Method and system for predicting residual power supply time of UPS (Uninterrupted Power Supply) of box transformer substation of wind power plant
By combining machine learning models with real-time UPS data and environmental factors, the problem of accurate prediction of remaining UPS power supply time in wind farms has been solved, achieving efficient operation and maintenance management and stable equipment operation.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- XIAN THERMAL POWER RES INST CO LTD
- Filing Date
- 2026-02-02
- Publication Date
- 2026-05-26
AI Technical Summary
Existing technologies struggle to accurately predict the remaining power supply time of UPS systems in wind farms. Traditional methods fail to adequately consider the impact of factors such as ambient temperature and charging/discharging frequency, resulting in significant deviations in the calculation of remaining power capacity and a lack of comprehensive utilization of multi-source information.
By employing a machine learning model that combines real-time remaining UPS power, power supply equipment power information, and UPS characteristic information, and by detecting changes in discharge current in real time and calculating corrected voltage, while taking into account the effects of different discharge modes and temperature, an accurate prediction model for remaining power supply time is established.
It improves the accuracy of UPS remaining power supply time prediction, provides timely basis for operation and maintenance decisions, reduces operation and maintenance costs, and ensures stable operation of wind farm equipment and continuous power generation.
Smart Images
Figure CN122085121A_ABST
Abstract
Description
Technical Field
[0001] This invention belongs to the field of UPS power supply monitoring technology for wind farm transformer substations, and relates to a method and system for predicting the remaining power supply time of UPS power supply in wind farm transformer substations. Background Technology
[0002] In the field of wind power generation, wind farm transformer substations serve as crucial hubs for power transmission and distribution, and their stable operation directly impacts the overall power generation efficiency and safety of the wind farm. UPS (Uninterruptible Power Supply) plays an indispensable role in wind farm transformer substations. When the main power supply is interrupted due to faults, extreme weather, routine maintenance, or other reasons, the UPS, with its rapid and seamless switching capabilities, can continuously supply stable power to the transformer substation, preventing serious consequences such as equipment damage, data loss, and production interruptions caused by power outages. It is of great significance for ensuring the stable operation of wind farm equipment and the continuity of power generation. Currently, real-time monitoring and remaining power supply time prediction are crucial for the reliable operation and maintenance of UPS power supplies. Real-time monitoring of the UPS's operating status allows for the timely detection and handling of potential faults, reducing the risk of production interruptions due to UPS failures. Simultaneously, accurately predicting the UPS's remaining power supply time provides maintenance personnel with scientific operational guidance, enabling them to plan contingency measures in advance and ensure the UPS can meet power supply needs in emergency situations. However, in practical applications, obtaining accurate remaining power of the UPS in wind farms and predicting future power consumption still faces many challenges. Traditional power calculation methods are difficult to accurately adapt to the complex and ever-changing operating environment of wind farms, and cannot fully consider the impact of factors such as ambient temperature and charging / discharging frequency on battery performance, resulting in significant deviations in remaining power calculations. Furthermore, existing UPS remaining power time prediction methods mostly rely on simple empirical formulas or single-data-dimensional analysis, lacking comprehensive utilization of multi-source information such as UPS monitoring information, SCADA system data, historical operating data, power supply equipment power, and UPS characteristics, making it difficult to achieve high-precision remaining power time prediction. Therefore, there is an urgent need to establish more scientific and accurate battery power consumption algorithm models and UPS remaining power time prediction models and algorithms to meet the growing demand for efficient operation and maintenance management in wind farms. Summary of the Invention
[0003] The purpose of this invention is to solve the technical problem of large prediction deviation of the remaining power of UPS in the existing technology, and to provide a method and system for predicting the remaining power supply time of UPS in wind farm substations.
[0004] To achieve the above objectives, the present invention employs the following technical solution: The first aspect of this invention provides a method for predicting the remaining power supply time of a UPS power supply in a wind farm transformer substation, comprising the following steps: S1, after discharge begins, input the real-time remaining power of the UPS, the power information of the power supply equipment, and the UPS characteristic information into the machine learning model to predict the current remaining power. S2, based on the predicted remaining power, determine whether the UPS needs charging: if charging is needed, end the prediction; otherwise, check if the UPS discharge current changes. If the discharge current at the current moment changes compared to the current at the previous moment, the voltage at the previous moment is recorded, and after a delay, the corrected voltage is calculated and the voltage corresponding to the discharge current at the current moment is updated in the prediction table. The corrected voltage is obtained by obtaining the difference between the actual voltage and the theoretical voltage at different voltage points through a reference current discharge experiment. The prediction table records the remaining power supply time to reach a specific current and a specific voltage. If the discharge current at the current moment has not changed compared to the current at the previous moment, record the current discharge time, calculate the remaining power supply time based on the current discharge time, and update the prediction table. S3, check in real time whether the current voltage has reached the cutoff value: if it has not reached it, return to S2; otherwise, select the corresponding correction method according to the discharge mode to predict the remaining power supply time and update the remaining power supply time in the prediction table.
[0005] Furthermore, the machine learning model employs linear regression, support vector machine, or neural network models.
[0006] Furthermore, the corrected voltage is specifically as follows:
[0007] in, voltage point The actual voltage; voltage point The theoretical voltage.
[0008] Furthermore, the prediction table is initially set as follows: The current is divided into multiple levels, and the voltage is divided into levels according to a set value. The initial data is obtained by using the current integration method, the open-circuit voltage method, or a combination of the two to fill the prediction table.
[0009] Furthermore, the method of combining the two specifically includes: The charging and discharging status of the battery is recorded using the current integration method, and the current SOC1 is obtained. The open-circuit voltage of the battery is measured using the open-circuit voltage method, and the current SOC2 is calculated based on the SOC-open-circuit voltage curve. Calculate the current SOC based on SOC1 and SOC2; SOC = k1SOC1 + k2SOC2; Where k1 and k2 are weighting coefficients.
[0010] Furthermore, the step of selecting the corresponding correction method based on the discharge mode to predict the remaining power supply time specifically involves: During constant current discharge, the remaining power supply time corresponding to the current-cutoff voltage in the prediction table is updated with the actual discharge time; During multi-current discharge, according to the formula Calculate the remaining power supply time; among which, To correct the voltage, Select the known range that is closest to the current. When temperature affects, according to After correcting the voltage, adjust the remaining power supply time according to the voltage-time relationship.
[0011] A second aspect of the present invention provides a computer program product, the computer program product including computer instructions, the computer instructions instructing a computer to execute the method for predicting the remaining power supply time of the UPS power supply in the wind farm transformer substation.
[0012] A third aspect of the present invention provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the method for predicting the remaining power supply time of the UPS power supply of the wind farm transformer substation.
[0013] A fourth aspect of the present invention provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the method for predicting the remaining power supply time of a UPS power supply in a wind farm transformer substation.
[0014] The fifth aspect of this invention provides a system for predicting the remaining power supply time of a UPS power supply in a wind farm transformer substation, comprising: The remaining power prediction module, after starting to discharge, inputs the real-time remaining power of the UPS, the power information of the power supply equipment, and the UPS characteristic information into the machine learning model to predict the current remaining power. The correction module determines whether the UPS needs charging based on the predicted remaining power. If charging is required, the prediction ends; otherwise, it checks whether the UPS discharge current has changed. If the discharge current at the current moment changes compared to the current at the previous moment, the voltage at the previous moment is recorded, and after a delay, the corrected voltage is calculated and the voltage corresponding to the discharge current at the current moment is updated in the prediction table. The corrected voltage is obtained by obtaining the difference between the actual voltage and the theoretical voltage at different voltage points through a reference current discharge experiment. The prediction table records the remaining power supply time to reach a specific current and a specific voltage. If the discharge current at the current moment has not changed compared to the current at the previous moment, record the current discharge time, calculate the remaining power supply time based on the current discharge time, and update the prediction table. The time prediction module checks in real time whether the current voltage has reached the cutoff value: if it has not reached it, it returns to S2; otherwise, it selects the corresponding correction method according to the discharge mode to predict the remaining power supply time and updates the remaining power supply time in the prediction table.
[0015] This invention primarily introduces the application effect analysis of UPS power monitoring technology for wind farm transformer substations and the establishment and application of a remaining power supply time prediction model. Using algorithms such as the Ah integral method and open-circuit voltage method, combined with environmental factors of the transformer substation, a battery power consumption calculation model is established to obtain real-time monitoring data of the remaining UPS power. Simultaneously, a UPS testing system is established to simulate the UPS load environment, ensuring that the power supply is in an effective state and meeting the closing requirements of the high-voltage load switch under off-grid conditions. Furthermore, based on UPS monitoring information and SCADA (Wind Turbine Monitoring and Control System) data, and using historical data and machine learning algorithms, combined with UPS remaining power, power supply equipment power information, and UPS characteristic information, a UPS remaining power supply time prediction model and algorithm are developed to predict the UPS power consumption in real time over a future period. A monitoring interface and alarm mechanism are also designed to allow administrators to view the UPS remaining power and operating status in real time and receive timely alarms when power is insufficient or abnormalities occur.
[0016] Compared with the prior art, the present invention has the following beneficial effects: This invention discloses a method for predicting the remaining power supply time of a UPS (Uninterruptible Power Supply) in a wind farm substation. It employs machine learning models (such as linear regression, support vector machines, or neural network models) combined with real-time remaining UPS power, power information of the power supply equipment, and UPS characteristic information to predict the remaining power. These models can learn complex patterns and relationships in historical data, thus more accurately predicting the current remaining power and providing a reliable basis for predicting the remaining power supply time. By real-time detection of discharge current changes and calculation of correction voltage, the influence of battery internal resistance and polarization voltage under different discharge currents is considered. Simultaneously, when the voltage reaches the cutoff value, the corresponding correction method is selected to predict the remaining power supply time based on different discharge modes (constant current discharge, multi-current discharge, temperature influence), allowing the prediction results to adapt to changes in the actual battery operating state in a timely manner, further improving the accuracy of the prediction. Real-time prediction of the remaining UPS power supply time provides timely decision-making basis for wind farm operation and maintenance personnel. Maintenance personnel can rationally arrange equipment operation, maintenance, and emergency handling measures based on the prediction results, avoiding equipment failures and production losses caused by insufficient UPS power, and improving the operational reliability and stability of the wind farm. The prediction table is continuously updated during the prediction process, ensuring that the prediction table reflects the actual usage and performance changes of the battery. This helps to track the battery's condition over a long period, providing a scientific basis for battery maintenance and replacement, and reducing maintenance costs.
[0017] Furthermore, this method can adapt to various operating conditions of UPS power supplies in wind farms. Whether it's changes in charging and discharging current, or the influence of factors such as temperature, it can accurately predict these conditions through corresponding detection and correction mechanisms. For example, under multi-current discharge conditions, it accurately predicts the remaining power supply time using a specific calculation formula, meeting the complex and ever-changing actual operational needs.
[0018] Furthermore, during the prediction table initialization phase, initialization data is obtained by comprehensively employing the current integration method, the open-circuit voltage method, or a combination of both. When the two methods are combined, the current SOC is calculated based on SOC1 and SOC2, fully considering the dynamic process and static characteristics of battery charging and discharging, making the initialization data more consistent with actual conditions, and enhancing the method's adaptability to different battery states and operating conditions. Attached Figure Description
[0019] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of the present invention and should not be regarded as a limitation on the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.
[0020] Figure 1This is a flowchart of the method for predicting the remaining power supply time of a UPS power supply in a wind farm transformer substation according to the present invention. Figure 2 This is the circuit diagram of the UPS power supply for the wind farm transformer substation of the present invention; Figure 3 As an embodiment of the present invention, the open-circuit voltage-SOC curve is shown. Detailed Implementation
[0021] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. The components of the embodiments of the present invention described and marked in the accompanying drawings can generally be arranged and designed in various different configurations.
[0022] Therefore, the following detailed description of the embodiments of the invention provided in the accompanying drawings is not intended to limit the scope of the claimed invention, but merely to illustrate selected embodiments of the invention. All other embodiments obtained by those skilled in the art based on the embodiments of the invention without inventive effort are within the scope of protection of the invention.
[0023] It should be noted that similar labels and letters in the following figures indicate similar items. Therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures.
[0024] The present invention will now be described in further detail with reference to the accompanying drawings: See Figure 1 This invention discloses a method for predicting the remaining power supply time of a UPS power supply in a wind farm transformer substation, specifically including: S1. First, the purpose of predicting the remaining battery time is to obtain information about the working time that the battery system can provide. Therefore, in practice, we only need to know the working time that the battery system can still provide under the current conditions (voltage, current, temperature). At a certain moment, the voltage, current, and temperature values can be measured. Then, we can predict the duration of constant current discharge of the battery at this current. That is, the system has a table that divides the voltage into several levels and the current into several levels, as shown in Table 1.
[0025] Table 1 Battery Capacity Prediction Table Io Ix Im
[0026] In the table, t(n, m) represents the time remaining until the voltage reaches Vn after discharging with Im.
[0027] 1) Table Initialization: Initialization data does not need to fill the table completely, but the amount of data determines the accuracy of the remaining capacity prediction in the early stages of system operation. We divide the current into four levels: 0.05C / 0.1C / 0.15C / 0.2C, and the voltage into 0.1V increments. During initialization, algorithms such as the Ah integration method and open-circuit voltage method are used, combined with factors related to the substation's operating environment, to establish a battery power consumption calculation model and obtain real-time monitoring data of the remaining UPS power in the substation. (1) Current integration method The current integration method is a method for calculating the state of charge (SOC) of a battery by integrating the charging and discharging current. Its advantage lies in its simplicity and the fact that calibration is not required. The specific steps of the current integration method are as follows: Step 1: Record the SOC of the battery when it begins to charge or discharge.
[0028] Step 2: Record the current during battery charging and discharging.
[0029] Step 3: Integrate the battery current to obtain the change in battery charge.
[0030] Step 4: Calculate the current SOC based on the change in power and the initial SOC.
[0031] formula:
[0032] In the formula, SOC t This is the estimated SOC value at the current moment, SOC0 is the initial value, C is the nominal capacity of the battery, η is the battery charge / discharge efficiency, I is the charge / discharge current (discharge is positive), and dt represents the time interval.
[0033] (2) Open-circuit voltage method The open-circuit voltage method calculates the battery's state of charge (SOC) by measuring its open-circuit voltage (i.e., the voltage under no load). The advantage of the open-circuit voltage method is its simplicity, as it eliminates the need for current measurement. The specific steps of the open-circuit voltage method are as follows: Step 1: Based on the battery model and the SOC-open circuit voltage curve provided by the manufacturer, establish the relationship between SOC and open circuit voltage.
[0034] formula:
[0035] Among them, V B0 denoted as , where a is the open-circuit voltage of the battery, and b is the open-circuit voltage when fully charged and fully discharged.
[0036] Step 2: Measure the open-circuit voltage of the battery.
[0037] Step 3: Calculate the current SOC based on the SOC-open circuit voltage curve established in Step 1.
[0038] It is important to note that the SOC-open circuit voltage curve of a battery changes with the use and lifespan of the battery. Therefore, it is necessary to perform calibration regularly to ensure the accuracy of the calculation results.
[0039] (3) Combine the current integration method and the open-circuit voltage method The current integration method and the open-circuit voltage method each have their own advantages and disadvantages. Therefore, in practical applications, the two methods are often combined to improve the accuracy of the SOC.
[0040] The specific steps for combining the current integration method and the open-circuit voltage method are as follows: Step 1: Use the current integration method to record the charging and discharging status of the battery and obtain the current SOC1.
[0041] Step 2: Measure the open-circuit voltage of the battery using the open-circuit voltage method, and calculate the current SOC2 based on the SOC-open-circuit voltage curve.
[0042] Step 3: Calculate the current SOC based on SOC1 and SOC2.
[0043] Formula: SOC = k1SOC1 + k2SOC2 Here, k1 and k2 are weighting coefficients, and their sum equals 1. The selection of these weighting coefficients needs to take into account factors such as battery usage, test time, and test accuracy. Generally speaking, if the battery has undergone a longer charge-discharge test, then the value of k1 should be larger; if the battery's open-circuit voltage measurement accuracy is high, then the value of k2 should be larger.
[0044] Based on the initialization results, a portion of the data in the prediction table is obtained. If t(V1, I2) is known from the prediction table, the remaining time when I1 discharges to reach V1 is predicted using the conversion formula (2).
[0045] t(V1,I1)=t(V1-Vx2+Vx1,I2)×I2 / I1(2) In the formula: t(V1, I1) is the remaining time when the battery voltage reaches V1 after discharging with I1, and Vxn is the corresponding correction voltage for each current.
[0046] This method predicts the remaining time with an error of less than 15 minutes. In practical applications, the accuracy of the prediction can be further improved by increasing the measurement precision of time and voltage. The time prediction value obtained at this point can be used as a preliminary result of the remaining power supply time of the UPS. Subsequently, based on the dynamic changes of parameters such as current and voltage during the actual discharge process, the remaining power supply time prediction value can be updated and optimized in real time according to the judgment and correction steps in the flowchart.
[0047] S2, Start of Discharge and Remaining Power Prediction: After the UPS starts discharging, it uses a model based on historical data and machine learning algorithms, combined with the UPS remaining power, power supply equipment power information and UPS characteristic information, to predict the remaining power at the prediction point.
[0048] S3, Determine if charging is required: Based on the predicted remaining power, check if the UPS is charging. If the power is insufficient and charging is required, and the UPS is in a charging state (judgment result is "Y"), then the prediction of the remaining power supply time ends directly; if not charging (judgment result is "N"), continue to the next judgment step.
[0049] S4, Determine if the current has changed: Detect whether the current discharge current has changed. If the current has changed (the result is "Y"), first record the voltage before the change, then perform a delay operation, and then perform a voltage correction operation; if the current has not changed (the result is "N"), continue to the next step.
[0050] Corrected Voltage: Since the battery's internal resistance and polarization voltage differ at different discharge currents, a corrected voltage is needed for each discharge current. Discharge experiments were conducted using 0.05C as a baseline, with voltage increments of 0.1V. Measurements were recorded at different voltage points. For example, at a specific voltage point V0, when discharging at a current of 0.05C, the measured actual battery voltage was V. 0_actual1 The theoretical voltage is V 0_theory Then the corrected voltage V at this time x1 =V 0_actual1 -V 0_theory Following this method, measurements and calculations are performed sequentially at different voltage points to obtain the corrected voltages for each voltage point relative to a 0.05C discharge at different discharge currents. These corrected voltages are used to establish a precise mapping between "different discharge currents - voltages - remaining time," supporting subsequent "discharge time prediction" and "prediction table correction." Their specific functions include: (1) Eliminate the influence of current difference: Under different discharge currents, the battery internal resistance and polarization effect cause the actual voltage to deviate from the theoretical value (such as the voltage drop when discharging at a large current). By correcting the voltage, the discharge data of "non-reference current (0.05C)" can be converted into the "reference current system", so that the discharge time prediction of different currents has a unified reference standard.
[0051] (2) Filling / correcting the prediction table: The prediction table t(n,m) records the "remaining time under a specific current Im and a specific voltage Vn". The corrected voltage can be used for time conversion across current ranges (such as converting the remaining time of 0.2C discharge from 0.1C discharge data), allowing the prediction table to cover more current-voltage combinations, or correcting errors in existing data.
[0052] S5, Record Discharge Time: Record the current discharge time while the current remains unchanged. The key value of this step is accumulating real data from the "constant current discharge stage," which is used for subsequent "cutoff voltage determination" and "prediction table correction." (1) Establishing a benchmark in the time dimension: When the current is stable, the "discharge time" is the direct basis for measuring the battery capacity consumption (the power consumption can be calculated by combining the current integral). Recording this time can provide segmented data for the "total discharge time when the voltage reaches the cutoff value".
[0053] (2) Time conversion for supporting multiple current stages: If the current changes during the discharge process (e.g., switching from 0.1C to 0.2C), the "discharge time + correction voltage" of different stages can be converted into the equivalent time under a unified current by the formula t(V1, I1)=t(V1-Vx2+Vx1, I2)×I2 / I1, and finally corrected in the prediction table.
[0054] S6, Determine if the voltage has reached the cutoff value: Check if the current voltage has reached the cutoff value. If the cutoff value has not been reached (the result is "N"), return to continue determining whether to charge, and repeat the above process; if the cutoff value has been reached (the result is "Y"), proceed to the next step.
[0055] S7, correct the prediction time and end: Corrected prediction time: When the voltage reaches the cutoff value, the prediction time corresponding to each discharge current is corrected according to different situations.
[0056] Scenario 1: If the battery remains in a constant current discharge state until the discharge ends, the discharge prediction table for that constant current value is directly corrected. For example, if the battery discharges at a constant current of 0.1C until the cutoff voltage, the prediction table data corresponding to the 0.1C current is adjusted. The actual recorded discharge start time, cutoff time, and cutoff voltage are compared with the predicted time corresponding to that current and cutoff voltage in the prediction table. If there is a discrepancy, the actual discharge time is used as the accurate data, and the remaining time data for that current-voltage combination in the prediction table is updated.
[0057] Scenario 2: If the battery undergoes several discharge currents before complete discharge, the time conversion formula can be used to calculate the discharge rate. t(V1, I1) = t(V1 - Vx2 +V x1 The correction is performed using I2)×I2 / I1. Here, t(V1, I1) represents the remaining time when the battery voltage reaches V1 after discharging at I1. xn This is the corresponding correction voltage for each current; the selection of I2 follows the principle of proximity to ensure prediction accuracy. For example, if the previous discharge was at 0.1C, and this time we want to predict the remaining time for a 0.2C discharge, then I2... _2 Take 0.1C. In specific operation, first determine the correction voltage V corresponding to the current discharge of each segment. x1 V x2 Then, based on the formula, the predicted remaining time for different currents at corresponding voltages is calculated, thereby correcting the predicted data for these currents and corresponding voltages in the prediction table.
[0058] Scenario 3: Considering temperature factors, a correction of -4mV / ℃ is applied at different temperatures. The ambient temperature is recorded simultaneously with the discharge data. Assuming the initial prediction time is t0, the initial measured voltage is V0, the actual temperature is T, and the reference temperature is T0, the voltage change ΔV = (T - T0) × (-4mV / ℃) is first calculated based on the effect of temperature change on voltage. Then, the prediction time is adjusted according to the relationship between voltage and remaining time to obtain the corrected prediction time t. new For example, if the remaining battery capacity is approximately linearly related to the voltage, it can be determined according to... This is used to correct the predicted time and update the remaining time data for the corresponding temperature, current, and voltage combinations in the prediction table.
[0059] After completing the above correction operations, the entire UPS remaining power supply time prediction process ends.
[0060] One embodiment of the present invention provides a system for predicting the remaining power supply time of a UPS power supply in a wind farm transformer substation, comprising: The remaining power prediction module, after starting to discharge, inputs the real-time remaining power of the UPS, the power information of the power supply equipment, and the UPS characteristic information into the machine learning model to predict the current remaining power. The correction module determines whether the UPS needs charging based on the predicted remaining power. If charging is required, the prediction ends; otherwise, it checks whether the UPS discharge current has changed. If the discharge current at the current moment changes compared to the current at the previous moment, the voltage at the previous moment is recorded, and after a delay, the corrected voltage is calculated and the voltage corresponding to the discharge current at the current moment is updated in the prediction table. The corrected voltage is obtained by obtaining the difference between the actual voltage and the theoretical voltage at different voltage points through a reference current discharge experiment. The prediction table records the remaining power supply time to reach a specific current and a specific voltage. If the discharge current at the current moment has not changed compared to the current at the previous moment, record the current discharge time, calculate the remaining power supply time based on the current discharge time, and update the prediction table. The time prediction module checks in real time whether the current voltage has reached the cutoff value: if it has not reached it, it returns to S2; otherwise, it selects the corresponding correction method according to the discharge mode to predict the remaining power supply time and updates the remaining power supply time in the prediction table.
[0061] In another embodiment of the present invention, an electronic device is provided, comprising a processor and a memory. The memory stores a computer program, which includes program instructions. The processor executes the program instructions stored in the computer storage medium. The processor may be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. It is the computing and control core of the terminal, suitable for implementing one or more instructions, specifically suitable for loading and executing one or more instructions to achieve a corresponding method flow or corresponding function. The processor described in this embodiment of the present invention can be used in the operation of a method for predicting the remaining power supply time of a UPS power supply in a wind farm transformer substation.
[0062] In another embodiment of the present invention, a storage medium is provided, specifically a computer-readable storage medium (Memory), which is a memory device in a terminal device for storing programs and data. It is understood that the computer-readable storage medium here can include both the built-in storage medium in the terminal device and extended storage media supported by the terminal device; it can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device. The computer-readable storage medium provides storage space that stores the terminal's operating system. Furthermore, the storage space also stores one or more instructions suitable for loading and execution by a processor, which can be one or more computer programs (including program code). It should be noted that more specific examples (a non-exhaustive list) of the computer-readable storage medium include: an electrical connection having one or more wires, a portable disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.
[0063] Computer-readable storage media also include data signals propagated in baseband or as part of a carrier wave, carrying readable program code. Such propagated data signals can take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. A readable storage medium can also be any readable medium other than a readable storage medium that can send, propagate, or transmit a program for use by or in connection with an instruction execution system, apparatus, or device. The program code contained on the readable storage medium can be transmitted using any suitable medium, including but not limited to wireless, wired, optical fiber, RF, etc., or any suitable combination thereof.
[0064] Program code for performing the operations of this invention can be written in any combination of one or more programming languages, including object-oriented programming languages such as Java and C++, and conventional procedural programming languages such as C or similar languages. The program code can execute entirely on the user's computing device, partially on the user's device, as a standalone software package, partially on the user's computing device and partially on a remote computing device, or entirely on a remote computing device or server. In cases involving remote computing devices, the remote computing device can be connected to the user's computing device via any type of network, including a local area network (LAN) or a wide area network (WAN), or it can be connected to an external computing device (e.g., via the Internet using an Internet service provider).
[0065] One or more instructions stored in a computer-readable storage medium can be loaded and executed by a processor to implement the corresponding steps of the method for predicting the remaining power supply time of the UPS power supply in the wind farm transformer substation in the above embodiments.
[0066] The above are merely preferred embodiments of the present invention and are not intended to limit the present invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.
Claims
1. A method for predicting the remaining power supply time of a UPS power supply in a wind farm transformer substation, characterized in that, Includes the following steps: S1, after discharge begins, input the real-time remaining power of the UPS, the power information of the power supply equipment, and the UPS characteristic information into the machine learning model to predict the current remaining power. S2, based on the predicted remaining power, determine whether the UPS needs charging: if charging is needed, end the prediction; otherwise, check if the UPS discharge current changes. If the discharge current at the current moment changes compared to the current at the previous moment, record the voltage at the previous moment, calculate the corrected voltage after a delay, and update the voltage corresponding to the discharge current at the current moment in the prediction table. The corrected voltage is obtained by using a reference current discharge experiment to determine the difference between the actual voltage and the theoretical voltage at different voltage points. The prediction table records the remaining power supply time to reach a specific current and voltage. If the discharge current at the current moment has not changed compared to the current at the previous moment, record the current discharge time, calculate the remaining power supply time based on the current discharge time, and update the prediction table. S3, check in real time whether the current voltage has reached the cutoff value: if it has not reached it, return to S2; otherwise, select the corresponding correction method according to the discharge mode to predict the remaining power supply time and update the remaining power supply time in the prediction table.
2. The method for predicting the remaining power supply time of a UPS power supply in a wind farm transformer substation according to claim 1, characterized in that, The machine learning model used is a linear regression, support vector machine, or neural network model.
3. The method for predicting the remaining power supply time of a UPS power supply in a wind farm transformer substation according to claim 1, characterized in that, The corrected voltage is specifically: in, voltage point The actual voltage; voltage point The theoretical voltage.
4. The method for predicting the remaining power supply time of a UPS power supply in a wind farm transformer substation according to claim 1, characterized in that, The prediction table is initially set as follows: The current is divided into multiple levels, and the voltage is divided into levels according to a set value. The initial data is obtained by using the current integration method, the open-circuit voltage method, or a combination of the two to fill the prediction table.
5. The method for predicting the remaining power supply time of a UPS power supply in a wind farm transformer substation according to claim 4, characterized in that, The method of combining the two specifically includes: The charging and discharging status of the battery is recorded using the current integration method, and the current SOC1 is obtained. The open-circuit voltage of the battery is measured using the open-circuit voltage method, and the current SOC2 is calculated based on the SOC-open-circuit voltage curve. Calculate the current SOC based on SOC1 and SOC2; SOC = k1SOC1 + k2SOC2; Where k1 and k2 are weighting coefficients.
6. The method for predicting the remaining power supply time of a UPS power supply in a wind farm transformer substation according to claim 1, characterized in that, The step of selecting the corresponding correction method based on the discharge mode to predict the remaining power supply time is as follows: During constant current discharge, the remaining power supply time corresponding to the current-cutoff voltage in the prediction table is updated with the actual discharge time; During multi-current discharge, according to the formula Calculate the remaining power supply time; among which, To correct the voltage, Select the known range that is closest to the current. When temperature affects, according to After correcting the voltage, adjust the remaining power supply time according to the voltage-time relationship.
7. A computer program product, said computer program product comprising computer instructions, characterized in that, The computer instructions instruct the computer to execute the method for predicting the remaining power supply time of the UPS power supply in a wind farm transformer substation as described in any one of claims 1-6.
8. An electronic device, characterized in that, The method includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the method for predicting the remaining power supply time of a UPS power supply in a wind farm transformer substation as described in any one of claims 1-6.
9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements the method for predicting the remaining power supply time of a UPS power supply in a wind farm transformer substation as described in any one of claims 1-6.
10. A system for predicting the remaining power supply time of a UPS power supply in a wind farm transformer substation, characterized in that, include: The remaining power prediction module, after starting to discharge, inputs the real-time remaining power of the UPS, the power information of the power supply equipment, and the UPS characteristic information into the machine learning model to predict the current remaining power. The correction module determines whether the UPS needs charging based on the predicted remaining power. If charging is required, the prediction ends; otherwise, it checks whether the UPS discharge current has changed. If the discharge current at the current moment changes compared to the current at the previous moment, record the voltage at the previous moment, calculate the corrected voltage after a delay, and update the voltage corresponding to the discharge current at the current moment in the prediction table. The corrected voltage is obtained by using a reference current discharge experiment to determine the difference between the actual voltage and the theoretical voltage at different voltage points. The prediction table records the remaining power supply time to reach a specific current and voltage. If the discharge current at the current moment has not changed compared to the current at the previous moment, record the current discharge time, calculate the remaining power supply time based on the current discharge time, and update the prediction table. The time prediction module checks in real time whether the current voltage has reached the cutoff value: if it has not reached it, it returns to S2; otherwise, it selects the corresponding correction method according to the discharge mode to predict the remaining power supply time and updates the remaining power supply time in the prediction table.