Method for calculating power generation consumption rate of photovoltaic system based on short-term monitoring of power consumption

Abstract

The invention discloses a photovoltaic system power generation consumption rate calculation method based on short-term monitoring of power consumption, and the method comprises the following steps: monitoring the power consumption per hour of a power consumption unit in a period of time, and obtaining the power consumption curve of each type of natural day and the power consumption proportion of each type of natural day; inputting each type of natural daily electricity consumption curve into an ARMA model, and calculating the 24-hour electricity consumption distribution proportion of each type of natural daily; collecting electricity bills of a power consumption unit in the last year, and distributing the electricity consumption per hour every day in the next year in combination with a 24-hour electricity consumption distribution proportion; the photovoltaic power generation amount per hour every day in the next year is calculated through simulation of PVSYST software, the calculated power consumption per hour every day in the next year is subtracted to obtain the non-consumed power amount of photovoltaic power generation, the non-consumed power amount is subtracted from the total photovoltaic power generation amount to obtain the total consumed power amount, and the total consumed power amount is divided by the total power generation amount to obtain the photovoltaic consumption rate. And the photovoltaic consumption rate is utilized to calculate and predict the economic benefit of the photovoltaic system in combination with a local electricity price policy.

Description

Technical Field

[0001] The invention relates to a method for calculating a power generation absorption rate of a photovoltaic system based on short-term monitoring of electric energy consumption. Background Art

[0002] Distributed photovoltaic projects are generally self-generated and self-used, with surplus power connected to the grid or not connected to the grid.

[0003] Photovoltaic power generation has certain regularity but is unstable. It has the characteristics of generating electricity during the day and no electricity at night, high power generation efficiency at noon and low power generation efficiency in the morning and evening, long power generation time in summer and short power generation time in winter.

[0004] For distributed photovoltaic power users, their hourly electricity consumption is also unstable. For example, different production arrangements will result in working days and non-working days. There may be equipment with long-term insurance coverage that has a relatively stable electricity consumption, or there may be unstable characteristics such as working from nine to five, with electricity consumption after get off work much lower than during work hours.

[0005] There are also differences in electricity prices between provinces, peaks and valleys, and differences in electricity prices in different months. At the same time, there are differences between the grid-connected electricity price and the self-use electricity price of photovoltaic power. In the early stage of photovoltaic project construction, if the economic feasibility of photovoltaic projects needs to be evaluated, it is necessary to combine the uncertainty of power generation and electricity consumption to carry out more accurate calculations and simulations.

[0006] There are usually two ways to calculate the absorption rate of photovoltaic systems in existing technical solutions: 1. Obtain the production load curve of the production enterprise throughout the year, and then compare and calculate it with the photovoltaic power generation curve to obtain the absorption rate; 2. Use electricity consumption data to compare and calculate with photovoltaic power generation to obtain the absorption rate.

[0007] The above two methods have the following disadvantages: 1. Many production enterprises do not have the historical production load curve of the past year, and it takes too long to monitor the production load curve of one year; 2. The consumption rate results obtained by comparing the power generation are highly biased; 3. It is impossible to analyze the electricity consumption patterns or curves of the enterprises; 4. The floating characteristics of the electricity consumption of the electricity-consuming units are not calculated. Summary of the invention

[0008] The object of the present invention is to provide a calculation method for the power generation and consumption rate of a photovoltaic system based on short-term monitored power consumption, aiming at the problems existing in the existing calculation methods for the consumption rate of photovoltaic systems. The present invention is applicable to enterprises with stable production and power consumption, where the power consumption has periodicity, or several types of periodicity, such as the power consumption per hour of each day on weekdays has a cycle, and the power consumption per hour of each day on non-working days has a cycle. In the early stage, the future level can be predicted through short-term monitoring and calculation to meet the needs of project planning and construction. By adding a white noise sequence to the predicted power consumption combined with the variance obtained from data analysis, the characteristic of certain fluctuations in power consumption can be better simulated. Through the power consumption and power generation with randomness and in line with the power consumption nature of the power-consuming unit, accurate to each hour of the whole year, the consumption rate and economic benefits can be better calculated in combination with the local power grid policy (peak, valley, and flat) or EMC agreement.

[0009] The technical solution adopted by the present invention is as follows: A calculation method for the power generation and consumption rate of a photovoltaic system based on short-term monitored power consumption, characterized by including the following steps: Step A: Monitor the power consumption per hour of the power-consuming unit during a time period. The time period should include various types of natural days under various production types. Each type of natural day includes weekdays, semi-weekdays, and non-working days, and obtain the power consumption curves of each type of natural day; input the power consumption curves of each type of natural day obtained by monitoring into the ARMA model, and calculate the 24-hour power consumption distribution ratio of each type of natural day; Step B: Collect the annual electricity bills of the power-consuming unit for the previous year. The annual electricity bills should include the total power consumption and time-of-use electricity prices for each month, and calculate the electricity fees for each day of the previous year; through the annual electricity bills of the previous year, on the premise of keeping the annual and monthly total power consumption unchanged, distribute the daily power consumption of the previous year to each day of the whole year of the next year, and combine the 24-hour power consumption distribution ratio of each type of natural day calculated by the ARMA model to calculate the power consumption per hour of each day of the next year; Step C: Simulate and calculate the photovoltaic power generation per hour of each day of the year through PVSYST software; Step D: Divide the power consumption per hour of each day of the next year calculated in Step B by the photovoltaic power generation per hour of each day of the year simulated and calculated by the PVSYST software in Step C to obtain the photovoltaic consumption rate.

[0010] The specific method of Step A is as follows: Monitor / collect the current hourly power consumption or power consumption from a power quantity monitor, the APP of the State Grid Corporation of China, or other power consumption monitoring systems; combine the production situation of the power-consuming unit, and distinguish the natural days monitored into weekdays, semi-weekdays, and non-working days, and form respective sequences for the hourly power consumption of each day with the same power consumption type; The daily power consumption sequences collected are: The daily electricity consumption sequences collected are as follows:

[0011] Calculation of the hourly electricity consumption allocation ratio for each type of natural day:

[0012] Calculation of the hourly electricity consumption allocation ratio for each type of natural day: Check the 24-hour curves of each sequence to see if they conform to the same trend. Exclude the monitoring sequences that do not conform to the obvious abnormal trend. Perform parameter estimation, order determination, and model testing of the ARMA model for each type of natural day. If the test conforms to the ARMA model, then obtain an ARMA model for the hourly electricity consumption of each type of natural day belonging to this type of natural day; a and b are ARMA model parameters; c is a constant; {Xt} is the electricity consumption in the t-th hour period of a working day; e is the residual; p is the order of the autoregressive term; q is the order of the moving average term; Xt is a column of the electricity consumption of a working day collected The predicted electricity consumption Xt for a day has only 24 numbers; Similarly, the {Yt} of a non-working day, the electricity consumption in the t-th hour of a non-working day, and the daily electricity consumption sequence {Zt} of a semi-working day can be predicted; The ARMA model is a mature model and is used in various fields, so no more introduction is given.

[0013] After completing the ARMA model, the hourly electricity consumption allocation ratios for 24 hours of each type of natural day are obtained. This electricity consumption allocation ratio is used to allocate the daily electricity consumption calculated in the next year proportionally to each hour of each day; Through electricity consumption monitoring, the electricity consumption ratios of each type of natural day are also obtained.

[0014] The specific method of step B is as follows: Collect the electricity bills for the whole year of the previous year to estimate the total electricity consumption and the electricity consumption ratio of each month: Combine the electricity bills for the whole year of the previous year collected, that is, the electricity consumption of each month , D0i is the electricity consumption monitored in each month from January to December, and Di is the electricity consumption of each month used for later prediction, that is, keep the electricity consumption of each month unchanged; Allocate and predict the electricity consumption for the next year: Combine the calendar or the production and work arrangements of this electricity-consuming unit to divide the working days and non-working days of each month. For example, if there are m working days and n non-working days in the i-th month, then D i Allocate to the electricity consumption of each day in the i-th month; A is the total value of the electricity consumption of the monitored working days; B is the total electricity consumption on non - working days monitored; m is the number of working days in month i; n is the number of non - working days in month i; In this way, the electricity consumption d for each day of each month is obtained ij , where i is the month and j is the date; Then, according to {Xt} and {Yt} simulated by the ARMA model, the total daily electricity consumption d ij is allocated to each hour of this day to obtain {X ijt}, {Y ijt};

[0015] X ijt is the electricity consumption at the t - th hour of the j - th day in month i on a working day; Y ijt is the electricity consumption at the t - th hour of the j - th day in month i on a non - working day; In this way, the predicted electricity consumption for each hour of each day of the whole year is obtained; Then, according to the variance of the electricity consumption per hour of each type of natural day obtained from the previous monitoring , a white noise sequence with as the variance is generated ; Finally, the electricity consumption sequence for each hour of each day of the whole year predicted or simulated is:

[0016] Then the predicted electricity consumption {U ijt} for each hour of each day of the whole year is obtained.

[0017] The specific method of step C is as follows: The photovoltaic power generation E P The formula is:

[0018] Among them, after determining the photovoltaic system design and project location, P AZ is the photovoltaic capacity, E S is the irradiance under standard conditions, K is the comprehensive efficiency. After determining the specific project location, electrical configuration, and system capacity, P AZ , E S , and K are all constant values, H A is the solar irradiance. The irradiance for each hour of each day of the whole year can be simulated through the PVSYST software using meteorological station data. In this way, the predicted power generation for each hour of each day of the specific photovoltaic system project throughout the year is obtained through the PVSYST software.

[0019] In step A, the period for monitoring the electricity - using unit is 10 - 30 days.

[0020] Through simple prediction of power generation and power consumption per hour every day throughout the year, the present invention can calculate the photovoltaic power generation and consumption rate per hour every day in the future throughout the year. Furthermore, according to the electricity price policy of the region where it is located, the photovoltaic income can be calculated. Description of the Drawings

[0021] Figure 1 It is the predicted power consumption curve graph for a working day in the embodiment of the present invention.

[0022] Figure 2 It is the predicted power consumption curve graph for a non - working day in the embodiment of the present invention. Detailed Embodiment

[0023] The following further elaborates on the present invention with reference to specific embodiments for a clear understanding of the present invention, but they do not limit the present invention.

[0024] Monitor the electricity meters of an enterprise and collect the electricity consumption data monitored for 10 days, including 7 working - day data as shown in Table 1 and 3 non - working - day data as shown in Table 2.

[0025] Table 1 Monitoring Table of Electricity Consumption on Working Days

[0026] Table 2 Monitoring Table of Electricity Consumption on Non - working Days

[0027] Use MATLAB to analyze the ARMA model for the two types of electricity consumption situations with the data in Table 1 and Table 2 respectively, and obtain the corresponding predicted curve graphs as Figure 1 and Figure 2 shown.

[0028] Convert the calculated predicted curve into the electricity consumption ratio within 24 hours; The average daily electricity consumption monitored on working days is 41,640.6 degrees, and the average daily electricity consumption on non - working days is 9,404.4 degrees.

[0029] That is, the electricity consumption on working days: the electricity consumption on non - working days A:B = 41,640.6:9,404.4 = 4.428:1.

[0030] Collect the annual electricity bills. The annual electricity consumption is approximately 12 million degrees, and the electricity consumption from January to December is 967,000 degrees, 843,000 degrees, 964,000 degrees, 1,003,000 degrees, 992,000 degrees, 976,000 degrees, 1,074,000 degrees, 1,142,000 degrees, 1,026,000 degrees, 998,000 degrees, 956,000 degrees, 1,005,000 degrees respectively.

[0031] Use the total monthly electricity consumption as the total monthly electricity consumption for the next year. Combine the working days and non - working days arrangements of this electricity - consuming unit, and allocate the monthly electricity consumption to each day of each month. The allocation ratio is based on the monitored ratio of electricity consumption on working days to non - working days, which is A:B, and the calculation is performed using EXCEL as shown in Table 3.

[0032] Table 3:

[0033] Allocate the daily electricity consumption ratio to each hour of each day according to the production types of each natural day simulated by the ARMA model. The calculation is performed using EXCEL as shown in Table 4 and Table 5.

[0034] Table 4: Hourly Electricity Consumption on Working Days of Each Month

[0035] Table 5: Hourly Electricity Consumption on Non - working Days of Each Month

[0036] Calculate the power generation of the planned photovoltaic power station system. The calculation uses PVSYST software. This PVSYST software can combine meteorological conditions to generate data on various sunny, cloudy, rainy, and snowy days throughout the year, and simulate the hourly power generation of the photovoltaic system for each of the 365 days of the year according to the photovoltaic system configuration. The planned scale of this photovoltaic power station is 10 MW.

[0037] Write the previously calculated electricity consumption data into the annual hourly photovoltaic power generation table exported by PVSYST, and randomly generate a white noise sequence for addition operation to simulate the instability of electricity consumption in practice.

[0038] The grid - connected photovoltaic power is the photovoltaic power generation minus the self - consumption electricity (take 0 if less than 0). At the same time, the self - consumption electricity is the photovoltaic power generation minus the grid - connected power. All of this photovoltaic power generation, self - consumption electricity, and grid - connected power are data for each hour of each day of the year.

[0039] Sum up the annual data to obtain the total annual self - consumption electricity, total power generation, and total grid - connected power. The self - consumption rate is the self - consumption electricity divided by the power generation.

[0040] Additionally, according to the electricity price policies of each region or EMC agreements, an electricity price sequence for each month and each time period can be generated. This sequence is also for each hour of each day of the year. Multiply the hourly self - consumption electricity or grid - connected power of the photovoltaic system throughout the year by the corresponding electricity price of its hour period and sum them up to obtain the annual photovoltaic income, as shown in Table 6.

[0041] Table VI: Calculation Table of Annual Absorption Rate and Yield

[0042] The annual hourly absorption rate is accurately predicted, and the yield can be calculated for different policies. The accuracy reaches every hour of every day.

[0043] Through the simulation of the ARMA model, the power consumption curve of each type of natural day can be predicted according to the production conditions of different electricity users, which is more in line with the actual situation.

[0044] Shorten the time for collecting / monitoring the hourly power consumption in the early stage. For electricity users with stable production, the future level can be predicted through a short period of time, without the need for long-term or even annual monitoring, meeting the needs of project planning and construction.

[0045] By adding a white noise sequence to the predicted power consumption combined with the variance obtained from data analysis, the characteristics of certain fluctuations in power consumption can be better simulated.

[0046] Through the power consumption and power generation with randomness and in line with the electricity consumption nature of electricity users, accurate to every hour of the whole year, the absorption rate and economic yield can be better calculated according to the local power grid policy (peak, valley, and flat).

[0047] Since the industrial electricity policies of each province are different, not only are the peak, valley, and flat periods with different electricity prices divided daily, but also the division of peak, valley, and flat periods may be different for each month. In addition, the construction party of the photovoltaic project and the electricity user of photovoltaic power generation may sign an EMC energy management agreement, resulting in differences in electricity prices at all times of the year. The calculation method of the present invention not only calculates the overall annual absorption rate, but also includes the absorption rate of every hour of every day of the whole year, facilitating the calculation of economic yield according to different electricity price policies or EMC agreements, and facilitating better planning and decision-making of photovoltaic projects in the early stage of construction.

Claims

1. A calculation method for the power generation and consumption rate of a photovoltaic system based on short-term monitored power consumption, characterized in that It includes the following steps: Step A: Monitor the hourly electricity consumption of the electricity-consuming unit for a period of time. The time period should include various types of natural days under various production types. The various types of natural days include working days, semi-working days, and non-working days, so as to obtain the electricity consumption curves and electricity consumption ratios of various types of natural days; input the electricity consumption curves of various types of natural days obtained by monitoring into the ARMA model, and calculate the 24-hour electricity consumption distribution ratios of various types of natural days; Step B: Collect the annual electricity bills of the electricity-consuming unit for the previous year. The annual electricity bills should include the total electricity consumption and time-of-use electricity prices for each month; through the annual electricity bills of the previous year, on the premise of keeping the annual and monthly total electricity consumption unchanged, distribute the daily electricity consumption of the previous year to each day of the whole year of the next year, and combine the 24-hour electricity consumption distribution ratios of various types of natural days calculated by the ARMA model and the daily electricity consumption distribution ratios of various types of natural days monitored to calculate the electricity consumption for each hour of each day of the next year; Step C: Use PVSYST software to simulate and calculate the photovoltaic power generation for each hour of each day of the year; Step D: Subtract the electricity consumption for each hour of each day of the next year calculated in Step B from the photovoltaic power generation for each hour of each day of the year simulated by PVSYST software in Step C to obtain the unconsumed electricity of photovoltaic power generation. The total photovoltaic power generation minus the unconsumed electricity is the total consumed electricity, and the total consumed electricity divided by the total power generation gives the photovoltaic consumption rate.

2. The calculation method of the power generation and consumption rate of the photovoltaic system based on short-term monitored power consumption according to claim 1, characterized in that: The specific method of Step A is as follows: Monitor / collect the current hourly electricity power or electricity consumption from the electricity monitor, the State Grid Electric Power Company APP, or other electricity monitoring systems; combine the production situation of the electricity-consuming unit, distinguish the natural days monitored into working days, semi-working days, and non-working days, and form respective sequences for the hourly electricity consumption of each day with the same electricity consumption type; The daily electricity sequences collected are: Calculation of the hourly power allocation ratio for each type of natural day: Check whether the 24-hour curves of each sequence conform to the same trend. Eliminate the monitoring sequences that do not conform to the obvious abnormal trend. Conduct parameter estimation, order determination, and model testing of the ARMA model for various types of natural days. If the test conforms to the ARMA model, then obtain an ARMA model for the hourly electricity consumption of each type of natural day belonging to this type of natural day; a and b are ARMA model parameters; c is a constant; {Xt} is the electricity consumption in the t-th hour period of a working day; e is the residual; p is the order of the autoregressive term; q is the order of the moving average term; Xt is a column of electricity consumption for a predicted day collected from working days Xt contains 24 numbers, representing the electricity consumption of a predicted day Similarly, the {Yt} of non-working days, the electricity consumption for the t-th hour of a non-working day, and the daily electricity sequence {Zt} of semi-working days can be predicted; After completing the ARMA model, the 24-hour electricity consumption distribution ratios of each type of natural day in a day are obtained; Through electricity monitoring, the electricity consumption ratios of various types of natural days are also obtained.

3. The calculation method of the power generation and consumption rate of the photovoltaic system based on short-term monitored power consumption according to claim 1, wherein: The specific method of Step B is as follows: Collect the electricity bills for the whole last year to estimate the total electricity consumption and the electricity consumption ratio for each month: Combine the collected electricity bills for the whole last year, that is, the electricity consumption for each month , D0i is the electricity consumption monitored for each month from January to December, and Di is the electricity consumption for each month used for later prediction, that is, keep the electricity consumption for each month unchanged; Distribute and predict the electricity consumption for the next year: Combine with the calendar or the production and work arrangements of this electricity-consuming unit to divide the working days and non-working days of each month. For example, in the i-th month, there are m working days and n non-working days, and then distribute D i to the daily electricity consumption of each day in the i-th month; Let A be the total electricity consumption on weekdays detected; B be the total electricity consumption on non - weekdays detected; m be the number of weekdays in the i - th month; n be the number of non - weekdays in the i - th month; thus, the electricity consumption d for each day of each month is obtained ij , where i is the month and j is the date; then, according to {Xt} and {Yt} simulated by the ARMA model, the total daily electricity consumption d ij is allocated to each hour of this day to obtain {X ijt} and {Y ijt}; X ijt is the electricity consumption at the t-th hour on the j-th day of the i-th month which is a working day; Y ijt is the electricity consumption at the t-th hour on the j-th day of the i-th month which is a non-working day; In this way, the predicted electricity consumption per hour for each day of the whole year is obtained; Then, based on the variances of the hourly electricity consumption of each type of natural day obtained from the previous monitoring , a white noise sequence with the variance of is generated ; Finally, the predicted or simulated electricity consumption sequence for each hour of each day of the whole year is: Then the predicted power consumption {U for each hour of each day of the whole year is obtained. ijt} 4. The calculation method of the power generation and consumption rate of the photovoltaic system based on short-term monitored power consumption according to claim 1, wherein: The specific method of Step C is as follows: Photovoltaic power generation E P The formula is as follows: After determining the design of the PV system and the project location, P AZ is the PV capacity, and E S is the irradiance under standard conditions, and K is the comprehensive efficiency. After determining the specific project location, electrical configuration, and system capacity, P AZ , E S , and K are all fixed values. H A is the solar irradiance. Through the data of the weather station, the irradiance per hour of each day throughout the year can be simulated by the PVSYST software. In this way, the predicted power generation per hour of each day throughout the year for a specific PV system project can be obtained through the PVSYST software.

5. The calculation method of the power generation and consumption rate of the photovoltaic system based on short-term monitored power consumption according to claim 1, wherein: The period of time for monitoring the electricity-consuming unit in Step A is 10 - 30 days.

Images