Bus load prediction method and device, storage medium and electronic device

Abstract

The invention discloses a bus load prediction method and device, a storage medium and an electronic device, and relates to the technical field of power system bus load prediction. The method comprises the following steps: acquiring a target reference curve of a bus load in a first preset time period; fitting the target reference curve to obtain at least one type of fitting curve; determining an initial prediction result based on the at least one type of fitted curve; and performing error adjustment on the initial prediction result based on the weight optimization curve to obtain a target prediction result. According to the invention, the technical problem that the load prediction result cannot be obtained in time due to the fact that a large amount of data needs to be collected and when the bus ultra-short-term load prediction is carried out, the requirement on hardware performance is high and the situation that the data is not collected in time easily occurs in the prior art is solved.

Description

Technical Field

[0001] The present invention relates to the technical field of power system bus load prediction, and in particular to a bus load prediction method, device, storage medium and electronic device. Background Art

[0002] Bus ultra-short-term load forecasting is a crucial part of power system dispatching and operation. It mainly forecasts bus loads in the short term in the future, providing an important basis for real-time control and dispatching of power systems. In the prior art, when conducting bus load forecasting, a large amount of data needs to be collected. When conducting bus ultra-short-term load forecasting, the hardware performance requirements are high, and data collection is prone to untimely, resulting in the inability to obtain load forecast results in a timely manner.

[0003] There is no effective solution to the above problems yet. Summary of the invention

[0004] The embodiments of the present invention provide a bus load forecasting method, device, storage medium and electronic device to at least solve the technical problem that the related technology needs to collect a large amount of data, has high requirements on hardware performance when performing ultra-short-term bus load forecasting, and is prone to untimely data collection, resulting in the inability to obtain load forecasting results in a timely manner.

[0005] According to one embodiment of the present invention, a bus load prediction method is provided, the method comprising: obtaining a target reference curve of the bus load within a first preset time period, wherein the target reference curve is used to reflect the changing trend of the bus load within the first preset time period; fitting the target reference curve to obtain at least one type of fitting curve, wherein different types of fitting curves have different load change rules; determining an initial prediction result based on at least one type of fitting curve, wherein the initial prediction result is used to represent the changing trend of the bus load within a second preset time period, and the second preset time period is after the first preset time period; performing error adjustment on the initial prediction result based on a weight optimization curve to obtain a target prediction result, wherein the weight optimization curve is used to simulate and reflect the changing trend of the bus load at the current moment.

[0006] Optionally, obtaining a target reference curve of the bus load within a first preset time period includes: obtaining historical bus load data within the first preset time period; constructing an initial reference curve of the bus load based on the historical bus load data; and performing data smoothing processing on the initial reference curve based on a preset data smoothing strategy to obtain a target reference curve.

[0007] Optionally, fitting the target reference curve to obtain at least one type of fitting curve includes: determining the similarity between the target reference curves based on a preset similarity measurement strategy; classifying the target reference curves whose similarity is greater than a preset similarity threshold into the same type to obtain at least one type of target reference curve; fitting at least one type of target reference curve to obtain at least one type of fitting curve.

[0008] Optionally, determining the initial prediction result based on at least one type of fitting curve includes: obtaining current bus load data within a third preset time period, wherein the third preset time period is between the first preset time period and the second preset time period; constructing a comparison reference curve based on the current bus load data, wherein the comparison reference curve is used to reflect the changing trend of the bus load within the third preset time period; determining a first fitting curve with the highest similarity to the comparison reference curve among at least one type of fitting curve; and determining the initial prediction result based on the first fitting curve.

[0009] Optionally, error adjustment is performed on the initial prediction result based on the weight optimization curve to obtain a target prediction result, including: determining a first segmentation curve based on the target reference curve, a preset segmentation duration and a target tracing point, and determining a second segmentation curve based on the comparison reference curve, the preset segmentation duration and the target tracing point, wherein the target tracing point is the time point of the initial prediction result; determining a weight optimization curve based on the first segmentation curve and the second segmentation curve; and error adjustment is performed on the initial prediction result based on the weight optimization curve to obtain a target prediction result.

[0010] Optionally, determining a first segmented curve based on a target reference curve, a preset segmentation duration and a target tracing point, and determining a second segmented curve based on a comparison reference curve, a preset segmentation duration and a target tracing point includes: segmenting the target reference curve based on the preset segmentation duration and the target tracing point to obtain a first segmented curve, and segmenting the comparison reference curve based on the preset segmentation duration and the target tracing point to obtain a second segmented curve.

[0011] Optionally, determining the weight optimization curve based on the first segmented curve and the second segmented curve includes: performing a similarity comparison on the first segmented curve and the second segmented curve to obtain a target segmented curve in the first segmented curve, wherein the target segmented curve has the highest similarity to the corresponding segmented curve in the second segmented curve; determining a second fitting curve in at least one type of fitting curve based on the target segmented curve; and determining the weight optimization curve based on the second fitting curve.

[0012] According to one embodiment of the present invention, a bus load prediction device is also provided, the device comprising: an acquisition module, used to acquire a target reference curve of the bus load within a first preset time period, wherein the target reference curve is used to reflect the changing trend of the bus load within the first preset time period; a fitting module, used to fit the target reference curve to obtain at least one type of fitting curve, wherein different types of fitting curves have different load change rules; a determination module, used to determine an initial prediction result based on at least one type of fitting curve, wherein the initial prediction result is used to represent the changing trend of the bus load within a second preset time period, and the second preset time period is after the first preset time period; an adjustment module, used to perform error adjustment on the initial prediction result based on a weight optimization curve to obtain a target prediction result, wherein the weight optimization curve is used to simulate and reflect the changing trend of the bus load at the current moment.

[0013] Optionally, the acquisition module is also used to acquire historical bus load data within a first preset time period; construct an initial reference curve of the bus load based on the historical bus load data; and perform data smoothing on the initial reference curve based on a preset data smoothing strategy to obtain a target reference curve.

[0014] Optionally, the fitting module is also used to determine the similarity between target reference curves based on a preset similarity measurement strategy; classify target reference curves whose similarity is greater than a preset similarity threshold into the same type to obtain at least one type of target reference curve; and perform fitting processing on at least one type of target reference curve to obtain at least one type of fitting curve.

[0015] Optionally, the determination module is also used to obtain current bus load data within a third preset time period, wherein the third preset time period is between the first preset time period and the second preset time period; construct a comparison reference curve based on the current bus load data, wherein the comparison reference curve is used to reflect the changing trend of the bus load within the third preset time period; determine a first fitting curve with the highest similarity to the comparison reference curve among at least one type of fitting curves; and determine an initial prediction result based on the first fitting curve.

[0016] Optionally, the adjustment module is also used to determine a first segmentation curve based on a target reference curve, a preset segmentation duration and a target tracing point, and to determine a second segmentation curve based on a comparison reference curve, a preset segmentation duration and a target tracing point, wherein the target tracing point is the time point of an initial prediction result; determine a weight optimization curve based on the first segmentation curve and the second segmentation curve; perform error adjustment on the initial prediction result based on the weight optimization curve to obtain a target prediction result.

[0017] Optionally, the determination module is also used to segment the target reference curve based on the preset segmentation duration and the target tracing point to obtain a first segmentation curve, and to segment the comparison reference curve based on the preset segmentation duration and the target tracing point to obtain a second segmentation curve.

[0018] Optionally, the determination module is also used to perform a similarity comparison between the first segmented curve and the second segmented curve to obtain a target segmented curve in the first segmented curve, wherein the target segmented curve has the highest similarity to the corresponding segmented curve in the second segmented curve; determine a second fitting curve in at least one type of fitting curve based on the target segmented curve; and determine a weight optimization curve based on the second fitting curve.

[0019] According to one embodiment of the present invention, there is further provided a computer-readable storage medium, in which a computer program is stored, wherein the computer program is configured to execute any of the above bus load prediction methods when running on a computer or a processor.

[0020] According to one embodiment of the present invention, there is also provided an electronic device, including a memory and a processor, wherein a computer program is stored in the memory, and the processor is configured to run the computer program to execute any of the above bus load prediction methods.

[0021] According to one embodiment of the present invention, there is further provided a computer program product, comprising a computer program, which implements any of the above bus load prediction methods when executed by a processor.

[0022] In an embodiment of the present invention, a target reference curve of the bus load within a first preset time period is obtained, wherein the target reference curve is used to reflect the change trend of the bus load within the first preset time period; the target reference curve is fitted to obtain at least one type of fitting curve, wherein different types of fitting curves have different load change laws; an initial prediction result is determined based on at least one type of fitting curve, wherein the initial prediction result is used to represent the change trend of the bus load within a second preset time period, and the second preset time period is after the first preset time period; the initial prediction result is error-adjusted based on a weight optimization curve to obtain a target prediction result, wherein the weight optimization curve is used to simulate the change trend of the bus load at the current moment, thereby achieving the purpose of accurately predicting the ultra-short-term bus load without collecting a large amount of power grid data, thereby achieving the technical effect of reducing the requirements for hardware performance while being able to adapt to a wider range of load predictions, thereby solving the technical problem that the related technology needs to collect a large amount of data, has high requirements for hardware performance when performing ultra-short-term bus load prediction, and is prone to untimely data collection, resulting in the inability to obtain load prediction results in a timely manner. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] The drawings described herein are used to provide a further understanding of the present invention and constitute a part of this application. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention. In the drawings:

[0024] Figure 1 is a flow chart of a bus load prediction method according to one embodiment of the present invention;

[0025] Figure 2 is a flow chart of ultra-short-term bus load forecasting according to one embodiment of the present invention;

[0026] Figure 3 is a structural diagram of a bus ultra-short-term load forecasting system according to one embodiment of the present invention;

[0027] Figure 4 It is a structural block diagram of a bus load prediction device according to one embodiment of the present invention. DETAILED DESCRIPTION

[0028] To facilitate understanding, some descriptions of concepts related to the embodiments of the present invention are exemplarily provided for reference.

[0029] As shown below:

[0030] Busbar: In the power system, busbar refers to a common conductor system connecting multiple electrical devices, which plays the role of distributing and collecting electric energy. Busbar can be a conductor used to connect different circuits and equipment in a substation, or it can be a node used to collect and distribute electric energy in a power grid. Busbar load forecasting is of great significance to the operation and dispatching, load management and safety control of the power system.

[0031] Ultra-Short Term Load Forecasting: This is a power system load forecasting technology used to predict power demand in the next few hours or even minutes. Ultra-short term load forecasting is usually based on the analysis of real-time data and historical data, using statistics, time series analysis, machine learning and other methods to make predictions. It plays a key role in power dispatching, adjustment of generator sets, optimization of power grid operation, etc., and helps to improve the stability and economy of the power system.

[0032] Curve Fitting: In mathematics and statistics, curve fitting is a process of finding the curve that best describes the observed data by performing mathematical calculations on the data. In this application, curve fitting technology is used to process historical load data, and a mathematical model (such as polynomial regression, exponential function, etc.) is established to describe the trend of historical load changes, thereby generating a fitting curve. These fitting curves can help with initial bus ultra-short-term load forecasting and subsequent small trend optimization.

[0033] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings 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 should fall within the scope of protection of the present invention.

[0034] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In the description of this embodiment, unless otherwise specified, the meaning of "multiple" is two or more. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0035] According to one embodiment of the present invention, an embodiment of a bus load forecasting method is provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0036] The method embodiment can be executed in an electronic device including a memory and a processor, a similar control device or a system. Taking an electronic device as an example, the electronic device may include one or more processors and a memory for storing data. Optionally, the electronic device may also include a communication device and a display device for communication functions. It can be understood by a person of ordinary skill in the art that the above structural description is only illustrative and does not limit the structure of the above electronic device. For example, the electronic device may also include more or fewer components than the above structural description, or have a configuration different from the above structural description.

[0037] The processor may include one or more processing units. For example, the processor may include a central processing unit (CPU), a graphics processing unit (GPU), a digital signal processing (DSP) chip, a microcontroller unit (MCU), a field-programmable gate array (FPGA), a neural network processor (NPU), a tensor processing unit (TPU), an artificial intelligence (AI) type processor, and the like. Among them, different processing units may be independent components or integrated into one or more processors. In some instances, the electronic device may also include one or more processors.

[0038] The memory can be used to store computer programs, such as storing a computer program corresponding to the bus load prediction method in the embodiment of the present invention. The processor implements the above-mentioned bus load prediction method by running the computer program stored in the memory. The memory may include a high-speed random access memory, and may also include a non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some examples, the memory may further include a memory remotely arranged relative to the processor, and these remote memories can be connected to the electronic device via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0039] The communication device is used to receive or send data via a network. The specific example of the above network may include a wireless network provided by a communication provider of the mobile terminal. In one example, the communication device includes a network adapter (network interface controller, NIC), which can be connected to other network devices through a base station so as to communicate with the Internet. In one example, the communication device can be a radio frequency (RF) module, which is used to communicate with the Internet wirelessly.

[0040] The display device may be, for example, a touch screen type liquid crystal display (LCD) and a touch display (also referred to as a "touch screen" or "touch display screen"). The liquid crystal display may enable a user to interact with a user interface of a mobile terminal. In some embodiments, the mobile terminal has a graphical user interface (GUI), and a user may interact with the GUI by finger contact and / or gestures on a touch-sensitive surface, wherein the human-computer interaction functions here may optionally include the following interactions: creating web pages, drawing, word processing, making electronic documents, games, video conferencing, instant messaging, sending and receiving emails, call interfaces, playing digital videos, playing digital music and / or web browsing, etc. The executable instructions for executing the above human-computer interaction functions are configured / stored in a computer program product or a readable storage medium executable by one or more processors.

[0041] In this embodiment, a bus load prediction method running on an electronic device is provided. Figure 1 is a flow chart of a bus load prediction method according to one embodiment of the present invention. Figure 1 As shown, the process includes the following steps:

[0042] Step S10, obtaining a target reference curve of the bus load within a first preset time period, wherein the target reference curve is used to reflect a change trend of the bus load within the first preset time period;

[0043] In an embodiment of the present invention, the first preset time can be understood as a specific time range selected in the bus ultra-short-term load forecast, which is used to analyze and predict the changing trend of the bus load. Exemplarily, the first preset time can be a historical single day, which is not limited here.

[0044] Bus load can be understood as the electrical energy demand or consumption carried by the bus node.

[0045] The target reference curve can be understood as a curve generated by analyzing historical bus load data within a first preset time period, and is used to reflect the typical change trend of the bus load within the time period.

[0046] Obtaining the target reference curve of the bus load in the first preset time period can be understood as a curve fitted by using data processing and analysis technology by analyzing the historical data of the bus load in the first preset time period when performing ultra-short-term bus load forecasting.

[0047] In an embodiment of the present invention, by obtaining a target reference curve, a standard or baseline can be provided for ultra-short-term bus load forecasting, so that the forecasting process is not only based on current real-time data, but also takes into account historical load change patterns, thereby improving the accuracy and reliability of the forecast.

[0048] Step S11, performing fitting processing on the target reference curve to obtain at least one type of fitting curve, wherein different types of fitting curves have different load variation rules;

[0049] In the embodiment of the present invention, the fitting process can be understood as the process performed on the target reference curve, the purpose of which is to describe the variation pattern of the bus load within the first preset time period through a mathematical model.

[0050] At least one type of fitting curve can be understood as one or more groups of fitting curves obtained by analyzing and fitting historical bus load data, each group of curves representing the variation pattern of bus load under specific conditions.

[0051] The load variation law can be understood as the pattern or trend of bus load changes over time.

[0052] Fitting the target reference curve to obtain at least one type of fitting curve can be understood as mathematical modeling of typical historical load data to capture and quantify different modes or trends of load changes.

[0053] The different load variation patterns of different types of fitting curves can be understood as the ability to identify and quantify a variety of different patterns or trends of bus load variation over time through in-depth analysis of historical load data and model fitting.

[0054] In an embodiment of the present invention, by fitting the target reference curve into different types of fitting curves, a more flexible and comprehensive bus ultra-short-term load forecasting model is constructed, so as to more accurately predict load changes, reduce prediction errors, and improve the efficiency and safety of power system scheduling and operation.

[0055] Step S12, determining an initial prediction result based on at least one type of fitting curve, wherein the initial prediction result is used to represent a change trend of the bus load within a second preset time period, and the second preset time period is after the first preset time period;

[0056] In the embodiment of the present invention, the initial prediction result can be understood as a preliminary prediction value of the bus load obtained after prediction based on the selected fitting curve.

[0057] The second preset time can be understood as a shorter time interval after the first preset time period, which is used for ultra-short-term load forecasting. For example, the second preset time period may be the next 5 minutes to 1 hour, which is not limited here.

[0058] Determining the initial prediction result based on at least one type of fitting curve can be understood as using fitting curves of different modes extracted from historical load data to preliminarily predict the changing trend of future bus loads.

[0059] The initial prediction result is used to represent the changing trend of the bus load within the second preset time period. It can be understood as a preliminary load forecast value generated by the prediction system based on the analysis of historical load data and model fitting, which is used to describe and predict the dynamic changes of the bus load within the second preset time period.

[0060] In the embodiment of the present invention, based on the selected fitting curve, the prediction system can generate a preliminary prediction of the bus load within the second preset time period, that is, an initial prediction result. By fitting the historical load data into different types of curves, and then selecting the most matching curve for prediction according to the current load trend, the reliance on a large amount of real-time data collection is reduced, and the hardware performance requirements are reduced.

[0061] Step S13, performing error adjustment on the initial prediction result based on the weight optimization curve to obtain the target prediction result, wherein the weight optimization curve is used to simulate the change trend of the bus load at the current moment.

[0062] In the embodiment of the present invention, the weight optimization curve can be understood as a curve used to reflect the changing trend of the bus load at the current time and its nearby time points. Exemplarily, the weight optimization curve can be calculated by the Pearson correlation coefficient method, which is not limited here.

[0063] The target prediction result can be understood as the final prediction value after optimization and adjustment based on the initial prediction result.

[0064] Based on the weight optimization curve, the initial prediction result is adjusted for error to obtain the target prediction result. It can be understood that after obtaining the initial prediction result, the prediction system will use the weight optimization curve to evaluate the error range of the initial prediction and fine-tune the initial prediction result by calculating the error protection range, so as to obtain the final prediction value after error adjustment, that is, the target prediction result.

[0065] The weight optimization curve is used to simulate the changing trend of the bus load at the current moment. It can be understood that the weight optimization curve is generated by a specific mathematical model and algorithm after the prediction system analyzes the instant trend of the bus load at the current moment and compares it with the reference curve of similar trend in historical data. It aims to capture and simulate the specific changes of the bus load at the current moment, including the increase, decrease or stable state of the load. That is, the weight optimization curve is essentially a reference curve selected from historical data that is most similar to the current instant trend curve, which is not restricted here.

[0066] In the embodiment of the present invention, the error of the initial prediction result is adjusted based on the weight optimization curve to obtain the target prediction result, aiming to improve the accuracy and real-time performance of the ultra-short-term load prediction of the bus. The weight optimization curve provides real-time data correction for the prediction model by simulating the load change trend at the current moment, making the prediction result closer to the actual load situation.

[0067] By obtaining the target reference curve, a standard or baseline can be provided for the ultra-short-term load forecasting of the bus, so that the forecasting process is not only based on the current real-time data, but also takes into account the pattern of historical load changes, thereby improving the accuracy and reliability of the forecast. By fitting the target reference curve into different types of fitting curves, it is intended to build a more flexible and comprehensive ultra-short-term load forecasting model for the bus, so as to more accurately predict load changes, reduce prediction errors, and improve the efficiency and safety of power system dispatching and operation. Based on the selected fitting curve, the forecasting system can generate a preliminary forecast of the bus load in the second preset time period, that is, the initial forecast result. By fitting the historical load data into different types of curves, and then selecting the most matching curve for forecasting according to the current load trend, the dependence on a large amount of real-time data collection is reduced, and the hardware performance requirements are reduced. The error adjustment of the initial forecast result is performed based on the weight optimization curve to obtain the target forecast result, aiming to improve the accuracy and real-time performance of the ultra-short-term load forecasting of the bus. The weight optimization curve provides real-time data correction for the forecasting model by simulating the load change trend at the current moment, so that the forecast result is closer to the actual load situation.

[0068] Through the above steps, by obtaining the target reference curve of the bus load in the first preset time period, wherein the target reference curve is used to reflect the change trend of the bus load in the first preset time period; fitting processing is performed on the target reference curve to obtain at least one type of fitting curve, wherein different types of fitting curves have different load change laws; an initial prediction result is determined based on at least one type of fitting curve, wherein the initial prediction result is used to represent the change trend of the bus load in the second preset time period, and the second preset time period is after the first preset time period; error adjustment is performed on the initial prediction result based on the weight optimization curve to obtain the target prediction result, wherein the weight optimization curve is used to simulate the change trend of the bus load at the current moment, thereby achieving the purpose of accurately predicting the ultra-short-term load of the bus without collecting a large amount of power grid data, thereby achieving the technical effect of reducing the requirements for hardware performance while being able to adapt to a wider range of load predictions, thereby solving the technical problem that the related technology needs to collect a large amount of data, has high requirements for hardware performance when predicting the ultra-short-term load of the bus, and is prone to untimely data collection, resulting in the inability to obtain the load prediction results in time.

[0069] Optionally, in step S10, obtaining a target reference curve of bus load within a first preset time period may include the following execution steps:

[0070] Step S101, obtaining historical bus load data within a first preset time period;

[0071] Step S102, constructing an initial reference curve of bus load based on historical bus load data;

[0072] Step S103: performing data smoothing processing on the initial reference curve based on a preset data smoothing strategy to obtain a target reference curve.

[0073] In the embodiment of the present invention, the historical bus load data can be understood as a collection of all load data recorded on the bus within the first preset time period. Exemplarily, the historical bus load data generally includes a detailed record of load changes over time, such as load values ​​every 5 minutes, 10 minutes or 1 hour, which is not limited here.

[0074] The initial reference curve can be understood as a trend line constructed based on historical bus load data, which is used to summarize the basic change trend of the bus load within the first preset time period.

[0075] The preset data smoothing strategy can be understood as a statistical or mathematical method for reducing data noise and highlighting data trends. For example, the preset data smoothing strategy means that a specific data processing algorithm has been set in the forecasting system, such as the moving average method, exponential smoothing method or Lorenz curve smoothing, etc., which is used to process historical load data to make it smoother and reduce the impact of short-term fluctuations and outliers, which is not limited here.

[0076] Data smoothing can be understood as the process of operating the initial reference curve using a preset data smoothing strategy, that is, by removing or reducing short-term fluctuations and noise on the initial reference curve, the curve is made smoother, thereby more clearly presenting the load change pattern, which is not limited here.

[0077] Obtaining historical bus load data within the first preset time period can be understood as obtaining a collection of load information collected from the bus within the first preset time period. Exemplarily, historical bus load data generally includes the load values ​​of the bus at different time points (such as every 5 minutes, every hour, etc.), which is not limited here.

[0078] Constructing the initial reference curve of bus load based on historical bus load data can be understood as constructing a curve that can reflect the basic trend of load change by analyzing and summarizing historical bus load data.

[0079] Performing data smoothing on the initial reference curve based on a preset data smoothing strategy to obtain a target reference curve can be understood as a process of applying the preset data smoothing strategy to process the initial reference curve to obtain a smoother target reference curve.

[0080] In the embodiment of the present invention, the accuracy, real-time and flexibility of bus ultra-short-term load forecasting can be significantly improved by acquiring historical bus load data, constructing an initial reference curve and applying a data smoothing strategy to generate a target reference curve.

[0081] Optionally, in step S11, performing fitting processing on the target reference curve to obtain at least one type of fitting curve may include the following execution steps:

[0082] Step S111, determining the similarity between target reference curves based on a preset similarity measurement strategy;

[0083] Step S112, classifying target reference curves whose similarity is greater than a preset similarity threshold into the same type, and obtaining at least one type of target reference curve;

[0084] Step S113: performing fitting processing on at least one type of target reference curve to obtain at least one type of fitting curve.

[0085] In the embodiment of the present invention, the preset similarity measurement strategy can be understood as a mathematical or statistical method for quantifying the similarity between two target reference curves. Exemplarily, the preset similarity measurement strategy can include cosine similarity, Pearson correlation coefficient, Euclidean distance or Manhattan distance, etc., which are not limited here.

[0086] The similarity between the target reference curves can be understood as the similarity between two or more target reference curves in load trends calculated by a preset similarity measurement strategy.

[0087] The preset similarity threshold can be understood as a pre-set value, which is used to classify whether the target reference curves belong to the same type.

[0088] Determining the similarity between target reference curves based on a preset similarity measurement strategy can be understood as that after selecting a similarity measurement method, the system will compare the target reference curves based on this strategy, thereby calculating the similarity between the target reference curves.

[0089] Classifying the target reference curves whose similarity is greater than a preset similarity threshold into the same type, and obtaining at least one type of target reference curve can be understood as calculating the similarity between the target reference curves and comparing these similarities with the preset threshold, and the prediction system will identify which curves have similarities exceeding the set threshold. All curves whose similarity exceeds the threshold will be classified into the same type.

[0090] Performing fitting processing on at least one type of target reference curve to obtain at least one type of fitting curve can be understood as that for each type of target reference curve, the prediction system will perform fitting processing to obtain at least one type of fitting curve.

[0091] In the embodiment of the present invention, the similarity between target reference curves is determined based on a preset similarity measurement strategy, and curves with similarity greater than a preset threshold are classified into the same type and fitted, which can significantly reduce the amount of data used for model training and prediction. The similarity measurement and classification steps ensure that only the data most relevant to the prediction is processed, thereby reducing the demand for computing resources.

[0092] Optionally, in step S12, determining the initial prediction result based on at least one type of fitting curve may include the following execution steps:

[0093] Step S121, obtaining current bus load data within a third preset time period, wherein the third preset time period is between the first preset time period and the second preset time period;

[0094] Step S122, constructing a comparison reference curve based on the current bus load data, wherein the comparison reference curve is used to reflect the change trend of the bus load within a third preset time period;

[0095] Step S123, determining a first fitting curve having the highest similarity to the comparison reference curve among at least one type of fitting curves;

[0096] Step S124: determining an initial prediction result based on the first fitting curve.

[0097] In the embodiment of the present invention, the third preset time period can be understood as a time interval defined in the prediction process, which is used for real-time collection and analysis of current bus load data.

[0098] The comparison reference curve can be understood as a curve constructed based on the current bus load data collected in real time within the third preset time period through data smoothing, trend extraction and other methods, which is used to reflect the changing trend of the current bus load within this time period.

[0099] The first fitting curve can be understood as a curve having the highest similarity to the comparison reference curve, that is, a type that is closest to the current bus load change trend within the third preset time period.

[0100] Acquiring the current bus load data within the third preset time period can be understood as collecting the bus load data in real time from the data collection point of the power system within the third preset time period. Exemplarily, the current bus load data may be located at multiple locations on the bus, including but not limited to power transmission lines, substations, and power plants, etc., which are not limited here.

[0101] Constructing a comparative reference curve based on the current bus load data can be understood as connecting the processed and analyzed instantaneous load data points to form a comparative reference curve. Exemplarily, the curve reflects the changing trend of the bus load in the third preset time period, including the peak value, valley value, fluctuation frequency and amplitude of the load, which are not limited here.

[0102] The comparison reference curve is used to reflect the change trend of the bus load within the third preset time period. It can be understood that the comparison reference curve is constructed based on the instant bus load data within the third preset time period, reflecting the instant load change trend.

[0103] Determining the first fitting curve with the highest similarity to the comparison reference curve among at least one type of fitting curve can be understood as that among all types of fitting curves, the prediction system will identify a curve with the highest similarity to the comparison reference curve, that is, the selection of the first fitting curve is based on its most similar match to the real-time data trend.

[0104] Determining the initial prediction result based on the first fitting curve can be understood as comparing the comparison reference curve constructed using the current bus load data with these historical fitting curves during the instantaneous prediction, finding the first fitting curve with the highest similarity to the comparison reference curve, and determining the initial prediction result based on the pattern and trend provided by the first fitting curve.

[0105] In an embodiment of the present invention, by acquiring the current bus load data within a third preset time period and constructing a comparison reference curve, determining a first fitting curve with the highest similarity to the comparison reference curve, and then determining an initial prediction result based on the first fitting curve, the real-time and accuracy of the prediction can be significantly enhanced, the demand for hardware resources can be reduced, and the flexibility and efficiency of power system scheduling and operation can be improved.

[0106] Optionally, in step S13, performing error adjustment on the initial prediction result based on the weight optimization curve to obtain the target prediction result may include the following execution steps:

[0107] Step S131, determining a first segmentation curve based on a target reference curve, a preset segmentation duration, and a target tracing point, and determining a second segmentation curve based on a comparison reference curve, a preset segmentation duration, and a target tracing point, wherein the target tracing point is a time point of an initial prediction result;

[0108] Step S132, determining a weight optimization curve based on the first piecewise curve and the second piecewise curve;

[0109] Step S133, performing error adjustment on the initial prediction result based on the weight optimization curve to obtain the target prediction result.

[0110] In the embodiment of the present invention, the preset segmentation time length can be understood as a fixed time length for dividing the time series data into a series of smaller time segments in order to facilitate the analysis of load change trends.

[0111] The target tracing point can be understood as the current time point used as a starting point or reference point when predicting future loads. For example, the number of target tracing points can be set to 3-5, which is not limited here.

[0112] The first segmented curve can be understood as being determined based on the target reference curve, the preset segment duration and the target tracing point, and represents the bus load change trend within a fixed time length before the target tracing point.

[0113] The second segmented curve can be understood as being determined based on the comparison reference curve, the same preset segment duration and the target tracing point, reflecting the instantaneous bus load changes within the same time length before the target tracing point, that is, the current actual load trend.

[0114] Determining the first segmented curve based on the target reference curve, the preset segmented duration and the target traceability point, and determining the second segmented curve based on the comparison reference curve, the preset segmented duration and the target traceability point can be understood as extracting historical load data based on the target reference curve within the preset segmented duration before the target traceability point, and constructing a curve reflecting the change trend of the historical pattern within the time period, i.e., the first segmented curve. Based on the comparison reference curve, within the same preset segmented duration as the first segmented curve, a curve reflecting the change trend of the immediate load is constructed, i.e., the second segmented curve.

[0115] Determining the weight optimization curve based on the first segmented curve and the second segmented curve can be understood as determining a weight allocation scheme based on the comparison result of the first segmented curve and the second segmented curve for subsequent prediction optimization. Exemplarily, the weight optimization curve is actually a scheme based on the comparison result of the two segmented curves to guide how to balance between historical patterns and immediate changes, which is not limited here.

[0116] Based on the weight optimization curve, the initial prediction result is adjusted for error to obtain the target prediction result. It can be understood that based on the weight optimization curve, the prediction system will implement an error adjustment strategy, which may include adjusting the size of the prediction value, correcting the shape of the prediction curve, or increasing the confidence interval of the prediction result, to generate the final target prediction result, which is not limited here.

[0117] In an embodiment of the present invention, a first segmented curve is determined based on a target reference curve, a preset segmented duration and a target traceability point, and a second segmented curve is determined based on a comparison reference curve, a preset segmented duration and a target traceability point, and then a weight optimization curve is determined based on the first segmented curve and the second segmented curve, and then an error adjustment is performed on the initial prediction result to obtain a target prediction result, which not only improves the prediction accuracy and real-time performance, but also reduces the demand for data collection and enhances the robustness of the prediction system.

[0118] Optionally, in step S131, determining the first segment curve based on the target reference curve, the preset segment duration and the target traceability point, and determining the second segment curve based on the comparison reference curve, the preset segment duration and the target traceability point may include the following execution steps:

[0119] Step S1311, segmenting the target reference curve based on the preset segmentation time and the target tracing point to obtain a first segmentation curve, and segmenting the comparison reference curve based on the preset segmentation time and the target tracing point to obtain a second segmentation curve.

[0120] In an embodiment of the present invention, the target reference curve is segmented based on the preset segmentation time and the target tracing point, and the first segmented curve can be understood as taking the target tracing point as the starting point, tracing back the preset segmentation time, and determining a historical record time window for analysis. From the target reference curve, historical load data within a specific time period is extracted according to the above time window and used to construct the first segmented curve.

[0121] The comparison reference curve is segmented based on the preset segmentation time and the target tracing point to obtain the second segmentation curve, which can be understood as tracing back the preset segmentation time based on the target tracing point, determining a time interval for analyzing the instantaneous load changes, and extracting the instantaneous load data within a specific time period from the comparison reference curve according to the above time interval, and using the extracted instantaneous load data to construct the second segmentation curve.

[0122] In an embodiment of the present invention, the target reference curve and the comparison reference curve are segmented based on the preset segmentation duration and the target tracing point to obtain a first segmented curve and a second segmented curve, which not only enhances the real-time and adaptability of the prediction, but also improves the prediction accuracy, reduces the demand for computing resources, and promotes the adaptive optimization of the prediction model.

[0123] Optionally, in step S132, determining the weight optimization curve based on the first piecewise curve and the second piecewise curve may include the following execution steps:

[0124] Step S1321, performing a similarity comparison on the first segmented curve and the second segmented curve to obtain a target segmented curve in the first segmented curve, wherein the target segmented curve has the highest similarity with the corresponding segmented curve in the second segmented curve;

[0125] Step S1322, determining a second fitting curve in at least one type of fitting curve based on the target segmented curve;

[0126] Step S1323: determining a weight optimization curve based on the second fitting curve.

[0127] In the embodiment of the present invention, the target segmented curve may be understood as a specific curve segment in the first segmented curve that has the highest similarity with the corresponding segmented curve in the second segmented curve.

[0128] The second fitting curve may be understood as a curve in a set of at least one type of fitting curves that best matches the local trend of the target segmented curve.

[0129] Comparing the first segmented curve with the second segmented curve to obtain the target segmented curve in the first segmented curve can be understood as finding the curve segment that best matches the current instantaneous load change trend from the first segmented curve set by performing a similarity comparison with the second segmented curve.

[0130] The second fitting curve in at least one type of fitting curve determined based on the target segmented curve can be understood as a mathematical model curve selected from the fitting curve types associated with the target segmented curve and best matching the local load variation trend of the target segmented curve.

[0131] Determining the weight optimization curve based on the second fitting curve can be understood as the weight optimization curve being generated based on a comprehensive analysis of the second fitting curve and real-time data.

[0132] In an embodiment of the present invention, by performing a similarity comparison between the first segmented curve and the second segmented curve, a target segmented curve in the first segmented curve is obtained, and a second fitting curve in at least one type of fitting curve is determined based on the target segmented curve and a weight optimization curve is determined, thereby enabling accurate diagnosis and rapid response to instant load changes, and providing accurate, real-time, and highly adaptable prediction results.

[0133] Figure 2 is a flow chart of ultra-short-term bus load forecasting according to one embodiment of the present invention, such as Figure 2 As shown, the ultra-short-term bus load forecasting process mainly includes obtaining the trend curve of bus load changes, obtaining the fitting curve, initial bus load forecasting, small trend optimization and outputting the final ultra-short-term bus load forecasting results. Figure 3 is a structural diagram of a bus ultra-short-term load forecasting system according to one embodiment of the present invention. Figure 3 As shown in the figure, the bus ultra-short-term load forecasting method mainly includes: large trend forecasting and small trend optimization. The bus ultra-short-term load forecasting system consists of a historical load trend analysis module, a curve fitting module, an initial load forecasting module and a forecasting optimization module. Among them, the historical load trend analysis module is responsible for obtaining the reference curve corresponding to the historical single-day bus load; the curve fitting module is responsible for fitting the reference curve to obtain different types of fitting curves; the initial load forecasting module is responsible for making an initial forecast of the bus ultra-short-term load through the fitting curve according to the current bus load change trend; the forecasting optimization module is responsible for adding an error range value to the initial forecasted load to ensure the accuracy of the forecasting result.

[0134] Combination Figure 2 and Figure 3The present invention is further explained that the macro trend prediction in the bus ultra-short-term load prediction method mainly includes: (1) Obtaining the trend curve of bus load change: using data smoothing technology (such as moving average, exponential smoothing, etc.) to reduce noise, highlight the trend of load change, and extract the trend curve of bus load change every day from historical data. These curves reflect the law of load change over time, and obtain the bus load trend curve of each day. These curves are used as references for subsequent analysis and are recorded as reference curves. (2) Obtaining fitting curves: using similarity measurement methods (such as cosine similarity, dynamic time warping, etc.) to calculate the similarity between reference curves of different days, set a similarity threshold, and classify reference curves with similarity higher than the threshold into the same type, and then perform fitting processing on the reference curves of the same type (such as polynomial regression, time series analysis and other models) to obtain the fitting curve of this type. Among them, each type represents a specific load change law. (3) Initial bus load prediction: Collect bus load data in real time, construct an immediate trend curve, determine which types of fitting curves the immediate trend curve is most similar to, select the fitting curve with the highest similarity, and predict the bus load in the future based on the trend changes of the fitting curve to obtain the initial prediction result.

[0135] The small trend optimization in the bus ultra-short-term load forecasting method mainly includes: defining the ultra-short-term duration as the segment duration, segmenting the trend curve to obtain several segmented small trend curves; determining the time point of the initial prediction result as the tracing point; extracting a set number (such as 3-5) of small trend curves before the tracing point in different reference curves; comparing the similarity of these small trend curves with the corresponding small trend curves in the immediate trend curve; finding the reference curve corresponding to the fitting curve with the highest similarity, and using it as the weight optimization curve; using the weight optimization curve to adjust and optimize the initial prediction result to obtain the final ultra-short-term bus load forecasting result.

[0136] The similarity threshold is manually set, and the cosine similarity comparison method is used, that is, the similarity of the two reference curves is calculated using formula (1):

[0137]

[0138] In formula (1), A is the vector corresponding to a point on a reference curve, and B is the vector corresponding to a point on another reference curve. i is the corresponding element of vector A, B i is the corresponding element of vector B, n is the length of the vector, and C is the cosine similarity.

[0139] At this time, when the similarity threshold is set to 0.9-1.0, two reference curves whose cosine similarity calculation results are within the range of the similarity threshold can be classified as the same type.

[0140] The process of fitting a reference curve of the same type using polynomial regression:

[0141] For the reference curve coordinates (x i ,y i ), where i = 1, 2, ..., n, and a p-order polynomial, namely formula (2), is used to fit these coordinate data:

[0142] y i =β0+β1x1+β2x2 2 +…+β ρ x i ρ +∈ i Formula (2)

[0143]

[0144] In formula (2) and formula (3), ∈ i is the error term, R is the residual sum of squares, β ρ As coefficient, obtain the coefficient β that minimizes the residual sum of squares R ρ .

[0145] When comparing the small trend curve with the corresponding small trend curve in the immediate trend curve, the Pearson coefficient method is used to calculate the similarity, as shown in formula (4):

[0146]

[0147] In the formula, x i is the coordinate of the time point corresponding to an endpoint of the small trend curve in the real-time trend curve. is the mean value of the coordinates of all the endpoints of the corresponding small trend curve in the instant trend curve at the corresponding time points, y i is the coordinate of the time point corresponding to an endpoint of a small trend curve in a reference curve of the same type, It is the mean of the coordinates of all the endpoints of a small trend curve of a reference curve of the same type at corresponding time points, P∈[-1,1], which is an indicator of the degree of linear relationship between the two reference curves.

[0148] According to the indicator calculation results, the calculation result closest to 1 is selected as the reference curve with the highest similarity, and it is used as the weight optimization curve. The weight optimization curve is used to adjust and optimize the initial prediction results. The process is to convert the indicator calculation result corresponding to the weight optimization curve into the error protection range, that is, according to the indicator calculation result Z of the weight optimization curve, the error protection range (1-Z)*100% is obtained, and then added to the initial bus load prediction result to obtain the final ultra-short-term bus load prediction result.

[0149] In the embodiment of the present application, historical load data of the bus is collected to obtain a reference curve, and the fitting of the same type of reference curve is facilitated by similarity comparison and time classification of the reference curve. As long as the instantaneous load data of the bus is collected, the instantaneous trend curve is generated and compared with the fitting curve, the initial prediction of the ultra-short-term load of the bus can be achieved through the change law of the fitting curve. In order to improve the accuracy of the prediction, an error range value is added to the initial prediction through small trend optimization to ensure the accuracy of the ultra-short-term load prediction of the bus.

[0150] Through the description of the above implementation methods, those skilled in the art can clearly understand that the method according to the above embodiment can be implemented by means of software plus a necessary general hardware platform, and of course can also be implemented by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present invention, or the part that contributes to the prior art, can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, a magnetic disk, or an optical disk), and includes a number of instructions for enabling a terminal device (which can be a mobile phone, a computer, a server, or a network device, etc.) to execute the methods described in each embodiment of the present invention.

[0151] In this embodiment, a bus load prediction device is also provided, which is used to implement the above-mentioned embodiments and preferred implementation modes, and the descriptions that have been made will not be repeated. As used below, the term "module" can be a combination of software and / or hardware that implements a predetermined function. Although the devices described in the following embodiments are preferably implemented in software, the implementation of hardware, or a combination of software and hardware, is also possible and conceivable.

[0152] Figure 4 is a structural block diagram of a bus load prediction device according to one embodiment of the present invention. Figure 4 As shown, a bus load prediction device 400 is taken as an example, and the device includes: an acquisition module 401, which is used to acquire a target reference curve of the bus load within a first preset time period, wherein the target reference curve is used to reflect the change trend of the bus load within the first preset time period; a fitting module 402, which is used to fit the target reference curve to obtain at least one type of fitting curve, wherein different types of fitting curves have different load change laws; a determination module 403, which is used to determine an initial prediction result based on at least one type of fitting curve, wherein the initial prediction result is used to represent the change trend of the bus load within a second preset time period, and the second preset time period is after the first preset time period; an adjustment module 404, which is used to perform error adjustment on the initial prediction result based on a weight optimization curve to obtain a target prediction result, wherein the weight optimization curve is used to simulate the change trend of the bus load at the current moment.

[0153] Optionally, the acquisition module 401 is also used to obtain historical bus load data within a first preset time period; construct an initial reference curve of the bus load based on the historical bus load data; and perform data smoothing on the initial reference curve based on a preset data smoothing strategy to obtain a target reference curve.

[0154] Optionally, the fitting module 402 is also used to determine the similarity between target reference curves based on a preset similarity measurement strategy; classify target reference curves whose similarity is greater than a preset similarity threshold into the same type to obtain at least one type of target reference curve; and perform fitting processing on at least one type of target reference curve to obtain at least one type of fitting curve.

[0155] Optionally, the determination module 403 is also used to obtain current bus load data within a third preset time period, wherein the third preset time period is between the first preset time period and the second preset time period; construct a comparison reference curve based on the current bus load data, wherein the comparison reference curve is used to reflect the changing trend of the bus load within the third preset time period; determine a first fitting curve with the highest similarity to the comparison reference curve among at least one type of fitting curves; and determine an initial prediction result based on the first fitting curve.

[0156] Optionally, the adjustment module 404 is also used to determine a first segmented curve based on a target reference curve, a preset segmented duration and a target traceability point, and to determine a second segmented curve based on a comparison reference curve, a preset segmented duration and a target traceability point, wherein the target traceability point is the time point of an initial prediction result; determine a weight optimization curve based on the first segmented curve and the second segmented curve; and perform error adjustment on the initial prediction result based on the weight optimization curve to obtain a target prediction result.

[0157] Optionally, the determination module 403 is also used to segment the target reference curve based on the preset segmentation duration and the target tracing point to obtain a first segmentation curve, and to segment the comparison reference curve based on the preset segmentation duration and the target tracing point to obtain a second segmentation curve.

[0158] Optionally, the determination module 403 is also used to perform a similarity comparison between the first segmented curve and the second segmented curve to obtain a target segmented curve in the first segmented curve, wherein the target segmented curve has the highest similarity to the corresponding segmented curve in the second segmented curve; determine a second fitting curve in at least one type of fitting curve based on the target segmented curve; and determine a weight optimization curve based on the second fitting curve.

[0159] It should be noted that the above modules can be implemented by software or hardware. For the latter, it can be implemented in the following ways, but not limited to: the above modules are all located in the same processor; or the above modules are located in different processors in any combination.

[0160] An embodiment of the present invention further provides a computer-readable storage medium, in which a computer program is stored. The computer program is configured to execute the steps of any of the above method embodiments when running on a computer or a processor.

[0161] Optionally, in this embodiment, the computer-readable storage medium may be configured to store a computer program for performing the following steps:

[0162] Step S10, obtaining a target reference curve of the bus load within a first preset time period, wherein the target reference curve is used to reflect a change trend of the bus load within the first preset time period;

[0163] Step S11, performing fitting processing on the target reference curve to obtain at least one type of fitting curve, wherein different types of fitting curves have different load variation rules;

[0164] Step S12, determining an initial prediction result based on at least one type of fitting curve, wherein the initial prediction result is used to represent a change trend of the bus load within a second preset time period, and the second preset time period is after the first preset time period;

[0165] Step S13, performing error adjustment on the initial prediction result based on the weight optimization curve to obtain the target prediction result, wherein the weight optimization curve is used to simulate the change trend of the bus load at the current moment.

[0166] Optionally, in this embodiment, the above-mentioned computer-readable storage medium may include but is not limited to: a USB flash drive, a read-only memory (ROM), a random access memory (RAM), a mobile hard disk, a magnetic disk or an optical disk, and other media that can store computer programs.

[0167] An embodiment of the present invention further provides an electronic device, including a memory and a processor, wherein a computer program is stored in the memory, and the processor is configured to run the computer program to execute the steps in any one of the above method embodiments.

[0168] Optionally, in this embodiment, the processor in the electronic device may be configured to run a computer program to perform the following steps:

[0169] Step S10, obtaining a target reference curve of the bus load within a first preset time period, wherein the target reference curve is used to reflect a change trend of the bus load within the first preset time period;

[0170] Step S11, performing fitting processing on the target reference curve to obtain at least one type of fitting curve, wherein different types of fitting curves have different load variation rules;

[0171] Step S12, determining an initial prediction result based on at least one type of fitting curve, wherein the initial prediction result is used to represent a change trend of the bus load within a second preset time period, and the second preset time period is after the first preset time period;

[0172] Step S13, performing error adjustment on the initial prediction result based on the weight optimization curve to obtain the target prediction result, wherein the weight optimization curve is used to simulate the change trend of the bus load at the current moment.

[0173] An embodiment of the present invention further provides a computer program product, including a computer program, which implements the steps in any of the above method embodiments when executed by a processor.

[0174] Optionally, in this embodiment, the computer program in the above computer program product may be configured to perform the following steps when executed by a processor:

[0175] Step S10, obtaining a target reference curve of the bus load within a first preset time period, wherein the target reference curve is used to reflect a change trend of the bus load within the first preset time period;

[0176] Step S11, performing fitting processing on the target reference curve to obtain at least one type of fitting curve, wherein different types of fitting curves have different load variation rules;

[0177] Step S12, determining an initial prediction result based on at least one type of fitting curve, wherein the initial prediction result is used to represent a change trend of the bus load within a second preset time period, and the second preset time period is after the first preset time period;

[0178] Step S13, performing error adjustment on the initial prediction result based on the weight optimization curve to obtain the target prediction result, wherein the weight optimization curve is used to simulate the change trend of the bus load at the current moment.

[0179] Optionally, the specific examples in this embodiment may refer to the examples described in the above embodiments and optional implementation modes, and this embodiment will not be described in detail here.

[0180] The serial numbers of the above embodiments of the present invention are only for description and do not represent the advantages or disadvantages of the embodiments.

[0181] In the above embodiments of the present invention, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0182] In the several embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways. Among them, the device embodiments described above are only schematic. For example, the division of the units can be a logical function division. There may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of units or modules, which can be electrical or other forms.

[0183] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple units. Some or all of the units may be selected according to actual needs to achieve the purpose of the present embodiment.

[0184] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0185] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions for a computer device (which can be a personal computer, a server or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, read-only memory (ROM), random access memory (RAM), mobile hard disk, magnetic disk or optical disk, etc., various media that can store program codes.

[0186] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principle of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A bus load prediction method, characterized in that: The method comprises: Acquire a target reference curve of the bus load within a first preset time period, wherein the target reference curve is used to reflect a change trend of the bus load within the first preset time period; Performing fitting processing on the target reference curve to obtain at least one type of fitting curve, wherein different types of fitting curves have different load variation rules; Determining an initial prediction result based on the at least one type of fitting curve, wherein the initial prediction result is used to represent a change trend of the bus load within a second preset time period, and the second preset time period is after the first preset time period; The initial prediction result is error-adjusted based on the weight optimization curve to obtain a target prediction result, wherein the weight optimization curve is used to simulate the change trend of the bus load at the current moment.

2. The method according to claim 1, characterized in that The step of obtaining a target reference curve of bus load within a first preset time period includes: Acquire historical bus load data within the first preset time period; Constructing an initial reference curve of the bus load based on the historical bus load data; The initial reference curve is subjected to data smoothing processing based on a preset data smoothing strategy to obtain the target reference curve.

3. The method according to claim 1, characterized in that The performing fitting processing on the target reference curve to obtain at least one type of fitting curve comprises: Determine the similarity between the target reference curves based on a preset similarity measurement strategy; Classifying the target reference curves whose similarity is greater than a preset similarity threshold into the same type, to obtain at least one type of target reference curve; The at least one type of target reference curve is fitted to obtain the at least one type of fitting curve.

4. The method according to any one of claims 1 to 3, characterized in that Determining the initial prediction result based on the at least one type of fitting curve comprises: Acquire current bus load data within a third preset time period, wherein the third preset time period is between the first preset time period and the second preset time period; Constructing a comparison reference curve based on the current bus load data, wherein the comparison reference curve is used to reflect the change trend of the bus load within the third preset time period; Determine a first fitting curve among the at least one type of fitting curves that has the highest similarity to the comparison reference curve; The initial prediction result is determined based on the first fitting curve.

5. The method according to claim 4, characterized in that The performing error adjustment on the initial prediction result based on the weight optimization curve to obtain the target prediction result comprises: Determine a first segmented curve based on the target reference curve, the preset segment duration and the target tracing point, and determine a second segmented curve based on the comparison reference curve, the preset segment duration and the target tracing point, wherein the target tracing point is the time point of the initial prediction result; Determining the weight optimization curve based on the first piecewise curve and the second piecewise curve; The initial prediction result is error-adjusted based on the weight optimization curve to obtain the target prediction result.

6. The method according to claim 5, characterized in that The determining of the first segment curve based on the target reference curve, the preset segment duration and the target tracing point, and the determining of the second segment curve based on the comparison reference curve, the preset segment duration and the target tracing point include: The target reference curve is segmented based on the preset segmentation time and the target tracing point to obtain the first segmentation curve, and the comparison reference curve is segmented based on the preset segmentation time and the target tracing point to obtain the second segmentation curve.

7. The method according to claim 5, characterized in that The determining the weight optimization curve based on the first piecewise curve and the second piecewise curve comprises: Performing a similarity comparison on the first segmented curve and the second segmented curve to obtain a target segmented curve in the first segmented curve, wherein the target segmented curve has the highest similarity with the corresponding segmented curve in the second segmented curve; determining a second fitting curve of the at least one type of fitting curve based on the target piecewise curve; The weight optimization curve is determined based on the second fitting curve.

8. A bus load prediction device, characterized in that: The device comprises: An acquisition module, used to acquire a target reference curve of the bus load within a first preset time period, wherein the target reference curve is used to reflect a change trend of the bus load within the first preset time period; A fitting module, used for fitting the target reference curve to obtain at least one type of fitting curve, wherein different types of fitting curves have different load variation rules; a determination module, configured to determine an initial prediction result based on the at least one type of fitting curve, wherein the initial prediction result is used to represent a change trend of the bus load within a second preset time period, the second preset time period being after the first preset time period; The adjustment module is used to perform error adjustment on the initial prediction result based on the weight optimization curve to obtain the target prediction result, wherein the weight optimization curve is used to simulate the change trend of the bus load at the current moment.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, wherein the computer program is configured to execute the bus load prediction method described in any one of claims 1 to 7 when running on a computer or a processor.

10. An electronic device comprising a memory and a processor, characterized in that: A computer program is stored in the memory, and the processor is configured to run the computer program to execute the bus load prediction method described in any one of claims 1 to 7.

Images