Parameter correction method and device based on load prediction, electronic equipment and storage medium

Abstract

The invention provides a parameter correction method based on load prediction. The method comprises the following steps: acquiring electrical load information of a power distribution network area; an electrical load prediction model is constructed, the electrical load prediction model obtains electrical prediction data according to the electrical load information, and self-correction operation is carried out on input parameters of the electrical load prediction model; acquiring three-phase voltage information of the power distribution network area and a charge state of a battery, and calculating first reactive regulation power and first active regulation power of the power distribution network area by a preset area power regulation model according to the power utilization prediction data, the three-phase voltage information and the charge state; and adjusting the charging time period and the discharging time period of the power distribution network according to the first reactive adjustment power and the first active adjustment power. According to the technical scheme of the embodiment, the charging and discharging time period of the power distribution network is flexibly adjusted by dynamically calculating the reactive power adjusting power and the active power adjusting power of the PCS, so that the electric energy quality of the power distribution network is actively improved.

Description

Technical Field

[0001] The present invention relates to the technical field of electrochemical energy storage, and in particular to a parameter correction method, device, electronic equipment and storage medium based on load prediction. Background Art

[0002] At present, most of the energy storage systems in the power system substations are based on manual configuration of charging and discharging strategy parameters. The configuration of strategy parameters does not achieve self-correction and self-adaptation or only partially achieves self-correction and self-adaptation. Most of the power loads only achieve monitoring functions without data prediction functions. The self-correction and self-adaptation of the power quality management are mostly independent modes and cannot be effectively coordinated, resulting in a high technical threshold for parameter adjustment and greater difficulty in operation and maintenance. Summary of the invention

[0003] The present invention aims to solve at least one of the technical problems existing in the prior art. To this end, the present invention proposes a parameter correction method, device, electronic device, and storage medium based on load forecasting, which can effectively coordinate self-correction and self-adaptation, and reduce the technical threshold and operation and maintenance difficulty of parameter adjustment.

[0004] In a first aspect, an embodiment of the present invention provides a parameter correction method based on load forecasting, comprising:

[0005] Obtain power load information of distribution network substations;

[0006] Constructing a power load prediction model, wherein the power load prediction model obtains power consumption prediction data according to the power load information, and performs a self-correction operation on input parameters of the power load prediction model;

[0007] The three-phase voltage information of the distribution network substation and the state of charge of the battery are obtained, and the preset substation power regulation model calculates the first reactive power regulation power and the first active power regulation power of the distribution network substation according to the power consumption forecast data, the three-phase voltage information and the state of charge;

[0008] The charging time period and the discharging time period of the power distribution network are adjusted according to the first reactive power adjustment power and the first active power adjustment power.

[0009] In some embodiments of the present invention, a power load sample data source, a weather data sample data source, and a holiday sample data source of the distribution network substation are obtained, and a standard data set of the power load prediction model is constructed according to the power load sample data source, the weather data sample data source, and the holiday sample data source, wherein the power load data sample source includes at least a time value and power load historical data, the weather data sample source includes at least time data and weather historical data, and the holiday data sample source includes holiday time data and holiday historical data;

[0010] Performing data cleaning on the power load sample data source, the weather data sample data source, and the holiday sample data source;

[0011] Merging the cleaned power load sample data source, the weather data sample data source and the holiday sample data source into the standard data set;

[0012] Obtaining a time series of the power load forecasting model, and sorting data samples in the standard data set according to the time series;

[0013] Discretizing the weather historical data, holiday historical data, and power load historical data in the standard data set;

[0014] Normalizing the power load history data in the standard data set to obtain trend characteristics and random characteristics in the time series;

[0015] Constructing a feature matrix according to the time series, the power load historical data, the weather historical data and the holiday historical data;

[0016] The standard data set is iterated until the power load prediction model meets the training requirements.

[0017] In some embodiments of the present invention, the calculating of the first reactive power regulation power and the first active power regulation power of the distribution network station area includes:

[0018] Obtaining a first apparent power, a first power factor, and a first phase difference, calculating a first active power according to the first apparent power and the first power factor, and calculating a first reactive power according to the first apparent power and the first phase difference;

[0019] calculating a first compensation power of the first active power, and compensating the first active power to obtain a first target power;

[0020] A second compensation power of the first reactive power is calculated, and the first reactive power is compensated to obtain a second target power.

[0021] In some embodiments of the present invention, calculating a first compensation power of the first active power, compensating the first active power, and obtaining a first target power includes:

[0022] Obtaining apparent power before compensation and power factor before compensation, and calculating active power before compensation according to the apparent power before compensation and the power factor before compensation;

[0023] Obtaining the transformer low-load active power value and the power load active power value, and calculating the energy storage active power value according to the transformer low-load active power value and the power load active power value;

[0024] Obtaining the heavy-load active power value of the transformer, and calculating the energy storage discharge active power value according to the heavy-load active power value of the transformer and the active power value of the electric load;

[0025] Obtaining the energy storage A-phase active power value, B-phase active power value and C-phase active power value, and calculating the energy storage active power value according to the energy storage A-phase active power value, B-phase active power value and C-phase active power value;

[0026] The pre-compensation active power is compensated and adjusted according to the energy storage active power value, the energy storage discharge active power value and the energy storage active power value to obtain a first target power.

[0027] In some embodiments of the present invention, the calculating the second compensation power of the first reactive power, compensating the first reactive power, and obtaining the second target power includes:

[0028] Acquire a phase difference, and calculate a reactive power value before compensation according to the apparent power before compensation and the phase difference;

[0029] Obtaining the energy storage A-phase reactive power value, B-phase reactive power value and C-phase reactive power value, and calculating the energy storage reactive power compensation value according to the energy storage A-phase reactive power value, B-phase reactive power value and C-phase reactive power value;

[0030] The reactive power compensation value to be stored is compensated and adjusted according to the reactive power value before compensation to obtain a second target power.

[0031] In some embodiments of the present invention, the power load model predicts the power load data according to the power load information and the power data information, including:

[0032] Obtaining power load data of the distribution network area, holiday data at the forecast time node, and weather data at the forecast time node;

[0033] The power load data, the holiday data and the weather data are input into the power load prediction model, and the power load prediction model outputs the power load data at the prediction time node.

[0034] In some embodiments of the present invention, the model constraints of the power load forecasting model include:

[0035] The current total active power of energy storage is less than or equal to the active power threshold of energy storage system charging and discharging;

[0036] The current total active power of energy storage is less than or equal to the current data value of the charging and discharging active power of the energy storage system;

[0037] The current charge is equal to the minimum charge threshold of the energy storage system.

[0038] In a second aspect, an embodiment of the present invention provides a parameter correction device based on load forecasting, comprising at least one control processor and a memory for communicating with the at least one control processor; the memory stores instructions executable by the at least one control processor, and the instructions are executed by the at least one control processor so that the at least one control processor can execute the parameter correction method based on load forecasting as described in the first aspect above.

[0039] In a third aspect, an embodiment of the present invention provides an electronic device, comprising a parameter correction device based on load prediction as described in the second aspect above.

[0040] In a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium storing computer-executable instructions, wherein the computer-executable instructions are used to execute the parameter correction method based on load forecasting as described in the first aspect above.

[0041] The parameter correction method based on load forecasting according to the embodiment of the present invention has at least the following beneficial effects:

[0042] Obtain the power load information of the distribution network substation; construct a power load prediction model, the power load prediction model obtains the power consumption prediction data according to the power load information, and performs self-correction operations on the input parameters of the power load prediction model; obtain the three-phase voltage information of the distribution network substation and the charge state of the battery, and the preset substation power regulation model calculates the first reactive power and the first active power of the distribution network substation according to the power consumption prediction data, the three-phase voltage information and the charge state; and adjusts the charging time period and the discharging time period of the distribution network according to the first reactive power and the first active power. According to the technical solution of this embodiment, through the preset substation power regulation model, the power consumption prediction data or the data of the planned charging and discharging and power and other influencing factors are added to predict and determine whether three-phase imbalance and power quality problems occur, dynamically calculate the PCS reactive power and active power, and flexibly adjust the charging and discharging time period of the distribution network, so as to actively realize the improvement of the power quality of the distribution network. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] Figure 1 is a flow chart of a parameter correction method based on load forecasting provided by an embodiment of the present invention;

[0044] Figure 2 It is a flow chart of constructing a power load prediction model provided by the present invention;

[0045] Figure 3 It is a flow chart of calculating the first reactive power regulation power and the first active power regulation power of the distribution network area provided by the present invention;

[0046] Figure 4 is a flow chart of calculating the first target power provided by the present invention;

[0047] Figure 5 is a flow chart of calculating the second target power provided by the present invention;

[0048] Figure 6 It is a flow chart of the power load model provided by the present invention predicting power load data according to power load information and power data information;

[0049] Figure 7 It is a structural diagram of a parameter correction device based on load forecasting provided in another embodiment of the present invention. DETAILED DESCRIPTION

[0050] Embodiments of the present invention are described in detail below, examples of which are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present invention, and cannot be understood as limiting the present invention.

[0051] In the description of the present invention, it should be understood that descriptions involving orientations, such as up, down, front, back, left, right, etc., and orientations or positional relationships indicated are based on the orientations or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be understood as a limitation on the present invention.

[0052] In the description of the present invention, "several" means one or more, "more" means more than two, "greater than", "less than", "exceed" etc. are understood as not including the number itself, and "above", "below", "within" etc. are understood as including the number itself. If there is a description of "first" or "second", it is only used for the purpose of distinguishing the technical features, and cannot be understood as indicating or implying the relative importance or implicitly indicating the number of the indicated technical features or implicitly indicating the order of the indicated technical features.

[0053] In the description of the present invention, unless otherwise clearly defined, terms such as setting, installing, connecting, etc. should be understood in a broad sense, and technicians in the relevant technical field can reasonably determine the specific meanings of the above terms in the present invention based on the specific content of the technical solution.

[0054] The embodiment of the present invention provides a parameter correction method based on load forecasting, including obtaining power load information of a distribution network substation; constructing a power load forecasting model, the power load forecasting model obtains power forecasting data according to the power load information, and performs self-correction operations on the input parameters of the power load forecasting model; obtaining the three-phase voltage information of the distribution network substation and the charge state of the battery, and the preset substation power regulation model calculates the first reactive regulation power and the first active regulation power of the distribution network substation according to the power forecasting data, the three-phase voltage information and the charge state; and adjusting the charging time period and the discharging time period of the distribution network according to the first reactive regulation power and the first active regulation power. According to the technical solution of this embodiment, through the preset substation power regulation model, power forecasting data or data of factors such as planned charging and discharging and power are added to predict and determine whether three-phase imbalance and power quality problems occur, dynamically calculate the PCS reactive regulation power and active regulation power, and flexibly adjust the charging and discharging time period of the distribution network, so as to actively realize the improvement of the power quality of the distribution network.

[0055] It should be noted that the strategies used in the substation power regulation model include platform planned charging and discharging strategy, substation heavy overload and energy consumption strategy, reactive power compensation strategy, three-phase imbalance management strategy, line high and low voltage management strategy, platform planned charging and discharging strategy, related control parameters include enablement status, time period, charging and discharging mode, active power, battery SOC; substation heavy overload and energy consumption strategy related control parameters include enablement status, transformer rated capacity, load rate, battery SOC; reactive power compensation strategy control parameters include power factor, battery SOC; three-phase imbalance management strategy control parameters include three-phase imbalance degree, battery SOC; line high and low voltage management strategy related control parameters include voltage deviation degree, battery SOC.

[0056] The specific power regulation model of the substation is as follows: switching and launching the charging and discharging mode according to the model strategy; obtaining the relevant thresholds of the model strategy and the activation flags of each strategy, and calculating the active power threshold, reactive power threshold, battery SOC threshold, and charging and discharging satisfaction and exit conditions of the current node energy storage according to the model strategy threshold and activation representation; obtaining the total capacity of the substation energy storage configuration, power load forecast data, energy storage data, and power grid data; inputting the active power threshold, reactive power threshold, charging and discharging satisfaction and exit conditions, power load forecast data, energy storage data, and power grid data of the energy storage, and calculating the active power and reactive power that should be output for charging and discharging.

[0057] The control method of the embodiment of the present invention is further described below based on the accompanying drawings.

[0058] Reference Figure 1 , Figure 1A flow chart of a parameter correction method based on load forecasting provided by an embodiment of the present invention, the parameter correction method based on load forecasting includes but is not limited to the following steps:

[0059] Step S11, obtaining power load information of the distribution network area;

[0060] It should be noted that when the distribution network substation is connected to new power load or power load data with sudden changes, the power load information of the distribution network substation is obtained, and the current multi-dimensional power consumption data of the distribution network substation is obtained based on the power load information. Through real-time monitoring and analysis of the power load information, it is possible to promptly discover whether there are power quality problems such as three-phase imbalance, low power factor or low voltage in the distribution network substation.

[0061] Step S12, constructing a power load prediction model, the power load prediction model obtains power consumption prediction data according to the power load information, and performs a self-correction operation on the input parameters of the power load prediction model;

[0062] It should be noted that the power load prediction model is constructed, the power load model is trained based on the convolutional network deep learning model, and the state probability compensation factor is weighted to improve the accuracy of the prediction. The energy storage EMS server uses big data technology to sub-correct the power load model of the distribution network area. The corrected power load prediction model is synchronized to the edge energy storage EMS gateway device at a regular interval, and the edge energy storage EMS gateway device updates its parameters.

[0063] Specifically, based on the convolutional neural network deep learning model, the following steps are included:

[0064] Construct a standard data set for training the power load forecasting model:

[0065] Three sample data sets are collected, and the data set sources include power load data sample source, weather data sample source, and holiday data sample source; the power load data samples include time value and power load historical data, the weather data samples include time and weather historical data, and the holiday data samples include time and holiday historical data.

[0066] Start to clean the data of each data sample source. The cleaning method includes: filling missing values ​​in the data sample, correcting or deleting outliers in the data sample, and deleting duplicate records in the data sample. The outliers in the power load data include the processing of power load data with sudden changes.

[0067] Merge the data, integrate the data from the three sources into a unified data sample standard set, and delete redundant data; sort the data samples by time (the time of entering the database); discretize the weather historical data, holiday historical data, and power load historical data in the standard data set, and extract the periodicity and seasonality characteristics in the time series data; normalize the power load historical data in the standard data set again, and extract the trend and randomness characteristics in the time series data;

[0068] Construct time value, user load data, weather data, and holiday data as feature matrix;

[0069] Use the standard data set to iteratively update the model until it meets the training requirements.

[0070] Step S13, obtaining the three-phase voltage information of the distribution network substation and the state of charge of the battery, and the preset substation power regulation model calculates the first reactive power regulation power and the first active power regulation power of the distribution network substation according to the power consumption forecast data, the three-phase voltage information and the state of charge;

[0071] It should be noted that the power load model is synchronously sent to the edge computing energy storage EMS gateway device through the periodic data collection communication collection machine system. The edge computing energy storage EMS gateway device adds power load forecast data such as electricity prices or planned charging and discharging time periods and power quality problems such as low power factor or low voltage to the power regulation model of the substation based on the power load forecast model and the preset substation power regulation model, and dynamically calculates the reactive regulation power and active regulation power of the distribution network substation to obtain the first reactive regulation power and the first active regulation power.

[0072] Step S14: adjusting the charging time period and the discharging time period of the power distribution network according to the first reactive power adjustment power and the first active power adjustment power.

[0073] It should be noted that the charging time period and discharging time period of the distribution network substation are adjusted according to the first reactive power regulation power and the first active power regulation power, so as to realize the intelligent charging and discharging strategy, flexibly adjust the charging and discharging time periods of the distribution network substation, and thus improve the power quality of the distribution network substation.

[0074] Furthermore, the intelligent charging and discharging strategy is verified periodically to obtain verification results, and the optimization scheduling strategy is optimized based on the verification results. The power load prediction model and the substation power regulation model are regularly synchronized and trained based on the verification results.

[0075] It should be noted that by deploying the energy storage EMS server in the intranet of the distribution network, the acquisition communication front-end system and the edge computing energy storage EMS gateway device are connected. When the energy storage EMS server receives the power load information processed and forwarded by the acquisition communication front-end, the power load prediction model is trained according to the power load information, thereby completing the adjustment of relevant parameters. The power load prediction model adopts a convolutional neural network deep learning model and a weighted state probability compensation factor to improve the prediction accuracy. The energy storage EMS server uses big data technology to self-correct the power load model parameters of each distribution network area.

[0076] In addition, in one embodiment, referring to Figure 2 ,exist Figure 1 Step S12 of the illustrated embodiment also includes but is not limited to the following steps:

[0077] Step S21, obtaining a power load sample data source, a weather data sample data source, and a holiday sample data source of a distribution network area, and constructing a standard data set of a power load prediction model according to the power load sample data source, the weather data sample data source, and the holiday sample data source, wherein the power load data sample source at least includes time value and power load historical data, the weather data sample source at least includes time data and weather historical data, and the holiday data sample source includes holiday time data and holiday historical data;

[0078] Step S22, cleaning the power load sample data source, the weather data sample data source, and the holiday sample data source;

[0079] Step S23, merging the cleaned power load sample data source, weather data sample data source and holiday sample data source into a standard data set;

[0080] Step S24, obtaining the time series of the power load prediction model, and sorting the data samples in the standard data set according to the time series;

[0081] Step S25, discretizing the weather historical data, holiday historical data, and power load historical data in the standard data set;

[0082] Step S26, normalizing the historical data of power load in the standard data set to obtain trend characteristics and random characteristics in the time series;

[0083] Step S27, constructing a feature matrix based on the time series, power load historical data, weather historical data and holiday historical data;

[0084] Step S28, iterating the standard data set until the power load prediction model meets the training requirements.

[0085] It should be noted that, in order to construct a standard data set for training the power load prediction model, three sample data sets are collected. The data set sources include power load data sample source, weather data sample source and holiday data sample source; the power load data source includes time value and power load historical data, the weather data sample source includes time, weather and historical data, and the holiday data sample source includes holiday time and holiday historical data. Data cleaning is performed on each data sample source, and the cleaning method includes: filling missing values ​​on the data sample source; correcting or deleting outliers on the data sample source; deleting duplicate records on the data sample, wherein the power load data outliers include the processing of power load data with data mutations, and the data sets of the power load data sample source, weather data sample source and holiday data sample source are obtained. Merge data, integrate the data of the power load data sample source, weather data sample source and holiday data sample source into a unified data sample standard set, and delete redundant data; sort the data samples by the time of entering the database to obtain a standard data set; discretize the weather historical data, holiday historical data, and power load historical data in the standard data set to extract the periodicity and seasonality characteristics in the time series data; normalize the power load historical data in the standard data set to extract the trend and randomness characteristics in the time series data; construct time value, user load data, weather data and holiday data as feature matrices. Use the standard data set to iteratively update until the model meets the training requirements.

[0086] In addition, in one embodiment, referring to Figure 3 ,exist Figure 1 Step S13 of the illustrated embodiment also includes but is not limited to the following steps:

[0087] Step S31, obtaining a first apparent power, a first power factor and a first phase difference, calculating a first active power according to the first apparent power and the first power factor, and calculating a first reactive power according to the first apparent power and the first phase difference;

[0088] Step S32, calculating a first compensation power of the first active power, compensating the first active power, and obtaining a first target power;

[0089] Step S33, calculating a second compensation power for the first reactive power, compensating the first reactive power, and obtaining a second target power.

[0090] It should be noted that the calculation of the first reactive power can be expressed by the following formula:

[0091]

[0092] Among them, St1 is the first apparent power, is the first power factor;

[0093] In addition, in one embodiment, referring to Figure 4 ,exist Figure 3 Step S32 of the illustrated embodiment also includes but is not limited to the following steps:

[0094] Step S41, obtaining the apparent power before compensation and the power factor before compensation, and calculating the active power before compensation according to the apparent power before compensation and the power factor before compensation;

[0095] Step S42, obtaining the transformer low-load active power value and the power load active power value, and calculating the energy storage active power value according to the transformer low-load active power value and the power load active power value;

[0096] Step S43, obtaining the transformer heavy load active power value, and calculating the energy storage discharge active power value according to the transformer heavy load active power value and the electric load active power value;

[0097] Step S44, obtaining the energy storage A phase active power value, B phase active power value and C phase active power value, and calculating the energy storage active power value according to the energy storage A phase active power value, B phase active power value and C phase active power value;

[0098] Step S45, compensating and adjusting the active power before compensation according to the energy storage active power value, the energy storage discharge active power value and the energy storage active power value to obtain the first target power.

[0099] It should be noted that the power demand of the power grid is balanced by calculating the active power value and discharge power value of the distribution network substation. When the transformer is underloaded, the energy storage system absorbs excess power to avoid energy waste. When the transformer is overloaded, the energy storage system discharges to supplement power to ensure the stable operation of the power grid. Through compensation and adjustment, the power factor of the power grid is optimized, the utilization efficiency of electric energy is improved, and finally a target reactive power (i.e., the first target power) that better meets the actual needs of the power grid is obtained, thereby reducing the consumption and operating costs of the distribution network substation.

[0100] The specific formula is as follows:

[0101]

[0102] Among them, S1 is the apparent power value before compensation, is the apparent power value before compensation, and P1 is the active power value before compensation.

[0103]

[0104] in, is the energy storage active power value, is the transformer low load active power value, Pload It is the active power value of the electrical load.

[0105]

[0106] in, is the energy storage active power value, It is the transformer heavy load active power value.

[0107] P total (t) = P A (t)+P B (t)+P c (t);

[0108] Among them, P total is the energy storage active power value, P A is the active power value of energy storage phase A, P B is the active power value of energy storage phase B, P C is the active power value of energy storage phase C.

[0109] In addition, in one embodiment, referring to Figure 5 ,exist Figure 3 Step S33 of the illustrated embodiment also includes but is not limited to the following steps:

[0110] Step S51, obtaining the phase difference, and calculating the reactive power value before compensation according to the apparent power before compensation and the phase difference;

[0111] Step S52, obtaining the energy storage A phase reactive power value, the B phase reactive power value and the C phase reactive power value, and calculating the energy storage reactive power compensation value according to the energy storage A phase reactive power value, the B phase reactive power value and the C phase reactive power value;

[0112] Step S53: compensating and adjusting the reactive power compensation value of the energy storage to be used according to the reactive power value before compensation to obtain a second target power.

[0113] It should be noted that the specific formula of this embodiment is as follows:

[0114]

[0115] Among them, Q t is the reactive power value before compensation, S t To compensate for the forward apparent power, is the phase difference.

[0116]

[0117] Among them, S t1 is the target apparent power, is the target power factor.

[0118]

[0119] Among them, Q t1 is the target reactive power value.

[0120]

[0121] Among them, Q t is the reactive power value to be compensated.

[0122] Q total (t) = Q A (t)+Q B (t)+Q C (t);

[0123] Among them, Q total is the reactive power value of energy storage, Q A is the reactive power value of phase A of energy storage, Q B is the reactive power value of energy storage phase B, Q C is the reactive power value of phase C of energy storage.

[0124] The parameter compensation and correction steps are as follows: the edge computing energy storage gateway sends a periodic synchronization request to the energy storage EMS model; the energy storage EMS identifies the substation and obtains the model parameters; the energy storage EMS returns the model parameters to the edge computing energy storage gateway; the edge computing energy storage gateway stores the local model parameters locally; the edge computing energy storage gateway inputs the model data to verify the model accuracy, and starts local correction after meeting the model accuracy requirements.

[0125] In addition, in one embodiment, referring to Figure 6 ,exist Figure 1 Step S12 of the illustrated embodiment also includes but is not limited to the following steps:

[0126] Step S61, obtaining power load data of the distribution network area, holiday data at the prediction time node, and weather data at the prediction time node;

[0127] Step S62: inputting power load data, holiday data and weather data into a power load prediction model, and the power load prediction model outputs the power load data at the prediction time node.

[0128] It should be noted that the power load data is used to reflect the power consumption of the distribution network area in different time periods. By analyzing the historical power load data, the law and trend of load changes can be revealed, providing a basis for load forecasting; by obtaining holiday data, the load changes during holidays are predicted for the power load forecasting model; weather data is introduced as an input variable into the power load forecasting model, thereby improving the model's prediction accuracy for load changes. In this embodiment, by obtaining the current power load data, the holiday data at the prediction time node, and the weather data at the prediction time node, the power load forecasting model is input with the current power load data, the holiday data at the prediction time node, and the weather data at the prediction time node, and the power load data at the prediction time node is output.

[0129] like Figure 7 As shown, Figure 7 : is a structural diagram of a parameter correction device based on load prediction provided by an embodiment of the present invention. The present invention also provides a parameter correction device based on load prediction, comprising:

[0130] The processor 701 may be implemented by a general-purpose central processing unit (CPU), a microprocessor, an application-specific integrated circuit (ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solutions provided in the embodiments of the present application;

[0131] The memory 702 can be implemented in the form of a read-only memory (ROM), a static storage device, a dynamic storage device, or a random access memory (RAM). The memory 702 can store an operating system and other application programs. When the technical solution provided in the embodiment of this specification is implemented by software or firmware, the relevant program code is stored in the memory 702, and the processor 701 calls and executes the parameter correction method based on load forecasting in the embodiment of this application;

[0132] Input / output interface 703, used to implement information input and output;

[0133] Communication interface 704, used to realize communication interaction between the present apparatus and other devices, which can be realized by wired means (such as USB, network cable, etc.) or wireless means (such as mobile network, WI FI, Bluetooth, etc.);

[0134] A bus 705 that transmits information between the various components of the device (e.g., the processor 701, the memory 702, the input / output interface 703, and the communication interface 704);

[0135] The processor 701 , the memory 702 , the input / output interface 703 and the communication interface 704 are connected to each other in communication within the device via a bus 705 .

[0136] An embodiment of the present application also provides an electronic device, comprising the load prediction-based parameter correction device as described above.

[0137] An embodiment of the present application further provides a storage medium, which is a computer-readable storage medium and stores a computer program. When the computer program is executed by a processor, the above-mentioned parameter correction method based on load forecasting is implemented.

[0138] As a non-transient computer-readable storage medium, the memory can be used to store non-transient software programs and non-transient computer executable programs. In addition, the memory may include a high-speed random access memory, and may also include a non-transient memory, such as at least one disk storage device, a flash memory device, or other non-transient solid-state storage devices. In some embodiments, the memory may optionally include a memory remotely arranged relative to the processor, and these remote memories may be connected to the processor via a network. Examples of the above-mentioned networks include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and a combination thereof. The device embodiments described above are merely schematic, wherein the units described as separate components may or may not be physically separated, and are implemented to be located in one place, or may also be distributed to multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the present embodiment.

[0139] It will be appreciated by those skilled in the art that all or some of the steps and systems in the disclosed method above may be implemented as software, firmware, hardware and appropriate combinations thereof. Some physical components or all physical components may be implemented as software executed by a processor, such as a central processing unit, a digital signal processor or a microprocessor, or may be implemented as hardware, or may be implemented as an integrated circuit, such as an application specific integrated circuit. Such software may be distributed on a computer-readable medium, which may include a computer storage medium (or a non-transitory medium) and a communication medium (or a temporary medium). As known to those skilled in the art, the term computer storage medium includes volatile and non-volatile, removable and non-removable media implemented in any method or technology for storing information (such as computer-readable instructions, data structures, program modules or other data). Computer storage media include, but are not limited to, RAM, ROM, EEPROM, flash memory or other memory technologies, CD-ROM, digital versatile disks (DVD) or other optical disk storage, magnetic cassettes, magnetic tapes, disk storage or other magnetic storage devices, or any other medium that may be used to store desired information and may be accessed by a computer. Furthermore, it is well known to those skilled in the art that communication media generally include computer readable instructions, data structures, program modules, or other data in a modulated data signal such as a carrier wave or other transport mechanism, and may include any information delivery media.

[0140] The above is a specific description of the preferred implementation of the present invention, but the present invention is not limited to the above-mentioned implementation mode. Technical personnel familiar with the field can also make various equivalent deformations or substitutions under the shared conditions without violating the spirit of the present invention. These equivalent deformations or substitutions are all included in the scope defined by the claims of the present invention.

Claims

1. A parameter correction method based on load forecasting, characterized in that: include: Obtain power load information of distribution network substations; Constructing a power load prediction model, wherein the power load prediction model obtains power consumption prediction data according to the power load information, and performs a self-correction operation on input parameters of the power load prediction model; The three-phase voltage information of the distribution network substation and the state of charge of the battery are obtained, and the preset substation power regulation model calculates the first reactive power regulation power and the first active power regulation power of the distribution network substation according to the power consumption forecast data, the three-phase voltage information and the state of charge; The charging time period and the discharging time period of the power distribution network are adjusted according to the first reactive power adjustment power and the first active power adjustment power.

2. The parameter correction method based on load forecasting according to claim 1 is characterized in that: The construction of the power load prediction model includes: Obtain a power load sample data source, a weather data sample data source, and a holiday sample data source of the distribution network substation, and construct a standard data set for the power load prediction model according to the power load sample data source, the weather data sample data source, and the holiday sample data source, wherein the power load data sample source includes at least a time value and power load historical data, the weather data sample source includes at least time data and weather historical data, and the holiday data sample source includes holiday time data and holiday historical data; Performing data cleaning on the power load sample data source, the weather data sample data source, and the holiday sample data source; Merging the cleaned power load sample data source, the weather data sample data source and the holiday sample data source into the standard data set; Obtaining a time series of the power load forecasting model, and sorting data samples in the standard data set according to the time series; Discretizing the weather historical data, holiday historical data, and power load historical data in the standard data set; Normalizing the power load history data in the standard data set to obtain trend characteristics and random characteristics in the time series; Constructing a feature matrix according to the time series, the power load historical data, the weather historical data and the holiday historical data; The standard data set is iterated until the power load prediction model meets the training requirements.

3. The parameter correction method based on load forecasting according to claim 1 is characterized in that: The calculating of the first reactive power regulation power and the first active power regulation power of the distribution network station area includes: Obtaining a first apparent power, a first power factor, and a first phase difference, calculating a first active power according to the first apparent power and the first power factor, and calculating a first reactive power according to the first apparent power and the first phase difference; calculating a first compensation power of the first active power, and compensating the first active power to obtain a first target power; A second compensation power of the first reactive power is calculated, and the first reactive power is compensated to obtain a second target power.

4. The parameter correction method based on load forecasting according to claim 3 is characterized in that: The calculating a first compensation power of the first active power and compensating the first active power to obtain a first target power includes: Obtaining apparent power before compensation and power factor before compensation, and calculating active power before compensation according to the apparent power before compensation and the power factor before compensation; Obtaining the transformer low-load active power value and the power load active power value, and calculating the energy storage active power value according to the transformer low-load active power value and the power load active power value; Obtaining the heavy-load active power value of the transformer, and calculating the energy storage discharge active power value according to the heavy-load active power value of the transformer and the active power value of the electric load; Obtaining the energy storage A-phase active power value, B-phase active power value and C-phase active power value, and calculating the energy storage active power value according to the energy storage A-phase active power value, B-phase active power value and C-phase active power value; The pre-compensation active power is compensated and adjusted according to the energy storage active power value, the energy storage discharge active power value and the energy storage active power value to obtain a first target power.

5. The parameter correction method based on load forecasting according to claim 3 is characterized in that: The calculating the second compensation power of the first reactive power and compensating the first reactive power to obtain the second target power includes: Acquire a phase difference, and calculate a reactive power value before compensation according to the apparent power before compensation and the phase difference; Obtaining the energy storage A-phase reactive power value, B-phase reactive power value and C-phase reactive power value, and calculating the energy storage reactive power compensation value according to the energy storage A-phase reactive power value, B-phase reactive power value and C-phase reactive power value; The reactive power compensation value to be stored is compensated and adjusted according to the reactive power value before compensation to obtain a second target power.

6. The parameter correction method based on load forecasting according to claim 1, characterized in that: The power load model predicts power load data according to the power load information and the power data information, including: Obtaining power load data of the distribution network area, holiday data at the forecast time node, and weather data at the forecast time node; The power load data, the holiday data and the weather data are input into the power load prediction model, and the power load prediction model outputs the power load data at the prediction time node.

7. The parameter correction method based on load forecasting according to claim 1 is characterized in that: The model constraints of the power load forecasting model include: The current total active power of energy storage is less than or equal to the active power threshold of energy storage system charging and discharging; The current total active power of energy storage is less than or equal to the current data value of the charging and discharging active power of the energy storage system; The current charge is equal to the minimum charge threshold of the energy storage system.

8. A parameter correction device based on load forecasting, characterized in that: It includes at least one control processor and a memory for communicating with the at least one control processor; the memory stores instructions that can be executed by the at least one control processor, and the instructions are executed by the at least one control processor to enable the at least one control processor to execute the parameter correction method based on load forecasting as described in any one of claims 1 to 7.

9. An electronic device, characterized in that: It includes the parameter correction device based on load forecasting as described in claim 8.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer-executable instructions, and the computer-executable instructions are used to enable a computer to execute the parameter correction method based on load forecasting according to any one of claims 1 to 7.

Images