A cloud processing-based reactive power regulation method and system for electric vehicle charging piles

Abstract

The present application relates to a kind of reactive power regulation method and system based on cloud processing electric vehicle charging pile, the method includes: through intelligent electric meter and charging pile, real-time electrical quantity and operating condition data are collected;Using historical data and machine learning algorithm, the reactive power regulation range satisfying voltage safety constraint is predicted;Based on target power factor or grid regulation instruction, reactive power optimization model is established, and the reactive power distribution scheme of charging pile is calculated;Optimization instruction is issued to each charging pile to execute, and strategy is dynamically adjusted.The system includes data acquisition unit, communication unit, cloud processing unit, instruction issuing unit and user interface unit, realizes the centralized management and optimization of charging pile reactive power regulation.Compared with traditional mode, the present application makes full use of the bidirectional reactive power regulation capability of charging pile, reduces equipment investment, improves energy efficiency, and has high efficiency and flexibility, adapts to the demand of future power grid for reactive power regulation, and has wide application prospect.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of charging piles, and particularly relates to a reactive power regulation method and system for electric vehicle charging piles based on cloud processing. BACKGROUND

[0002] With the widespread application of distributed photovoltaic power generation and electric vehicle charging piles, the power consumption characteristics of the user side have changed significantly. The connection of distributed photovoltaic power generation makes the power supply of the user side have strong randomness and volatility; at the same time, the electric vehicle charging load has the characteristics of fast instantaneous change and large power demand. These factors lead to rapid changes in the user side load and frequent voltage fluctuations, bringing great challenges to the stable operation of power equipment.

[0003] The traditional voltage and reactive power regulation method mainly relies on transformer tap adjustment and capacitor switching, but these solutions usually have the problems of slow response speed, large size and high cost, and are not suitable for the scene of photovoltaic power generation uncertainty and electric vehicle charging load rapid change. At the same time, because the user side is generally only configured with capacitors, it can only provide reactive power but cannot absorb reactive power, and cannot regulate the overvoltage in the case of high photovoltaic generation or low district load.

[0004] In addition, when issuing electricity bills, the power supply company will adjust the electricity bill according to the average power factor during the billing period for power users with installed capacity exceeding a certain value, i.e. power regulation electricity. If the power factor is lower than the evaluation standard, the power regulation electricity penalty will be added, otherwise part of the electricity will be exempted as power regulation electricity reward.

[0005] At the same time, with the increasing demand for reactive power regulation of the power grid, the power user side may gradually be required to participate in the reactive power regulation of the power grid in the future. By responding to the demand for reactive power regulation of the power grid, power users can obtain regulation benefits.

[0006] Although electric vehicle charging piles have the potential for bidirectional reactive power regulation due to the use of power electronic charging modules, there is currently a lack of effective methods to integrate and utilize these dispersed reactive power resources. The current technical defects not only limit the efficiency of power grid management, but also fail to fully utilize the potential of electric vehicle charging facilities in reactive power regulation and voltage quality improvement. Therefore, there is an urgent need for a new reactive power regulation method that can fully integrate and utilize the reactive power regulation capabilities of electric vehicle charging piles to improve the reactive power management capabilities of the power user side and directly reflect in the electricity bill savings and voltage quality improvement of users. SUMMARY

[0007] The application relates to a reactive power optimization method and system for electric vehicle charging piles based on cloud processing, aiming to improve the power factor of power users, improve voltage quality, and respond to the demand for reactive power regulation of the power grid.

[0008] The first aspect of the present application provides a cloud-based reactive power optimization method for electric vehicle charging piles, the method comprising the following steps:

[0009] Data acquisition step: real-time acquisition of electrical quantities including power factor, active power, reactive power and voltage through smart meters, while collecting operating state data of each charging pile, the operating state data including the working state of the charging pile, real-time active power and real-time reactive power, and uploading the electrical quantities and the operating state data to a cloud processing unit for storage, forming historical data;

[0010] Reactive power regulation range prediction step: based on the historical data, a relationship model between active power, reactive power, voltage, voltage difference and reactive power regulation amount is established through a machine learning algorithm, the relationship model including:

[0011] According to the pre-set upper limit value and lower limit value of voltage, the voltage difference between the current voltage and the upper limit value and lower limit value is calculated;

[0012] The voltage difference, current active power and current reactive power are input into the relationship model to predict the upper limit and lower limit of reactive power regulation that meets the voltage safety constraint;

[0013] Reactive power optimization calculation step:

[0014] According to the user-set power factor target value or the received grid reactive power regulation instruction, the target reactive power is determined;

[0015] Based on the real-time power and maximum apparent power of each charging pile, the upper and lower limits of reactive power regulation of each charging pile are calculated;

[0016] Under the condition of meeting the reactive power regulation constraints of each charging pile, the reactive power output distribution scheme of each charging pile is calculated through an optimization algorithm;

[0017] Instruction execution step:

[0018] The reactive power output distribution scheme is converted into a reactive power regulation instruction and issued to each charging pile;

[0019] Real-time monitoring of the execution of the charging pile and collection of the electrical quantities after reactive power regulation;

[0020] According to the monitoring results, it is judged whether the target requirement is met, if not, the reactive power optimization calculation step is returned to recalculate the distribution scheme.

[0021] Further, the reactive power regulation range prediction utilizes a machine learning algorithm to model the historical data to accurately predict the reactive power regulation range that meets the voltage safety constraint condition.

[0022] Further, the objective function is:

[0023]

[0024] Wherein, Q target is the total reactive power target value of the charging pile, Q i is the reactive power output of the i-th charging pile, and Z is the absolute value of the maximum reactive power adjustment of the charging pile.

[0025] The second aspect of the application also provides a cloud-based electric vehicle charging pile reactive power adjustment system, which comprises:

[0026] a data acquisition unit for acquiring electrical quantity and charging pile operating state data;

[0027] a communication unit for uploading data and receiving instructions;

[0028] a cloud processing unit comprising a data storage and preprocessing module, a reactive power adjustment range prediction module, and a reactive power optimization module;

[0029] a control instruction issuing unit for sending reactive power adjustment instructions to the charging pile;

[0030] a user interface unit for visualizing the system operating state and configuration parameters.

[0031] Further, the cloud processing unit utilizes machine learning algorithms to predict the reactive power adjustment range and adopts mathematical optimization methods to calculate the optimal reactive power distribution scheme.

[0032] Through the above method and system, the application realizes centralized management and optimized utilization of the reactive power adjustment capability of electric vehicle charging piles, improves the reactive power management capability of power users, improves the voltage quality, reduces the electricity cost caused by low power factor, and can respond to the reactive power adjustment demand of the power grid, and has good application prospect.

[0033] Compared with the prior art, the application has the following advantages:

[0034] The application unifies the dispersed reactive power adjustment capability of charging piles through cloud processing technology, effectively utilizes the application value of electric vehicle charging piles in reactive power adjustment through machine learning technology and advanced reactive power optimization algorithm. By adjusting the reactive power, the application can significantly improve the power factor of power users, improve the voltage quality, and enhance the overall stability of the power grid. Compared with the traditional reactive power compensation device, the application fully utilizes the reactive power adjustment potential of the electric vehicle charging pile itself, reduces or replaces the construction demand of the conventional reactive power compensation equipment, thereby reducing the equipment investment cost.

[0035] In addition, the cloud-based centralized management and optimization function provided by the present application makes the reactive power regulation more flexible and efficient. Users can not only monitor the system operation status in real time, but also flexibly adjust the reactive power regulation strategy according to actual needs to cope with various complex power grid environments. This efficient and intelligent regulation method not only helps power users reduce electricity bills due to low power factor, but also improves the opportunity for power users to obtain reactive power compensation, thereby significantly reducing electricity costs.

[0036] In the scenario of multiple charging stations working together, the present application can coordinate the reactive power output of multiple charging stations through the cloud system to achieve overall optimization of the regional power grid. This not only improves the operation efficiency of the regional power grid, but also provides an effective means of reactive power regulation for the power grid, which helps to stabilize the operation of the power grid and reduce the risk of voltage fluctuation.

[0037] As the demand for reactive power regulation of future power grids increases, the system of the present application can also interface with the power grid dispatch center to respond to external reactive power regulation instructions and provide fast and accurate reactive power regulation services for the power grid. This highly automated and intelligent regulation system will play an important role in the future power market, helping power users and power grid operators achieve higher operating benefits.

[0038] In summary, the present application not only significantly improves the reactive power regulation capability of electric vehicle charging piles, but also reduces equipment investment and operating costs through centralized management and optimization of the cloud, improving the electricity utilization efficiency of power users and the overall stability of the power grid. The present application has wide application prospects and is particularly suitable for various power grid environments that need to optimize power factor and improve energy efficiency. BRIEF DESCRIPTION OF DRAWINGS

[0039] Figure 1 is a flowchart of the electric vehicle charging pile reactive power regulation method based on cloud processing proposed by the present application

[0040] Figure 2 is a system architecture diagram of the electric vehicle charging pile reactive power regulation system based on cloud processing proposed by the present application

[0041] Figure 3 is a flowchart of Example 1 DETAILED DESCRIPTION

[0042] The electric vehicle charging pile reactive power optimization method and system based on cloud processing of the present application will be described in detail below in conjunction with the drawings and examples.

[0043] Example 1: Cloud-based reactive power optimization method

[0044] The present embodiment provides an electric vehicle charging pile reactive power optimization method based on cloud processing, the specific steps of which are as follows:

[0045] Data Collection

[0046] Install smart meters at the grid gateway to collect the following electrical quantities in real time:

[0047] Active Power P: Represents the real-time active power consumption of the power user.

[0048] Reactive Power Q: Represents the real-time reactive power consumption of the power user.

[0049] Voltage V: Real-time voltage value at the grid gateway.

[0050] Power Factor PF: Represents the power factor of the power user.

[0051] From each electric vehicle charging pile, obtain the following data:

[0052] Running Status: Whether the charging pile is in running, idle or fault state.

[0053] Output Power: The current charging pile's output active power and reactive power.

[0054] Reactive Power Adjustment Range Prediction

[0055] Using the collected historical data, establish a relationship model between the grid parameters and the reactive power adjustment. The specific implementation is as follows:

[0056] Data Preprocessing: Clean the historical data, eliminate outliers and missing values, and ensure data quality.

[0057] Feature Extraction: Select key features such as active power P, reactive power Q, voltage V, voltage change ΔV (difference from the previous time point).

[0058] Model Training: Use machine learning algorithms (such as LightGBM) to train the prediction model, establish the relationship between active power, reactive power, voltage, voltage change and reactive power adjustment.

[0059] Real-time Prediction: Input the current real-time data, use the trained model to predict the reactive power adjustment range under the condition of meeting the voltage safety constraint, i.e. the upper and lower limits of reactive power Q min and Q max .

[0060] According to the set voltage upper limit Vmax and Vmin, calculate the voltage difference

[0061] ΔV up = V max -V current

[0062] ΔV down = Vcurrent -V min

[0063] The above data constitutes a feature vector, which is input into the trained machine learning model to predict the reactive power up and down adjustment limit ΔQup,limit,ΔQdown,limit that meets the voltage safety constraint in the current state, and then the upper and lower limits of the reactive power are calculated according to the current reactive power

[0064] Q max = Q current + ΔQ up,limit

[0065] Q min = Q current - ΔQ down,limit

[0066] Reactive power optimization calculation

[0067] According to the predicted reactive power adjustment range, the reactive power optimization calculation is performed:

[0068] Determine the target reactive power Q target :

[0069] Power factor control mode: according to the set power factor target PF target and the current active power P, the target reactive power is calculated:

[0070] Q target = P x tan(arccos(PF target ))

[0071] Grid reactive power adjustment mode: if the grid reactive power adjustment instruction is received, set Q target to the instruction value.

[0072] Range constraint: ensure that Q target is within the predicted reactive power adjustment range:

[0073] Q min ≤ Q target ≤ Q max

[0074] If it exceeds the range, take the corresponding upper and lower limit value.

[0075] Establish an optimization model to optimize the distribution of the reactive power output of each charging pile:

[0076] Objective function: the difference between the total reactive power of the charging pile and the target reactive power is as small as possible, and in order to prevent the optimization result from being unstable and leading to frequent adjustment, the maximum reactive power adjustment amount absolute value Z is introduced:

[0077]

[0078] wherein Q target is the total reactive power target value of the charging piles, Q i is the reactive power output of the i-th charging pile, and Z is the absolute value of the maximum reactive power adjustment amount:

[0079] Z = max(|ΔQ i |)

[0080] Constraints:

[0081] Charging pile capacity limit:

[0082]

[0083] According to the power circle characteristics of AC / DC:

[0084]

[0085] Q i,min = -Q i,max

[0086] wherein Q i,min and Q i,max are the lower and upper limits of the reactive power of the charging pile, S i,max is the maximum apparent power of the charging pile, P i is the real-time power of the charging pile.

[0087] Solution method: the above optimization problem is solved by using the SCIP solver to obtain the optimal reactive power output distribution scheme of each charging pile.

[0088] Instruction issuance and execution

[0089] Instruction generation: according to the optimization calculation result, generate reactive power adjustment instructions, including charging pile number and corresponding reactive power set value.

[0090] Instruction issuance: through the communication unit, issue the instructions to each charging pile.

[0091] Execution monitoring: monitor the execution of the charging pile to ensure that the instructions are correctly executed.

[0092] Dynamic adjustment

[0093] According to the real-time monitoring results and new collected data, the above steps are executed in a loop to dynamically adjust the reactive power adjustment strategy to adapt to the changes of the power grid and the load.

[0094] Example two: reactive power adjustment system based on cloud

[0095] The embodiment provides a reactive power adjustment system for electric vehicle charging piles based on cloud processing, and the structure is as follows:

[0096] Data acquisition unit

[0097] Gateway smart meter: installed at the connection between the power user and the power grid, real-time acquisition of electrical parameters.

[0098] Charging pile data acquisition module: embedded in the charging pile, collecting the running state and power data of the charging pile.

[0099] Communication unit

[0100] Deployed in the charging pile controller, using wired or wireless communication mode, data transmission and instruction transceiver with the cloud server.

[0101] Cloud processing unit

[0102] Data storage and preprocessing module: store historical data, data cleaning and preprocessing.

[0103] Reactive power regulation range prediction module: based on historical data and real-time data, predict the reactive power regulation range.

[0104] Reactive power optimization module: perform reactive power optimization calculation, generate optimal regulation scheme.

[0105] Control instruction issuing unit

[0106] Send the reactive power regulation instruction of cloud computing to each charging pile to control its reactive power output.

[0107] User interface unit

[0108] Web visualization interface: for users to view system running status, set power factor target, monitor voltage quality, etc.

[0109] Parameter configuration interface: allows users to configure system parameters such as power factor target value, voltage safety range, etc.

[0110] Example three: respond to grid reactive power regulation demand

[0111] When the grid appears reactive power shortage, the grid dispatching center sends reactive power regulation instructions to power users. Using the system of the application, power users can respond to the grid's reactive power regulation demand, the specific implementation steps are as follows:

[0112] Receive instructions

[0113] The system receives the reactive power regulation instructions issued by the grid dispatching center through the communication unit, such as increasing the reactive power by 100kVar.

[0114] Adjust target value

[0115] Set Q target to the current reactive power plus the adjustment amount required by the grid.

[0116] Re-calculate the reactive power output distribution scheme of each charging pile according to the step in the reference embodiment one of the reactive power optimization calculation.

[0117] Execution and feedback

[0118] Issue adjustment instructions to the charging pile to adjust the reactive power output.

[0119] The execution result and the current reactive power are fed back to the power grid dispatching center.

Claims

1. A reactive power regulation method for electric vehicle charging piles based on cloud processing, characterized in that, The method comprises the following steps: Data acquisition step: real-time acquisition of electrical quantities including power factor, active power, reactive power and voltage by smart meters, while collecting the operating state data of each charging pile, including the working state, real-time active power and real-time reactive power, and uploading the electrical quantities and operating state data to the cloud processing unit for storage, forming historical data; Reactive power regulation range prediction step: based on the historical data, a relationship model between active power, reactive power, voltage, voltage difference and reactive power regulation is established by a machine learning algorithm, which includes: According to the pre-set upper and lower voltage limits, the voltage difference between the current voltage and the upper and lower limits is calculated; The voltage difference, current active power and current reactive power are input into the relationship model to predict the upper and lower limits of reactive power regulation that meet the voltage safety constraint; Reactive power optimization calculation step: Determine the target reactive power according to the user-set power factor target value or the received grid reactive power regulation instruction; Based on the real-time power and maximum apparent power of each charging pile, calculate the upper and lower limits of reactive power regulation for each charging pile; Under the condition of meeting the reactive power regulation constraints of each charging pile, calculate the reactive power output distribution scheme of each charging pile by optimization algorithm; Instruction execution step: Convert the reactive power output distribution scheme into a reactive power regulation instruction and issue it to each charging pile; Real-time monitoring of the execution of the charging pile and collecting the electrical quantities after reactive power regulation; According to the monitoring results, judge whether the target requirements are met, if not, return to the reactive power optimization calculation step to recalculate the distribution scheme; where the objective function is: wherein, is the total reactive power target value of the charging pile, is the reactive power output of the charging pile, is the reactive power output of the charging pile, is the absolute value of the maximum reactive power adjustment amount of the charging pile. 2.The cloud processing based reactive power adjustment method for electric vehicle charging piles according to claim 1, wherein, The reactive power regulation range prediction utilizes machine learning algorithm to model the historical data to accurately predict the reactive power regulation range that meets the voltage safety constraint.

3. A cloud processing based reactive power regulation system for electric vehicle charging station, characterized in that, The system is used to execute the method as claimed in any one of claims 1-2, comprising: a data acquisition unit for collecting electrical quantities and charging pile operating state data; a communication unit for uploading data and receiving instructions; a cloud processing unit including data storage and preprocessing module, reactive power regulation range prediction module and reactive power optimization module; a control instruction issuing unit for sending reactive power regulation instructions to charging piles; a user interface unit for visualizing system operating state and configuration parameters.

4. The reactive power regulation system for electric vehicle charging piles based on cloud processing according to claim 3, characterized in that, The cloud processing unit utilizes machine learning algorithm to predict reactive power regulation range, and adopts mathematical optimization method to calculate optimal reactive power distribution scheme.

Images