Power load prediction method and device for metallurgical enterprise, and medium

Abstract

The embodiment of the invention provides a power load prediction method and device for a metallurgical enterprise and a medium, and belongs to the technical field of power. The method comprises the following steps: classifying each production unit of the metallurgical enterprise into a stable load unit and an impact load unit, wherein the impact strength of the impact load unit on a power grid is greater than that of the stable load unit; for the stable load unit and the impact load unit, sequentially executing load static prediction based on historical energy data and load dynamic correction based on real-time energy data to obtain respective load prediction results; and integrating respective load prediction results of the stable load unit and the impact load unit to obtain a power load prediction result of the metallurgical enterprise. According to the embodiment of the invention, the load static prediction and the load dynamic correction are respectively carried out on the stable load unit and the impact load unit, the overall power load prediction is finally realized, and the safety production is ensured.

Description

Technical Field

[0001] The present invention relates to the field of electric power technology, and in particular to a method, a device and a medium for predicting electric power load of a metallurgical enterprise. Background Art

[0002] With the continuous development of industrialization, metallurgical enterprises need a lot of electricity to support their production. Their electricity consumption accounts for about 10% of the total electricity consumption in the country, and the proportion of industrial electricity consumption is as high as about 15%. Therefore, electricity costs have become an important part of the production costs of metallurgical enterprises. In the face of fierce market competition, how to effectively reduce the cost of purchasing electricity has become a key task for metallurgical enterprises to improve their competitiveness and economic benefits.

[0003] Considering the difference between the cost of self-generated electricity and the cost of purchased electricity, metallurgical enterprises, as both power generation units and power users, can reduce the cost of purchased electricity by increasing the proportion of self-generated electricity during peak electricity price periods and increasing the proportion of purchased electricity during flat periods. Therefore, it is particularly important to formulate a reasonable power generation plan.

[0004] However, metallurgical enterprises, especially steel enterprises, generally have impact loads such as refining furnaces and rolling mills. The actual working conditions in the production process are complex and changeable, which can easily lead to large fluctuations in power loads and affect the dispatching control and load regulation of generator sets. Therefore, power load forecasting has very important application value in ensuring the safe operation of the system, achieving energy conservation and maximizing production benefits.

[0005] At present, enterprises usually predict the daily electricity load, but because the time of impact load generation is uncertain and there is no obvious pattern, they usually choose to ignore the load impact and only predict the electricity load of each production unit based on the historical average of the stable load, and then adjust the final load curve based on experience. Taking steel enterprises as an example, they usually consider the maintenance plan and production plan comprehensively, predict the daily load based on the stable load average of each production unit, and then adjust it based on experience by relevant personnel. In the actual production process, the electricity load is also adjusted based on the experience and judgment of the dispatcher.

[0006] However, daily load forecasting cannot fully consider the impact of shock loads and relies on manual experience for adjustment and optimization. The accuracy of the results is affected by personal ability. In the actual production process, there is a lack of ultra-short-term load forecasting. Dispatchers only evaluate subsequent load changes based on experience, which lacks stability and reliability.

[0007] Therefore, this application proposes a new power load forecasting scheme for metallurgical enterprises. Summary of the invention

[0008] The purpose of the embodiments of the present invention is to provide a method, device and medium for predicting power load of a metallurgical enterprise, so as to at least partially solve the above technical problems.

[0009] In order to achieve the above-mentioned purpose, an embodiment of the present invention provides a method for predicting the power load of a metallurgical enterprise, comprising: classifying each production unit of the metallurgical enterprise into a stable load unit and an impact load unit, wherein the impact intensity of the impact load unit on the power grid is greater than that of the stable load unit; for the stable load unit and the impact load unit, performing load static prediction based on historical energy data and load dynamic correction based on real-time energy data in sequence to obtain respective load prediction results; and integrating the respective load prediction results of the stable load unit and the impact load unit to obtain the power load prediction result of the metallurgical enterprise.

[0010] Optionally, classifying the production units of the metallurgical enterprise into stable load units and impact load units includes: classifying the various production units of the metallurgical enterprise according to the production process characteristics and / or load fluctuation characteristics of each production unit to determine the stable load units and the impact load units.

[0011] Optionally, both the historical energy data and the real-time energy data include planning data, production condition data and equipment operation data, and the planning data and the production condition data are obtained from the production management system of the metallurgical enterprise, while the equipment operation data is obtained from the equipment management system of the metallurgical enterprise.

[0012] Optionally, before executing the static load forecasting, the power load forecasting method further includes: performing data cleaning and / or outlier correction on the historical energy data.

[0013] Optionally, for the stable load unit, the static load prediction based on historical energy data includes: extracting load characteristic parameters that respectively characterize the normal state and maintenance state of the stable load unit from the historical energy data; and determining the respective load mean and load fluctuation range of the stable load unit in the normal state and the maintenance state based on the load characteristic parameters; and in combination with future maintenance plans, performing load prediction on the stable load unit based on the determined load mean and load fluctuation range to obtain a first prediction result.

[0014] Optionally, for the stable load unit, the dynamic load correction based on the real-time energy data includes: tracking the state change of the stable load unit based on the real-time energy data to dynamically correct the first prediction result to obtain a second prediction result.

[0015] Optionally, for the impact load unit, the static load prediction based on historical energy data includes: extracting load characteristic parameters associated with the production plan of the prediction period from the historical energy data; determining the load mean and load fluctuation range of the impact load unit in the impact period and non-impact period under normal conditions according to the load characteristic parameters; performing load prediction for a single production cycle for the impact load unit based on the determined load mean and load fluctuation range in combination with the load fluctuation information of the previous cycle adjacent to the current production cycle; and splicing the load prediction results of multiple production cycles according to the scheduling plan associated with the prediction period to obtain a third prediction result.

[0016] Optionally, for the impact load unit, the dynamic load correction based on real-time energy data includes: tracking the state change of the impact load unit based on the real-time energy data to dynamically correct the third prediction result to obtain a fourth prediction result.

[0017] On the other hand, an embodiment of the present invention also provides an electric power load forecasting device for a metallurgical enterprise, comprising: a memory configured to store instructions; and a processor configured to call the instructions from the memory and to implement any of the above-mentioned electric power load forecasting methods when executing the instructions.

[0018] On the other hand, an embodiment of the present invention further provides a machine-readable storage medium, on which instructions are stored, and the instructions are used to enable a machine to execute any of the above-mentioned power load forecasting methods.

[0019] Through the above technical scheme, the embodiment of the present invention divides the stable load units and impact load units with different fluctuation characteristics, and performs static load prediction and dynamic load correction on the two respectively, and finally realizes the overall power load prediction, which helps to ensure safe production.

[0020] Other features and advantages of the embodiments of the present invention will be described in detail in the subsequent detailed description. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] The accompanying drawings are used to provide a further understanding of the embodiments of the present invention and constitute a part of the specification. Together with the following specific implementations, they are used to explain the embodiments of the present invention, but do not constitute a limitation on the embodiments of the present invention. In the accompanying drawings:

[0022] Figure 1 It is a flow chart of a method for predicting power load of a metallurgical enterprise according to an embodiment of the present invention;

[0023] Figure 2 is a schematic diagram of a flow chart of load forecasting for a stable load unit in Example 1 of an embodiment of the present invention;

[0024] Figure 3 is a schematic diagram of a flow chart of load prediction for an impact load unit in Example 2 of an embodiment of the present invention; and

[0025] Figure 4 It is a structural schematic diagram of a power load forecasting device for a metallurgical enterprise according to an embodiment of the present invention. DETAILED DESCRIPTION

[0026] The specific implementation of the embodiment of the present invention is described in detail below in conjunction with the accompanying drawings. It should be understood that the specific implementation described here is only used to illustrate and explain the embodiment of the present invention, and is not used to limit the embodiment of the present invention.

[0027] It should be noted that the acquisition, transmission, storage, use, and processing of data in the technical solution of this application are in compliance with the relevant provisions of national laws and regulations. In the embodiments of the present invention, some existing solutions in the industry such as certain software, components, and models may be mentioned, which should be considered as exemplary. Their purpose is only to illustrate the feasibility of implementing the technical solution of this application, but it does not mean that the applicant has or will necessarily use the solution.

[0028] Figure 1 FIG. 1 is a flow chart of a method for predicting power load of a metallurgical enterprise according to an embodiment of the present invention, wherein a steel enterprise is taken as an example of a metallurgical enterprise. Figure 1 As shown, the power load prediction method may include the following steps S100-S300.

[0029] Step S100, classifying the various production units of the metallurgical enterprise into stable load units and impact load units, wherein the impact intensity of the impact load unit on the power grid is greater than that of the stable load unit.

[0030] In a preferred embodiment, the various production units of the metallurgical enterprise can be classified according to the production process characteristics and / or load fluctuation characteristics of each production unit to determine the stable load unit and the impact load unit. The impact load unit is a unit with a relatively large power load that is prone to impact the power grid, while the stable load unit is a unit with a relatively stable power load.

[0031] On the one hand, based on a large amount of historical energy data of each production unit, the load fluctuation characteristics of each production unit can be analyzed and obtained, and then the stable load unit and the impact load unit can be determined accordingly.

[0032] On the other hand, according to the production process characteristics of each production unit, for example, the production units are classified into sintering, blast furnace, refining furnace, continuous casting, converter, steel rolling, etc. Taking steel rolling as an example, its production is continuous, and the production rhythm is fast, the load fluctuation is large, and the impact on the power grid is large, so it belongs to the impact load unit. Similarly, according to the production process characteristics of other production units, it can be determined whether they belong to stable load units or impact load units.

[0033] Step S200, for the stable load unit and the impact load unit, static load prediction based on historical energy data and dynamic load correction based on real-time energy data are sequentially performed to obtain respective load prediction results.

[0034] In a preferred embodiment, both the historical energy data and the real-time energy data include planning data, production condition data and equipment operation data, and the planning data and the production condition data are obtained from the production management system of the metallurgical enterprise, while the equipment operation data is obtained from the equipment management system of the metallurgical enterprise.

[0035] For example, real-time and historical data such as production plans, maintenance plans, scheduling plans, and production condition signals are obtained through the production management system, and real-time and historical data on the operating status of key equipment are obtained through the equipment management system.

[0036] In addition, in the example, the collection of energy data can be achieved by setting up multiple metering points for data collection, where the metering point refers to the location where the metering device should be installed. In addition, the metering point data can be classified according to multiple dimensions and angles, such as dividing the metering points into data belonging to the production workshop and data belonging to the public auxiliary workshop according to the workshop to which they belong, where the public auxiliary workshop refers to the public auxiliary workshop where industrial enterprises provide water, electricity, gas, cold, heat, etc. for direct production activities; according to the different production process units, the data of the production process units are divided into sintering, blast furnace, refining furnace, continuous casting, converter, steel rolling and other level data; according to the attributes of the metering points, they are divided into instantaneous flow, cumulative flow, active power, reactive power, temperature, pressure, current and other data. In particular, according to the time series, it is divided into historical data and real-time data, that is, historical energy data and real-time energy data are formed.

[0037] The above step 200 first performs static load forecasting based on historical energy data, but historical energy data has defects such as large data volume and inaccurate data information. Therefore, the historical energy data can be preprocessed first, and the preprocessing preferably includes: data cleaning and / or outlier correction of the historical energy data.

[0038] Among them, data cleaning is a common data processing method, which will not be elaborated here. As for the correction of abnormal values, in the application scenario of the embodiment of the present invention, for example, for the metering points of some stable load units belonging to the public auxiliary workshop, if there is maintenance in a certain period of history, data correction is required to eliminate the influence of maintenance. Therefore, data preprocessing methods such as data cleaning and abnormal value detection can help ensure the accuracy and reliability of the data.

[0039] The above step S200 performs load prediction on the stable load unit and the impact load unit respectively. The load prediction processes of the two are specifically introduced below by way of examples.

[0040] Figure 2 This is a flow chart of load forecasting for stable load units in Example 1 of an embodiment of the present invention. For example, for stable load units with basically stable power loads such as water treatment units and office areas, load fluctuations occur only due to the influence of maintenance plans. In this regard, Example 1 mainly considers the maintenance plan to forecast the load, and then corrects the forecast results in combination with real-time energy data.

[0041] In Example 1, Figure 2 As shown, the load prediction for the stable load unit includes static load prediction corresponding to steps S210A-S230A and dynamic load correction corresponding to step S240A. Each step is described in detail as follows.

[0042] Step S210A, extracting load characteristic parameters respectively representing the normal state and the maintenance state of the stable load unit from the historical energy data.

[0043] Step S220A, determining the respective load mean values ​​and load fluctuation ranges of the stable load unit in the normal state and the maintenance state according to the load characteristic parameters.

[0044] Step S230A, in combination with the future maintenance plan, based on the determined load mean and load fluctuation range, load forecasting is performed on the stable load unit to obtain a first forecasting result.

[0045] For steps S210A-S230A, for example, load characteristic parameters such as instantaneous flow, cumulative flow, active power, reactive power, temperature, pressure, and current are extracted from the collected historical energy data about the corresponding stable load unit. Taking current as an example, the current values ​​in the normal state and the maintenance state are obviously different. Then, the extracted load characteristic parameters are statistically processed to obtain the respective load mean values ​​and load fluctuation ranges in the normal state and the maintenance state. Finally, the load impact amount (such as the shutdown time of the production unit shown in the maintenance plan) is determined in combination with the maintenance plan, and a daily load static prediction is performed based on the determined load mean value and load fluctuation range to obtain a daily load static prediction result, which is the first prediction result recorded in the embodiment of the present invention.

[0046] Among them, the daily load static prediction can be performed by a neural network. For example, the BP neural network, through the back propagation algorithm, adjusts the weight and threshold of the network according to the error between the input sample and the prediction result, thereby realizing load prediction. In the load prediction of the power system, the BP neural network can learn and capture the pattern of load change, such as periodicity, trend and abnormal situation, and the BP neural network can also consider the influence of various load factors on the load, thereby improving the accuracy of the prediction. In this example one, the BP neural network model is trained based on historical energy data, and the production plan information (including product planned steel type, specification, number of rolling mills, planned rolling time, etc.) is used as the input of the BP neural network model as the load influence quantity, and the corresponding static load prediction result can be output. In addition, the least square method, support vector machine (Support Vector Machine, SVM), etc. are also commonly used load prediction algorithms. In particular, the SVM algorithm has the advantages of fast learning speed, global optimization and strong promotion ability, and can achieve a relatively ideal effect in the load prediction of the embodiment of the present application.

[0047] Step S240A, tracking the state change of the stable load unit based on the real-time energy data to dynamically correct the first prediction result to obtain a second prediction result.

[0048] For example, on the basis of obtaining the first prediction result, according to the real-time energy data obtained, such as real-time production condition signals and maintenance plans, the state changes of the production units are tracked in real time, and the first prediction result is continuously and dynamically corrected to obtain an ultra-short-term load dynamic prediction result, which is recorded as the second prediction result in the embodiment of the present invention, to form a load prediction trend.

[0049] Among them, the dynamic correction is aimed at obtaining the load dynamic prediction result, so it is actually a dynamic prediction, which can also be performed through the BP neural network. In this example 1, for the constructed BP neural network model, the actual steel grade, specification, rolling time and working condition parameters (including rolling force, rolling torque, rolling speed) of the product are used as input to output the load dynamic prediction result, that is, the second prediction result.

[0050] It should be noted that, in the embodiment of the present invention, load forecasting is classified into the following three types according to time:

[0051] (1) Ultra-short-term load forecasting: refers to the prediction of load values ​​within tens of minutes, minutes or even seconds in the future. The corresponding prediction results can be used to control the power grid online, reasonably dispatch real-time power generation capacity, and minimize power generation costs.

[0052] (2) Short-term load forecasting: refers to the load forecasting from one day to one week, usually in hours. The embodiment of the present invention mainly focuses on daily load forecasting. The corresponding forecasting results are mainly used for power distribution and coordination, unit combination and other plans.

[0053] (3) Medium- and long-term load forecast: refers to the forecast for more than one year. The corresponding forecast results mainly provide a basis for the development of power, scale construction, power dispatching, capital and other supply and demand balance. The embodiment of the present invention does not consider medium- and long-term load forecasting.

[0054] Among them, the time span of short-term load forecasting is short, and the impact of external unstable factors on it is more obvious than that of medium- and long-term load forecasting, and the difficulty of forecasting is also greater; and the current metallurgical enterprises lack ultra-short-term load forecasting in the actual production process. Thus, through the above steps S210A-S240A, the embodiment of the present invention provides an automated and process-based load forecasting solution for stable load units through static load forecasting and dynamic load correction, and can realize ultra-short-term load forecasting of the actual production process, thereby improving the stability and reliability of load forecasting of stable load units.

[0055] Figure 3 This is a flow chart of load forecasting for impact load units in Example 2 of an embodiment of the present invention. For example, for steel rolling equipment, it belongs to an impact load unit, and load fluctuations are affected by multiple factors. In this regard, Example 2 mainly considers combining production plans, impact periods and non-impact periods with different load characteristics to forecast the load, and then combines real-time energy data to correct the forecast results.

[0056] In Example 2, if Figure 3As shown, the load prediction for the stable load unit includes static load prediction corresponding to steps S210B-S240B and dynamic load correction corresponding to step S250B, and each step is described in detail as follows.

[0057] Step S210B: extracting load characteristic parameters associated with the production plan for the forecast period from the historical energy data.

[0058] For example, first obtain the production plan for the forecast period, and then extract the load characteristic parameters under normal production conditions with the same product types and specifications as the planned production from the historical energy data. The load characteristic parameters are also instantaneous flow, cumulative flow, active power, reactive power, temperature, pressure, current, etc.

[0059] Step S220B: determining the load mean and load fluctuation range of the impact load unit in the impact period and the non-impact period of the normal state according to the load characteristic parameters.

[0060] For example, combined with the time series information of the extracted load characteristic parameters, the corresponding normal state impact period and non-impact period can be determined, and then the load characteristic parameters of the impact period and non-impact period are statistically processed respectively to obtain the load mean and load fluctuation range of the impact period and non-impact period.

[0061] Step S230B, combining the load fluctuation information of the previous cycle adjacent to the current production cycle, based on the determined load mean and load fluctuation range, performs load forecasting for a single production cycle on the impact load unit.

[0062] Step S240B, according to the scheduling plan associated with the forecast period, the load forecast results of multiple production cycles are spliced ​​to obtain a third forecast result.

[0063] For step S230B and step S240B, for example, combined with the load fluctuation of the most recent rolling cycle under normal production conditions, a static prediction result of the power load in units of rolling cycles is obtained, and then each fragment is spliced ​​according to the scheduling plan to obtain the corresponding static prediction result of the daily power load, which is recorded as the third prediction result in the embodiment of the present invention.

[0064] In addition, similar to step S230A, the load forecasting here can also be performed by a BP neural network.

[0065] Step S250B: Track the state change of the impact load unit based on the real-time energy data to dynamically correct the third prediction result to obtain a fourth prediction result.

[0066] For example, based on the third prediction result, the real-time operation status of the rolling equipment is monitored according to the real-time production condition signal obtained by docking with the production management system. Once a change in the signal is detected, the load characteristic parameters are immediately recorded and recalculated, new process parameters are generated, the first prediction result is dynamically corrected, and the corresponding fourth prediction result is obtained, refreshing the power load prediction trend.

[0067] In addition, similar to step S240A, the load dynamic correction here can also be performed by a BP neural network.

[0068] Current metallurgical enterprises often neglect the impact load with a relatively high impact intensity on the power grid. However, through the above steps S210B - S250B, the embodiments of the present invention provide an automated and process-based load prediction solution for the impact load unit via load static prediction and load dynamic correction, and can achieve ultra-short-term load prediction for the actual production process, improving the stability and reliability of the load prediction for the impact load unit.

[0069] After obtaining the second prediction result corresponding to the stable load unit and the fourth prediction result corresponding to the impact load unit, return to Figure 1 step S300 to integrate the load prediction results.

[0070] Step S300: Integrate the load prediction results of the stable load unit and the impact load unit respectively to obtain the power load prediction result of the metallurgical enterprise.

[0071] For example, for the second prediction result corresponding to the stable load unit and the fourth prediction result corresponding to the impact load unit, the two are integrated according to a preset rule to obtain the overall power load prediction result, form a prediction trend, and adjust the overall result obtained here in a timely manner according to the dynamic correction situation of the above results. Among them, the rules for integration include but are not limited to weighted average, equal-weight average, etc.

[0072] In other examples, after obtaining the power load prediction result of the metallurgical enterprise through step S300, the following operations can be performed:

[0073] 1) Adjust the enterprise's electricity load curve through the obtained prediction result to reduce the impact of the load peak;

[0074] 2) Provide a decision-making reference to the dispatcher through this prediction result to assist them in making timely decisions during the production process, ensuring the overall balance of the power system load and guaranteeing safe production;

[0075] 3) The prediction result can be combined with the gas balance dispatching and peak-shaving power generation optimization to achieve efficient use of energy and reduce energy consumption costs.

[0076] In summary, the embodiment of the present invention conducts a qualitative and quantitative analysis of the power load of a metallurgical enterprise, fully considers the various factors that affect load fluctuations in the production process, divides stable load units and impact load units with different fluctuation characteristics, and performs static load prediction and dynamic load correction on the two, ultimately achieving overall power load prediction, achieving the purpose of ensuring the safe operation of the system, and on this premise, achieving energy conservation and maximizing production benefits.

[0077] Figure 4 Schematic diagram of the structure of a power load forecasting device for a metallurgical enterprise according to an embodiment of the present invention. Figure 4 As shown, the power load forecasting device includes: a memory configured to store instructions; and a processor configured to call the instructions from the memory and implement any of the above-mentioned power load forecasting methods when executing the instructions.

[0078] The processor includes a kernel, and the kernel retrieves the corresponding program unit from the memory. One or more kernels may be provided, and the power load prediction method of the embodiment of the present invention is implemented by adjusting kernel parameters.

[0079] The memory may include non-permanent memory in a computer-readable medium, random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM, and the memory includes at least one memory chip.

[0080] The power load prediction device of the embodiment of the present invention may be a server, PC, PAD, mobile phone, etc. For more implementation details and effects of the power load prediction device, please refer to the above embodiment of the power load prediction method, which will not be repeated here.

[0081] An embodiment of the present invention provides a machine-readable storage medium, on which instructions are stored, and the instructions are used to enable a machine to execute any of the above-mentioned power load forecasting methods.

[0082] An embodiment of the present invention provides a processor, which is used to run a program, wherein the program executes any of the above-mentioned power load forecasting methods when running.

[0083] An embodiment of the present invention further provides a computer program product, which, when executed on a data processing device, is suitable for executing the steps of initializing any of the above-mentioned power load forecasting methods.

[0084] Those skilled in the art will appreciate that the embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application may adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that contain computer-usable program code.

[0085] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of the processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0086] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0087] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0088] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0089] The memory may include non-permanent memory in a computer-readable medium, random access memory (RAM) and / or non-volatile memory in the form of read-only memory (ROM) or flash RAM. The memory is an example of a computer-readable medium.

[0090] Computer readable media include permanent and non-permanent, removable and non-removable media that can be implemented by any method or technology to store information. Information can be computer readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic tape disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer readable media does not include temporary computer readable media (transitory media), such as modulated data signals and carrier waves.

[0091] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.

[0092] The above are only embodiments of the present application and are not intended to limit the present application. For those skilled in the art, the present application may have various changes and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included within the scope of the claims of the present application.

Claims

1. A method for predicting power load of a metallurgical enterprise, characterized in that: include: Classifying the various production units of the metallurgical enterprise into stable load units and impact load units, wherein the impact intensity of the impact load unit on the power grid is greater than that of the stable load unit; For the stable load unit and the impact load unit, static load prediction based on historical energy data and dynamic load correction based on real-time energy data are sequentially performed to obtain respective load prediction results; as well as The load forecast results of the stable load unit and the impact load unit are integrated to obtain the power load forecast result of the metallurgical enterprise.

2. The power load forecasting method according to claim 1, characterized in that: The classification of the production units of the metallurgical enterprise into stable load units and impact load units includes: According to the production process characteristics and / or load fluctuation characteristics of each production unit, each production unit of the metallurgical enterprise is classified to determine the stable load unit and the impact load unit.

3. The power load forecasting method according to claim 1, characterized in that: Both the historical energy data and the real-time energy data include planning data, production condition data and equipment operation data, and the planning data and the production condition data are obtained from the production management system of the metallurgical enterprise, while the equipment operation data is obtained from the equipment management system of the metallurgical enterprise.

4. The power load forecasting method according to claim 1, characterized in that: Before executing the static load prediction, the power load prediction method further includes: The historical energy data is cleaned and / or outlier corrected.

5. The power load forecasting method according to claim 1, characterized in that: For the stable load unit, the load static prediction based on historical energy data includes: Extracting load characteristic parameters respectively representing the normal state and the maintenance state of the stable load unit from the historical energy data; and Determining the load mean and load fluctuation range of the stable load unit in the normal state and the maintenance state according to the load characteristic parameters; and In combination with future maintenance plans, based on the determined load mean and load fluctuation range, load forecasting is performed on the stable load unit to obtain a first forecasting result.

6. The power load forecasting method according to claim 5, characterized in that: For the stable load unit, the load dynamic correction based on real-time energy data includes: Based on the real-time energy data, the state change of the stable load unit is tracked to dynamically correct the first prediction result to obtain a second prediction result.

7. The power load forecasting method according to claim 1, characterized in that: For the impact load unit, the load static prediction based on historical energy data includes: Extracting load characteristic parameters associated with the production plan for the forecast period from the historical energy data; Determine the load mean and load fluctuation range of the impact load unit in the impact period and the non-impact period of the normal state according to the load characteristic parameters; Combining the load fluctuation information of a previous cycle adjacent to the current production cycle, based on the determined load mean and load fluctuation range, predicting the load of a single production cycle for the impact load unit; and According to the scheduling plan associated with the forecast period, the load forecast results of multiple production cycles are spliced ​​to obtain a third forecast result.

8. The power load forecasting method according to claim 7, characterized in that: For the impact load unit, the load dynamic correction based on real-time energy data includes: The state change of the impact load unit is tracked based on the real-time energy data to dynamically correct the third prediction result to obtain a fourth prediction result.

9. A power load forecasting device for a metallurgical enterprise, characterized in that: include: a memory configured to store instructions; as well as A processor is configured to call the instructions from the memory and implement the power load forecasting method according to any one of claims 1 to 7 when executing the instructions.

10. A machine-readable storage medium, characterized in that: The machine-readable storage medium stores instructions for enabling a machine to execute the power load forecasting method described in any one of claims 1 to 7.

Images