Industrial power supply energy storage optimization method and system adopting dynamic prediction

Abstract

The invention discloses an industrial power supply energy storage optimization method and system adopting dynamic prediction, and relates to the technical field of power supply energy storage, and the method comprises the steps: building prediction models of four seasons according to historical power consumption data and weather information, and obtaining first predicted power consumption; calculating prediction errors of the first five workdays, and carrying out weighted average to obtain second predicted electricity consumption; combining the first predicted electricity consumption with the second predicted electricity consumption to obtain final predicted electricity consumption; in the energy storage area, a plurality of electric energy storage blocks are divided to store predicted electricity consumption of a plurality of workdays in the future, and charging is carried out according to an estimated value; a buffer electric energy storage block is arranged, and when the predicted value is insufficient, a buffer area provides standby electric energy; in addition, daily residual electric energy is transferred to a buffer area, the electric energy is kept constant, and the residual electric energy is stored in the next workday. Waste of electric energy is avoided, and stable supply and efficient utilization of the industrial power supply are ensured.

Description

Technical Field

[0001] The present invention relates to the technical field of power supply energy storage, and in particular to an industrial power supply energy storage optimization method and system using dynamic prediction. Background Art

[0002] Among many industries, manufacturing is the main force of energy consumption. The stability and reliability of industrial power supply are the cornerstones to ensure the orderly production of manufacturing industry. However, due to various factors, such as power shortage and waste of electric energy, the stability of power supply may be affected, and even lead to power outage. Such results will directly affect the normal operation of the production line. Therefore, finding and optimizing the energy storage method of industrial power supply to ensure its stability and continuity has become a major issue facing my country's manufacturing industry. The development of dynamic prediction technology provides us with a possible solution to solve these problems that affect the stability of industrial production.

[0003] In the Chinese invention application with application publication number CN116562921A, a method, system and medium for optimizing industrial power storage based on dynamic prediction are disclosed. The method obtains the energy storage efficiency information of the energy storage power supply in the current industrial area, and generates a corresponding energy storage plan based on the energy storage efficiency information of the energy storage power supply in the current industrial area and the power supply demand characteristic data information of the current industrial area. Then, the power generation efficiency information of the power generation equipment is obtained, and the corresponding energy storage plan is corrected according to the power generation efficiency information of the power generation equipment to generate a corrected energy storage plan. Finally, energy storage measures are generated based on the corrected energy storage plan. The application fully considers the power supply loss of the energy storage power supply in the industrial area, making it more reasonable to select the energy storage power supply for energy storage.

[0004] In the above invention application, the influence of temperature on the energy storage efficiency of the energy storage power supply is fully considered, so that the accurate estimation of the energy storage situation of the energy storage power supply can be improved, which is conducive to the formulation of a more accurate energy storage plan; however, the overall electricity consumption of each day is not predicted, which may result in untimely power supply; in addition, the existing industrial power supply energy storage optimization method does not set up multiple storage blocks and buffer power, which will result in that if there is a shortage of electricity during electricity use, it will not be processed in time, thereby affecting the processing of the entire industrial process, and will cause waste of electricity during charging and discharging. Summary of the invention

[0005] 1. Technical issues to be resolved

[0006] In response to the technical problems in the background technology, the present invention proposes an industrial power supply energy storage optimization method and system using dynamic prediction, which obtains a first predicted power consumption and a second predicted power consumption through prediction and further combines them to obtain a final estimated power consumption; divides a plurality of energy storage blocks in the energy storage area to store the predicted power consumption for several future working days, and charges according to the estimated value; sets a buffer energy storage block, and when the predicted value is insufficient, the buffer zone provides backup power; avoids waste of electric energy, ensures a stable supply and efficient use of industrial power, and thus solves the technical problems recorded in the background technology.

[0007] (II) Technical solution

[0008] To achieve the above objectives, the present invention is implemented through the following technical solutions:

[0009] An industrial power storage optimization method using dynamic prediction includes:

[0010] Obtain historical industrial electricity consumption data, establish electricity consumption prediction models for the four seasons based on the daily workload and weather data of each historical working day, predict the electricity consumption of the next working day based on the electricity consumption prediction model, and use the prediction result as the first predicted electricity consumption Pe1 for the next working day;

[0011] Calculate the difference between the first predicted power consumption and the actual power consumption for each of the first five working days, and use the weighted average of these differences to obtain the error prediction value as the second predicted power consumption Pe2 for the next working day;

[0012] The calculated first predicted power consumption Pe1 and the second predicted power consumption Pe2 are combined to obtain the estimated power consumption Pe for the next working day; a region block is divided in the industrial power energy storage area as an energy storage block for storing the predicted power consumption for the next working day, and charging is performed according to the estimated power consumption Pe for the next working day; the industrial power energy storage area can have multiple energy storage blocks at the same time, which are used to store the predicted power consumption for several future working days;

[0013] A buffer electric energy storage block is set in the industrial power supply energy storage area to store backup electric energy when the predicted power consumption does not meet the current power demand, and the remaining electric energy in the electric energy storage block of each working day is transferred to the buffer electric energy storage block; the electric energy in the buffer electric energy storage block is maintained within a preset constant range, and the excess electric energy is used as one of the sources of electric energy in the electric energy storage block in the next time period. When the electric energy is insufficient, it is charged to reach the constant range.

[0014] Specifically, historical data is obtained, including daily workload data, weather data, and industrial electricity consumption data for each working day in the past year, and data is aggregated by season;

[0015] According to the historical data obtained from the data sets of different seasons, electricity consumption forecasting models for four seasons are established respectively, namely spring electricity consumption forecasting model, summer electricity consumption forecasting model, autumn electricity consumption forecasting model and winter electricity consumption forecasting model.

[0016] Specifically, the method for establishing the electricity consumption prediction model for the four seasons is as follows:

[0017] Industrial electricity consumption is affected by workload, season and weather by default; daily workload in spring, summer, autumn and winter is calculated separately Daily electricity consumption , where sea represents the season, represented by 1, 2, 3, and 4, corresponding to the four seasons of spring, summer, autumn, and winter; the data of cloudy or overcast weather in each season are taken out, and cloudy and overcast weather are taken as standard weather; the relationship between daily electricity consumption and daily workload under standard weather is fitted by a univariate linear regression equation, and for multiple groups of data with the same daily workload, the mean of the daily electricity consumption in these data is taken as the corresponding value of the daily electricity consumption of the daily workload, and the corresponding linear slope k of the daily electricity consumption and daily workload under standard weather in each season is calculated by the linear fitting method sea ; Then the corresponding equation Ec=k for daily electricity consumption Ec and daily workload Wo in standard weather in each season was established sea *Wo.

[0018] Furthermore, record the daily electricity consumption in different weather conditions in each season. Daily workload The comprehensive corresponding equation is where μ j The weather index of the jth weather type is 1 for cloudy and overcast weather. The data of other weather types are gathered into a set, and the daily electricity consumption and daily workload are recorded. Multiple data of the same weather type are brought into the comprehensive corresponding equation of daily electricity consumption and daily workload. , calculate multiple weather indexes μ under this weather j Values, these weather indices μ j Mean of the values As the final value of the weather index for that weather; Calculate the weather index values ​​for all weathers according to the above method; Obtain the daily power consumption for different weathers in each season according to the above method Daily workload The comprehensive corresponding equation is

[0019] Substitute the target workload data and weather data of the next working day into the electricity consumption prediction model of the season of the working day to obtain the electricity consumption prediction value of the next working day, and record it as the first predicted electricity consumption Pe1 of the next working day.

[0020] Specifically, the first predicted power consumption and actual power consumption data of the previous five working days are taken from the historical data, and the difference between the first predicted power consumption and the actual power consumption data in each working day, i.e., the error err, is calculated. o , where o represents the oth working day before;

[0021] Perform weighted average calculation on the five groups of differences and use Indicates the weight l of the previous o-th working day o , then the error prediction value err of the electricity consumption on the next working day is calculated according to the following formula:

[0022]

[0023] The calculated error prediction value of the electricity consumption on the next working day is used as the second predicted electricity consumption Pe2 for the next working day, and the value of Pe2 can be positive or negative.

[0024] Furthermore, the total power obtained by adding the calculated first predicted power consumption Pe1 and the second predicted power consumption Pe2 is used as the estimated power consumption Pe for the next working day;

[0025] The industrial power storage area stores the corresponding electric energy in blocks according to the estimated power consumption of each day. Specifically, according to the size of the estimated power consumption Pe of the next working day, the industrial power storage system divides the area blocks of corresponding sizes as the energy storage blocks for storing the predicted power consumption of the next working day;

[0026] The industrial power storage area can have multiple energy storage blocks at the same time to store the predicted power consumption for several future working days. Each storage block is marked with a corresponding date; and the charging and discharging processes of each energy storage block do not affect each other.

[0027] Specifically, when the electric energy stored in the electric energy storage block on a certain working day is insufficient to meet the power supply demand of the day, after the electric energy stored in the electric energy storage block of the day is used up, the buffer electric energy storage block is used as the power supply source of the day, and it is calculated whether the electric energy stored in the current cache electric energy storage block can meet the remaining power demand of the day. The specific calculation method is: obtain the estimated power consumption Pe, target workload Wo1 and remaining workload Wo0 of the day, and calculate the remaining power consumption Pe0 of the day based on these data. The expression is:

[0028] Furthermore, the remaining power Pe0 is compared with the remaining power Ce in the buffer power storage block. If Pe0≤Ce, it means that the power in the current buffer power storage block can meet the power demand of the day, and there is no need to recharge the power storage block of the day.

[0029] If Pe0>Ce, it means that the current power in the buffer energy storage block cannot meet the power demand for the day, and the energy storage block for the day needs to be charged again. The required power is obtained by subtracting the remaining power consumption Pe0 from the remaining power Ce in the buffer energy storage block. When the power in the buffer energy storage block is 0, the energy storage block for the day will continue to be used for power supply operations.

[0030] Furthermore, when the power in the buffer energy storage block is used up, a charging operation will be performed immediately to keep the power therein within a preset constant range; if the power in the energy storage block after recharging on that day has not yet met the power demand for that day, the power in the buffer energy storage block will continue to be used in the same way; if the power in the energy storage block after recharging on that day meets the power demand for that day and there is some power remaining, this part of power will be transferred to the buffer energy storage block; when establishing a new day's energy storage block, if the power in the buffer energy storage block exceeds the preset constant range, first take out some power from the buffer energy storage block and put it into the energy storage block for that day, so that the power in the buffer energy storage block is maintained within the said constant range.

[0031] An industrial power supply energy storage optimization system using dynamic prediction, comprising:

[0032] The data acquisition module acquires historical data, including daily workload data, weather data, and industrial power consumption data for each working day in the past year; and acquires the estimated power consumption data, target workload data, and remaining workload data for the day when the power stored in the power storage block of the day is used up;

[0033] The electricity consumption prediction module analyzes and predicts the electricity consumption of the next working day, including a first electricity consumption prediction unit, a second electricity consumption prediction unit and a comprehensive electricity consumption prediction unit; wherein the first electricity consumption prediction unit establishes electricity consumption prediction models for four seasons according to the daily workload and weather data of each historical working day, and predicts the electricity consumption of the next working day based on the electricity consumption prediction model; the second electricity consumption prediction unit predicts the electricity consumption error of the next working day according to the difference between the first predicted electricity consumption and the actual electricity consumption of each day in the previous five working days; the comprehensive electricity consumption prediction unit combines the prediction results of the first electricity consumption prediction unit and the second electricity consumption prediction unit to obtain the estimated electricity consumption Pe for the next working day;

[0034] The electric energy block storage module stores the industrial electric energy in blocks in the industrial power storage system, including a conventional electric energy storage unit and a buffer electric energy storage unit; wherein the conventional electric energy storage unit is used for block storage of the predicted power consumption for several working days; the buffer electric energy storage unit is used for storing buffer electric energy, which is used as a backup power source in case of emergencies or when the power consumption does not match the predicted power consumption.

[0035] (III) Beneficial effects

[0036] The present invention provides an industrial power supply energy storage optimization method and system using dynamic prediction, which has the following beneficial effects:

[0037] 1. Based on the actual data of the past year, a power consumption prediction model under the influence of different seasons, different weather and different workloads was established. This fully takes into account the impact of seasons, weather and workload on power consumption, and can more accurately predict the power consumption of the next working day. When the power consumption of the next working day is predicted, the historical data based on it will change accordingly, realizing the dynamic prediction of power consumption;

[0038] 2. When predicting the error of the next day, the data of the past few days still have a certain time continuity with the next day. Especially in a short period of time, the trend and pattern of the data may be relatively stable. Therefore, the errors of the past few days can be used as a reference factor for predicting the error of the next day. Using the error data of the past 5 days can help capture this dynamic change and better predict the error of the next day.

[0039] 3. Combining the first predicted power consumption and the second predicted power consumption, various influencing factors can be taken into account more comprehensively, thereby improving the accuracy of power consumption prediction; by storing electric energy in blocks according to the estimated power consumption every day, the industrial power storage area can be managed more finely to avoid unnecessary waste; multiple power storage blocks can exist at the same time to store the predicted power consumption for several working days in the future, and the charging and discharging processes of each power storage block do not affect each other, which can make more efficient use of energy and reduce energy waste;

[0040] 4. By setting up a buffer electric energy storage block, the remaining electric energy in the storage block of each working day can be better utilized, thus avoiding energy waste and improving energy utilization efficiency; when the electric energy stored in the electric energy storage block of a certain working day is not enough to meet the power supply demand of that day, the electric energy in the buffer electric energy storage block can be used to supplement it, making the power supply process more flexible and able to better cope with changes in electricity demand. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] Figure 1 A flowchart of the steps of the industrial power supply energy storage optimization method using dynamic prediction provided by the present invention;

[0042] Figure 2 A schematic diagram of the structure of an industrial power supply energy storage optimization system using dynamic prediction provided by the present invention. DETAILED DESCRIPTION

[0043] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0044] refer to Figure 1 The present invention provides an industrial power supply energy storage optimization method using dynamic prediction, comprising:

[0045] Step 1: Obtain historical industrial electricity consumption data, establish electricity consumption prediction models for the four seasons based on the daily workload and weather data of each historical working day, predict the electricity consumption of the next working day based on the electricity consumption prediction model, and use the prediction result as the first predicted electricity consumption Pe1 for the next working day;

[0046] The step 1 comprises the following steps:

[0047] Step 101: Obtain historical data, including daily workload data, weather data, and industrial power consumption data for each working day in the past year, and aggregate the data according to seasons, such as aggregating data from December to February in the spring historical data set; wherein, daily workload data refers to the total amount of work done by the factory every day; weather data refers to the weather of each day, such as sunny, rainy, or snowy; and industrial power consumption data refers to the total amount of power consumed by the factory every day;

[0048] Step 102: Establish electricity consumption prediction models for four seasons based on historical data obtained from data sets of different seasons, namely, a spring electricity consumption prediction model, a summer electricity consumption prediction model, an autumn electricity consumption prediction model, and a winter electricity consumption prediction model;

[0049] Industrial electricity consumption is affected by workload, season and weather by default; daily workload in spring, summer, autumn and winter is calculated separately Daily electricity consumption , where sea represents the season, represented by 1, 2, 3, and 4, corresponding to the four seasons of spring, summer, autumn, and winter; the data of cloudy or overcast weather in each season are taken out. Cloudy and overcast days are not affected by sunlight or bad weather such as rain, so the impact of weather on electricity consumption at this time can be ignored. At this time, the daily electricity consumption and the daily workload are linearly related, so cloudy and overcast weather are taken as standard weather; a univariate linear regression equation is used to fit the relationship between daily electricity consumption and daily workload under standard weather. For multiple groups of data with the same daily workload, the mean of daily electricity consumption in these data is taken as the corresponding value of daily electricity consumption for the daily workload. The corresponding linear slope k of the daily electricity consumption and daily workload in standard weather in each season is calculated by the linear fitting method. sea ; Then the corresponding equation Ec=k for daily electricity consumption Ec and daily workload Wo in standard weather in each season was established sea *Wo;

[0050] Record the daily electricity consumption in different weather conditions in each season Daily workload The comprehensive corresponding equation is where μ j The weather index of the jth weather type is 1 for cloudy and overcast weather. The data of other weather types are gathered into a set, and the daily electricity consumption and daily workload are recorded. Multiple data of the same weather type are brought into the comprehensive corresponding equation of daily electricity consumption and daily workload. , calculate multiple weather indexes μ under this weather j Values, these weather indices μ j Mean of the values As the final value of the weather index for that weather; Calculate the weather index values ​​for all weathers according to the above method; Obtain the daily power consumption for different weathers in each season according to the above method Daily workload The comprehensive corresponding equation is

[0051] Step 103: Substitute the target workload data and weather data of the next working day into the power consumption prediction model of the season of the working day to obtain the power consumption prediction value of the next working day, and record it as the first predicted power consumption Pe1 of the next working day;

[0052] When using, combine the contents in steps 101 to 103:

[0053] Through the actual data of one year in history, a power consumption prediction model under the influence of different seasons, different weather and different workloads was established. This fully takes into account the impact of seasons, weather and workload on power consumption, and can more accurately predict the power consumption of the next working day. Each time the power consumption of the next working day is predicted, the historical data based on it will change accordingly, realizing dynamic prediction of power consumption.

[0054] Step 2: Calculate the difference between the first predicted power consumption and the actual power consumption for each of the first five working days, and use the weighted average of these differences to obtain the error prediction value as the second predicted power consumption Pe2 for the next working day;

[0055] The step 2 comprises the following steps:

[0056] Step 201: Take the first predicted power consumption and actual power consumption data of the previous five working days from the historical data, and calculate the difference between the first predicted power consumption and the actual power consumption data in each working day, i.e., the error err o , where o represents the oth working day before;

[0057] Step 202: Perform a weighted average operation on the five groups of calculated differences. The closer to the next working day, the greater the reference value of the calculated difference for the error prediction of the next working day. represents the weight l of the previous o-th working day o , then the error prediction value err of the electricity consumption on the next working day is calculated according to the following formula:

[0058]

[0059] The calculated error prediction value of the electricity consumption on the next working day is used as the second predicted electricity consumption Pe2 for the next working day, and the value of Pe2 can be positive or negative.

[0060] When using, combine the contents in steps 201 to 202:

[0061] When predicting the error of the next day, the data of the past few days still have a certain time continuity with the next day. Especially in a short period of time, the trend and pattern of the data may be relatively stable. Therefore, the errors of the past few days can be used as a reference factor for predicting the error of the next day; using the error data of the past 5 days can help capture this dynamic change and better predict the error of the next day.

[0062] Step 3: Combine the calculated first predicted power consumption Pe1 and the second predicted power consumption Pe2 to obtain the estimated power consumption Pe for the next working day; divide an area block in the industrial power energy storage area as an energy storage block for storing the predicted power consumption for the next working day; the industrial power energy storage area can have multiple energy storage blocks at the same time, which are used to store the predicted power consumption for several future working days;

[0063] The step three comprises the following steps:

[0064] Step 301: The total power obtained by adding the calculated first predicted power consumption Pe1 and the second predicted power consumption Pe2 is used as the estimated power consumption Pe for the next working day;

[0065] Step 302: The industrial power energy storage area stores the corresponding electric energy in blocks according to the estimated power consumption of each day. Specifically, according to the size of the estimated power consumption Pe of the next working day, the industrial power energy storage system divides the area blocks of corresponding sizes as the electric energy storage blocks for storing the predicted power consumption of the next working day;

[0066] Step 303: There may be multiple energy storage blocks in the industrial power energy storage area at the same time, which are used to store the predicted power consumption for several future working days. Each storage block is marked with the corresponding date; and the charging and discharging processes of each energy storage block do not affect each other. For example, when the energy storage block corresponding to a certain working day is being discharged, the charging operation of the energy storage block of the next working day can also be carried out at this time.

[0067] When using, combine the contents in steps 301 to 303:

[0068] Combining the first predicted electricity consumption and the second predicted electricity consumption, various influencing factors can be taken into account more comprehensively, thereby improving the accuracy of electricity consumption forecasting; by storing electricity in blocks according to the estimated daily electricity consumption, the industrial power storage area can be managed more finely to avoid unnecessary waste; multiple energy storage blocks can exist at the same time to store the predicted electricity consumption for several working days in the future, and the charging and discharging processes of each energy storage block do not affect each other, which can make more efficient use of energy and reduce energy waste.

[0069] Step 4: a buffer electric energy storage block is set in the industrial power energy storage area to store backup electric energy when the predicted power consumption does not meet the current power demand, and the remaining electric energy in the electric energy storage block of each working day is transferred to the buffer electric energy storage block; the electric energy in the buffer electric energy storage block is maintained within a preset constant range, and the excess electric energy is used as one of the sources of electric energy in the electric energy storage block in the next time period, and when the electric energy is insufficient, it is charged to reach the constant range;

[0070] The step 4 comprises the following steps:

[0071] Step 401: a region is divided in the industrial power energy storage region for storing buffer power, which is called a buffer power storage block. The buffer power can be used as a backup power source in case of emergencies or when the power consumption does not match the predicted power consumption. After the buffer power storage block is established, the storage block is charged so that it is within a preset constant range.

[0072] Step 402: When the electric energy stored in the electric energy storage block on a certain working day is insufficient to meet the power supply demand of the day, after the electric energy stored in the electric energy storage block of the day is used up, the buffer electric energy storage block is used as the power supply source of the day, and it is calculated whether the electric energy stored in the current buffer electric energy storage block can meet the remaining power demand of the day. The specific calculation method is: obtain the estimated power consumption Pe, the target workload Wo1 and the remaining workload Wo0 of the day, and calculate the remaining power consumption Pe0 of the day based on these data. The expression is:

[0073] Compare the remaining power consumption Pe0 with the remaining power Ce in the buffer power storage block. If Pe0≤Ce, it means that the power in the current buffer power storage block can meet the power demand of the day, and the power storage block of the day does not need to be recharged; if Pe0>Ce, it means that the power in the current buffer power storage block cannot meet the power demand of the day, and the power storage block of the day needs to be recharged. The required power is obtained by subtracting the remaining power consumption Pe0 from the remaining power Ce in the buffer power storage block. When the power in the buffer power storage block is 0, the power storage block of the day will continue to be used for power supply operation;

[0074] Step 403, when the power in the buffer electric energy storage block is used up, a charging operation will be performed immediately to keep the electric energy therein within a preset constant range; if the electric energy in the electric energy storage block after recharging on that day has not yet met the power demand on that day, the electric energy in the buffer electric energy storage block will continue to be used according to the method of step 402; if the electric energy in the electric energy storage block after recharging on that day meets the power demand on that day and there is some electric energy remaining, this part of the electric energy will be transferred to the buffer electric energy storage block; when establishing a new day's energy storage block, if the electric energy in the buffer electric energy storage block exceeds the preset constant range, first take out some electric energy from the buffer electric energy storage block and put it into the electric energy storage block of that day, so that the electric energy in the buffer electric energy storage block is maintained within the said constant range.

[0075] When using, combine the contents in steps 401 to 403:

[0076] By setting up a buffer electric energy storage block, the remaining electric energy in the storage block of each working day can be better utilized, thus avoiding energy waste and improving energy utilization efficiency; when the electric energy stored in the electric energy storage block of a certain working day is not enough to meet the power supply demand of that day, the electric energy in the buffer electric energy storage block can be used to supplement it, thus making the power supply process more flexible and able to better cope with changes in electricity demand.

[0077] refer to Figure 2 The present invention also provides an industrial power supply energy storage optimization system using dynamic prediction, comprising:

[0078] The data acquisition module acquires historical data, including daily workload data, weather data, and industrial power consumption data for each working day in the past year; and acquires the estimated power consumption data, target workload data, and remaining workload data for the day when the power stored in the power storage block of the day is used up;

[0079] The electricity consumption prediction module analyzes and predicts the electricity consumption of the next working day, including a first electricity consumption prediction unit, a second electricity consumption prediction unit and a comprehensive electricity consumption prediction unit; wherein the first electricity consumption prediction unit establishes electricity consumption prediction models for four seasons according to the daily workload and weather data of each historical working day, and predicts the electricity consumption of the next working day based on the electricity consumption prediction model; the second electricity consumption prediction unit predicts the electricity consumption error of the next working day according to the difference between the first predicted electricity consumption and the actual electricity consumption of each day in the previous five working days; the comprehensive electricity consumption prediction unit combines the prediction results of the first electricity consumption prediction unit and the second electricity consumption prediction unit to obtain the estimated electricity consumption Pe for the next working day;

[0080] The electric energy block storage module stores the industrial electric energy in blocks in the industrial power storage system, including a conventional electric energy storage unit and a buffer electric energy storage unit; wherein the conventional electric energy storage unit is used for block storage of the predicted power consumption for several working days; the buffer electric energy storage unit is used for storing buffer electric energy, which is used as a backup power source in case of emergencies or when the power consumption does not match the predicted power consumption.

[0081] In the above embodiments, all or part of the embodiments may be implemented by software, hardware, firmware or any combination thereof. When implemented by software, all or part of the embodiments may be implemented in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the processes or functions described in the embodiments of the present invention are generated. The computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions may be stored in a computer storage medium or transmitted via a computer storage medium.

[0082] Computer instructions can be transmitted from one website, computer, server or data center to another website, computer, server or data center by wired (e.g., coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means. Computer storage media can be any available medium that can be accessed by a computer or a data storage device such as a server or data center that includes one or more available media. Available media can be magnetic media (e.g., floppy disks, hard disks, tapes), optical media (e.g., DVDs), or semiconductor media (e.g., solid state drives (SSDs)).

[0083] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed by the present invention, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. An industrial power supply energy storage optimization method using dynamic prediction, characterized in that: include: Obtain historical industrial electricity consumption data, establish electricity consumption prediction models for the four seasons based on the daily workload and weather data of each historical working day, predict the electricity consumption of the next working day based on the electricity consumption prediction model, and use the prediction result as the first predicted electricity consumption Pe1 for the next working day; Calculate the difference between the first predicted power consumption and the actual power consumption for each of the first five working days, and use the weighted average of these differences to obtain the error prediction value as the second predicted power consumption Pe2 for the next working day; The calculated first predicted power consumption Pe1 and the second predicted power consumption Pe2 are combined to obtain the estimated power consumption Pe for the next working day; a region block is divided in the industrial power energy storage area as an energy storage block for storing the predicted power consumption for the next working day, and charging is performed according to the estimated power consumption Pe for the next working day; the industrial power energy storage area can have multiple energy storage blocks at the same time, which are used to store the predicted power consumption for several future working days; A buffer electric energy storage block is set in the industrial power supply energy storage area to store backup electric energy when the predicted power consumption does not meet the current power demand, and the remaining electric energy in the electric energy storage block of each working day is transferred to the buffer electric energy storage block; The electric energy in the buffer electric energy storage block is maintained within a preset constant range, and the excess electric energy is used as the electric energy source in the electric energy storage block for the next time period. When the electric energy is insufficient, it is charged to reach the preset constant range.

2. The method for optimizing industrial power storage using dynamic prediction according to claim 1, characterized in that: Obtain historical data, including daily workload data, weather data, and industrial electricity consumption data for each working day in the past year, and aggregate the data by season; According to the historical data obtained from the data sets of different seasons, electricity consumption forecasting models for four seasons are established respectively, namely spring electricity consumption forecasting model, summer electricity consumption forecasting model, autumn electricity consumption forecasting model and winter electricity consumption forecasting model.

3. The method for optimizing industrial power storage using dynamic prediction as claimed in claim 2, characterized in that: The method for establishing the electricity consumption prediction model for the four seasons is as follows: Industrial electricity consumption is affected by workload, season and weather by default; daily workload in spring, summer, autumn and winter is calculated separately Daily electricity consumption , where sea represents the season, represented by 1, 2, 3, and 4, corresponding to the four seasons of spring, summer, autumn, and winter; the data of cloudy or overcast weather in each season is taken out, and the cloudy and overcast weather is taken as the standard weather; The relationship between daily electricity consumption and daily workload under standard weather conditions is fitted using a univariate linear regression equation. For multiple groups of data with the same daily workload, the mean of the daily electricity consumption in these data is taken as the corresponding value of the daily electricity consumption for the daily workload. The corresponding linear slope k between the daily electricity consumption and the daily workload under standard weather conditions in each season is calculated using the linear fitting method. sea ; Then the corresponding equation Ec=k for daily electricity consumption Ec and daily workload Wo in standard weather in each season was established sea *W O .

4. The method for optimizing industrial power storage using dynamic prediction as claimed in claim 3, characterized in that: Record the daily electricity consumption in different weather conditions in each season Daily workload The comprehensive corresponding equation is where μ j The weather index of the jth weather type is 1 for cloudy and overcast weather. The data of other weather types are gathered into a set, and the daily electricity consumption and daily workload are recorded. Multiple data of the same weather type are brought into the comprehensive corresponding equation of daily electricity consumption and daily workload. , calculate multiple weather indexes μ under this weather j Value, weather index μ j Mean of the values As the final value of the weather index for that weather; Calculate the weather index values ​​for all weather conditions and obtain the daily electricity consumption for different weather conditions in each season. Daily workload The comprehensive corresponding equation is Substitute the target workload data and weather data of the next working day into the electricity consumption prediction model of the season of the working day to obtain the electricity consumption prediction value of the next working day, and record it as the first predicted electricity consumption Pe1 of the next working day.

5. The method for optimizing industrial power storage using dynamic prediction according to claim 4, characterized in that: Take the first predicted power consumption and actual power consumption data of the previous 5 working days from the historical data, and calculate the difference between the first predicted power consumption and the actual power consumption data on each working day, i.e., the error err o , where o represents the oth working day before; Perform weighted average calculation on the five groups of differences and use Indicates the weight l of the previous o-th working day o , then the error prediction value err of the electricity consumption on the next working day is calculated according to the following formula: The calculated error prediction value of the electricity consumption on the next working day is used as the second predicted electricity consumption Pe2 for the next working day, and the value of Pe2 can be positive or negative.

6. The method for optimizing industrial power storage using dynamic prediction according to claim 5, characterized in that: The total power obtained by adding the calculated first predicted power consumption Pe1 and the second predicted power consumption Pe2 is used as the estimated power consumption Pe for the next working day; The industrial power storage area stores the corresponding electric energy in blocks according to the estimated power consumption of each day. Specifically, according to the size of the estimated power consumption Pe of the next working day, the industrial power storage system divides the area blocks of corresponding sizes as the energy storage blocks for storing the predicted power consumption of the next working day; The industrial power storage area can have multiple energy storage blocks at the same time, which are used to store the predicted electricity consumption for several future working days. Each storage block is marked with the corresponding date.

7. The method for optimizing industrial power storage using dynamic prediction according to claim 6, characterized in that: When the electric energy stored in the electric energy storage block on a certain working day is insufficient to meet the power supply demand of the day, after the electric energy stored in the electric energy storage block of the day is used up, the buffer electric energy storage block is used as the power supply source of the day, and it is calculated whether the electric energy stored in the current cache electric energy storage block can meet the remaining power demand of the day. The specific calculation method is: obtain the estimated power consumption Pe, target workload Wo1 and remaining workload Wo0 of the day, and calculate the remaining power consumption Pe0 of the day based on these data. The expression is:

8. The method for optimizing industrial power storage using dynamic prediction according to claim 7, characterized in that: Compare the remaining power consumption Pe0 with the remaining power Ce in the buffer power storage block. If Pe0≤Ce, the power storage block for that day will not be recharged. If Pe0>Ce, the energy storage block of the day will be charged again. The required amount of electricity is obtained by subtracting the remaining power consumption Pe0 from the remaining energy Ce in the buffer energy storage block. When the energy in the buffer energy storage block is 0, the energy storage block of the day will continue to be used for power supply operations.

9. The method for optimizing industrial power storage using dynamic prediction according to claim 8, characterized in that: When the power in the buffer power storage block is used up, the charging operation will be immediately performed to keep the power therein within a preset constant range; if the power in the power storage block after recharging on that day does not meet the power demand on that day, the power in the buffer power storage block will continue to be used in the same way; if the power in the power storage block after recharging on that day meets the power demand on that day and there is some power left, this part of power will be transferred to the buffer power storage block; When establishing an energy storage block for a new day, if the electric energy in the buffer energy storage block exceeds a preset constant range, part of the electric energy is first taken out of the buffer energy storage block and put into the energy storage block for this day, so that the electric energy in the buffer energy storage block is maintained within the constant range.

10. An industrial power supply energy storage optimization system using dynamic prediction, characterized in that: include: Data acquisition module, which acquires historical data, including daily workload data, weather data, and industrial electricity consumption data for each working day in the past year; and obtaining the estimated power consumption data, target workload data, and remaining workload data for the day when the power stored in the power storage block for the day is used up; The electricity consumption prediction module analyzes and predicts the electricity consumption of the next working day, including a first electricity consumption prediction unit, a second electricity consumption prediction unit and a comprehensive electricity consumption prediction unit; wherein the first electricity consumption prediction unit establishes electricity consumption prediction models for four seasons according to the daily workload and weather data of each historical working day, and predicts the electricity consumption of the next working day based on the electricity consumption prediction model; the second electricity consumption prediction unit predicts the electricity consumption error of the next working day according to the difference between the first predicted electricity consumption and the actual electricity consumption of each day in the previous five working days; the comprehensive electricity consumption prediction unit combines the prediction results of the first electricity consumption prediction unit and the second electricity consumption prediction unit to obtain the estimated electricity consumption Pe for the next working day; The electric energy block storage module stores the industrial electric energy in blocks in the industrial power storage system, including a conventional electric energy storage unit and a buffer electric energy storage unit; wherein the conventional electric energy storage unit is used for block storage of the predicted power consumption for several working days; the buffer electric energy storage unit is used for storing buffer electric energy, which is used as a backup power source in case of emergencies or when the power consumption does not match the predicted power consumption.

Images