In the rapidly evolving landscape of electric vehicle (EV) infrastructure, the reliability and accuracy of EV charging stations are paramount. As a researcher focused on metering technologies, I have observed that faults in EV charging stations not only affect operational efficiency but also compromise metering fairness, user safety, and the sustainable growth of the新能源 industry. The EV charging station serves as the core metering device during charging processes, and any inaccuracies can lead to financial discrepancies, erode user trust, and trigger disputes. Therefore, developing robust remote measurement solutions is critical to ensuring transparent energy transactions and enhancing the intelligence of EV charging networks. This article presents a comprehensive study on a remote calibration device for EV charging stations, incorporating non-contact measurement techniques, wireless communication modules, and a cloud-based data analysis system. Through experimental validation, the device demonstrates high accuracy and stability, aligning with industry standards and paving the way for widespread application in EV charging station metering.
The importance of fault analysis in EV charging stations cannot be overstated. From a metering fairness perspective, the EV charging station directly determines the accuracy of energy measurements and the integrity of billing systems. In commercial settings, such as shared EV charging stations or corporate facilities, even minor deviations can result in overcharging or undercharging, undermining consumer rights. Moreover, operational failures in EV charging stations can interrupt services, leading to reputational damage for operators. Thus, establishing efficient monitoring mechanisms is essential for maintaining equitable energy exchange. This research addresses these challenges by proposing a remote measurement scheme that leverages additional diagnostic modules, avoiding the pitfalls of AI-based approaches. By focusing on hardware-based solutions, we aim to provide a reliable, cost-effective method for real-time monitoring and calibration of EV charging stations.
Remote measurement methods for EV charging stations generally fall into two categories: those based on artificial intelligence (AI) and big data, and those involving the installation of additional diagnostic modules. AI-based methods rely on historical data and machine learning algorithms to identify complex patterns and anomalies in parameters like voltage, current, temperature, and power fluctuations. For instance, AI models can detect subtle fault indicators that traditional methods might miss, enabling rapid response and predictive maintenance. However, these approaches face significant limitations. Training AI models requires vast amounts of data, which is often imbalanced—common faults may be well-represented, but critical, rare failures are scarce. This data imbalance reduces the model’s generalization ability, making it less effective in diverse real-world environments. Additionally, AI systems are probabilistic and may struggle with outlier events, such as those caused by extreme weather or grid instability, leading to unreliable fault detection. Environmental variability further complicates AI performance, as models trained on one dataset may not adapt to different regional conditions, such as coastal corrosion or freezing temperatures affecting EV charging stations.
In contrast, the method of adding extra diagnostic modules offers a more straightforward and stable approach. These modules directly monitor key parameters like current, voltage, and temperature, providing real-time alerts and shutdown capabilities in case of abnormalities. Since they operate independently of the EV charging station’s main control system, they remain functional even during software failures, enhancing safety and reliability. Hardware-based modules are less susceptible to environmental interference and do not require extensive data processing, resulting in consistent performance. Although the initial cost per module is low, scaling up to large networks of EV charging stations can incur significant expenses. Maintenance is also a consideration, as these modules may degrade over time due to factors like humidity or aging, necessitating periodic calibration. From a metrological standpoint, however, this method ensures trade settlement fairness and protects consumer interests, making it the preferred choice for this study. The following sections detail the design and implementation of a remote calibration device based on this approach.
The remote communication scheme for the calibration device is designed for versatility and ease of installation across various EV charging station scenarios. It incorporates both Wi-Fi and LTE modules to facilitate seamless data transmission. The Wi-Fi module enables local communication with external devices like tablets or computers, while the LTE module provides internet connectivity for remote monitoring via cloud platforms. This dual-module approach allows users to access real-time data from EV charging stations from anywhere, enhancing operational efficiency. The connection setup involves linking the calibration device to the EV charging station’s circuitry without disrupting existing operations, as illustrated in the schematic diagram. This ensures that the device can be deployed in diverse environments, from urban charging hubs to remote EV charging stations, without requiring major modifications.
The backend data acquisition and analysis system forms the core of the remote calibration framework for EV charging stations. It begins by collecting initial energy measurement data from the EV charging station. In the data acquisition layer, issues related to data governance are addressed to ensure quality, including accuracy, completeness, consistency, validity, timeliness, and accessibility. The data layer then processes the raw energy data through analog-to-digital conversion, followed by resource comparison and integration. This involves aggregating data records and performing unified analysis against a database to yield high-accuracy measurement results. In the capability layer, the processed data is categorized, analyzed, and stored in a distributed manner, building a support system for generating calibration reports. Finally, in the application layer, the data is formatted and consolidated to produce technical reports that comply with metrological requirements for EV charging stations. This structured approach ensures that the system can handle the dynamic charging profiles of EV charging stations, such as rapid current changes during start and end phases, while maintaining precision.
The metering module of the calibration device is composed of several key components: current input module, voltage input module, multiplier module, digital signal processing unit, error processing module, and data output module. According to existing metrological verification protocols for EV charging stations, the energy error is determined based on the indicated energy value. The calibration device calculates this error to perform verification. Given that the highest accuracy class for EV charging stations is typically level 1, the calibration device must achieve at least level 0.2 accuracy. Since DC EV charging stations can output currents up to 300 A, the metering module employs an indirect measurement method for current. The current input module uses a wide-range current/voltage conversion device to prevent measurement deviations caused by frequent range switching during charging. To capture transient current variations—such as those occurring at the beginning and end of charging—the module incorporates a response time buffer. A high-precision DC current transformer based on non-contact measurement principles is selected for the current detection module to meet the demanding accuracy needs of EV charging stations.
Additionally, the device is equipped with a DC comparator, a non-contact current measurement tool based on magnetic induction. This comparator consists of a环形磁芯 with wound coils and can be clamped around the conductor without electrical connection, making it ideal for EV charging station applications where circuit integrity is crucial. It features a dual-core structure: core A is made of high-permeability soft magnetic material, and core B uses low-permeability ferrite. This design cancels out odd harmonic components while amplifying even harmonics, allowing for precise extraction of signal-related even harmonics. The excitation winding is connected to a square-wave oscillator and wound separately on core A, while the detection winding Ws and balance winding W2 are wound on both cores, and the measured winding W1 passes through them. The working principle can be summarized using the following equations related to magnetic modulation:
$$ \Phi_A = k_A \cdot I_{\text{dc}} $$
$$ \Phi_B = k_B \cdot I_{\text{dc}} $$
where $\Phi_A$ and $\Phi_B$ represent the magnetic fluxes in cores A and B, respectively, $k_A$ and $k_B$ are constants dependent on the core materials, and $I_{\text{dc}}$ is the DC current being measured. The even harmonic voltage $V_{\text{even}}$ induced in the detection winding is proportional to the current:
$$ V_{\text{even}} = C \cdot I_{\text{dc}} $$
where $C$ is a calibration constant. This setup ensures high accuracy and minimal interference, which is vital for the precise operation of EV charging stations.
Considering the charging characteristics of EVs, where power gradually changes, optimizing the energy integration method is essential for reducing measurement errors. The trapezoidal integration method, combined with an increased number of sampling points, effectively approximates the power curve. The energy $E$ can be calculated as:
$$ E = \sum_{i=1}^{n} \frac{P_i + P_{i-1}}{2} \cdot \Delta t $$
where $P_i$ is the power at sample $i$, and $\Delta t$ is the sampling interval. This approach enhances measurement accuracy for EV charging stations by better fitting the dynamic charging profiles.
To validate the performance of the developed remote calibration device for EV charging stations, comparative experiments were conducted alongside a TD1320 EV charging station field tester. Both devices were used simultaneously on five different EV charging stations, with tests performed under identical conditions of voltage and current. The results, summarized in the table below, demonstrate the accuracy and reliability of our device. The error percentage is computed as:
$$ \text{Error} = \frac{E_{\text{measured}} – E_{\text{reference}}}{E_{\text{reference}}} \times 100\% $$
where $E_{\text{reference}}$ is the energy value from the EV charging station, and $E_{\text{measured}}$ is the value obtained by the test device. The En value, derived from JJF 1117—2010 (Metrological Comparison), is used to assess consistency, with |En| ≤ 1 indicating acceptable results.
| No. | Voltage (V) | Current (A) | Charging Station Energy (kWh) | TD1320 Measurement (kWh) | TD1320 Error (%) | Our Device Measurement (kWh) | Our Device Error (%) | En Value |
|---|---|---|---|---|---|---|---|---|
| 1 | 600 | 100 | 3.05 | 3.041 | 0.30 | 3.040 | 0.33 | -0.2 |
| 2 | 600 | 100 | 3.09 | 3.075 | 0.50 | 3.074 | 0.52 | -0.1 |
| 3 | 600 | 100 | 3.06 | 3.036 | 0.79 | 3.037 | 0.76 | 0.2 |
| 4 | 600 | 100 | 3.13 | 3.067 | 2.06 | 3.066 | 2.10 | -0.3 |
| 5 | 600 | 100 | 3.05 | 3.044 | 0.20 | 3.043 | 0.22 | -0.2 |
The experimental data shows that all |En| values are within the acceptable range of 1, confirming that the measurements from our device are consistent with the reference standard. This validates the effectiveness of the remote calibration scheme for EV charging stations, highlighting its potential for widespread adoption in metrological verification.
In conclusion, the remote calibration device developed in this study offers a robust solution for enhancing the accuracy and reliability of EV charging stations. By integrating Wi-Fi and LTE modules, it enables seamless remote monitoring and data transmission, while the non-contact DC current measurement technique avoids interference with existing circuits. The experimental results affirm its high precision and stability, meeting the requirements of relevant metrological standards. Looking ahead, future work will focus on incorporating AI technologies to optimize fault diagnosis models, improving module performance to reduce operational costs, and exploring low-cost industrialization strategies for large-scale deployment. These efforts will contribute to the digital transformation of EV charging station networks, supporting the growth of sustainable transportation infrastructure. As the demand for EV charging stations continues to rise, such advancements will play a crucial role in ensuring fair and efficient energy management.