New energy power station electric energy quality detection device

Abstract

The invention relates to a new energy power station electric energy quality detection device. According to the technical scheme, the system comprises a multi-channel data acquisition module, a data analysis module based on deep learning, a 5G communication module and a touch display module; the multi-channel data acquisition module is used for simultaneously acquiring voltage, current, frequency, harmonic wave and other electric energy parameters of different areas of the new energy power station; the data analysis module based on deep learning performs feature extraction and anomaly recognition on the acquired multi-channel data to generate an electric energy quality evaluation report; the 5G communication module transmits the evaluation report to a remote monitoring center in real time; the touch display module is used for displaying electric energy parameters and evaluation reports and supporting manual operation instruction input, and the fault diagnosis module is used for monitoring hardware and software of the device in real time and sending out an alarm when a fault is found. The method has the beneficial effects that the data acquisition capability and precision are improved, and a complex scene is accurately identified.

Description

Technical Field

[0001] This invention belongs to the field of power quality testing technology and relates to a power quality testing device for new energy power plants. Background Technology

[0002] As the global energy structure shifts towards clean energy, the installed capacity of new energy power plants continues to climb, and their power quality has an increasingly significant impact on the safe and stable operation of the power grid. Power quality problems such as voltage sags and dips, three-phase imbalance, and harmonic distortion can lead to shortened lifespan of grid-connected equipment, malfunctions of relay protection systems, and even regional power grid failures.

[0003] Existing power quality detection devices have the following limitations:

[0004] Most devices only support single-channel or 3-5 limited-channel acquisition, which cannot cover the multi-area monitoring needs of the distributed layout of new energy power plants, resulting in insufficient acquisition capabilities, such as different collector areas of photovoltaic arrays and different wind turbines in wind farms. Relying on traditional algorithms such as Fourier transform and wavelet analysis, the accuracy of identifying non-steady-state power quality problems (such as random fluctuations in wind power and intermittent output of photovoltaics) is low, and the analysis precision is poor. Using 4G or wired Ethernet transmission, there are problems such as weak signal coverage and transmission delays exceeding 100ms in remote areas. Moreover, data encryption often uses non-national cryptographic algorithms such as DES, which have security vulnerabilities and poor communication performance.

[0005] Therefore, the present invention provides a power quality detection device for new energy power plants to solve the above problems. Summary of the Invention

[0006] In view of the problems existing in the prior art, the present invention discloses a power quality detection device for a new energy power station. The technical solution adopted includes a multi-channel data acquisition module, a deep learning-based data analysis module, a 5G communication module, and a touch display module. The multi-channel data acquisition module is used to simultaneously collect power parameters such as voltage, current, frequency, and harmonics in different areas of the new energy power station. The deep learning-based data analysis module performs feature extraction and anomaly identification on the collected multi-channel data and generates a power quality assessment report. The 5G communication module transmits the assessment report to a remote monitoring center in real time. The touch display module is used to display the power parameters and the assessment report, and supports manual operation command input.

[0007] As a preferred embodiment of the present invention, the multi-channel data acquisition module includes multiple high-precision Hall sensors, which are used to acquire voltage and current signals of different phase sequences, supporting 16-32 channels of synchronous acquisition with an acquisition frequency of up to 20kHz; by using multiple high-precision Hall sensors, the accuracy and capability of data acquisition can be effectively improved.

[0008] As a preferred embodiment of the present invention, the data analysis module based on deep learning adopts an improved convolutional neural network model, which is trained by a large amount of historical power quality data; by adopting the improved convolutional neural network model, the accuracy of identifying transient power quality events can be effectively improved.

[0009] As a preferred embodiment of the present invention, the 5G communication module uses a national cryptographic algorithm to encrypt the transmitted data.

[0010] As a preferred embodiment of the present invention, it further includes a fault diagnosis module, which is used to monitor the hardware and software of the device itself in real time and issue an alarm signal when a fault is detected.

[0011] As a preferred embodiment of the present invention, the touch display module supports multi-window display, which can simultaneously display power parameters and evaluation reports of different areas.

[0012] As a preferred embodiment of the present invention, it also includes a large-capacity storage module for storing the collected power parameters and the generated evaluation report, with a storage time of not less than one year.

[0013] As a preferred embodiment of the present invention, the operating temperature range of the device is -40°C to 70°C, and the relative humidity range is 5% to 95%.

[0014] The beneficial effects of this invention are:

[0015] 1. With a 16-32 channel synchronous acquisition design, it covers the multi-area monitoring needs of the distributed layout of new energy power plants. The 20kHz acquisition frequency is 4 times higher than that of traditional devices, ensuring that no transient power events are missed.

[0016] 2. The improved convolutional neural network model can improve the accuracy of identifying non-steady-state power quality problems and can accurately identify complex scenarios such as random fluctuations in wind power and intermittent output of photovoltaic power.

[0017] 3. Adopting 5G SA mode and the national cryptographic SM4 encryption algorithm, the transmission delay is controlled within 20ms, meeting the real-time and security requirements of the power system;

[0018] 4. It integrates fault self-diagnosis, 1-year data storage, and multi-window interactive functions, effectively shortening the time for locating on-site problems and significantly reducing operation and maintenance costs. Attached Figure Description

[0019] Figure 1 This is a schematic diagram of the overall framework of the present invention; Detailed Implementation

[0020] Example 1

[0021] like Figure 1 As shown, the power quality detection device for a new energy power station according to the present invention adopts a technical solution including a multi-channel data acquisition module, a deep learning-based data analysis module, a 5G communication module, a touch display module, a fault diagnosis module, a large-capacity storage module, and an environment adaptation unit.

[0022] The multi-channel data acquisition module is equipped with 24 HCS1300 Hall sensors (range ±1000A / ±1000V), of which 16 channels acquire DC parameters of each string of the photovoltaic array and 8 channels acquire AC parameters output by the inverter. After the sensor signals are conditioned by the AD8221 instrumentation amplifier, they are converted from analog to digital by the ADS8688 16-bit ADC chip with a sampling rate of 20kHz and a synchronization error between channels ≤1μs.

[0023] The deep learning-based data analysis module uses an NVIDIA Jetson Xavier NX processor and runs an improved CNN model: the input layer receives 24 channels × 1024 points of temporal data, which is processed alternately by convolutional layers (3×3 convolutional kernels, stride 1), SE attention mechanism layers, and pooling layers (2×2 pooling kernels), and then outputs 12 types of event labels through 3 fully connected layers; the model is trained on 1 million historical data of photovoltaic power plants (2018-2023), and the validation set accuracy reaches 98.7% after 100 iterations.

[0024] The 5G communication module is equipped with Huawei MH5000-31 module, supports NR NSA / SA dual mode, and uses SM4 algorithm encryption: the evaluation report is grouped into 128 bits, and a dynamic key is generated by Huada HC32F460 hardware encryption chip, which is updated every 10 minutes. It is transmitted to the remote monitoring center through 5G power dedicated slice, with an average transmission latency of 18ms.

[0025] The touch display module uses a 10.1-inch industrial screen, which displays in four windows: inverter A-phase voltage waveform (refresh rate 50Hz), array 1-8 series current bar chart, harmonic distortion rate trend curve (past 24 hours), and real-time evaluation report (including 12 types of event warning levels), and supports multi-touch operation.

[0026] The fault diagnosis module monitors the status of each module every 100ms using an STM32H743 microcontroller. When the sensor power supply voltage deviates from 3.3V±5%, the CPU temperature exceeds 60℃, or the signal strength is <-90dBm, an 80dB buzzer alarm is triggered and a red LED flashes, and the fault code is stored.

[0027] The working principle of this invention is as follows: After the device is powered on, the multi-channel data acquisition module synchronously collects the power parameters of each area of ​​the photovoltaic power station through Hall sensors, and transmits them to the data analysis module after analog-to-digital conversion; the improved CNN model extracts data features in real time and identifies anomalies, generating an evaluation report containing 12 types of events; the report is encrypted by the 5G module and transmitted to the monitoring center, and is simultaneously displayed in a split window on the touch screen; the fault diagnosis module monitors the equipment status throughout the process and triggers an alarm when an anomaly occurs; all data is stored on an SSD hard drive and can be exported via USB interface.

[0028] Example 2

[0029] The difference between this embodiment and Embodiment 1 is that the multi-channel data acquisition module is equipped with 16 LEM LAH 50-P sensors (range ±50A / ±690V), 8 channels for acquiring wind turbine generator output parameters, 8 channels for acquiring transformer low-voltage side parameters, a sampling rate of 10kHz, and 5kV optocoupler isolation to prevent ground loop interference.

[0030] The data analysis module uses the Ascend 310 edge chip and runs a lightweight CNN model (depthseparable convolution, reducing computation by 30%), focusing on identifying wind power-specific events such as "voltage fluctuations caused by turbulence", with an accuracy of 97.2%.

[0031] The 5G module incorporates SM2 digital signature to ensure data integrity, and its outer casing is coated with an anti-salt spray coating (compliant with ASTM B117 standards) to adapt to coastal wind farm environments.

[0032] Electrical connection methods or structures not described in detail in this article are existing technologies.

[0033] While the specific embodiments of the present invention have been described in detail above, the present invention is not limited to the above embodiments. Within the scope of knowledge possessed by those skilled in the art, various changes can be made without departing from the spirit of the present invention, and modifications or variations without creative effort are still within the protection scope of the present invention.

Claims

1. A power quality detection device for a new energy power station, characterized in that: The system includes a multi-channel data acquisition module, a deep learning-based data analysis module, a 5G communication module, and a touch display module. The multi-channel data acquisition module is used to simultaneously collect power parameters such as voltage, current, frequency, and harmonics from different areas of the new energy power plant. The deep learning-based data analysis module performs feature extraction and anomaly identification on the collected multi-channel data to generate a power quality assessment report. The 5G communication module transmits the assessment report to a remote monitoring center in real time. The touch display module is used to display the power parameters and assessment report, and supports manual operation command input.

2. The power quality detection device for a new energy power station according to claim 1, characterized in that: The multi-channel data acquisition module includes multiple high-precision Hall sensors, which are used to acquire voltage and current signals with different phase sequences. It supports 16-32 channels of simultaneous acquisition and the acquisition frequency reaches 20kHz.

3. The power quality detection device for a new energy power station according to claim 1, characterized in that: The deep learning-based data analysis module employs an improved convolutional neural network model, which is trained using a large amount of historical power quality data.

4. The power quality detection device for a new energy power station according to claim 1, characterized in that: The 5G communication module uses national cryptographic algorithms to encrypt the transmitted data.

5. The power quality detection device for a new energy power station according to claim 1, characterized in that: It also includes a fault diagnosis module, which is used to monitor the device's own hardware and software in real time and issue an alarm signal when a fault is detected.

6. The power quality detection device for a new energy power station according to claim 1, characterized in that: The touch display module supports multi-window display, which can simultaneously display power parameters and evaluation reports from different areas.

7. The power quality detection device for a new energy power station according to claim 1, characterized in that: It also includes a large-capacity storage module for storing the collected power parameters and the generated evaluation reports, with a storage time of no less than one year.

8. The power quality detection device for a new energy power station according to claim 1, characterized in that: The device operates in a temperature range of -40°C to 70°C and a relative humidity range of 5% to 95%.

Images