Industrial electricity demand prediction method and system considering energy consumption characteristicsTechnical Field
The invention relates to the technical field of electricity demand prediction, in particular to an industrial electricity demand prediction method and system considering energy consumption characteristics.
Background
With the vigorous development of various industries in the current society, energy consumption, especially power consumption, becomes a key component in the industrial operation cost. The electricity requirements of different industries present complex and variable characteristics due to the differences of the production process, the equipment type and the production schedule.
On the one hand, the traditional electricity demand prediction method is often too general, and most of the traditional electricity demand prediction methods only conduct simple trend extrapolation according to historical electricity consumption data and cannot deeply analyze actual operation conditions of production equipment in industry. For example, in the manufacturing industry, rated powers of various processing devices are different, operation time is also different every day due to order tasks and scheduling arrangement, future demands are estimated by simply relying on past electricity consumption, difference factors of device layers are ignored, prediction deviation is quite easy to cause, and actual electricity consumption planning demands of industry cannot be accurately met.
On the other hand, the time factor is significant for industrial electric effects. During weekdays, weekends and holidays, the production activity of the industry is greatly different, and the fluctuation of electricity load is obvious. In addition, in daily, there are different electricity consumption characteristics such as peak, flat peak, low valley, etc. in different periods, for example, the chemical industry greatly increases the electricity consumption in the daytime production peak period, and the equipment part in the night low valley period is shut down to reduce the electricity consumption suddenly. If the electricity utilization rules of the time dimensions cannot be accurately mastered, enterprises are difficult to reasonably use An Paicuo peak electricity, so that the power shortage and the insufficient power supply in the peak period can possibly cause the influence on production, the power resource is easily wasted in the valley period, and the unnecessary electricity utilization cost is increased.
Moreover, yield variation is a common dynamic factor in industrial operations, closely linked to electricity demand. When the enterprises receive emergency orders and the output is required to be increased, the equipment combination and the operation time length which participate in production are correspondingly adjusted, so that the electricity consumption requirement is directly influenced. The traditional prediction method seldom carefully integrates yield association, so that under a yield fluctuation scene, electricity consumption prediction cannot be adjusted in time, reliable basis cannot be provided for enterprise production scheduling and power resource allocation, and efficient operation and cost control of enterprises are prevented.
In view of the above, in view of the shortcomings of the existing electricity demand prediction methods in terms of considering the industrial energy characteristics, a new electricity demand prediction method and system capable of comprehensively considering the conditions of production equipment, time factors and yield association in the industry are urgently needed, so as to meet the actual demands of industry for accurate electricity management, optimizing the power resource allocation and reducing the electricity cost.
Disclosure of Invention
The invention aims to provide an industrial electricity demand prediction method and system considering energy consumption characteristics, and solves the technical problems in the background technology.
The aim of the invention can be achieved by the following technical scheme:
An industrial electricity demand prediction method considering energy consumption characteristics, comprising:
first step, industrial data acquisition
Basic data in the target industry is collected, wherein the basic data comprises rated power of various production equipment in the industry, daily operation time of the production equipment, historical electricity consumption data of the industry and production scheduling plans of the industry;
second step, equipment electricity demand assessment
Calculating the daily theoretical electricity consumption of a single device according to the collected basic data, summarizing the daily theoretical electricity consumption of all devices in the industry, and determining the total daily theoretical electricity consumption;
Third step, time factor adjustment
Step D1, analyzing a time distribution rule of historical electricity consumption data:
Step D2, adjusting the total amount of theoretical electricity consumption per day according to the time type of the current date:
Fourth step, yield correlation analysis
Step G1, establishing an association relation between equipment power consumption and yield:
And G2, combining the expected output in the production scheduling plan, and determining the electricity consumption increment corresponding to the output change:
And G3, accumulating the electricity consumption increment to the adjusted daily electricity consumption predicted value to obtain a daily electricity demand predicted value.
The invention further provides a scheme, wherein rated power of various production devices in industry is recorded as Pi, wherein i represents the number of different production devices, i=1, 2, &..;
Recording the daily operation time length of the production equipment as Ti,j, j representing the date, j=1, 2.
Historical electricity usage data recorded in time series form is noted as Lk, k represents the electricity usage period, k=1, 2. P represents the number of power usage periods within a specified observation period, and Lk represents the power usage amount of the kth power usage period;
Industrial production scheduling plan that specifies the combination of devices involved in production and the expected yield in different scheduling periods, and records the expected yield as Qr,t, r representing the product line, i.e., the combination of devices involved in production, r=1, 2.
As a further scheme of the invention, the calculation mode of the daily theoretical power consumption of the single equipment is as follows:
by:
And calculating the daily theoretical power consumption E1i,j of each device in the observation period.
As a further scheme of the invention, the total daily theoretical electricity consumption is determined by the following steps:
Then by:
The theoretical total electricity consumption E2j of each date in the observation period is calculated.
In the step D1, the time distribution rule analysis mode is as follows:
Dividing the electricity consumption period in a specified observation period into a plurality of different types of electricity consumption periods according to the types of preset time periods, and marking the electricity consumption periods as g, wherein g=1, 2,..;
Selecting a type of electricity utilization period;
then pass through:
Calculating the average electricity utilization ratio Rg of different types to the application electricity period;
In the formula,Representing the sum of the power usage of one type corresponding to the applied power period.
As a further aspect of the invention, the time period types include weekdays, weekends, holidays, and peak, peak and valley periods of each day.
In the step D2, the total daily theoretical electricity consumption is adjusted as follows:
by:
And calculating a daily electricity consumption estimated value E3j after each date adjustment in the observation period.
The establishment mode of the association relation in the step G1 is as follows:
selecting a product line in a production scheduling plan;
Then, the electricity consumption of each electric equipment on the product line in a plurality of corresponding shift arrangement time periods is obtained and is recorded as E4i,t;
simultaneously extracting the expected output Qr,t of the product line respectively corresponding to a plurality of corresponding scheduling periods;
Then by:
Calculating the equipment electricity utilization coefficient Cr of the product line corresponding to the expected output;
in the step G2, the electricity consumption increment is calculated as follows:
by:
calculating electricity consumption increment ZEr,k caused by the change of the yield;
In the step G3, the calculation formula of the daily electricity demand predicted value is as follows:
Wherein EYj is a predicted daily electricity demand value.
An industrial electricity demand prediction system considering energy consumption characteristics, the system for implementing an industrial electricity demand prediction method considering energy consumption characteristics, the system comprising:
the data acquisition unit is used for collecting basic data in the target industry;
The electricity utilization evaluation unit is used for determining the total daily theoretical electricity utilization amount according to the collected basic data;
The factor adjusting unit is used for analyzing the time distribution rule of the historical electricity consumption data and adjusting the total daily theoretical electricity consumption according to the time type of the current date:
And the association analysis unit is used for establishing an association relation between equipment electricity consumption and yield, determining an electricity consumption increment corresponding to the yield change according to the corresponding association relation and the expected yield in the production scheduling plan, and accumulating the electricity consumption increment to the adjusted daily electricity consumption predicted value to obtain a daily electricity demand predicted value.
The invention has the beneficial effects that:
The method comprises the steps of collecting industrial data, collecting basic data covering various aspects of rated power, daily operation time, historical electricity consumption data, production scheduling plans and the like of various production equipment in the industry, fully considering key elements such as equipment attribute, actual operation condition, past electricity consumption appearance, production scheduling and the like related to industrial electricity consumption, laying a foundation for the follow-up accurate electricity consumption demand prediction to comprehensively and attach to the actual operation of the industry, and avoiding inaccurate prediction results caused by data loss or one-sided.
And in the equipment electricity demand evaluation link, based on the collected abundant basic data, the daily theoretical electricity consumption of a single equipment can be calculated scientifically and reasonably, the daily theoretical electricity consumption total amount of all the equipment in the industry is further obtained by summarizing, and the whole theoretical electricity consumption scale of the industry in a conventional state is clearly defined, so that the general level of the industrial electricity consumption can be mastered from a basic level, an important reference is provided for subsequent finer adjustment and analysis, and the basic accuracy of electricity demand prediction is improved.
The time factor adjustment link comprises the steps of carefully analyzing the time distribution rule of historical electricity data, dividing the electricity consumption time period according to diversified time period types such as working days, weekends, holidays, daily peak, flat peak, low valley time periods and the like, calculating the average electricity consumption proportion of each type of electricity consumption time period, accurately capturing the difference of electricity consumption conditions under different time characteristics, and correspondingly adjusting the total amount of electricity consumption of a daily theory according to the time type of the current date, so that a prediction result can be fully suitable for the change characteristics of electricity consumption requirements under different time scenes, and the accuracy and rationality of prediction when facing the influence of the time factor are effectively improved.
And in the yield association analysis link, firstly, an association relation between equipment electricity consumption and yield is established, equipment electricity consumption coefficient is determined by analyzing equipment electricity consumption and predicted yield on a product line, then electricity consumption increment corresponding to yield change is accurately determined by combining the predicted yield in a production scheduling plan, and a final daily electricity consumption demand predicted value is obtained by accumulating the electricity consumption increment to corresponding values, the direct influence of the key variable of the production yield on the electricity consumption demand is fully considered, the fact that the electricity consumption demand prediction can be attached to the actual production condition under different yield plans is ensured, and the accuracy of integral prediction is further enhanced.
The system architecture power-assisted efficient implementation comprises a specially designed industrial electricity demand prediction system considering energy consumption characteristics, and comprises a data acquisition unit, an electricity evaluation unit, a factor adjustment unit and an association analysis unit, wherein each unit is explicitly and cooperatively matched in a dividing way and is respectively responsible for the work of data collection, theoretical electricity total amount determination, time factor adjustment, yield association analysis and the like of corresponding links, so that the efficient and orderly operation of the whole prediction method is realized, the implementation of landing in actual industrial application is facilitated, the efficiency and quality of industrial electricity demand prediction work are improved, the industry is helped to better plan electricity utilization arrangement, the electric power resource is reasonably allocated, the electricity utilization cost is effectively controlled and the like.
In summary, the invention can provide accurate, reliable and practical electricity demand prediction for industry through comprehensive multi-aspect factors, multi-loop fine analysis and design of matched system architecture, and has positive and important effects on scientific electricity management of industry.
Drawings
The invention is further described below with reference to the accompanying drawings.
FIG. 1 is a system block diagram of an industrial electricity demand prediction system that takes into account energy usage characteristics in accordance with the present invention.
FIG. 2 is a flow chart of an industrial electricity demand prediction method taking into account energy consumption characteristics according to the present invention.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
Examples
Referring to fig. 1 and 2, the present invention is an industrial electricity demand prediction method considering energy consumption characteristics, comprising:
first step, industrial data acquisition
For the target industry, the following underlying data is collected:
The power rating of various production facilities within the industry is noted as Pi, where i represents the number of the different production facilities, i=1, 2. N represents the number of different production devices, and rated power is obtained from a device nameplate or a product specification, and the unit is kilowatt;
the production facility was run for a daily period of time and noted as Ti,j, j representing the date, j=1, 2. The daily operation time length is recorded by an operation monitoring device which is pre-installed on the equipment, and the unit is hour;
Historical electricity consumption data of industry is recorded in a time series form and is recorded as Lk, k represents electricity consumption time period, k=1, 2......p..p, p representing the number of power usage periods within a given observation period, Lk represents the power consumption amount of the kth power consumption period, in kilowatt-hours, and historical electricity usage data is obtained from electricity meter records within the power sector or business, in this embodiment, each hour is used as an electricity usage period to record electricity usage data;
An industrial production scheduling plan which defines the equipment combination and the expected yield of the production in different scheduling periods, and marks the expected yield as Qr,t, r represents a product line, namely the equipment combination of the production, r=1, 2, &....u..u, u representing the number of product lines, t represents a shift period, t=1, 2..v., v represents the number of shift slots, and the units corresponding to the expected output are determined according to the characteristics of the industrial product, such as tons or pieces;
second step, equipment electricity demand assessment
Calculating the daily theoretical electricity consumption of a single device according to the collected basic data, summarizing the daily theoretical electricity consumption of all devices in the industry, and determining the total daily theoretical electricity consumption;
The method comprises the following steps:
Firstly, by:
Calculating the daily theoretical power consumption E1i,j of each device in the observation period;
Then by:
Calculating theoretical total electricity consumption E2j of each date in the observation period;
Third step, time factor adjustment
Step D1, analyzing a time distribution rule of historical electricity consumption data:
The method comprises the following steps:
Dividing the electricity consumption period in a specified observation period into a plurality of different types of electricity consumption periods according to the types of preset time periods, and marking the electricity consumption periods as g, wherein g=1, 2,..;
In this embodiment, the time period types such as weekdays, weekends, holidays, and peak, flat peak, valley periods of each day;
Selecting a type of electricity utilization period;
then pass through:
Calculating the average electricity utilization ratio Rg of different types to the application electricity period;
In the formula,Representing a sum of power usage of one type corresponding to the applied power period;
Step D2, adjusting the total daily theoretical electricity consumption of the industry according to the time type of the current date:
The method comprises the following steps:
by:
Calculating a daily electricity consumption estimated value E3j after each date adjustment in the observation period;
Fourth step, yield correlation analysis
Step G1, establishing an association relation between equipment power consumption and yield:
The method comprises the following steps:
selecting a product line in a production scheduling plan;
Then, the electricity consumption of each electric equipment on the product line in a plurality of corresponding shift arrangement time periods is obtained and is recorded as E4i,t;
simultaneously extracting the expected output Qr,t of the product line respectively corresponding to a plurality of corresponding scheduling periods;
Then by:
Calculating the equipment electricity utilization coefficient Cr of the product line corresponding to the expected output;
And G2, combining the expected output in the production scheduling plan, and determining the electricity consumption increment corresponding to the output change:
The method comprises the following steps:
by:
calculating electricity consumption increment ZEr,k caused by the change of the yield;
Step G3, accumulating the increment ZEr,k to the adjusted daily electricity consumption predicted value E3j to obtain a daily electricity consumption demand predicted value;
The formula is as follows:
Wherein EYj is a predicted daily electricity demand value.
According to the method, through collecting multi-dimensional basic data such as rated power, daily operation time, historical electricity consumption data of industry and production scheduling plans of various production devices in an industry, comprehensive and detailed information support is provided for subsequent accurate electricity demand prediction, multiple aspects such as self-characteristics, actual operation conditions, past electricity consumption conditions and production arrangement of the devices are covered, prediction can be based on actual and attached industrial actual operation conditions, daily theoretical electricity consumption of a single device can be scientifically calculated based on collected data, daily theoretical electricity consumption total amount of all the devices is summarized, the electricity consumption scale in the industry theory under normal conditions is defined, the basic electricity consumption level is helped to be grasped, the time distribution rule of the historical electricity consumption data is analyzed, different types of electricity consumption periods are distinguished, corresponding average electricity consumption proportion is calculated, the theoretical electricity consumption total amount is adjusted according to the time types of the current date, the influence of time factors on the electricity consumption demands is fully considered, the predicted result is attached to the electricity consumption change conditions under the actual time scenes, the association relation between equipment electricity consumption and the output is established, the equipment electricity consumption and the expected output can be scientifically calculated on the basis of the collected data, the electricity consumption coefficient is determined on the line, the electricity consumption of the device is predicted to have the electricity consumption demand is directly estimated according to the expected electricity consumption demand change when the electricity consumption is not considered, and the electricity consumption is not influenced on the electricity consumption plan is directly on the production schedule.
Examples
As an embodiment two of the present application, when the present application is implemented, compared with the embodiment one, the technical solution of the present embodiment differs from the embodiment one only in that the present embodiment further includes dynamic feedback adjustment, and the manner is as follows:
In the daily production process, acquiring daily actual electricity data in real time, and marking the data as Yj;
then pass through:
Calculating an error rate Wj between actual daily electricity consumption data and a predicted value of a corresponding daily electricity consumption demand;
The error rate Wj is then compared with a predetermined error threshold WY:
If Wj is more than WY, checking the equipment running state monitoring system to check whether equipment suddenly breaks down and stops;
if the electric equipment is found to be stopped, positioning the electric equipment number, and simultaneously obtaining a stopping time node, further calculating the originally predicted electricity demand in the residual operation time of the equipment, and calculating the part of electricity demand according to 0 at the moment to correct the follow-up prediction;
Meanwhile, the system communicates with a production scheduling department to confirm whether an urgent production task is temporarily added;
if the temporarily increased urgent production task exists, the information of equipment, yield requirement, estimated time consumption and the like related to the task is known in detail so as to re-evaluate the electricity increment;
Then rechecking the affected part according to the backtracking and checking result;
if the equipment failure is caused, the theoretical total power consumption of the rest normal operation equipment in the follow-up unfinished period is recalculated;
According to the method of the second step, the theoretical electricity consumption of the single equipment in the residual period is calculated firstly based on the residual operation time and the rated power, and then the new theoretical electricity consumption total of the industrial residual period is obtained by summarizing;
and repeating the third step, and adjusting time factors according to the time type of the current residual period and the recalculated theoretical total power consumption to obtain an adjusted power consumption estimated value.
Finally, as in the fourth step, consider yield-related variations;
if the emergency production task changes the yield layout, recalculating the equipment electricity utilization coefficient of unit yield, further calculating electricity utilization increment generated by yield fluctuation, accumulating the electricity utilization increment to the adjusted electricity utilization predicted value, and obtaining an updated daily electricity utilization demand predicted value for guiding the electricity utilization arrangement of subsequent production;
In the embodiment, actual electricity data is collected in real time in the daily production process, and compared with a predicted value to calculate an error rate, and a subsequent backtracking checking mechanism is triggered by comparing the error rate with a preset error threshold. When the error is larger, the influence factors such as sudden fault shutdown of the equipment and temporary increase of urgent production tasks can be purposefully checked, and the real-time dynamic feedback adjustment mechanism can timely find out the reasons of inconsistent prediction and reality, so that the adaptability and accuracy of the prediction method in practical application are enhanced. If a temporary emergency production task exists, the electricity increment is reevaluated by detailed knowledge of related tasks, and the electricity consumption coefficient of the equipment with unit yield is recalculated when necessary, so that an updated daily electricity demand forecast value is obtained, the forecast value can adapt to the conditions of yield layout change and the like caused by temporary production task fluctuation, the electricity consumption arrangement of the subsequent production is better guided, and the phenomena of electricity shortage, waste and the like caused by temporary task are avoided.
Examples
As an embodiment three of the present application, in the implementation of the present application, the technical solution of the present embodiment is to combine the solutions of the above embodiment one and embodiment two compared with the embodiment one and embodiment two.
The embodiment combines the schemes of the first embodiment and the second embodiment, and integrates the advantages of the comprehensive multi-factor prediction method of the first system and the real-time dynamic feedback adjustment mechanism of the second embodiment. The method can accurately predict the electricity demand based on multi-factor comprehensive analysis under normal production conditions, and can timely correct the prediction result through dynamic feedback when actual production change conditions such as equipment faults, temporary emergency production tasks and the like occur, so that the method can always fit the actual electricity consumption conditions, ensure the accuracy and adaptability of electricity demand prediction to the greatest extent and the effective guiding effect on industrial production electricity utilization arrangement, and realize reasonable electricity utilization, scientific arrangement of electricity resources, effective control of electricity utilization cost and the like for the omnibearing power-assisted industry.
Referring to fig. 1 and 2, the present invention further provides an industrial electricity demand prediction system considering energy consumption characteristics, where the system is used to implement an industrial electricity demand prediction method considering energy consumption characteristics, and the system includes:
the data acquisition unit is used for collecting basic data in the target industry;
The electricity utilization evaluation unit is used for determining the total daily theoretical electricity utilization amount according to the collected basic data;
The factor adjusting unit is used for analyzing the time distribution rule of the historical electricity consumption data and adjusting the total daily theoretical electricity consumption according to the time type of the current date:
And the association analysis unit is used for establishing an association relation between equipment electricity consumption and yield, determining an electricity consumption increment corresponding to the yield change according to the corresponding association relation and the expected yield in the production scheduling plan, and accumulating the electricity consumption increment to the adjusted daily electricity consumption predicted value to obtain a daily electricity demand predicted value.
The above formulas are all formulas with dimensionality removed and numerical calculation, the formulas are formulas with the latest real situation obtained by software simulation through collecting a large amount of data, and preset parameters and threshold selection in the formulas are set by those skilled in the art according to the actual situation.
The foregoing is merely illustrative of the present application, and the present application is not limited thereto, and any person skilled in the art will readily recognize that variations or substitutions are within the scope of the present application. Therefore, the protection scope of the present application shall be subject to the protection scope of the claims.