Electric power spot market transaction electricity price prediction method

Abstract

The invention relates to the field of transaction electricity price prediction, in particular to an electricity spot market transaction electricity price prediction method. According to the invention, on the basis of the power capacity data of the power spot market, the historical accidental outage capacity of each unit time period in each historical day is obtained and the periodicity is analyzed, and the total estimated available capacity is determined by combining the environmental parameters of each unit time period of the prediction day and the total load analysis accidental outage capacity of the system. Calculating a spinning reserve rate and obtaining a corresponding initial electricity price in combination with the total system load of each unit time period of the prediction day, constructing a temperature trajectory vector of each unit time period of each historical day according to the sensible temperature of the same month over the years of the prediction day, and counting the electricity price change rates corresponding to different temperature trajectory vectors; according to the sensible temperature variable quantity of each unit time period of the prediction day, all the time periods to be corrected are screened, the initial electricity price is corrected according to the electricity price change rate corresponding to the temperature trajectory vector of each time period to be corrected, and the accuracy of electricity price prediction in a complex environment is improved.

Description

Technical Field

[0001] This invention relates to the field of electricity price forecasting, and specifically to a method for forecasting electricity prices in the spot market. Background Technology

[0002] In the electricity spot market, price fluctuations directly affect the decisions and profits of market participants, and also relate to the stable operation of the electricity market. Accurately predicting transaction prices is of great significance for power companies' production scheduling, electricity users' electricity consumption planning, and market regulators' control and management. Therefore, it is essential to develop a more accurate method for predicting electricity spot market transaction prices.

[0003] However, existing technologies have the following problems: 1. Existing technologies rely heavily on historical electricity price data and simple load correlation analysis, ignoring the dynamic changes in unexpected outage capacity in the power system. The uncertainty of unexpected outage capacity will affect the power supply capacity, and thus affect the supply and demand balance, which is an important factor affecting electricity prices.

[0004] 2. Existing technologies predict electricity prices for future periods by combining temperature values ​​from different seasons with historical data. While considering that different temperatures affect users' electricity demand, they do not take into account the indirect impact of temperature change trajectories on the rate of change in electricity prices, and ignore the impact of sudden temperature changes on electricity prices. This can lead to some deviation between the predicted and actual electricity prices. Summary of the Invention

[0005] The purpose of this invention is to overcome the shortcomings of the prior art and provide a method for predicting electricity spot market transaction prices. This method calculates the spinning reserve rate by accurately analyzing unexpected outage capacity and combining it with the total system load to obtain the initial electricity price. Furthermore, it introduces a temperature trajectory vector to quantify the impact of temperature change trends on the rate of change of electricity prices and corrects the initial electricity price, thereby achieving accurate prediction of electricity spot market transaction prices.

[0006] The technical solution adopted by the present invention to solve its technical problem is: a method for predicting electricity prices in the electricity spot market, including: S1, obtaining the historical unexpected outage capacity of each unit time period on each historical day based on the electricity capacity data of the electricity spot market within a historical set time period.

[0007] S2. Analyze the periodicity of historical unexpected outage capacity for each unit time period, and combine the environmental parameters and total system load for each unit time period of the forecast day to analyze the unexpected outage capacity and determine the total estimated available capacity.

[0008] S3. Obtain the total system load for each unit time period on the forecast day based on the published system load forecast curve, calculate the spinning reserve rate in combination with the total estimated available capacity, and obtain the corresponding initial electricity price based on the spinning reserve rate for each unit time period on the forecast day.

[0009] S4. Based on the month of the predicted date, obtain the perceived temperature for each unit time period of each historical day in the same month of previous years, construct the temperature trajectory vector for each unit time period, and calculate the electricity price change rate corresponding to different temperature trajectory vectors.

[0010] S5. Based on the change in perceived temperature for each unit of time period of the predicted day, filter all time periods to be corrected, obtain the corresponding electricity price change rate based on the temperature trajectory vector of each time period to be corrected, and correct the initial electricity price of the time period to be corrected.

[0011] Compared with the prior art, the present invention has the following beneficial effects: (1) The present invention analyzes the periodicity of the historical unexpected outage capacity of each unit time period, and combines the environmental parameters and total system load of each unit time period of the prediction day to analyze the unexpected outage capacity and determine the total estimated available capacity, which provides key input for the accurate calculation of the subsequent initial electricity price and avoids the distortion of electricity price prediction due to the deviation of available capacity estimation.

[0012] (2) This invention obtains the total system load for each unit time period of the forecast day based on the published system load forecast curve, calculates the spinning reserve rate in combination with the total estimated available capacity, obtains the corresponding initial electricity price based on the spinning reserve rate for each unit time period of the forecast day, quantifies the supply and demand balance by calculating the spinning reserve rate, so that the electricity price forecast has clear supply and demand data support, avoids subjective estimation deviation, and ensures the rationality of the initial electricity price forecast.

[0013] (3) This invention obtains the perceived temperature of each unit time period of each historical day in the same month of the predicted date based on the month of the predicted date, constructs the temperature trajectory vector of each unit time period, and statistically analyzes the electricity price change rate corresponding to different temperature trajectory vectors. This breaks through the limitations of traditional single temperature value analysis and takes into account the electricity price change rate corresponding to the temperature change characteristics under different seasons, making the analysis of the impact of temperature on electricity price more targeted.

[0014] (4) This invention filters all time periods to be corrected based on the change in perceived temperature of each unit time period of the predicted day, obtains the corresponding electricity price change rate based on the temperature trajectory vector of each time period to be corrected, and corrects the initial electricity price of the time period to be corrected. This invention specifically addresses the impact of sudden temperature changes that were not considered in the initial electricity price, making the final prediction result more in line with the actual electricity consumption scenario and improving the accuracy and practicality of the prediction. Attached Figure Description

[0015] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0016] Figure 1 This is a schematic diagram of the method steps of the present invention.

[0017] Figure 2 This is a schematic diagram illustrating the steps for obtaining the initial electricity price for each time period in this invention.

[0018] Figure 3 This is a schematic diagram of the steps involved in the method for correcting the initial electricity price for the time period to be corrected in this invention. Detailed Implementation

[0019] Various exemplary embodiments of the present invention will now be described in detail with reference to the accompanying drawings. It should be noted that, unless otherwise specifically stated, the relative arrangement, numerical expressions, and values ​​of the components and steps set forth in these embodiments do not limit the scope of the invention. Furthermore, it should be understood that, for ease of description, the dimensions of the various parts shown in the drawings are not drawn to actual scale.

[0020] The following description of at least one exemplary embodiment is merely illustrative and is in no way intended to limit the invention or its application or use. Techniques, methods, and apparatus known to those skilled in the art may not be discussed in detail, but where appropriate, such techniques, methods, and apparatus should be considered part of the specification.

[0021] In all examples shown and discussed herein, any specific values ​​should be interpreted as merely exemplary and not as limitations. Therefore, other examples of exemplary embodiments may have different values.

[0022] Please see Figure 1 As shown, the present invention provides a method for predicting electricity prices in the electricity spot market, including: S1, obtaining the historical unexpected outage capacity of each unit time period on each historical day based on the electricity capacity data of the electricity spot market within a historical set time period.

[0023] Each time period is divided into 24 time periods per day, with each time period being one hour. For example, the 9 o'clock time period refers to the continuous one hour from 9:00:00 to 9:59:59.

[0024] Considering that unplanned outage capacity is the total power output of generator units in the power system that cannot operate due to unplanned faults, sudden problems, or other unexpected factors, its magnitude directly affects the actual power supply capacity of the power system.

[0025] Furthermore, considering that historical unexpected outage capacity is the key basis for analyzing and predicting daily unexpected outage capacity, all subsequent analyses must be based on the historical unexpected outage capacity for each historical day and corresponding time period. Only by accurately obtaining the historical unexpected outage capacity through difference calculation can we provide support for the accurate analysis of predicting daily unexpected outage capacity.

[0026] In a specific embodiment of the present invention, the power capacity data includes total installed capacity, total available capacity and total planned maintenance capacity, wherein the total installed capacity refers to the total power generation capacity of all generator units owned by all power companies in the power market under rated operating conditions, which is the maximum total amount of power that the power system can theoretically output.

[0027] Total available capacity refers to the total capacity of generator units that can actually be put into operation by each power company within a unit of time period. It only includes the capacity that is in normal operation, that is, the total amount of electricity that can be output at present.

[0028] Total planned maintenance capacity refers to the total amount of electricity that is not output due to the maintenance and repair of generator units within a unit of time, as the power company formulates a maintenance plan in advance.

[0029] The total installed capacity equals the sum of the total planned maintenance capacity, the total available capacity, and the capacity for unexpected outages.

[0030] This invention obtains the historical unexpected outage capacity of each unit time period on each historical day based on the total installed capacity, total available capacity, and total planned maintenance capacity within a historically set time period. This provides a direct reference for predicting the unexpected outage capacity of each unit time period on subsequent prediction days, enhancing the practicality of the prediction method and its engineering implementation capability.

[0031] S2. Analyze the periodicity of historical unexpected outage capacity for each unit time period, and combine the environmental parameters and total system load for each unit time period of the forecast day to analyze the unexpected outage capacity and determine the total estimated available capacity.

[0032] Considering that the unexpected outage capacity may have a periodic pattern due to equipment aging, if there is a periodic pattern, the unexpected outage capacity of the predicted day can be quickly determined based on the periodic pattern. Otherwise, it is necessary to consider key influencing factors such as environmental factors and total system load, combined with historical data, to comprehensively analyze the unexpected outage capacity of each unit time on the predicted day.

[0033] In a preferred embodiment of the present invention, the method for calculating the unexpected outage capacity for each unit time period of the predicted day includes: S21, extracting the total installed capacity, total available capacity and total planned maintenance capacity for each unit time period corresponding to each historical day from the power capacity data of the power spot market within a historical set time period, and calculating the historical unexpected outage capacity for each unit time period corresponding to each historical day by means of the difference.

[0034] Specifically, the historical unplanned outage capacity for each historical day is calculated by obtaining the difference between the total installed capacity and the total available capacity for each unit time period corresponding to each historical day, and then calculating the difference with the total planned maintenance capacity.

[0035] S22. Statistically analyze the historical unexpected outage capacity of each unit time period on each historical day, and construct a time series sequence of the historical unexpected outage capacity corresponding to each unit time period.

[0036] S23. Obtain the high value point of the historical unexpected outage capacity corresponding to each unit time period, and determine whether the historical unexpected outage capacity has periodicity.

[0037] The method for obtaining the high-value points includes: firstly, performing a difference operation on the historical accidental outage capacity time series corresponding to each unit time period to obtain the accidental outage capacity change difference sequence.

[0038] Next, we screen the turning points where the sign of the difference in unexpected shutdown capacity changes changes from positive to negative, and mark the historical dates corresponding to the turning points as candidate high value points.

[0039] Then, the historical unexpected outage capacity of the candidate high value point is compared with the historical unexpected outage capacity of its neighboring historical days. If the unexpected outage capacity of the candidate high value point is greater than the unexpected outage capacity of its neighboring historical days, it is recorded as a high value point.

[0040] In a specific embodiment of the present invention, at least two historical days before and after the current historical day can be selected as its neighboring historical day. Preferably, in this embodiment, the neighboring historical day is set to three historical days before and after the current historical day.

[0041] Finally, all high-value points in the historical accidental outage capacity time series corresponding to each unit time period are statistically analyzed.

[0042] In addition, it should be noted that the method for determining whether the historical unexpected outage capacity is periodic includes: selecting the maximum historical unexpected outage capacity from the historical unexpected outage capacity corresponding to all high-value points in each unit time period, and setting a capacity difference threshold based on the unit percentage of the maximum historical unexpected outage capacity.

[0043] To prevent individual abnormally high values ​​from affecting the overall periodicity assessment, a statistically based quartile method is used in this specific embodiment to identify and remove abnormally high values ​​from historical capacity data. The specific implementation process of the quartile method is as follows: First, the first and third quartiles of the historical unexpected outage capacity corresponding to all high-value points in each unit time period are found. The interquartile range is calculated using the difference between the first and third quartiles. The sum of the third quartile and 1.5 times the interquartile range is used as the outage boundary, and historical unexpected outage capacity exceeding the outage boundary is removed. Finally, the largest historical unexpected outage capacity is selected from the remaining historical unexpected outage capacity.

[0044] Specifically, in this embodiment, 1% of the maximum historical unexpected outage capacity is used as the capacity difference threshold. The implementer can also set other specific values. The purpose of setting the capacity difference threshold is to filter out high-value points similar to the maximum historical unexpected outage capacity.

[0045] Calculate the capacity difference between the historical unexpected outage capacity corresponding to all high value points and the maximum historical unexpected outage capacity, and record the days with a capacity difference less than the capacity difference threshold as high value days.

[0046] Understandably, when the difference between a certain high value point and the maximum historical unexpected outage capacity is less than the capacity difference threshold, it can be considered that the historical unexpected outage capacity of the high value point is similar to the maximum historical unexpected outage capacity. The similar historical unexpected outage capacity is an important basis for analyzing periodicity. Therefore, it can be preliminarily considered that there may be a periodic relationship between the high value day and the high value day corresponding to the maximum historical unexpected outage capacity, which has reference value for subsequent periodicity analysis.

[0047] The high-value days are grouped according to the number of adjacent high-value days, and a high-value day sequence corresponding to the group with different numbers of adjacent high-value days is constructed. The date difference of each high-value day in each high-value day sequence is compared to see if they are the same.

[0048] If all groups of date differences in a certain date difference sequence are the same, then the historical unexpected outage capacity is considered to have periodicity; otherwise, the historical unexpected outage capacity is considered not to have periodicity.

[0049] If the historical unexpected outage capacity is periodic, the period is formed by the first and last high-value days in each group, and the date difference between each group is used as the period length.

[0050] In a specific example of the present invention, it is assumed that the consecutive high-value days A, B, C, D, E, and F are grouped into groups of 3, that is, every three high-value days form a group, resulting in two groups, ABC and DEF. The date difference x between high-value days A and C, and the date difference y between D and F are calculated respectively. If the date differences x and y are equal, it is determined that the historical unexpected shutdown capacity has periodicity, and the corresponding period is formed between A and C, with a period length of x.

[0051] S24. If the historical unexpected outage capacity of a certain unit time period is periodic, based on the position of the prediction date within the corresponding period, obtain the historical unexpected outage capacity at the same position in each period of the historical unexpected outage capacity time series, and take the average of these as the unexpected outage capacity of that unit time period on the prediction date.

[0052] S25. Conversely, based on the environmental parameters and total system load of each unit time period in each historical day, a comprehensive analysis is conducted to predict the unexpected outage capacity of each unit time period on the day.

[0053] It should be noted that the specific analysis method for the unexpected outage capacity of each unit time period on the predicted day includes: obtaining the environmental parameters and total system load of each unit time period on each historical day from the power system database. The environmental parameters include average temperature, maximum wind speed and weather type.

[0054] Based on the environmental parameters and total system load for each time period of the forecast date, historical unexpected outage capacity for each historical day that is identical to the environmental parameters and total system load for each time period of the forecast date is selected.

[0055] It should be noted that if there is no historical data with identical environmental parameters and total system load, interpolation should be used to estimate the unexpected outage capacity.

[0056] Specifically, considering that the environment has a greater impact on unexpected outage capacity, when environmental parameters are the same, the operating environment and fault risk level of generator units are basically the same. Therefore, we prioritize finding historical days with the same environmental parameters as a certain unit period of the forecast date, and select the system total load values ​​of two historical days as the closest smaller and larger values ​​of the system total load for that unit period. We obtain the unexpected outage capacity corresponding to the two sets of data, and use the system total load value as the independent variable and the unexpected outage capacity as the dependent variable. We calculate the unexpected outage capacity of the target unit period using linear interpolation.

[0057] If there is no historical day with the exact same environmental parameters as a unit of time period on the predicted date, then historical days with the same weather type are selected first, and then other parameters are matched. Finally, the unexpected outage capacity of the target unit of time period is calculated by linear interpolation.

[0058] After removing the maximum and minimum values, the average of the historical unexpected outage capacity for each remaining historical day is taken as the unexpected outage capacity for each unit time period.

[0059] This invention analyzes the periodicity of historical unplanned outage capacity in each time period, and combines it with environmental parameters and total system load in each time period of the forecast day to determine the total estimated available capacity. This provides a key input for the accurate calculation of the subsequent initial electricity price and avoids distortion of electricity price prediction due to deviations in available capacity estimation.

[0060] S3. Obtain the total system load for each unit time period on the forecast day based on the published system load forecast curve, calculate the spinning reserve rate in combination with the total estimated available capacity, and obtain the corresponding initial electricity price based on the spinning reserve rate for each unit time period on the forecast day.

[0061] Given that the spinning reserve ratio is a crucial indicator for measuring the supply-demand balance and safety margin of a power system across different time periods, a higher spinning reserve ratio indicates a greater available generation capacity and lower electricity prices, while a lower spinning reserve ratio indicates a tighter generation capacity and higher electricity prices. Therefore, electricity prices and the spinning reserve ratio are inversely related. Thus, the spinning reserve ratio is a core characteristic parameter for predicting electricity prices; ignoring it will lead to inaccurate electricity price forecasts.

[0062] In a specific embodiment of the present invention, the method for calculating the spinning reserve rate includes: obtaining the difference between the total estimated available capacity and the total system load for each unit time period of the forecast day, and recording the ratio of the difference to the corresponding total system load as the spinning reserve rate.

[0063] like Figure 2 As shown, the method for obtaining the initial electricity price corresponding to each unit time period of the predicted day includes: W1, obtaining the spinning reserve ratio of each unit time period in each historical day within a historical set time period, dividing the spinning reserve ratio into partitions according to a set value to obtain each spinning reserve ratio interval.

[0064] The set value can be set to 5%, 10%, etc. In a preferred embodiment of the present invention, the rotational reserve rate is divided into 10%.

[0065] W2. Filter the electricity prices of each historical day within the same spinning reserve ratio range for each time period, and use the mode of these prices as the electricity price within the corresponding spinning reserve ratio range.

[0066] W3. Calculate the electricity price for each unit time period under different spinning reserve ratio ranges and construct an electricity price correlation table.

[0067] W4. Based on the spinning reserve ratio of each time period on the forecast date, query the electricity price correlation table and output the corresponding electricity price, which is then recorded as the initial electricity price.

[0068] This invention obtains the total system load for each unit time period of the forecast day based on the published system load forecast curve, calculates the spinning reserve rate in combination with the total estimated available capacity, obtains the corresponding initial electricity price based on the spinning reserve rate for each unit time period of the forecast day, and quantifies the degree of supply and demand balance by calculating the spinning reserve rate, so that the electricity price forecast has clear supply and demand data to support it, avoids subjective estimation bias, and ensures the rationality of the initial electricity price forecast.

[0069] S4. Based on the month of the predicted date, obtain the perceived temperature for each unit time period of each historical day in the same month of previous years, construct the temperature trajectory vector for each unit time period, and calculate the electricity price change rate corresponding to different temperature trajectory vectors.

[0070] Considering that the use of cooling and heating equipment will increase the total system load during the high-temperature period in summer and the cold period in winter, temperature is an important factor affecting the total system load and also an important factor affecting electricity prices.

[0071] Furthermore, considering that residents use cooling and heating equipment based on perceived temperature rather than the dry-bulb temperature reported in weather forecasts, and that factors affecting actual perceived temperature include relative humidity and wind speed, it is necessary to analyze the perceived temperature for each time period.

[0072] Based on this, in a preferred embodiment of the present invention, the method for obtaining the perceived temperature of each unit time period of each historical day in the same month of the same year includes: extracting the dry-bulb temperature, relative humidity and wind speed of each unit time period of each historical day in the same month of the same year from the meteorological data platform.

[0073] The dry-bulb temperature, relative humidity, and wind speed of each historical day in the same month of previous years are substituted into the formula for calculating the perceived temperature to obtain the perceived temperature for each unit of time period.

[0074] The formula for calculating perceived temperature is as follows: .

[0075] in Represents perceived temperature. Represents dry bulb temperature. Represents water vapor pressure. This represents wind speed. Values ​​such as 0.33, 0.70, and 4.00 in the formula are all set values ​​in the existing formula for calculating perceived temperature.

[0076] The formula for water vapor pressure is: .

[0077] in This represents the relative humidity of the air. Values ​​such as 100, 6.105, 17.27, and 237.7 in the formula are all set values ​​in the existing formula for calculating perceived temperature.

[0078] The statistical method for the electricity price change rate corresponding to each temperature trajectory vector includes: based on the perceived temperature of each unit time period of each historical day in the same month of previous years, constructing the temperature trajectory vector of each unit time period of each historical day and the perceived temperature of the previous unit time period.

[0079] For the 0:00 time period of a certain day, the corresponding perceived temperature T is... t The perceived temperature T at 11 PM the previous day t-1 The two-dimensional temperature trajectory vector is specifically expressed as [T t-1 T t ].

[0080] Calculate the difference between the electricity price for each unit period in each historical day and the electricity price for the previous unit period. The ratio of the electricity price difference to the electricity price for the previous unit period is recorded as the electricity price change rate.

[0081] Statistically calculate the rate of change of electricity price for all trajectories with the same temperature, and use their average value as the rate of change of electricity price for each temperature trajectory vector.

[0082] This invention obtains the perceived temperature for each unit time period of each historical day in the same month of previous years based on the month in which the prediction date is located, constructs a temperature trajectory vector for each unit time period, and statistically analyzes the electricity price change rate corresponding to different temperature trajectory vectors. This breaks through the limitations of traditional single temperature value analysis, takes into account the electricity price change rate corresponding to the temperature change characteristics under different seasons, and makes the analysis of the impact of temperature on electricity prices more targeted.

[0083] S5. Based on the change in perceived temperature for each unit of time period of the predicted day, filter all time periods to be corrected, obtain the corresponding electricity price change rate based on the temperature trajectory vector of each time period to be corrected, and correct the initial electricity price of the time period to be corrected.

[0084] In a preferred embodiment of the present invention, the temperature change refers to the temperature difference between each unit time period and the previous unit time period.

[0085] When the change in perceived temperature during a certain time period of a predicted day exceeds the set threshold for the change in perceived temperature, that time period is recorded as a period to be corrected.

[0086] The threshold for the change in perceived temperature is set at 3°C, but implementers can modify it to other specific values.

[0087] like Figure 3 As shown, the specific steps for correcting the initial electricity price of the period to be corrected include: S51, taking the initial electricity price of each non-period to be corrected period of the forecast date as the corresponding final forecast electricity price.

[0088] S52. Obtain the difference between the final predicted electricity price of the previous period and the initial electricity price of the corresponding period to be corrected for each period to be corrected. Record the ratio of this difference to the final predicted electricity price of the previous period as the initial electricity price change rate. Compare the initial electricity price change rate of each period to be corrected with the corresponding electricity price change rate.

[0089] In addition, it should be added that if the 0:00 time period of the forecast day is a time period to be revised, the final forecast electricity price of the previous time period is actually the actual electricity price announced at 23:00 of the previous day.

[0090] S53. If the initial rate of change of electricity price for a certain period to be corrected is different from the corresponding rate of change of electricity price, then the product of the final predicted electricity price of the previous period and the corresponding rate of change of electricity price shall be used as the electricity price correction value.

[0091] S54. The sum of the final predicted electricity price of the period preceding the period to be corrected and the corresponding electricity price correction value shall be used as the final predicted electricity price.

[0092] It should be noted that the electricity price correction process of the present invention is executed sequentially in unit time periods. When there are multiple consecutive time periods to be corrected, the correction process will be carried out in sequence. The final predicted electricity price after correction in the previous time period will be used as the benchmark for correction in the next time period to ensure the continuity of the electricity price change trajectory.

[0093] As an example, suppose the 10 AM time slot tomorrow is the period to be corrected, with an initial electricity price of 210 yuan / MWh. The final electricity price at 9 AM is 200 yuan / MWh. The initial change rate of the electricity price at 10 AM is calculated to be 5%. Assuming the change rate of the electricity price calculated based on the temperature trajectory vector at 10 AM is 8%, and comparing the initial change rate of the electricity price at 10 AM with the corresponding change rate, we find that they are different. Based on the final electricity price at 9 AM and the corresponding change rate, the calculated electricity price correction value is the product of the two, which is 16 yuan / MWh. Therefore, the final predicted electricity price for the 10 AM time slot is 216 yuan / MWh.

[0094] S55. Conversely, the initial electricity price for the period to be corrected is recorded as the final predicted electricity price.

[0095] Finally, the final predicted electricity prices for each time period on the forecast date are compiled to create a forecast price table for the forecast date, which is convenient for electricity market participants to access directly.

[0096] This invention filters all time periods to be corrected based on the perceived temperature changes in each unit of time on the predicted day, obtains the corresponding electricity price change rate based on the temperature trajectory vector of each time period to be corrected, and corrects the initial electricity price of the time period to be corrected. This invention specifically addresses the impact of sudden temperature changes that were not considered in the initial electricity price, making the final prediction result more consistent with actual electricity consumption scenarios and improving prediction accuracy and practicality.

[0097] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented, in whole or in part, in the form of a computer program product.

[0098] Those skilled in the art will recognize that the modules and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0099] In addition, the functional modules in the various embodiments of this application can be integrated into one processing module, or each module can exist physically separately, or two or more modules can be integrated into one module.

[0100] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

[0101] Finally, the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for predicting a power spot market transaction price, characterized by, The method comprises the following steps: S1, obtaining historical unexpected outage capacity of each unit period on each historical day based on power capacity data of the power spot market in a historical setting time period; S2, analyzing periodicity of the historical unexpected outage capacity of each unit period, combining environmental parameters and system total load of each unit period of the forecast day to analyze the unexpected outage capacity, and determining total estimated available capacity; S3, obtaining system total load of each unit period of the forecast day according to a published system load prediction curve, combining the total estimated available capacity to calculate spinning reserve rate, and obtaining corresponding initial electricity price according to the spinning reserve rate of each unit period of the forecast day; S4, obtaining the sensible temperature of each unit period of each historical day in the same period of the past years according to the month in which the forecast day is located, constructing a temperature trajectory vector of each unit period, and counting the electricity price change rate corresponding to different temperature trajectory vectors; S5, screening all to-be-corrected periods according to the sensible temperature change amount of each unit period of the forecast day, obtaining the corresponding electricity price change rate according to the temperature trajectory vector of each to-be-corrected period, and correcting the initial electricity price of the to-be-corrected period.

2. The method of claim 1, wherein, The method for calculating the unexpected outage capacity of each unit period of the forecast day comprises: extracting total installed capacity, total available capacity and total planned maintenance capacity of each unit period of each historical day from power capacity data of the power spot market in a historical setting time period, and obtaining the historical unexpected outage capacity of each unit period of each historical day by difference calculation; counting the historical unexpected outage capacity of each unit period on each historical day, and constructing a historical unexpected outage capacity time sequence corresponding to each unit period; obtaining high points of each unit period according to the historical unexpected outage capacity time sequence corresponding to each unit period, and judging whether the historical unexpected outage capacity has periodicity; if the historical unexpected outage capacity of a unit period has periodicity, obtaining the historical unexpected outage capacity of each cycle at the same position in the historical unexpected outage capacity time sequence according to the position of the forecast day in the corresponding cycle, and taking the mean value as the unexpected outage capacity of the unit period of the forecast day; otherwise, comprehensively analyzing the unexpected outage capacity of each unit period of the forecast day according to environmental parameters and system total load of each unit period on each historical day.

3. The method of claim 2, wherein, The method for obtaining the high points comprises: performing difference operation on the historical unexpected outage capacity time sequence corresponding to each unit period to obtain an unexpected outage capacity change difference value sequence; screening turning points corresponding to a change from positive to negative of the unexpected outage capacity change difference value, and marking the historical day corresponding to the turning point as a candidate high point; comparing the historical unexpected outage capacity of the candidate high point with the historical unexpected outage capacity of its adjacent historical day, respectively, if the unexpected outage capacity of the candidate high point is greater than the unexpected outage capacity of its adjacent historical day, then the candidate high point is recorded as a high point; counting all high points in the historical unexpected outage capacity time sequence corresponding to each unit period.

4. The method of claim 3, wherein, The method for judging whether the historical unexpected outage capacity has periodicity comprises: screening the maximum historical unexpected outage capacity from the historical unexpected outage capacity of all high points of each unit period, and setting a capacity difference threshold according to the unit percentage of the maximum historical unexpected outage capacity; Calculate the capacity difference value of the historical unexpected outage capacity corresponding to each high-value point and the maximum historical unexpected outage capacity, and record the capacity difference value less than the capacity difference value threshold as a high-value day; Group each high-value day according to the number of adjacent high-value days, construct a high-value day sequence corresponding to different adjacent high-value day groupings, and compare whether the date difference of each group of high-value days in each high-value day sequence is the same; If the date difference of each group in a certain date difference sequence is the same, it is judged that the historical unexpected outage capacity has periodicity, otherwise it is judged that the historical unexpected outage capacity does not have periodicity.

5. The method of claim 2, wherein, The specific analysis method of the unexpected outage capacity of each unit time period of the prediction day includes: Obtain the environmental parameters and system total load of each unit time period in each historical day from the power system database, and the environmental parameters include average temperature, maximum wind speed and weather type; According to the environmental parameters and system total load of each unit time period of the prediction day, the historical unexpected outage capacity of each historical day with the same environmental parameters and system total load as each unit time period of the prediction day is screened out; Remove the maximum and minimum values, and take the average of the historical unexpected outage capacity of the remaining historical days as the unexpected outage capacity of each unit time period.

6. The method of claim 1, wherein, The calculation method of the spinning reserve rate includes: Obtain the difference between the total estimated available capacity and the system total load of each unit time period of the prediction day, and record the ratio of the difference to the corresponding system total load as the spinning reserve rate.

7. The method of predicting a transaction price of an electricity spot market according to claim 6, wherein The method for obtaining the initial electricity price corresponding to each unit time period of the prediction day includes: Divide the spinning reserve rate into zones according to a set value, and obtain each spinning reserve rate interval; Screen the electricity prices of each historical day under the same spinning reserve rate interval, and take the mode as the electricity price under the corresponding spinning reserve rate interval; Statistical electricity prices under different spinning reserve rate intervals, and construct an electricity price correlation table; According to the spinning reserve rate of each unit time period of the prediction day, query the electricity price correlation table to output the corresponding electricity price, and record it as the initial electricity price.

8. The method of claim 1, wherein, The method for obtaining the apparent temperature of each unit time period of each historical day in the same period month of each year includes: Extract the dry-bulb temperature, air relative humidity and wind speed of each unit time period of each historical day in the same period month of each year from the meteorological data platform; Put the dry-bulb temperature, air relative humidity and wind speed of each unit time period of each historical day in the same period month of each year into the apparent temperature calculation formula to obtain the apparent temperature of each unit time period.

9. The method of claim 8, wherein, The statistical method of the electricity price change rate corresponding to each temperature trajectory vector includes: According to the apparent temperature of each unit time period of each historical day in the same period month of each year, the apparent temperature of each unit time period and its previous unit time period in each historical day forms a temperature trajectory vector of each unit time period; Calculate the electricity price difference value of each unit time period and its previous unit time period, and record the ratio of the electricity price difference value to the previous unit time period electricity price as the electricity price change rate; Statistical electricity price change rate of the same temperature trajectory vector, and take its average value as the electricity price change rate corresponding to each temperature trajectory vector.

10. The method of claim 9, wherein, The specific steps for correcting the initial electricity price of the to-be-corrected time period include: The initial electricity price of each non-revised time period of the predicted day is taken as the corresponding final predicted electricity price; The difference between the final predicted electricity price of the previous time period of each revised time period and the initial electricity price of the corresponding revised time period is obtained, and the ratio of the difference to the final predicted electricity price of the previous time period is taken as the initial electricity price change rate. The initial electricity price change rate of each revised time period is compared with the corresponding electricity price change rate; If the initial electricity price change rate of a certain revised time period is different from the corresponding electricity price change rate, the product of the final predicted electricity price of the previous time period of the revised time period and the corresponding electricity price change rate is taken as the electricity price correction value; The sum of the final predicted electricity price of the previous time period of the revised time period and the corresponding electricity price correction value is taken as the final predicted electricity price; Otherwise, the initial electricity price of the revised time period is taken as the final predicted electricity price.

Images