Distributed resource group scheduling and group control system

Abstract

A distributed resource group scheduling and group control system relates to the technical field of information transmission, and comprises a data acquisition module used for collecting operation parameters of distributed energy resources, including but not limited to generation power, voltage, current and temperature; the compression and encryption module is used for compressing and encrypting the collected data; the transmission network is used for transmitting the compressed and encrypted data and supporting efficient and safe data transmission; the receiving end decoding module is used for decrypting and decompressing the received data, recovering the original information, and checking and verifying to ensure the integrity and authenticity of the data; the compression and encryption module comprises an adaptive prediction model compression algorithm and a hybrid encryption mechanism; the group control decision module is used for performing group control on the distributed resources based on the received data; through a data compression algorithm, the transmission data volume is greatly reduced, so that the transmission speed is improved.

Description

Technical Field

[0001] The present invention relates to the field of information transmission technology, and in particular to a distributed resource group dispatching and control system. Background Art

[0002] With the growing global emphasis on and widespread adoption of renewable energy, the proportion of distributed energy resources (such as solar and wind) in power systems continues to increase. Distributed resource group dispatching and control systems, as a key means of managing and coordinating the generation and consumption of these distributed energy resources, are becoming increasingly important in power systems.

[0003] The distributed resource group dispatching and control system is primarily used to manage and coordinate the generation and consumption of distributed energy resources (such as solar, wind, and small hydropower). This system provides real-time monitoring and control of the operating status of each distributed resource, ensuring the stability and reliability of the power system through precise regulation of these resources. Specifically, the system obtains real-time operating parameters for the generated power, voltage, and current of distributed resources. Based on the real-time needs of the power system, it dynamically adjusts the generation and consumption of these resources to achieve a balanced supply and demand.

[0004] However, the current distributed resource group dispatching and control system has significant deficiencies in data transmission, specifically low data transmission efficiency, which leads to information delays and affects the system's real-time regulation of distributed resources. At the same time, the security of data transmission cannot be fully guaranteed, and there is a risk of data leakage and tampering, which seriously threatens the stable and reliable operation of the power system. Summary of the Invention

[0005] To overcome the shortcomings of the background technology, the present invention discloses a distributed resource group scheduling and control system. It aims to closely integrate the characteristics of the distributed resource group scheduling and control system and propose an efficient and secure data transmission system. To achieve the above-mentioned purpose, the present invention adopts the following technical solutions:

[0006] A distributed resource group dispatching and control system includes a data acquisition module for collecting operating parameters of distributed energy resources, including but not limited to generated power, voltage, current and temperature;

[0007] Compression and encryption module, used to compress and encrypt the collected data to improve data transmission efficiency and security;

[0008] Transmission network, used to transmit compressed and encrypted data, supporting efficient and secure data transmission;

[0009] A receiving-end decoding module is used to decrypt and decompress the received data, restore the original information, and perform verification and validation to ensure the integrity and authenticity of the data. The compression and encryption module includes an adaptive prediction model compression algorithm and a hybrid encryption mechanism.

[0010] The group control decision module performs group control of distributed resources based on received data; it includes:

[0011] The supply and demand balancing engine builds a regional power supply and demand model based on restored real-time data and generates an energy dispatch curve for the next 24 hours in combination with weather forecast data;

[0012] The dynamic control unit sends power adjustment instruction sets to each resource node based on the model calculation results, achieving frequency deviation control within the range of ±0.2Hz and response delay less than 50ms;

[0013] The closed-loop feedback module collects grid status data in real time after instruction execution, and corrects the control model parameters through the Kalman filter algorithm to form an end-to-end adaptive optimization loop.

[0014] Preferably, the adaptive prediction model compression algorithm includes:

[0015] S1, data segmentation and feature extraction, divide the original data into several small blocks and identify repeated patterns and redundant information in the data;

[0016] S2, dynamic prediction window adjustment and model training, dynamically adjust the prediction window size according to the changing trend of historical data, and use appropriate machine learning algorithms for model training;

[0017] S3, multi-level compression strategy execution step, uses lossless compression algorithm to perform preliminary compression on data, and uses lightweight lossy compression algorithm to further reduce the data volume for data with high redundancy.

[0018] Preferably, the hybrid encryption mechanism includes:

[0019] Symmetric encryption uses the AES algorithm to encrypt data and periodically changes session keys to reduce security risks;

[0020] Asymmetric encryption uses the RSA algorithm to generate and exchange session keys to ensure the security of key distribution;

[0021] Data integrity verification: Sign and verify data through asymmetric encryption algorithms to ensure data integrity and authenticity.

[0022] Preferably, the data acquisition module performs preliminary filtering and calibration preprocessing on the collected raw data; the receiving end decoding module is deployed in the control center and has a multi-threaded processor.

[0023] Due to the adoption of the above-mentioned technical solution, the present invention has the following beneficial effects:

[0024] The present invention discloses a distributed resource group dispatching and control system, which significantly reduces the amount of transmitted data through a data compression algorithm, thereby improving the transmission speed. The adaptive prediction model compression algorithm can flexibly adjust the compression method according to the characteristics of the data and changes in the transmission environment. Under the premise of ensuring data quality, it significantly improves the compression ratio and reduces data transmission time. The hybrid encryption mechanism is adopted to effectively prevent the risk of data leakage during transmission. The symmetric encryption algorithm ensures the efficiency of data encryption and decryption, and the asymmetric encryption algorithm ensures the security of key distribution and data integrity verification, providing multi-level security protection for data transmission. The hybrid encryption mechanism is adopted to effectively prevent the risk of data leakage during transmission. The symmetric encryption algorithm ensures the efficiency of data encryption and decryption, and the asymmetric encryption algorithm ensures the security of key distribution and data integrity verification, providing multi-level security protection for data transmission. It supports various types of distributed resources and is suitable for different application scenarios. The system can flexibly adjust data collection, compression, encryption and transmission strategies according to the characteristics and needs of different distributed resources, and has good compatibility and scalability. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] Figure 1 This is a system block diagram of the present invention. DETAILED DESCRIPTION

[0026] The present invention can be explained in detail through the following embodiments. The purpose of disclosing the present invention is to protect all technical improvements within the scope of the present invention. In the description of the present invention, it should be understood that if there are terms such as "upper", "lower", "front", "back", "left" and "right" indicating directions or positional relationships, they only correspond to the drawings of this application and are for the convenience of describing the present invention. They do not indicate or imply that the device or element referred to must have a specific direction.

[0027] Example 1, combined with the attached Figure 1 ,A distributed resource group scheduling and control system includes a data acquisition module, a compression and encryption module, a transmission network and a receiving end decoding module.

[0028] The data acquisition module is responsible for collecting raw data from distributed resources, including operating parameters such as power generation, voltage, current, and temperature. Deployed at each distributed resource site, it uses high-precision data acquisition equipment to monitor the operating status of the equipment in real time. The data acquisition module also includes data preprocessing capabilities, enabling preliminary filtering and calibration of the collected raw data.

[0029] The compression and encryption module preprocesses the collected data, including compression and encryption operations. By integrating an adaptive predictive model compression algorithm with a hybrid encryption mechanism, it ensures efficient data transmission and secure protection. The module utilizes high-performance computing equipment to implement compression and encryption functions, ensuring that the processing speed and capacity meet the real-time requirements of the system. Furthermore, the compression and encryption algorithms are optimized to improve their execution efficiency.

[0030] The transmission network supports efficient and secure data transmission and can be either a dedicated network or an optimized public network. Dedicated networks offer greater security and reliability, making them suitable for scenarios requiring high data security. Public networks offer wider coverage and lower costs, making them suitable for cost-sensitive scenarios. QoS (Quality of Service) technology is used to prioritize different types of data, ensuring the priority transmission of critical data. Real-time network monitoring and optimization are also implemented to improve network stability and reliability.

[0031] The receiving-end decoding module decrypts and decompresses the received data to restore the original information. It also verifies and validates the recovered original data to ensure its integrity and authenticity. Deployed in a control center, the receiving-end decoding module utilizes high-performance computing equipment to process and analyze received data in real time. The receiving-end decoding module also provides data storage and management capabilities, enabling long-term preservation and analysis of processed data.

[0032] The adaptive predictive model compression algorithm is a data compression method based on machine learning. Its core goal is to improve data transmission efficiency and reduce bandwidth usage. The algorithm is unique in that it dynamically adjusts the prediction window size and employs a multi-level compression strategy. This allows it to flexibly adjust the compression method based on data characteristics and changes in the transmission environment, effectively improving data transmission efficiency.

[0033] Data segmentation: Divide raw data into several small blocks according to specific rules. Each block has relatively independent characteristics and meaning, facilitating subsequent independent processing. For example, segmentation can be based on time series or data type. Feature extraction: Perform in-depth feature analysis on each block of data, applying data mining and machine learning techniques to identify repetitive patterns and redundant information. For example, periodic fluctuation patterns in power data can be effectively identified and extracted.

[0034] Dynamic Prediction Window: Dynamically adjusts the prediction window size based on historical data trends. When data changes are relatively stable, the prediction window is appropriately narrowed to improve prediction accuracy; when data changes are drastically, the prediction window is expanded to better capture the overall data characteristics. Model Training: Use historical data to train the prediction model, employing appropriate machine learning algorithms (such as linear regression and neural networks) to generate prediction results. By continuously optimizing model parameters, the accuracy and reliability of predictions are improved.

[0035] Lossless compression: Huffman coding is used for initial data compression. Lossless compression ensures no information is lost during the compression process and is suitable for scenarios requiring high data accuracy. Lossy compression: For highly redundant data, lightweight lossy compression algorithms (such as wavelet transform) are used to further reduce the data volume. Lossy compression sacrifices data accuracy to some extent, but can significantly improve the compression ratio and is suitable for scenarios requiring less stringent data accuracy.

[0036] Compression Execution: Based on the results of feature analysis, an appropriate compression algorithm is selected to compress the data and generate a compressed data packet. The selection of a compression algorithm takes into account the data characteristics, compression ratio, and compression speed. Real-time Optimization: During data transmission, network conditions and data characteristics are monitored in real time, and the compression strategy is dynamically adjusted based on actual conditions. For example, when network bandwidth is sufficient, the compression ratio can be appropriately lowered to improve real-time data transmission; when network bandwidth is limited, the compression ratio can be increased to reduce the amount of data transmitted.

[0037] The hybrid encryption mechanism includes both symmetric and asymmetric encryption. Symmetric encryption uses the Advanced Encryption Standard (AES) algorithm to encrypt the actual transmitted data. The AES algorithm offers fast encryption speed and high efficiency, ensuring data security while meeting the requirements of real-time data transmission. Key management: Session keys are regularly rotated to reduce the security risks associated with long-term use of the same key. A secure key distribution mechanism ensures the secure transmission and storage of session keys.

[0038] Asymmetric encryption uses the RSA algorithm to generate and exchange session keys. The RSA algorithm is a widely used asymmetric encryption algorithm that ensures secure key distribution. The secure exchange of session keys is achieved through the encryption and decryption process using public and private keys. Data integrity verification: Data is signed and verified using an asymmetric encryption algorithm to ensure data integrity and authenticity. The sender signs the data using the private key, and the receiver verifies the signature using the public key. Only data that passes verification is considered complete and authentic.

[0039] Specifically, a distributed photovoltaic power station deployed multiple data acquisition modules. Each node collects real-time operating data from the photovoltaic panels, including power generation, power factor, voltage and current parameters. This data is transmitted to the central control room in real time via wireless communication modules. These modules include LEM ITN 600-S ULTRASTAB series closed-loop Hall effect sensors (with an accuracy of ±0.2%) for current signals, PT100 platinum resistance thermometers coupled with a 24-bit analog-to-digital converter for temperature data, and a power calculation module integrated into a Xilinx Zynq-7000 FPGA, which simultaneously acquires generated power, three-phase voltage / current, and ambient temperature and humidity data at a 1kHz sampling rate. This node uses a three-stage filtering system (hardware RC low-pass filtering, digital median filtering, and Kalman dynamic correction) to cleanse data. Periodic self-calibration is performed based on the IEEE 1547 standard parameter table to ensure a data error rate of ≤0.5%.

[0040] The acquisition node transmits the pre-processed data to the compression and encryption module via the RS-485 bus (Modbus RTU protocol, 19200bps).

[0041] The compression and encryption module utilizes a layered processing architecture. Data units are first divided into 500ms time windows and labeled with resource IDs. An improved LSTM neural network then identifies the periodic fluctuations of power data. The dynamic prediction window automatically adjusts based on the sliding variance value: when the variance of 10 consecutive data points exceeds 100 V², a large 2-second window is used; otherwise, a small 200ms window is used. The window overlap ratio is fixed at 30% to prevent edge information loss. The compression execution phase employs a hybrid algorithm strategy, applying modified Huffman coding to the fundamental frequency component (with a code table dynamically updated every 5 minutes) and lossy compression using the discrete wavelet transform (DWT) for harmonic components. The energy preservation threshold is set at 95%, and the compression ratio is dynamically adjustable between 1:5 and 1:20. The encryption phase utilizes the National Security SM4 / SM9 coprocessor for layered encryption. At the session layer, a 256-bit AES key is negotiated using the ECDH protocol (rotated every 30 minutes). At the data layer, block encryption is performed using the SM4 algorithm. SHA-3-512 is used to generate a 64-byte hash value, and an EdDSA signature is appended for integrity. The module uses TSMC's 7nm process ASIC chip to achieve 16-way parallel processing of data streams, with a single-way delay of ≤5ms and an encryption and decryption throughput of ≥10Gbps.

[0042] The transmission network utilizes a dual-channel redundant design. The primary channel is an industrial Ethernet network based on the OPC UA protocol, and the backup channel is a 5G NR-U wireless network. Switching latency is ≤50ms. The network optimization system defines three levels of QoS priority: critical control instructions (latency ≤50ms), device status data (latency ≤200ms), and historical data (latency ≤1s). It also uses a reinforcement learning model to dynamically adjust transmission parameters. Real-time monitoring metrics include packet loss rate (threshold <0.1%), jitter (<10ms), and bandwidth utilization (target 80%-90%). When the Network Health Index (NHI) (=0.7 × bandwidth utilization + 0.3 × packet loss rate) exceeds 0.85, a token bucket traffic shaping mechanism (bucket capacity 1MB) is triggered. The system implements an exponential backoff strategy for disconnection and reconnection, with an initial retry interval of 1s and a maximum interval of 64s.

[0043] The receiving-end decoding module utilizes a distributed microservices architecture. The decryption service is deployed in a secure isolation zone and uses biometric authentication (fingerprint + iris) to access the 4096-bit RSA private key stored in the HSM. The decompression service runs in the data processing zone, performing inverse wavelet transform and Huffman decoding on the received data, and passing a CRC check (three failures trigger a retransmission of a 1MB data block). The data storage service uses the InfluxDB time series database to archive processing results and integrates the Isolation Forest algorithm for anomaly detection (sensitivity is controlled by a sliding window t-test with a p-value threshold of 0.01). The system performed excellently in a 50-node test: median end-to-end latency was 720ms (P95 ≤ 1.2s), compression ratio was 42.7% higher than Zlib, and bit error rate was less than 1×10 -10 .

[0044] The exception handling mechanism includes a multi-level fault-tolerance strategy: the data acquisition end detects sensor failure through Kalman filter residual analysis (residual > 3σ for 10 seconds, triggering an alarm) and enables cubic spline interpolation or 30 minutes of historical data backfill; the transport layer records logs and initiates ECDH key renegotiation (≤ 200ms) when decryption fails, and decompression CRC check abnormalities trigger upstream data retransmission.

[0045] The supply and demand balancing engine at the cluster control decision layer is built on an NVIDIA DGX A100 supercomputing cluster. It uses a LSTM-MILP hybrid model to generate a 24-hour dispatch curve by integrating real-time power data (100ms refresh rate), GRAPES high-precision weather forecasts (1km² grid), and power market signals. This model, pre-trained with years of historical data, uses the NSGA-III multi-objective optimization algorithm to achieve Pareto optimality for cost per kilowatt-hour (KWh), carbon emission intensity (≤0.35kg CO2 / kWh), and reserve capacity ratio (≥15%), while meeting the frequency deviation constraint of ±0.2Hz. The dynamic control unit issues structured commands (with Ed25519 signatures and CRC-32 checksums) over the lightweight MQTT protocol (QoS2). Leveraging 5G URLLC slicing, it achieves 38ms global command broadcast. Each node actuator utilizes dual-mode redundancy, combining PID control (steady-state error ≤0.05%) and local droop control (4% coefficient), ensuring grid stability even during a 200ms communication outage.

[0046] The closed-loop feedback module uses the WAMS wide-area measurement system (100Hz sampling) to obtain frequency fluctuations (μHz level), power tracking errors, and equipment health status, and constructs a dual-time-scale Kalman filter correction system: the fast loop (1s period) updates the noise covariance matrix online, and the slow loop (5-minute period) uses an elastic weight consolidation algorithm (α=0.02) to optimize the LSTM model weights.

[0047] The parts of the present invention that are not described in detail are prior art. It is obvious to those skilled in the art that the present invention is not limited to the details of the above-mentioned exemplary embodiments, and that the present invention can be implemented in other specific forms without departing from the spirit or basic characteristics of the present invention. Therefore, no matter from which point of view, the embodiments should be regarded as exemplary and non-restrictive, and all changes that fall within the meaning and scope of the same elements are intended to be included in the present invention.

Claims

1. A distributed resource group dispatching and control system, Its characteristics include: A data acquisition module for collecting operating parameters of distributed energy resources, including but not limited to generated power, voltage, current, and temperature; Compression and encryption module, used to compress and encrypt the collected data to improve data transmission efficiency and security; Transmission network, used to transmit compressed and encrypted data, supporting efficient and secure data transmission; A receiving-end decoding module is used to decrypt and decompress the received data, restore the original information, and perform verification and validation to ensure the integrity and authenticity of the data. The compression and encryption module includes an adaptive prediction model compression algorithm and a hybrid encryption mechanism. The group control decision module performs group control of distributed resources based on received data; it includes: The supply and demand balancing engine builds a regional power supply and demand model based on restored real-time data and generates an energy dispatch curve for the next 24 hours in combination with weather forecast data; The dynamic control unit sends power adjustment instruction sets to each resource node based on the model calculation results, achieving frequency deviation control within the range of ±0.2Hz and response delay less than 50ms; The closed-loop feedback module collects grid status data in real time after instruction execution, and corrects the control model parameters through the Kalman filter algorithm to form an end-to-end adaptive optimization loop.

2. The distributed resource group dispatching and control system according to claim 1, characterized in that: The adaptive prediction model compression algorithm includes: S1, data segmentation and feature extraction, divide the original data into several small blocks and identify repeated patterns and redundant information in the data; S2, dynamic prediction window adjustment and model training, dynamically adjust the prediction window size according to the changing trend of historical data, and use appropriate machine learning algorithms for model training; S3, multi-level compression strategy execution step, uses lossless compression algorithm to perform preliminary compression on data, and uses lightweight lossy compression algorithm to further reduce the data volume for data with high redundancy.

3. The distributed resource group dispatching and control system according to claim 2, characterized in that: The hybrid encryption mechanism includes: Symmetric encryption uses the AES algorithm to encrypt data and periodically changes session keys to reduce security risks; Asymmetric encryption uses the RSA algorithm to generate and exchange session keys to ensure the security of key distribution; Data integrity verification: Sign and verify data through asymmetric encryption algorithms to ensure data integrity and authenticity.

4. The distributed resource group scheduling and control system according to claim 1, wherein: The data acquisition module performs preliminary filtering and calibration preprocessing on the collected raw data; the receiving end decoding module is deployed in the control center and has a multi-threaded processor.

Images