Long-term electric power spot market electricity price prediction method based on thermal power load rate

Abstract

The invention discloses a thermal power load rate-based long-term electric power spot market electricity price prediction method, and the method comprises the steps: constructing an electricity price normal probability distribution model corresponding to a target load rate in combination with a load rate-electricity price fitting curve for any target month and the target load rate under the corresponding target month; according to a given wind and light output curve and a total load curve of a power system, combined with the operation domain constraint of the extraction condensing unit and the parameters of the extraction condensing unit and the straight condensing unit, the thermal power starting capacity is updated in real time, so that the actual thermal power load rate and the minimum load rate of the power system are calculated; outputting the electricity price according to the size relation between the actual load rate and the minimum load rate, and obtaining the electricity price in each time period of the whole year. According to the method, the problem that the electricity price simulation result of the long-term electric power spot market is inaccurate due to the fact that the thermal power load rates in different seasons and different time periods are obviously different due to the influence of factors such as power utilization requirements and unit starting conditions and the like and the difference of the electricity price in different seasons and different time periods is directly caused is solved.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of power system price simulation technology, and particularly relates to a long-term power spot market price prediction method based on thermal power load rate. BACKGROUND

[0002] With the deepening of the power market reform, accurate simulation of market electricity price is of great significance. The load rate of thermal power directly reflects the degree of system supply and demand tension, and is a key factor affecting market electricity price. However, due to the influence of factors such as electricity demand and unit start-up conditions, the thermal power load rate in different seasons and time periods shows significant differences, which will directly lead to differences in electricity prices in different seasons and time periods.

[0003] In power system planning and operation, it is necessary to determine a reasonable power system reserve capacity to ensure system reliability. However, such long-term technical decisions are highly dependent on the prediction of future power spot market prices. The traditional price prediction method often fails to fully consider the influence of thermal power load rate and extraction-condensing unit start-up capacity, resulting in large prediction deviation. The decision-making scheme based on this deviation may result in technical consequences of excess or insufficient reserve capacity. Therefore, the prior art does not provide a long-term price generation method that more accurately reflects the actual operation characteristics of the power system, and thus cannot make optimized long-term technical decisions based on it. There is an urgent need for a power spot market price prediction method that takes thermal power load rate as the dominant factor, fits the price curve according to the monthly characteristics of thermal power load rate, fully considers the unit operation constraints, and generates random fluctuations in electricity price, to provide effective support for power market-related decisions. SUMMARY

[0004] The present application provides a long-term power spot market price prediction method based on thermal power load rate to overcome the above technical problems.

[0005] In order to achieve the above purpose, the technical scheme of the present application is: A long-term power spot market price prediction method based on thermal power load rate, specifically comprising the following steps: S1: Obtain historical data and divide the historical data by month to fit 12 load rate-price fitting curves for the whole year; The historical data includes historical actual thermal power load rate and historical actual market electricity price; S2: For any target month and target load rate in the corresponding target month, the fitting curve value corresponding to the target load rate is taken as the mean value, and the standard deviation is calculated based on the historical actual electricity price corresponding to the target load rate as a sample to construct a price normal probability distribution model corresponding to the target load rate; S3: According to the given wind and light output curve and the total load curve of the power system, and combining the extraction condensing unit operation domain constraint and the extraction condensing unit and pure condensing unit parameter, the actual thermal power load rate and the minimum load rate of the power system are calculated by updating the thermal power start-up capacity. S4: Determine whether the actual thermal power load rate is greater than or equal to the minimum load rate of the power system; If not, define the current long-term electricity spot market price as 0; If yes, based on the normal probability distribution model of the price, a normal probability distribution price cluster corresponding to the actual thermal power load rate is obtained, and a price is randomly selected from the normal probability distribution price cluster according to the probability distribution as the final output price; S5: Based on S2 to S4, the monthly price is obtained, the long-term electricity spot market price based on the thermal power load rate is simulated, and the long-term electricity spot market price based on the thermal power load rate is predicted, so as to guide the price decision in the process of power system planning and operation.

[0006] Further, the fitting formula of the load rate-price fitting curve in S1 is: (1) In the formula: represents the fitting curve value of the load rate-price fitting curve; represents the month and takes the value of 1-12; represents the fitting coefficient of the month; represents the target load rate.

[0007] Further, the normal probability distribution model of the price corresponding to the target load rate constructed in S2 is expressed as: (2) (3) In the formula: represents the target price; represents the fitting curve value corresponding to the target load rate of the month; represents the standard deviation corresponding to the target load rate of the month; represents the number of historical actual market price samples corresponding to the target load rate of the month; represents the historical actual market price corresponding to the target load rate of the month.

[0008] ​​​​​Further, the S3 specifically comprises steps of: S31: According to the given wind and light output curve and the total load curve of the power system, the actual output of the thermal power is obtained, and the calculation formula of the actual output of the thermal power is: (4) In the formula: denotes the index of the divided period; denotes the period power supply; denotes the period power transmission; denotes the period wind power output; denotes the period photovoltaic output; denotes the actual output of the thermal power; S32: According to the peak heat load adjustable capacity, the number of units of the extraction condensing unit and the total capacity of the extraction condensing unit are determined, and the expression is: (5) (6) (7) In the formula: denotes the period peak heat load adjustable capacity; denotes the upward adjustment of the spinning reserve rate; denotes the period heat load; denotes the number of units of the extraction condensing unit in the period; denotes the maximum heat output of a single extraction condensing unit; denotes the total capacity of the extraction condensing unit in the period; denotes the rated capacity of a single extraction condensing unit; S33: According to the heat load and the extraction condensing unit operation domain constraint, the maximum and minimum electric output of the extraction condensing unit are determined, comprising steps of: S331: The extraction condensing unit operation domain constraint is obtained, and the expression is: (8) (9) In the formula: denotes the maximum electric output of the extraction condensing unit in the period; denotes the minimum electric output of the extraction condensing unit in the period; , , , , , Pmax, i represents the maximum electric output of the single extraction condensing unit; Pmax, i represents the maximum electric output of the single extraction condensing unit; Pmin, i represents the minimum electric output of the single extraction condensing unit; Pcrit represents the critical boundary value of the thermal load; S332: Based on the extraction condensing unit operation domain constraint, the total on-line capacity of the pure condensing unit is determined according to the peak electric load adjustable capacity and the maximum electric output of the extraction condensing unit, and the expression is: (10) (11) (12) In the formula: Ppeak, i represents the peak electric load adjustable capacity of the i th time period; Ppeak, i represents the peak electric load adjustable capacity of the i th time period; Ptotal, i represents the total on-line capacity of the pure condensing unit in the i th time period; Ptotal, i represents the total on-line capacity of the pure condensing unit in the i th time period; Ptotal, i represents the total on-line capacity of the pure condensing unit in the i th time period; Ptotal, i represents the total on-line capacity of the pure condensing unit in the i th time period; Ptotal, i represents the total on-line capacity of the pure condensing unit in the i th time period; S333: The thermal power on-line capacity is updated in real time according to the total on-line capacity of the pure condensing unit combined with step S32 to obtain the updated thermal power on-line capacity; And the expression for updating the thermal power on-line capacity in real time is: (13) S34: The actual thermal power load rate is obtained according to the updated thermal power on-line capacity and the actual thermal power output; And the actual thermal power load rate is obtained by the formula: (14) According to the actual thermal power load rate, the minimum electric output of the extraction condensing unit and the minimum electric output of the single pure condensing unit , the minimum load rate of the power system is obtained; And the minimum load rate of the power system is obtained by the formula: (15).

[0009] Further, the expression for randomly selecting an electricity price from the normal probability distribution electricity price cluster according to the probability distribution in S4 is: (16) In the formula: P represents the output electricity price; represents a random price generated by a normal distribution probability.

[0010] Beneficial effects: the application provides a long-term power spot market price prediction method based on thermal power load rate, which makes the load rate-price fitting curve of the price simulation more in line with the actual market law by fully considering the differences of different months, and constructs a normal probability distribution model based on historical actual market price samples to accurately restore the random fluctuation characteristics of the price and improve the authenticity of the simulation; meanwhile, considering the differences between the heating period and the non-heating period, combining the wind and light output and the real-time update of the start-up capacity of the extraction-condensing unit, the actual thermal power load rate and the minimum load rate of the power system are accurately calculated, and the actual market price is finally confirmed according to the actual thermal power load rate, which further improves the accuracy of the price simulation. The application fully considers the daily real-time update of the start-up capacity, increases the minimum load rate constraint, realizes the prediction of the power spot market price, and guides the price decision in the process of power system planning and operation. BRIEF DESCRIPTION OF DRAWINGS

[0011] In order to more clearly illustrate the technical solutions in the embodiments of the application or the prior art, the following will briefly introduce the drawings needed to be used in the embodiments or the prior art description. Obviously, the drawings in the following description are some embodiments of the application, and other drawings can also be obtained by those skilled in the art without creative labor.

[0012] Figure 1 The flowchart of the long-term power spot market price prediction method based on thermal power load rate of the application; Figure 2 The load rate-price fitting curve of the thermal power of a certain province in May-August 2021 in the embodiment; Figure 3 The load rate-price fitting curve of the thermal power of a certain province in September-December 2021 in the embodiment; Figure 4 The load rate-price fitting curve of the thermal power of a certain province in January-April 2022 in the embodiment; Figure 5 The normal distribution diagram of the price in the embodiment; Figure 6 The load rate and minimum load rate distribution curve diagram in the embodiment; Figure 7 The annual price change curve diagram in the embodiment; Figure 8 The core technical flow block diagram of the long-term power spot market price prediction method based on thermal power load rate in the embodiment. DETAILED DESCRIPTION

[0013] In order to make the purposes, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only some of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative labor fall within the scope of protection of the present application.

[0014] The present embodiment provides a long-term electricity spot market electricity price prediction method based on thermal power load rate, as shown in Figure 1 and Figure 8 , specifically comprising the steps of: S1: obtaining historical data and dividing the historical data by month to fit 12 load rate-price fitting curves corresponding to each month in a year; the historical data includes historical actual thermal power load rate and historical actual market electricity price; Specifically, the present embodiment obtains historical actual thermal power load rate data and historical actual market electricity price data of a province, divides the data by month, and obtains 12 load rate-price fitting curves for the whole year by monthly fitting, and the fitting formula of the load rate-price fitting curve is: (1) In the formula: represents the fitting curve value of the load rate-price fitting curve; represents the month and takes a value of 1-12; represents the fitting coefficient of the month, which is obtained by solving the polyfit function of MATLAB, and in the present embodiment, the monthly load rate-price fitting curves for the whole year are fitted according to the actual thermal power load rate and electricity price data of the power system of a province from May 2021 to April 2022, as shown in Figures 2 to 4 , which clearly presents the nonlinear relationship between load rate and electricity price in different months; S2: for any target month and target load rate in the corresponding target month, taking the fitting curve value corresponding to the target load rate as the mean value, and calculating the standard deviation of the historical actual electricity price corresponding to the target load rate as a sample to construct an electricity price normal probability distribution model corresponding to the target load rate; The constructed electricity price normal probability distribution model corresponding to the target load rate has the expression: (2) (3) In the formula: represents the target electricity price; represents the fitting curve value corresponding to the target load rate of the month; . represents the target load rate of the 1st month corresponding standard deviation; represents the target load rate of the 1st month corresponding number of historical actual market electricity price samples; represents the target load rate of the 1st month corresponding 1st historical actual market electricity price; in this embodiment, the load rate is 0.6 in December 2021, and the electricity price normal distribution diagram is as shown in Figure 5 ; S3: according to the given wind and light output curve and the total load curve of the power system, and combining the extraction condensing unit operation domain constraint, the extraction condensing unit and the pure condensing unit, the thermal power unit capacity is updated to calculate the actual thermal power load rate and the minimum load rate of the power system; Specifically, it includes the following steps: S31: according to the given wind and light output curve and the total load curve of the power system, the actual output of thermal power is obtained, and the calculation formula of the actual output of thermal power is: (4) In the formula: represents the index of the divided period; represents the period power supply; represents the period external power transmission; represents the period wind power output; represents the period photovoltaic output, while ensuring that the actual output of thermal power is non-negative; represents the actual output of thermal power; S32: according to the peak heat load adjustable capacity, the number of extraction condensing unit start-up units and the total start-up capacity are determined, and the expression is: (5) (6) (7) In the formula: represents the period peak heat load adjustable capacity; represents the upward adjustment of the spinning reserve rate; represents the period heat load; represents the number of extraction condensing unit start-up units in the period; represents the maximum heat output of the extraction condensing unit; represents Total on-line capacity of period condensing-extraction unit; Indicates rated capacity of single condensing-extraction unit; S33: determining maximum and minimum electric output of condensing-extraction unit according to thermal load and condensing-extraction unit operation domain constraint, comprising steps of: S331: obtaining condensing-extraction unit operation domain constraint, whose expression is: (8) (9) In the formula: Indicates Maximum electric output of period condensing-extraction unit; Indicates Minimum electric output of period condensing-extraction unit; , , , , , All indicate operation parameters of condensing-extraction unit; Indicates maximum electric output of single condensing-extraction unit; Indicates minimum electric output of single condensing-extraction unit; Indicates critical demarcation value of thermal load; S332: determining total on-line capacity of pure condensing unit according to peak electric load adjustable capacity and maximum electric output of condensing-extraction unit based on condensing-extraction unit operation domain constraint, whose expression is: (10) (11) (12) In the formula: Indicates Period peak electric load adjustable capacity; Indicates Total on-line capacity of period pure condensing unit; Indicates Number of on-line units of period pure condensing unit; Indicates rated capacity of single pure condensing unit; Indicates Minimum electric output of period pure condensing unit; Indicates minimum electric output of single pure condensing unit; S333: real-time updating thermal power on-line capacity to obtain updated thermal power on-line capacity according to total on-line capacity of pure condensing unit and step S32; And the expression for real-time updating thermal power on-line capacity is: (13) S34: Obtain the actual thermal power load rate according to the updated thermal power on-line capacity and the actual thermal power output; and the actual thermal power load rate The formula for obtaining the actual thermal power load rate is: (14) According to the actual thermal power load rate, the minimum power output of the extraction-condensing unit and the minimum power output of a single pure condensing unit , the minimum load rate of the power system is obtained; and the formula for obtaining the minimum load rate of the power system is: (15) Specifically, in this embodiment, the whole year is divided into 8 periods every day, and after calculation, the load rate and minimum load rate distribution of 2928 periods in the whole year are as shown in Figure 6 From the figure, it can be seen that the minimum load rate in the heating period is significantly higher than that in the non-heating period, mainly because the extraction-condensing unit is constrained by the heating load, the minimum power output is high, which leads to the increase of the minimum load rate of the system, which is consistent with the actual operation characteristics; S4: Determine whether the actual thermal power load rate is greater than or equal to the minimum load rate of the power system; If not, define the current long-term electricity spot market price as 0; (In the power system, the thermal power load rate usually refers to the ratio of the actual output of the thermal power unit to the maximum possible output, and the minimum load rate refers to the lowest load rate that the thermal power unit can reach in a certain period of time, such as a day or a year. When the actual thermal power load rate is less than the minimum load rate, it means that the utilization rate of the thermal power unit is lower than the designed or expected minimum level, and in this embodiment, the corresponding long-term electricity spot market price at this time is defined as 0); If yes, obtain the normal probability distribution price cluster corresponding to the actual thermal power load rate based on the price normal probability distribution model, and randomly select a price from the normal probability distribution price cluster according to the probability distribution as the final output price; Specifically, the expression for randomly selecting a price from the normal probability distribution price cluster according to the probability distribution is: (16) In the formula: represents the output price; represents a random price generated according to the normal distribution probability, while ensuring that the price is non-negative; S5: Obtain the price of each period in the whole year based on S2 to S4, realize the simulation of the long-term electricity spot market price based on the thermal power load rate, and realize the long-term electricity spot market price prediction based on the thermal power load rate, so as to guide the price decision in the process of power system planning and operation. ​

[0015] In this embodiment, the actual operation data of a certain provincial power system is taken as a reference, the unit parameters and system parameters, and the extraction condensing unit operation range are shown in Table 1 and Table 2: Table 1. Unit parameters and system parameters

[0016] Table 2. Extraction condensing unit operation range

[0017] In this embodiment, the above initial data is input into the model, and the electricity price output of all 2928 time periods is realized by daily rolling calculation, and the results are saved to an excel file, including the starting capacity, thermal power Q, minimum starting load rate, actual starting load rate and electricity price and other core indicators, as shown in Figure 7 The annual electricity price change curve diagram, through example analysis, it can be known that the method described in this embodiment establishes a load rate-price fitting model through monthly polynomial fitting, fully takes into account the differences between different months, and makes the electricity price simulation curve highly consistent with the operation law of the actual market; at the same time, based on the historical actual electricity price sample, a normal probability distribution model is constructed, which accurately reproduces the random fluctuation characteristics of the electricity price, distinguishes the heat load difference between the heating period and the non-heating period, combines the wind and light output data, and real-time updates the thermal power starting capacity according to the extraction condensing unit feasible operation domain constraint, realizes the accurate calculation of the actual load rate and the minimum load rate, and further strengthens the accuracy of the electricity price simulation result, which is used for electricity price decision guidance in the process of power system planning and operation.

[0018] Finally, it should be noted that: the above embodiments are only used to illustrate the technical solutions of the present application, but not to limit them; although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that they can still modify the technical solutions recorded in the foregoing embodiments, or make equivalent replacement for part or all of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the scope of the technical solutions of the embodiments of the present application.

Claims

1. A method for predicting long-term electricity spot market prices based on thermal power load factor, characterized in that, The specific steps include: S1: Obtain historical data and divide the historical data into months to fit 12 load factor-electricity price fitting curves for each month of the year; The historical data includes historical actual thermal power load rate and historical actual market electricity price; S2: For any target month and the target load rate under the corresponding target month, the fitted curve value corresponding to the target load rate is used as the mean, and the historical actual electricity price corresponding to the target load rate is used as the sample to calculate the standard deviation, so as to construct a normal probability distribution model of the electricity price corresponding to the target load rate. S3: Based on the given wind and solar power output curves and the total load curve of the power system, and combined with the operating domain constraints of extraction condensing units and the parameters of extraction condensing units and pure condensing units, update the thermal power operating capacity to calculate the actual thermal power load rate and the minimum load rate of the power system. S4: Determine whether the actual thermal power load rate is greater than or equal to the minimum load rate of the power system; If not, then the current long-term electricity spot market price is defined as 0; If so, then based on the normal probability distribution model of electricity prices, obtain the normal probability distribution electricity price cluster corresponding to the actual thermal power load rate, and randomly select an electricity price from the normal probability distribution electricity price cluster according to the probability distribution as the final output electricity price; S5: Based on S2 to S4, the electricity prices for each month of the year are obtained, and the long-term electricity spot market price based on the thermal power load factor is simulated. This enables the prediction of the long-term electricity spot market price based on the thermal power load factor, which can be used to guide electricity price decision-making during the planning and operation of the power system.

2. The method for predicting long-term electricity spot market prices based on thermal power load factor according to claim 1, characterized in that, The fitting formula for the load factor-electricity price fitting curve mentioned in S1 is: (1) In the formula: This represents the fitted curve value of the load factor-electricity price fitting curve; Represents the month and takes values ​​from 1 to 12; Indicates the first The fitting coefficient for the month; This indicates the target load factor.

3. The method for predicting long-term electricity spot market prices based on thermal power load factor according to claim 1, characterized in that, The expression for the electricity price normal probability distribution model corresponding to the target load factor constructed in S2 is as follows: (2) (3) In the formula: Indicates the target electricity price; Indicates the first Monthly target load factor The corresponding fitted curve value; Indicates the first Monthly target load rate The corresponding standard deviation; Indicates the first Monthly target load rate The corresponding number of historical actual market electricity price samples; Indicates the first Monthly target load rate The corresponding number The historical actual market electricity price.

4. The method for predicting long-term electricity spot market prices based on thermal power load factor according to claim 3, characterized in that, S3 specifically includes the following steps: S31: Based on the given wind and solar power output curves and the total load curve of the power system, obtain the actual output of thermal power, and the calculation formula for the actual output of thermal power is: (4) In the formula: Indicates the time period index of the segment; express Power supply during specific time periods; express Power transmission during specific time periods; express Wind power output during certain periods; express Solar power output during specific time periods; This indicates the actual output of thermal power; S32: The number of extraction condensing units and the total operating capacity are determined based on the adjustable capacity of peak heat load. The expression is as follows: (5) (6) (7) In the formula: express Adjustable capacity for peak heat load during specific time periods; This indicates an increase in the spinning reserve ratio; express Heat load during a given period; express Number of condensing turbine units in operation during a given time period; This indicates the maximum thermal output of a single condensing turbine unit; express Total operating capacity of the time-period extraction condensing unit; This indicates the rated capacity of a single condensing turbine unit; S33: Determine the maximum and minimum electrical output of the extraction-condensing turbine based on the heat load and the operating domain constraints of the extraction-condensing turbine, including the following steps: S331: Obtain the operating domain constraints of the extraction condensing unit, the expression of which is: (8) (9) In the formula: express Maximum electrical output of the time-phase extraction condensing unit; express Minimum electrical output of the time-phase extraction condensing unit; , , , , , All of these represent the operating parameters of the extraction condensing unit; This indicates the maximum electrical output of a single condensing turbine unit; This indicates the minimum electrical output of a single extraction condensing unit; Indicates the critical threshold value of heat load; S332: Based on the operating domain constraints of the extraction condensing unit, the total operating capacity of the pure condensing unit is determined according to the adjustable capacity of peak electrical load and the maximum electrical output of the extraction condensing unit. The expression is as follows: (10) (11) (12) In the formula: express Adjustable capacity for peak load during specific time periods; express Total operating capacity of pure condensing units during the time period; express Number of pure condensing units in operation during a given time period; This indicates the rated capacity of a single condensing unit; S333: Based on the total operating capacity of the pure condensing units and in conjunction with step S32, the operating capacity of the thermal power units is updated in real time to obtain the updated operating capacity of the thermal power units. The expression for real-time updating of thermal power plant operating capacity is as follows: (13) S34: Obtain the actual thermal power load rate based on the updated thermal power plant operating capacity and actual thermal power output; And the actual thermal power load rate The formula for obtaining it is: (14) Based on the actual thermal power load rate and the minimum electrical output of the extraction condensing unit Minimum electrical output of a single pure condensing unit To obtain the minimum load factor of the power system; And the minimum load rate of the power system The formula for obtaining it is: (15)。 5. The method for predicting long-term electricity spot market prices based on thermal power load factor according to claim 4, characterized in that, The expression for randomly selecting an electricity price from the normally distributed electricity price cluster according to the probability distribution in S4 is as follows: (16) In the formula: Indicates the output electricity price; This represents a random electricity price generated according to a normal distribution probability.

Images