; dividing each frequency of the representative point values by a total number of the normalized performance ratio values to obtain an empirical probability corresponding to each representative point value; and constructing a probability density function based on the representative point values and the corresponding empirical probabilities.

[0009] By adopting the above technical solutions, after the steps of calculating the ratio of the actual delivery electricity quantity to the contract electricity quantity for each historical discrete time unit to obtain a set of performance ratios and determining the performance ratios as the credibility, the normalized performance ratios are generated by normalizing the performance ratios, which is to enable performance ratios of different magnitudes and ranges to be compared and analyzed under the same standard and eliminate the influence of differences in the data itself. Then, a plurality of representative point values are selected from a value range of the normalized performance ratios according to a preset interval, which is a reasonable discretization processing of the data for subsequent operations. Then, each value in the normalized performance ratios is mapped to the nearest representative point value according to a minimum distance criterion, which realizes the classification and integration of the data. Then, the frequencies of the normalized performance ratio values of the representative point values are counted, and each frequency of the representative point values is divided by a total number of the normalized performance ratio values to obtain an empirical probability corresponding to each representative point value. Based on the probability information, a probability density function is constructed, so that in subsequent judgment of the electricity supply situation, the probability density function can be used to comprehensively consider the probability of different supply situations, the transaction decision based on the probability density function is more scientific and accurate, the supply risk misjudgment caused by not considering the probability factor is reduced, and the long-term transaction decision is better served.

[0010] In some embodiments of the first aspect, in some embodiments, the step of determining whether the current discrete time unit is in the state of sufficient supply or in the state of insufficient supply, and taking the next discrete time unit of the current discrete time unit as the current discrete time unit, comprises: applying the current expected power supply range as an input variable to the probability density function to obtain a probability distribution of the current expected power supply range; counting a first total probability of falling into the first range in the probability distribution of the current expected power supply range; if the first total probability is less than the importance threshold of the current region, determining that the current discrete time unit is in the state of sufficient supply, and taking the next discrete time unit of the current discrete time unit as the current discrete time unit; if the first total probability is not less than the importance threshold of the current region, determining that the current discrete time unit is in the state of insufficient supply, and taking the next discrete time unit of the current discrete time unit as the current discrete time unit.

[0011] By applying the current expected power supply range as an input variable to the probability density function, the probability distribution of the current expected power supply range can be obtained by means of the previously constructed probability density function. This probability distribution reflects the possibility of the supply range at different values. Then, the first total probability of falling into the first range in the probability distribution of the current expected power supply range is counted. By comparing the first total probability with the importance threshold of the current region, the supply state of the current discrete time unit is determined. The determination of whether the supply is sufficient or not is no longer a simple comparison based on fixed numerical values, but fully considers the uncertainty and probability characteristics of the actual power supply. The transaction decision made in this way is more in line with the actual situation of the complex and changeable power market, and can more accurately perceive the supply risk in advance, provide protection for reasonable arrangement of transactions and avoid economic losses caused by supply risks, and improve the ability of the entire transaction decision assistance method to cope with complex situations.

[0012] In some embodiments of the first aspect, after the step of extracting the current expected power supply range in the current discrete time unit in the total power supply range sequence and the current expected power demand range in the current discrete time unit in the power forecast demand range sequence, the current discrete time unit being any discrete time unit, the method further comprises: when the maximum value of the current expected power supply range falls within the current expected power demand range, determining a second range of the maximum value of the current expected power supply range and the minimum value of the current expected power demand range; calculating a second proportion of the second range in the current expected power demand range; if the first proportion is not less than the importance threshold of the current region, determining that the current discrete time unit is in a sufficient supply state, and regarding the next discrete time unit of the current discrete time unit as the current discrete time unit; if the first proportion is less than the importance threshold of the current region, determining that the current discrete time unit is in an insufficient supply state, and regarding the next discrete time unit of the current discrete time unit as the current discrete time unit.

[0013] By adopting the above technical solution, the previous determination logic for the case where the minimum value falls within the demand range is complemented. By adding an analysis path for this different critical case, the technical features work together to measure the matching degree of supply and demand from more angles, avoiding the limitations of relying on a single determination condition, making the determination of whether the supply is sufficient more comprehensive and accurate, providing richer and more detailed reference basis for transaction decision-making, and further more accurately controlling the supply risk in the power market, ensuring that the power transaction can be reasonably arranged according to a more practical judgment, reducing economic losses caused by misjudgment of supply risk.

[0014] In some embodiments of the first aspect, after the step of extracting the current expected power supply range in the current discrete time unit in the total power supply range sequence and the current expected power demand range in the current discrete time unit in the power forecast demand range sequence, the current discrete time unit being any discrete time unit, the method further comprises: when the minimum value of the current expected power supply range is greater than the maximum value of the current expected power demand range, determining that the current discrete time unit is in a sufficient supply state, and regarding the next discrete time unit of the current discrete time unit as the current discrete time unit.

[0015] By adopting the above technical solution, when the minimum value of the current expected power supply range is greater than the maximum value of the current expected power demand range, the current discrete time unit is directly determined to be in a sufficient supply state, and the next discrete time unit of the current discrete time unit is regarded as the current discrete time unit. When this obvious situation of supply greater than demand occurs, a determination of sufficient supply is quickly given, without the need for subsequent operations such as complex proportion calculation or probability analysis, improving the execution efficiency of the entire transaction decision-making assistance method.

[0016] In some embodiments of the first aspect, in some embodiments, after the step of extracting the current expected power supply range at the current discrete time unit in the total power supply range sequence and the current expected power demand range at the current discrete time unit in the power prediction demand range sequence, the current discrete time unit being any discrete time unit, the method further comprises: determining that the current discrete time unit is in a supply shortage state when the maximum value of the current expected power supply range is less than the minimum value of the current expected power demand range, and regarding the next discrete time unit under the current discrete time unit as the current discrete time unit.

[0017] By adopting the above technical solution, when the maximum value of the current expected power supply range is less than the minimum value of the current expected power demand range, it is determined that the current discrete time unit is in a supply shortage state, and the next discrete time unit under the current discrete time unit is regarded as the current discrete time unit. By extracting and directly comparing the supply and demand ranges, once such a situation that the supply is obviously less than the demand occurs, a determination of a supply shortage is made quickly, so that subsequent corresponding coping strategies can be immediately started around this state.

[0018] In a second aspect, the present application provides a power supply and demand balance determination system. The power supply and demand balance determination system comprises one or more processors and a memory. The memory is coupled to the one or more processors, and is configured to store computer program codes. The computer program codes comprise computer instructions. The one or more processors invoke the computer instructions to enable the power supply and demand balance determination system to perform the method described in the first aspect and any possible implementation manner of the first aspect.

[0019] In a third aspect, the present application provides a computer program product comprising instructions, which, when executed on a power supply and demand balance determination system, enable the power supply and demand balance determination system to perform the method described in the first aspect and any possible implementation manner of the first aspect.

[0020] In a fourth aspect, the present application provides a computer-readable storage medium comprising instructions, which, when executed on a power supply and demand balance determination system, enable the power supply and demand balance determination system to perform the method described in the first aspect and any possible implementation manner of the first aspect.

[0021] The one or more technical solutions provided in the embodiments of the present application have at least the following technical effects or advantages:

[0022] 1、Based on the current power trading scheme and the corresponding credibility, the power supply range sequence is generated. Each scheme is combined with its credibility to convert the expected supply into a time sequence. Then, the power supply range sequence of all power trading schemes is accumulated to obtain the total power supply range sequence. The historical power load time series data is input into the power load prediction model to generate the corresponding demand range sequence. The supply and demand can be compared in the same time sequence dimension. The pre-supply and demand situation can be extracted to avoid subsequent power supply and demand mismatch, power supply failure, and power outage. The stability of power supply is ensured, and various power activities in the region can be carried out in an orderly manner.

[0023] 2、In multiple historical discrete time units, the contract power and the actual delivery power are obtained, and the ratio is calculated to obtain the performance ratio. The actual performance is taken into account, which means that the subsequent generation of the power supply range sequence is not based on the contract alone, but also takes into account the actual performance. This avoids the problem of relying solely on contract estimates and makes the entire trading decision-making process more reliable. It reduces the risk of supply risk caused by inaccurate supply estimates and ensures smooth power trading.

[0024] 3、When the ratio of actual delivery power to contract power is calculated for each historical discrete time unit, a set of performance ratios is obtained. After determining the performance ratio as the credibility, the performance ratio is then normalized to generate a normalized performance ratio. This step is to compare and analyze the performance ratios of different magnitudes and ranges under the same standard, eliminating the influence of data differences. Then, from the value range of the normalized performance ratio, several representative point values are selected according to the preset interval. This is a reasonable discretization of data to facilitate subsequent operations. Each normalized performance ratio value is mapped to the nearest representative point value based on the minimum distance criterion, achieving data classification and integration. Then, the frequency of the normalized performance ratio value of the representative point value is counted, and the frequency of each representative point value is divided by the total number of normalized performance ratio values to obtain the empirical probability corresponding to each representative point value. Based on these probability information, a probability density function is constructed. In the subsequent judgment of power supply, the probability density function reflects the probability of different supply situations, which makes the trading decision-making based on this more scientific and accurate, reduces the risk of misjudgment due to not considering the probability factor, and better serves the long-term trading decision-making. BRIEF DESCRIPTION OF DRAWINGS

[0025] Figure 1 is a flowchart of the power supply and demand balance determination method in the embodiments of the present application;

[0026] Figure 2 is another flowchart of the method for determining the balance between supply and demand of power supply capacity according to an embodiment of the present application;

[0027] Figure 3 is an exemplary hardware structure diagram of the system for determining the balance between supply and demand of power supply capacity according to an embodiment of the present application. DETAILED DESCRIPTION

[0028] The terms used in the following embodiments of the present application are only for the purpose of describing specific embodiments and are not intended to be limiting of the present application. As used in the specification and the appended claims of the present application, the singular forms "a," "an" and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise. It will also be understood that the terms and / or phrases used in this application are intended to refer to any or all possible combinations of one or more of the listed items.

[0029] Hereinafter, the terms first and second are used only for the purpose of description and are not intended to imply or indicate relative importance or imply a specified number of technical features indicated. Therefore, the features defined with first and second can explicitly or implicitly include one or more of the features, and in the description of the embodiments of the present application, the meaning of multiple is two or more, unless otherwise specified.

[0030] Please refer to Figure 1 , Figure 1 is a flowchart of the method for determining the balance between supply and demand of power supply capacity according to an embodiment of the present application;

[0031] S101, obtaining all power transaction schemes of the current region; the power transaction scheme defines a set of one or more power purchase and sale information, including contract capacity and execution period;

[0032] Wherein, the current region: refers to a specific geographical area of interest, in which power transaction activities are carried out, which can be a city, a province or other specific power supply area. Power transaction scheme: refers to a set of one or more power purchase and sale information, including contract capacity (representing the amount of power purchased or sold in the power transaction contract) and execution period (referring to the time limit of transaction execution specified in the power transaction contract, such as monthly, quarterly or annually, etc.).

[0033] It should be noted that in the scenarios involved in the present application, there is a clear goal orientation, which is to maximize benefits. In this context, relevant subjects often take the initiative to make decisions, and the power trading decision-making process has a distinctive feature, that is, various contracts are signed in advance. Each contract here has important significance, which corresponds to the right to physically deliver electricity in the future, meaning that after the contract is signed, the actual delivery of electricity can be realized when the agreed conditions are met.

[0034] In order to deeply understand the overall situation of the current regional power trading, so as to carry out subsequent work such as power supply analysis, transaction strategy formulation, etc., it is necessary to obtain all the signed power trading schemes in the current region. And need to extract the power trading scheme corresponding to the current region from it.

[0035] S102, based on the current power trading scheme and the corresponding credibility, generate a power supply range sequence; the power supply range sequence is a time sequence of expected power supply ranges at discrete time units in a future preset period; the current power trading scheme is any power trading scheme;

[0036] Among them, the current power trading scheme refers to a selected scheme from all power trading schemes, which is used as the basis data for generating the power supply range sequence. Credibility: used to represent the reliability of the power supply information involved in the current power trading scheme, which may be a value or level evaluated based on historical data, market conditions, transaction party reputation, etc. Power supply range sequence: refers to a time sequence of expected power supply ranges at discrete time units in a future preset period, which describes the power range that can be supplied at different discrete time points according to the current power trading scheme and the credibility.

[0037] It should be noted that the credibility factor has different forms in different embodiments.

[0038] In some embodiments, the credibility is represented as a range data.

[0039] In some other embodiments, the credibility is initially in the form of a specific numerical value. Then by multiplying the corresponding range set in advance, it is converted into a range data.

[0040] For each current power transaction scheme, a theoretical time series that changes over time can be derived based on its execution period and contract power, two key pieces of information. This theoretical time series depicts the change in power supply over time as agreed in the contract under ideal conditions. On this basis, it is multiplied by the corresponding confidence level, i.e. the confidence level (whether it is range data or transformed range data), to obtain the power supply range sequence.

[0041] In some specific embodiments, the entire execution period is divided into a number of discrete time units according to the execution period specified in the power transaction scheme and a predetermined discrete time unit division rule. Then, taking advantage of the contract power and the confidence level, two important pieces of data information, the expected power supply range is calculated for each divided discrete time unit. Finally, the expected power supply range of each discrete time unit calculated is arranged in chronological order to form the power supply range sequence.

[0042] In some embodiments, the confidence level is calculated based on historical performance data.

[0043] When evaluating and analyzing a power transaction scheme, the past performance needs to be considered to determine the confidence level of the current scheme, at which point the relevant data is obtained and calculated in multiple historical discrete time units. Specifically, a number of historical discrete time units are first determined, for example, each month in the past year is selected as a discrete time unit. Then, for the current power transaction scheme, the contract power and the corresponding actual delivery power of each month are obtained. By comparing the actual delivery power and the contract power, the performance of the transaction scheme in the past is measured, and its confidence level is evaluated.

[0044] In other embodiments, the confidence level is calculated considering the reliability of the equipment.

[0045] When evaluating the reliability of power supply, the operating condition and failure probability of power generation equipment need to be considered, at which point the confidence level related to the reliability of the equipment is calculated. Specifically, the various types of power generation equipment in the power system are first sorted out to understand their technical parameters, service life, etc. to determine the failure rate of the equipment. Then, the available power generation capacity of the equipment is calculated based on the rated capacity and failure rate of the equipment. By comparing the available power generation capacity with the total demand of the system, the confidence level of the power supply is evaluated.

[0046] In some specific implementations: First, a detailed survey is conducted on all thermal power generation equipment in the power system, recording information such as their model, operating time, and maintenance records. Based on this information and the equipment manuals, their failure rates are determined. Then, based on the rated capacity and failure rate of the equipment, the available generating capacity of each piece of equipment is calculated. Finally, the available generating capacities of all thermal power generation equipment are summed to obtain the available capacity of the thermal power generation section, which is then compared with the system's projected load to assess its reliability.

[0047] In some specific implementations: First, historical fault data and operational data of the wind turbine generators at the wind farm are collected, and the failure rate is analyzed. Then, based on the rated power and failure rate of each wind turbine generator, the available power generation capacity of each unit is calculated. Considering the intermittency and uncertainty of wind power generation, and combining factors such as meteorological data and the geographical location of the power plant, the reliability of wind power generation for supplying the system is comprehensively evaluated.

[0048] In some embodiments, credibility is calculated based on the credit rating of the market entity;

[0049] In electricity trading, when considering the impact of counterparty credit risk on power supply, credibility is calculated based on the credit ratings of market participants. Specifically, the credit rating information of each market participant is first obtained. Then, based on the different roles and importance of each market participant in the electricity trading, corresponding weights are assigned. Finally, by combining the credit ratings and weights of all market participants, an overall credibility index is calculated to measure the reliability of the electricity trading scheme from a credit perspective.

[0050] In some specific implementations, firstly, credit rating reports of all power generation and supply companies in the region are collected; these reports are issued by professional credit rating agencies. Then, weights are assigned to each power generation and supply company based on their varying degrees of influence on power supply. Finally, a comprehensive credibility index is calculated using a weighted average method.

[0051] S103. Accumulate the power supply range sequence of all power trading schemes to obtain the total power supply range sequence;

[0052] Specifically, the power supply range sequences generated by each power trading scheme are summed or merged. For example, for each discrete time unit, the expected power supply ranges of all power trading schemes in that time unit are summed to obtain the total expected power supply range for the corresponding discrete time unit in the total power supply range sequence. This allows for a holistic understanding of the power supply capacity of the current region at each discrete time point within a preset future period.

[0053] S104, input the historical power load time series data of the current region into the power load prediction model to generate a power prediction demand range sequence corresponding to the discrete time units of the total power supply range sequence; the power prediction demand range sequence is a time sequence of the expected power demand range at the discrete time units in the future preset period;

[0054] The historical power load time series data refers to the power load data of the current region in the past period, arranged in chronological order to form a sequence, which reflects the power usage of the region at different time points in the past. The power load prediction model is a model based on mathematical algorithms and statistical analysis, which learns and analyzes historical power load time series data and other related factors (such as weather, economic development trends, etc.) to predict the range of future power demand. The power prediction demand range sequence refers to the time sequence of the expected power demand range at the discrete time units in the future preset period, which is the result of the power load prediction model based on the input historical data and related factors, used to represent the possible power demand range of the region at each discrete time point in the future.

[0055] In some embodiments, the power load is regarded as a random process that changes over time, and by analyzing the historical time series data, the autocorrelation (i.e. the correlation between load data at different time points), seasonality (such as summer peak electricity consumption, winter heating electricity consumption leading to periodic changes in load), trend (e.g. the trend of increasing power load over the years with urban development) and other characteristics are mined. For example, the commonly used autoregressive moving average model (ARMA) and its extended model (such as ARIMA, etc.), the ARMA model will be based on the autocorrelation in the historical load data, by constructing an autoregressive (AR) part to reflect the influence of past load values on current load values, and using a moving average (MA) part to process random error terms, after parameterizing these relationships in historical data, the future power load value can be predicted based on the current and past known data. ARIMA adds a difference operation to ARMA to handle non-stationary time series, making them stationary before modeling and predicting the relationships.

[0056] In some specific embodiments, first collect daily power load data for several years to form a time series. Perform stationarity test, if not stationary, use difference method to make it stationary, such as first-order difference (subtract the load value at the previous time from the load value at the next time). Then determine the order of the ARMA model (i.e. the number of autoregressive terms and moving average terms) by analyzing the autocorrelation function (ACF) and partial autocorrelation function (PACF) plots, and then build an ARIMA model for training, and use the trained model to predict future daily power load.

[0057] In some embodiments, the hourly power load data of the past year is selected as the time series, and the data is preprocessed, such as removing outliers, standardizing, and the like. Then, a seasonal autoregressive integrated moving average model (SARIMA) is used, which considers seasonal factors on the basis of ARIMA. By analyzing the load change law in different seasons and time periods within a year, the seasonal period (such as 24 hours within a day, 12 months within a year, etc.) and the corresponding model parameters are determined. After training, the hourly power load in the future is predicted, which is especially suitable for scenarios with obvious seasonal power consumption characteristics.

[0058] In some embodiments, it is assumed that there is a certain linear or nonlinear function relationship between the power load and the multiple independent variables. By collecting the power load values and the corresponding independent variable data in the historical data, a statistical method is used to fit the function relationship (i.e., the regression equation). For example, a multiple linear regression model assumes that there is a linear relationship between the power load (Y) and the multiple independent variables (X1, X2, …, Xn). The model expression is approximately Y = β0+ β1X1+ β2X2+ … + βnXn+ ε (where β0is the intercept term, β1, β2, …, βnare the regression coefficients, and ε is the random error term). By fitting the equation with historical data, that is, finding the set of regression coefficients that minimizes the error between the predicted power load value and the actual historical load value, the model is determined. After the model is determined, the corresponding power load prediction value can be obtained by inputting the future independent variable values (such as predicted air temperature, economic indicators, etc.).

[0059] In some embodiments, monthly power load data of the past three years is collected as the dependent variable, and average temperature, local industrial added value, and the number of working days per month are collected as the independent variables. The data is sorted and preprocessed, such as data standardization, outlier processing, and the like. Then, a stepwise regression method is used to select variables that have a significant impact on power load from the numerous independent variables, and a multiple linear regression model is constructed. The regression coefficients are determined by a least squares fitting algorithm, and after the model is trained, the future monthly power load is predicted according to the predicted values of the future independent variables (such as temperature forecasts from the meteorological department, economic indicators from the economic department, etc.).

[0060] In some specific embodiments, daily power load data in the past half year and corresponding daily maximum temperature, humidity, whether it is a holiday, etc. are selected. First, correlation analysis is performed on the data to preliminarily determine the correlation degree of each factor with the power load. Then, a polynomial regression model is used, considering that there may be a nonlinear relationship between temperature and power load (such as different load change characteristics at high and low temperatures), a polynomial regression equation containing the quadratic term and the cubic term of temperature is constructed, and after determining the coefficients of each term by fitting historical data, the future daily power load is predicted by using the predicted values of related independent variables such as future weather.

[0061] It should be noted that in the embodiments of the various power load prediction models mentioned above, the final output of these models after corresponding data processing, algorithm operation, etc. is only the power prediction demand sequence. The so-called power prediction demand sequence refers to the specific numerical prediction result of the power demand at each discrete time unit in the future arranged in chronological order, which presents a series of relatively certain prediction values, rather than a power prediction demand range sequence considering a certain fluctuation range.

[0062] In view of the above, in some embodiments, the result of the loss function of the last round of training model is used as the credibility. The loss function is an index used to measure the difference between the model prediction result and the actual historical data during the model training process, and its result reflects the accuracy and uncertainty of the model prediction.

[0063] Then, the method involved in the previous step S102, that is, the information related to the credibility is integrated into the adjustment process of the power prediction demand sequence, so as to convert the power prediction demand sequence into the power prediction demand range sequence, which is not limited here.

[0064] S105, extracting the current expected power supply range in the total power supply range sequence and the current expected power demand range in the power prediction demand range sequence under the current discrete time unit, the current discrete time unit being any discrete time unit;

[0065] Among them, the current discrete time unit refers to any discrete time point selected in the total power supply range sequence and the power prediction demand range sequence, which is used for subsequent supply and demand range comparison and analysis.

[0066] After obtaining the total power supply range sequence and the power prediction demand range sequence, in order to specifically analyze the supply and demand at each discrete time unit, the supply and demand range information corresponding to the current discrete time unit needs to be extracted from the two sequences. The scene is usually point-by-point analysis of power supply and demand, so as to more accurately evaluate whether the power supply at each time point can meet the demand.

[0067] It should be noted that after this step, the current expected power supply range and the current expected power demand range are compared respectively, and according to the comparison result, there are four cases, steps S106-S109, steps S110-S113, step S114, and step S115;

[0068] S106, when the minimum value of the current expected power supply range falls within the current expected power demand range, determining a first range between the maximum value of the current expected power demand range and the minimum value of the current expected power supply range;

[0069] First, it is determined whether the minimum value of the current expected power supply range is within the current expected power demand range. If so, the difference between the maximum value of the current expected power demand range and the minimum value of the current expected power supply range is calculated, and the range formed by this difference is the first range.

[0070] S107, calculating a first proportion of the first range in the current expected power demand range;

[0071] The first proportion refers to the proportion of the first range in the current expected power demand range, which is used to quantify the size of the first range relative to the power demand range, so as to compare with other standards (such as the importance threshold).

[0072] Specifically, the value of the first range is divided by the total value of the current expected power demand range (i.e. the maximum value minus the minimum value of the demand range), and the result is the first proportion.

[0073] S108, if the first proportion is less than the importance threshold of the current region; then determine that the current discrete time unit is in a supply sufficient state, and the next discrete time unit of the current discrete time unit is regarded as the current discrete time unit;

[0074] The supply sufficient state refers to a state in which the power supply of the current discrete time unit can meet the demand according to the comparison result of the first proportion and the importance threshold of the current region. This state information will be used as a reference basis for power trading decision, indicating that the power supply is relatively sufficient at this time point, and special supply guarantee measures may not be needed.

[0075] S109, if the first proportion is not less than the importance threshold of the current region; then determine that the current discrete time unit is in a supply insufficient state, and the next discrete time unit of the current discrete time unit is regarded as the current discrete time unit.

[0076] The supply insufficient state refers to a state that the power supply in the current discrete time unit is difficult to meet the demand according to the comparison result of the first proportion and the importance threshold of the current region. The state information serves as a reference basis for the power transaction decision, and prompts that the power supply is relatively tight at the time point, and relevant measures such as increasing power purchase, adjusting transaction strategy and the like need to be taken to ensure the supply.

[0077] It can be seen that, based on the current power transaction scheme and the corresponding credibility, the power supply range sequence is generated to combine each scheme with its credibility into the expected supply situation on the time sequence. Then, the total power supply range sequence is obtained by accumulating the power supply range sequences of all power transaction schemes, and the historical power load time sequence data is input into the power load prediction model to generate the corresponding demand range sequence, so that the supply and demand can be compared in the same time sequence dimension. The pre-supply and demand situation can be extracted, and the subsequent power supply and the power cut phenomenon caused by the mismatch between supply and demand can be avoided, the stability of power supply is ensured, and the orderly development of various power activities in the region is ensured.

[0078] In some embodiments, after step S105, the method further comprises:

[0079] Step S110, when the maximum value of the current expected power supply range falls within the current expected power demand range, determining a second range of the maximum value of the current expected power supply range and the minimum value of the current expected power demand range;

[0080] S111, calculating a second proportion of the second range in the current expected power demand range;

[0081] S112, if the first proportion is not less than the importance threshold of the current region, determining that the current discrete time unit is in a supply sufficient state, and regarding the next discrete time unit of the current discrete time unit as the current discrete time unit;

[0082] S113, if the first proportion is less than the importance threshold of the current region, determining that the current discrete time unit is in a supply insufficient state, and regarding the next discrete time unit of the current discrete time unit as the current discrete time unit.

[0083] It should be noted that in the framework of the entire power supply and demand state determination, the core target is to accurately measure the matching degree of power supply and demand in each discrete time unit, and then provide a reliable basis for power transaction decision to reasonably cope with the supply risk in the power market. Whether the situation involved in steps S106-S109 or the situation corresponding to steps S110-S113 is to analyze the relationship between the power supply range and the demand range from different angles, and to determine whether the supply is sufficient or insufficient state by setting reasonable determination conditions and thresholds.

[0084] In step S106, when the minimum value of the supply range falls into the demand range, it means that from the lower limit perspective, the supply has begun to touch the demand range. At this time, the first range is calculated, which is to quantify the size of the margin of demand relative to supply in this touching situation.

[0085] Then the first proportion is calculated. For example, if the first proportion value is small, it means that although the minimum value of the supply range falls within the demand range, the proportion in the demand range is small, which means that there is a relatively large accommodation space for demand relative to supply, that is, the power supply is relatively sufficient. Therefore, in step S108, when the first proportion is less than the importance threshold, it is determined that the current discrete time unit is in a sufficient supply state.

[0086] On the contrary, if the first proportion is not less than the importance threshold, it means that the minimum value of the supply range has occupied a large proportion in the demand range, and the accommodation margin of demand is small, which implies that the power supply is relatively tight. Therefore, in step S109, it is determined that the supply is insufficient.

[0087] In step S110, when the maximum value of the supply range falls into the demand range, it is from the upper limit perspective to observe the relationship between supply and demand. At this time, the second range determined reflects the exceeding situation of the upper limit of the supply in the demand range.

[0088] Then the second proportion is calculated. When the first proportion (here, the first proportion concept calculated in the previous step S107 is followed, because the overall is to consider different situations to determine the supply state) is not less than the importance threshold of the current area, it means that from the overall supply range (considering the minimum value falling and other comprehensive situations, the first proportion reflects a comprehensive measurement) and the current supply range maximum value in the demand range, the supply can relatively well cover the demand. Therefore, it is determined that the current discrete time unit is in a sufficient supply state, that is, the determination logic of step S112.

[0089] When the first proportion is less than the importance threshold of the current area, it means that under the comprehensive consideration, the supply may not be relatively good to meet the demand in the overall and from the analysis of the maximum value of the supply range in the demand range. There is a possibility of supply shortage, so in step S113, it is determined that the supply is insufficient.

[0090] The seemingly opposite determination logic is actually because the two situations are analyzed from two different perspectives of the lower limit and the upper limit of the relationship between the supply range and the demand range. Steps S106-S109 focus on the surplus condition after the minimum value of the supply range enters the demand range, and determine whether the supply is sufficient from the lower limit perspective. Steps S110-S113 focus on the excess degree when the maximum value of the supply range falls into the demand range, and comprehensively determine the supply state from the upper limit perspective.

[0091] It can be seen that the determination logic is complementary to the previous determination logic for the case that the minimum value falls into the demand range. By adding the analysis path of this different critical condition, the technical features work together to measure the matching degree of supply and demand from more angles, avoid the limitation of relying on a single determination condition, make the determination of whether the supply is sufficient more comprehensive and accurate, provide more rich and detailed reference basis for transaction decision, and further more accurately control the supply risk in the power market, ensure that the power transaction can be reasonably arranged according to the more actual judgment, and reduce the economic loss caused by the misjudgment of the supply risk.

[0092] After step S105, it further includes: S114, when the minimum value of the current expected power supply range is greater than the maximum value of the expected power demand range, determining that the current discrete time unit is in a sufficient supply state, and regarding the next discrete time unit of the current discrete time unit as the current discrete time unit.

[0093] It can be seen that when the minimum value of the current expected power supply range is greater than the maximum value of the current expected power demand range, it is directly determined that the current discrete time unit is in a sufficient supply state, and the next discrete time unit of the current discrete time unit is regarded as the current discrete time unit. When such an obvious condition of supply greater than demand occurs, a determination of sufficient supply is quickly given, without the need for subsequent operations such as complex proportion calculation or probability analysis, thereby improving the execution efficiency of the entire transaction decision assistance method.

[0094] After step S105, it further includes: S115, when the maximum value of the current expected power supply range is less than the minimum value of the current expected power demand range, determining that the current discrete time unit is in a insufficient supply state, and regarding the next discrete time unit of the current discrete time unit as the current discrete time unit.

[0095] It can be seen that when the maximum of the current expected power supply range is less than the minimum of the current expected power demand range, it is determined that the current discrete time unit is in a supply shortage state, and the next discrete time unit of the current discrete time unit is regarded as the current discrete time unit. By extracting and directly comparing the supply and demand ranges, as soon as such a situation of supply being obviously less than demand occurs, a determination of supply shortage is made quickly, so that subsequent corresponding coping strategies can be immediately started around this state.

[0096] Please refer to Figure 2 , Figure 2 is another flowchart of the power supply and demand balance determination method in the embodiments of the present application;

[0097] In some embodiments, step S102 specifically comprises:

[0098] S201, in a plurality of historical discrete time units, contract power and corresponding actual delivery power of a current power trading scheme are obtained;

[0099] Historical discrete time unit: refers to past time periods divided at certain time intervals, used for analyzing historical data of power trading. For example, it can be every month, every quarter, etc. in the past. Contract power: represents the power value agreed upon by both parties in advance in the power trading scheme, which is the specified power in the trading contract. Actual delivery power: refers to the power actually delivered in the actual power trading process, reflecting the actual power executed by the contract.

[0100] S202, for each historical discrete time unit, a ratio of actual delivery power to contract power is calculated to obtain a set of performance ratios; the performance ratio is determined as the credibility.

[0101] Performance ratio: refers to the ratio of actual delivery power to contract power, used to measure the degree of performance of the power trading contract in each historical discrete time unit.

[0102] Specifically, for each historical discrete time unit, the actual delivery power in the time unit is divided by the contract power to obtain a ratio, which is the performance ratio. By calculating the performance ratios of a plurality of historical discrete time units, a set of data is formed, and then the set of performance ratios is determined as the credibility.

[0103] It can be seen that the contract electricity and the actual delivery electricity in multiple historical discrete time units are obtained, the ratio is calculated to obtain the performance ratio, the performance ratio is determined as the credibility, the actual execution in the past is taken into account, which means that when the power supply range sequence is generated subsequently, the actual performance is taken into account instead of being based on the contract only, thereby avoiding the problem of deviating from the actual situation only by relying on the contract estimation, and further making the entire transaction decision assistance process based on this more reliable, reducing the probability of supply risk caused by inaccurate supply estimation, and ensuring smooth development of power transaction.

[0104] S203, normalizing the performance ratio to generate a normalized performance ratio;

[0105] After obtaining the performance ratio, in order to eliminate the magnitude difference between different data and facilitate subsequent analysis, the performance ratio needs to be normalized.

[0106] Specifically, a specific normalization algorithm or formula is used to process the performance ratio obtained in the foregoing. A common normalization method is the minimum-maximum normalization, which maps the value of each performance ratio to the interval [0, 1] to obtain the normalized performance ratio. In this way, different performance ratios can be compared, and the analysis result is not affected by the magnitude difference of the data itself.

[0107] S204, selecting a plurality of representative point values from the value range of the normalized performance ratio according to a preset interval;

[0108] Specifically, the value range of the normalized performance ratio is determined, and then a plurality of representative point values are selected in the value range according to a preset interval. For example, if the value range is [0, 1] and the preset interval is 0.1, the representative point values 0, 0.1, 0.2, 0.3, …, 0.9 and 1 are selected. These representative point values will be used as the basis for subsequent analysis and calculation, which can simplify the processing process of continuous data.

[0109] S205, mapping the value in each normalized performance ratio to the nearest representative point value according to the minimum distance criterion;

[0110] Minimum distance criterion: a data mapping rule for mapping the value of each normalized performance ratio to the nearest representative point value to realize data classification and simplification.

[0111] For each value of the normalized performance ratio, calculate the distance (usually the absolute value distance) between it and each representative point value, then find the representative point value with the smallest distance, and map the value of the normalized performance ratio to this nearest representative point value. In this way, all the values of the normalized performance ratio are classified into corresponding representative point values, facilitating subsequent statistics and analysis.

[0112] S206, count the frequency of the normalized performance ratio value of the representative point value;

[0113] For each representative point value, count the number of times it appears in all the mapped normalized performance ratio values, which is the frequency of the representative point value. By counting the frequency, the distribution of the performance ratio represented by different representative point values can be understood, providing data support for subsequent calculation of empirical probability.

[0114] S207, divide the frequency of each representative point value by the total number of normalized performance ratio values to obtain the empirical probability corresponding to each representative point value;

[0115] Divide the frequency of each representative point value by the total number of normalized performance ratio values to obtain the empirical probability corresponding to each representative point value. This empirical probability reflects the likelihood of the performance ratio corresponding to each representative point value appearing in past power transactions, providing key probability data for subsequent construction of the probability density function.

[0116] S208, based on the representative point values and their corresponding empirical probabilities, construct a probability density function.

[0117] Probability density function: a mathematical function used to describe the probability distribution of a random variable within a certain value range. Here, it is a function constructed based on representative point values and their corresponding empirical probabilities, used to describe the probability distribution of the performance ratio in power transactions.

[0118] With the representative point value as the independent variable and its corresponding empirical probability as the dependent variable, a function is constructed through certain mathematical methods (such as interpolation, fitting, etc.), which is the probability density function. It can intuitively show the probability distribution of different performance ratios (represented by representative point values), helping to analyze the credibility distribution characteristics of power transactions and providing a basis for subsequent decision-making.

[0119] It can be seen that when the ratio of actual delivery power to contract power is calculated for each historical discrete time unit, a set of performance ratio of fulfillment is obtained; after the step of determining the performance ratio of fulfillment as the credibility, the performance ratio of fulfillment is normalized to generate a normalized performance ratio of fulfillment, which is to enable the performance ratio of fulfillment of different orders of magnitude and different ranges to be compared and analyzed under the same standard, and to eliminate the difference of the data itself. Then, from the value range of the normalized performance ratio of fulfillment, a number of representative point values are selected according to the preset interval, which is a reasonable discretization processing of the data for subsequent operation. Then, the value in each normalized performance ratio of fulfillment is mapped to the nearest representative point value according to the minimum distance criterion, realizing the classification and integration of data. Then, the frequency of the normalized performance ratio of fulfillment value of the representative point value is counted, and the frequency of each representative point value is divided by the total number of the value of the normalized performance ratio of fulfillment, to obtain the empirical probability corresponding to each representative point value. Based on these probability information, a probability density function is constructed, so that in the subsequent judgment of power supply situation, the probability of different supply situations reflected by the probability density function can be considered comprehensively, so that the transaction decision based on this has more scientificity and accuracy, reduces the misjudgment of supply risk caused by not considering the probability factor, and better serves the medium and long term transaction decision.

[0120] Steps S107 to S109 are replaced by: step S209, applying the current expected power supply range as an input variable to the probability density function to obtain the probability distribution of the current expected power supply range;

[0121] The values of the current expected power supply range are substituted into the previously constructed probability density function, and the corresponding probability values are obtained by function calculation, so as to obtain the probability distribution of the current expected power supply range. This probability distribution can help understand the possibility of power supply at different values under the current expectation, and provide data support for subsequent judgment of power supply state.

[0122] S210, in the probability distribution of the current expected power supply range, the first total probability falling into the first range is counted;

[0123] In the probability distribution of the current expected power supply range, all probability values belonging to the first range are determined, and then the sum of these probability values is added to obtain the first total probability. Through the first total probability, the overall possibility of power supply in the set first range can be understood, which provides key probability information for judging the power supply state.

[0124] S211, if the first total probability is less than the importance threshold of the current region; then it is determined that the current discrete time unit is in a sufficient supply state, and the next discrete time unit under the current discrete time unit is regarded as the current discrete time unit;

[0125] S212, if the first total probability is not less than the importance threshold of the current region, determining that the current discrete time unit is in the insufficient supply state, and regarding the next discrete time unit of the current discrete time unit as the current discrete time unit.

[0126] It can be seen that, by applying the current expected power supply range as an input variable to the probability density function, the probability distribution of the current expected power supply range can be obtained by means of the previously constructed probability density function, and the probability distribution reflects the possibility of the supply range at different values. Then, the first total probability falling into the first range is counted in the probability distribution of the current expected power supply range, and the supply state of the current discrete time unit is determined by comparing the first total probability with the importance threshold of the current region. The determination of whether the supply is sufficient or not is no longer a simple comparison based on a fixed value, but fully considers the uncertainty and probability characteristics actually existing in the power supply. The transaction decision made in this way is more in line with the actual situation of the complex and changeable power market, and can more accurately perceive the supply risk in advance, provide protection for reasonable arrangement of transactions and avoid economic losses caused by supply risks, and improve the ability of the entire transaction decision assistance method to cope with complex situations.

[0127] The following introduces an example of a power supply and demand balance determination system according to the power supply amount provided by the embodiments of the present application. Figure 3 is an example of the hardware structure of the power supply and demand balance determination system according to the power supply amount provided by the embodiments of the present application.

[0128] In some embodiments, the power supply and demand balance determination system is a computer device or includes a computer device in the power supply and demand balance determination system. The computer device includes a processor, a memory and a network interface connected by a system bus. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store data. The network interface of the computer device is used to communicate with other terminals or servers outside through network connection. In some embodiments, the network interface can be a wired network interface, and in some embodiments, the network interface can also be a wireless network interface. The computer program is executed by the processor to implement the method in the embodiments of the present application.

[0129] Those skilled in the art can understand that, Figure 3The structure shown in the figure is only a block diagram of part of the structure related to the scheme of the present application, and does not constitute a limitation on the computer device to which the scheme of the present application is applied. A specific computer device can include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0130] The above-described embodiments are only used to illustrate the technical scheme of the present application, but not limit it; although the technical scheme of the present application is described in detail with reference to the foregoing embodiments, those skilled in the art should understand that the technical scheme recorded in the foregoing embodiments can be modified, or some technical features can be replaced by equivalents; and these modifications or replacements do not make the essence of the corresponding technical scheme deviate from the scope of the technical scheme of the embodiments of the present application.

[0131] In the above embodiments, according to the context, the term when can be interpreted as meaning if or after or in response to determining or in response to detecting. Similarly, according to the context, the phrase when determining or if detecting (a stated condition or event) can be interpreted as meaning if determining or in response to determining or in response to detecting (a stated condition or event).

[0132] In the above embodiments, all or part of the technical scheme can be realized by software, hardware, firmware or any combination thereof. When realized by software, all or part of the technical scheme can be realized in the form of a computer program product. The computer program product includes one or more computer instructions. When loaded and executed by a computer, the computer instructions produce all or part of the processes or functions described in the embodiments of the present application. The computer can be a general-purpose computer, a special-purpose computer, a computer network or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another, for example, the computer instructions can be transmitted from one website, computer, server or data center to another website, computer, server or data center through a wired (such as coaxial cable, optical fiber, digital subscriber line) or wireless (such as infrared, wireless, microwave, etc.) manner. The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server, data center, etc. integrated with one or more available media. The available medium can be a magnetic medium (for example, floppy disk, hard disk, magnetic tape), an optical medium (for example, DVD), or a semiconductor medium (for example, solid state disk) and the like.

[0133] Those skilled in the art can understand that all or part of the processes in the above-mentioned method embodiments can be implemented by a computer program instructing relevant hardware to complete, the program can be stored in a computer readable storage medium, and the program can include the processes of the above-mentioned method embodiments when executed. The aforementioned storage medium includes ROM or random storage memory RAM, magnetic disc or optical disc and various storage code medium.

Claims

1. A method for determining supply and demand balance based on power supply, characterized in that, include: Obtain all electricity trading schemes for the current region; The power trading scheme defines a set of one or more power purchase and sale information, including contracted electricity volume and execution period; Based on the current power trading scheme and its corresponding credibility, a power supply range sequence is generated; the power supply range sequence is a time series of the expected power supply range in discrete time units within a future preset period; the current power trading scheme is any power trading scheme; wherein, within multiple historical discrete time units, the contracted electricity volume and the corresponding actual delivered electricity volume of the current power trading scheme are obtained; For each historical discrete time unit, the ratio of the actual delivered volume to the contracted volume is calculated to obtain a set of performance ratios; The performance ratio is defined as the credibility. The total power supply range sequence is obtained by summing the power supply range sequences of all power trading schemes. The historical power load time series data of the current region is input into the power load forecasting model to generate a power forecast demand range sequence for discrete time units corresponding to the total power supply range sequence; the power forecast demand range sequence is a time series of the expected power demand range in discrete time units within a future preset period. Extract the current expected power supply range under the current discrete time unit from the total power supply range sequence and the current expected power demand range under the current discrete time unit from the power forecast demand range sequence, wherein the current discrete time unit is any of the discrete time units; When the minimum value of the current expected power supply range falls within the current expected power demand range, a first range is determined between the maximum value of the current expected power demand range and the minimum value of the current expected power supply range. Calculate the first proportion of the first range within the range of current expected electricity demand; If the first proportion is less than the importance threshold of the current region, then the current discrete time unit is determined to be in a state of sufficient supply, and the next discrete time unit is regarded as the current discrete time unit. If the first proportion is not less than the importance threshold of the current region, then the current discrete time unit is determined to be in a state of insufficient supply, and the next discrete time unit is regarded as the current discrete time unit.

2. The method according to claim 1, characterized in that, For each historical discrete time unit, the ratio of the actual delivered volume to the contracted volume is calculated to obtain a set of performance ratios. After determining the performance ratio as the credibility, the method further includes: The performance ratio is normalized to generate a normalized performance ratio. From the range of values ​​for the normalized performance ratio, select several representative point values ​​at preset intervals; Each value in the normalized performance ratio is mapped to the nearest representative point value according to the minimum distance criterion. Count the frequency of the normalized performance ratio values ​​of the representative point values; Divide the frequency of each representative point value by the total number of values ​​of the normalized performance ratio to obtain the empirical probability corresponding to each representative point value. Based on the numerical values ​​of each representative point and its corresponding empirical probability, a probability density function is constructed.

3. The method according to claim 2, characterized in that, The calculation is performed to determine the first percentage of the first range within the current expected electricity demand range. If the first proportion is less than the importance threshold of the current region, then the current discrete time unit is determined to be in a state of sufficient supply, and the next discrete time unit is regarded as the current discrete time unit. If the first proportion is not less than the importance threshold of the current region, then the step of determining that the current discrete time unit is in a state of insufficient supply and considering the next discrete time unit as the current discrete time unit specifically includes: The current expected power supply range is used as an input variable and applied to the probability density function to obtain the probability distribution of the current expected power supply range; The first total probability of falling into the first range is calculated from the probability distribution of the current expected power supply range; If the first total probability is less than the importance threshold of the current region, then the current discrete time unit is determined to be in a state of sufficient supply, and the next discrete time unit is regarded as the current discrete time unit. If the first total probability is not less than the importance threshold of the current region, then the current discrete time unit is determined to be in a state of insufficient supply, and the next discrete time unit is regarded as the current discrete time unit.

4. The method according to claim 1, characterized in that, After the step of extracting the current expected power supply range under the current discrete time unit in the total power supply range sequence and the current expected power demand range under the current discrete time unit in the power forecast demand range sequence, wherein the current discrete time unit is any of the discrete time units, the method further includes: When the maximum value of the current expected power supply range falls within the current expected power demand range, a second range is determined between the maximum value of the current expected power supply range and the minimum value of the current expected power demand range. Calculate the second proportion of the second range within the range of current expected electricity demand; If the first proportion is not less than the importance threshold of the current region, then the current discrete time unit is determined to be in a state of sufficient supply, and the next discrete time unit is regarded as the current discrete time unit. If the first proportion is less than the importance threshold of the current region, then the current discrete time unit is determined to be in a state of insufficient supply, and the next discrete time unit is regarded as the current discrete time unit.

5. The method according to claim 1, characterized in that, After the step of extracting the current expected power supply range under the current discrete time unit in the total power supply range sequence and the current expected power demand range under the current discrete time unit in the power forecast demand range sequence, wherein the current discrete time unit is any of the discrete time units, the method further includes: When the minimum value of the current expected power supply range is greater than the maximum value of the current expected power demand range, the current discrete time unit is determined to be in a state of sufficient supply, and the next discrete time unit is regarded as the current discrete time unit.

6. The method according to claim 1, characterized in that, After the step of extracting the current expected power supply range under the current discrete time unit in the total power supply range sequence and the current expected power demand range under the current discrete time unit in the power forecast demand range sequence, wherein the current discrete time unit is any of the discrete time units, the method further includes: If the maximum value of the current expected power supply range is less than the minimum value of the current expected power demand range, the current discrete time unit is determined to be in a state of insufficient supply, and the next discrete time unit is regarded as the current discrete time unit.

7. A system for determining supply and demand balance based on power supply, characterized in that, The system for determining the supply and demand balance based on power supply includes: One or more processors and memory; The memory is coupled to the one or more processors, and the memory is used to store computer program code, the computer program code including computer instructions, and the one or more processors call the computer instructions to cause the power supply and demand balance determination system to perform the method as described in any one of claims 1-6.

8. A computer program product containing instructions, characterized in that, When the computer program product is running on the power supply and demand balance determination system, the power supply and demand balance determination system performs the method as described in any one of claims 1-6.

9. A computer-readable storage medium comprising instructions, characterized in that, When the instruction is executed on the power supply and demand balance determination system, the power supply and demand balance determination system performs the method as described in any one of claims 1-6.

Images