16-channel high-precision synchronous data acquisition system based on multi-parameter model

Abstract

The invention discloses a 16-channel high-precision synchronous data acquisition system based on a multi-parameter model, which belongs to the technical field of data acquisition and comprises a multi-parameter model driving module, a self-adaptive channel scheduling module, a closed-loop calibration compensation module and a collaborative synchronous optimization module. The multi-parameter model driving module establishes a multi-dimensional optimization model fusing the sampling rate, the channel load and the error trend to output optimization parameters; the adaptive channel scheduling module performs real-time quality evaluation and dynamic resource allocation according to the optimization parameters to generate a scheduling instruction; the closed-loop calibration compensation module monitors a synchronization error in real time and calculates a compensation parameter; the collaborative synchronous optimization module realizes inter-channel parameter sharing and collaborative control to generate a synchronous control signal, and feeds back an optimization result to the multi-parameter model driving module to form a closed loop, and through deep coupling and closed loop feedback of the four modules, high-precision synchronization of multi-channel data acquisition is realized, and the synchronization precision reaches + / -35ns.

Description

Technical Field

[0001] This invention relates to the field of data acquisition technology, specifically to a 16-channel high-precision synchronous data acquisition system based on a multi-parameter model. This system achieves high-precision synchronous acquisition of multi-channel data through a deep coupling mechanism of multi-parameter driven model, adaptive channel scheduling, closed-loop calibration compensation, and collaborative synchronization optimization. It is particularly suitable for application scenarios that require multi-channel synchronous measurement, such as industrial monitoring, environmental detection, and scientific experiments. Background Technology

[0002] With the rapid development of industrial automation and intelligent measurement and control technologies, multi-channel data synchronous acquisition systems have become core equipment for modern industrial monitoring, environmental detection, and scientific experiments. These systems require real-time, synchronous, and high-precision acquisition of signals from multiple sensors to ensure high temporal correlation of the data, providing a reliable basis for subsequent data analysis and decision-making.

[0003] Existing multi-channel data acquisition systems mainly employ hardware synchronization schemes. For example, Chinese patent application CN110488718A discloses a multi-channel fully synchronous data acquisition system. This system uses a PXIe backplane as the communication and timing trigger bus, and manages multiple PXIe data acquisition cards through a PXIe controller to achieve multi-channel data acquisition. The main technical features of this system include: adopting a master-slave card architecture, with the master card inserted into the system timing slot and the slave card inserted into a hybrid slot or PXIe slot; simultaneously sending clock signals and trigger signals to the master and slave cards through the timing trigger bus in the PXIe backplane; the FPGAs on the master and slave cards performing phase adjustment on the received clock signals to synchronize the clock signals received by all ADCs; and the FPGAs on the master and slave cards sending synchronization trigger signals to the ADCs according to the trigger signals, enabling simultaneous sampling of all acquisition channels.

[0004] While the existing technical solution achieves clock and trigger synchronization at the hardware level, it still has the following shortcomings: First, the solution relies on a dedicated PXIe backplane and data acquisition card, resulting in high hardware costs, limited system scalability, and an inability to flexibly adapt to the needs of different application scenarios. Second, the solution only achieves synchronization at the hardware level, without considering dynamic error changes and channel differences during the acquisition process. When the external environment changes or channel characteristics are inconsistent, the synchronization accuracy will significantly decrease. Third, the clock signal in this solution is provided to different ADCs through equal-length traces, but board-level traces are difficult to be perfectly equal in length, and temperature drift and device aging will cause the clock phase offset to gradually accumulate, lacking a dynamic calibration and compensation mechanism. Fourth, each channel in this solution works independently, lacking inter-channel collaborative optimization. It cannot dynamically allocate resources based on the real-time quality status of each channel, resulting in the overall system performance being limited by the worst-performing channel. Fifth, the solution uses fixed sampling parameters and synchronization strategies, which cannot be adaptively adjusted according to the characteristics of the actual acquired data. Under conditions of signal frequency changes, load fluctuations, or noise interference, synchronization accuracy and data quality are difficult to guarantee.

[0005] The aforementioned problems are particularly prominent in industrial applications. Industrial monitoring systems typically require long-term continuous monitoring of various physical quantities such as temperature, pressure, flow rate, and vibration. Sensors exhibit diverse characteristics, wide signal frequency ranges, and complex environmental conditions. While existing hardware synchronization solutions can provide picosecond-level clock synchronization accuracy, in practical applications, due to inconsistent sensor response times, differences in signal transmission paths, non-ideal characteristics of analog front-end circuits, and minute deviations in ADC sampling timing, actual inter-channel synchronization errors often reach microsecond or even millisecond levels, far exceeding the accuracy of hardware clock synchronization. This severely impacts the temporal correlation of data and the accuracy of analysis.

[0006] Furthermore, the fixed parameter configurations in existing technical solutions are ill-suited to dynamically changing acquisition requirements. In practical applications, the signal quality of different channels may vary significantly. Some channels may experience data quality degradation due to sensor malfunctions, signal attenuation, or noise interference. However, the system cannot dynamically identify and address these issues, and can only passively accept low-quality data, impacting overall system performance. Simultaneously, the data generation rates and processing requirements of different channels may differ. Fixed resource allocation strategies cannot fully utilize system resources, resulting in high-priority channels not receiving sufficient sampling frequency and buffer space, while low-priority channels consume excessive resources, leading to resource waste and performance bottlenecks.

[0007] Therefore, existing multi-channel data acquisition systems urgently need a new technical solution that can overcome the limitations of hardware synchronization solutions. Through intelligent means of software algorithms and parameter models, it can achieve dynamic optimization and adaptive adjustment of the acquisition process, while establishing a collaborative mechanism and closed-loop feedback loop between channels. This will significantly improve the synchronization accuracy and overall performance of multi-channel data acquisition, while reducing hardware costs and increasing system flexibility. Summary of the Invention

[0008] The purpose of this invention is to solve the technical problems of existing multi-channel data acquisition systems, such as hardware limitations on synchronization accuracy, lack of dynamic error compensation, lack of collaborative optimization between channels, and inability to adaptively adjust parameter configuration, and to provide a 16-channel high-precision synchronous data acquisition system based on a multi-parameter model.

[0009] To achieve the above objectives, this invention provides a 16-channel high-precision synchronous data acquisition system based on a multi-parameter model. This system establishes a multi-dimensional optimization model integrating sampling rate, channel load, and error trends through a multi-parameter model-driven module, outputting optimized parameters to guide subsequent acquisition processes. An adaptive channel scheduling module performs real-time quality assessment and dynamic resource allocation for the 16 acquisition channels based on the optimized parameters, generating scheduling commands to control the acquisition process. A closed-loop calibration and compensation module monitors synchronization errors and drift trends in real time during data acquisition, calculates compensation amounts, and outputs compensation parameters. A collaborative synchronization optimization module enables parameter sharing and collaborative control among channels, generating synchronization control signals to control the synchronous acquisition of the 16 channels, and feeding the optimization results back to the multi-parameter model-driven module for model updates, forming a complete closed-loop feedback loop.

[0010] The core innovation of this system lies in the deep coupling between four modules: multi-parameter driving, adaptive scheduling, closed-loop compensation, and collaborative optimization. The output parameters of the multi-parameter model driving module directly serve as the key inputs of the adaptive channel scheduling module, achieving parameter-level coupling. The scheduling instructions generated by the adaptive channel scheduling module control the error monitoring and compensation calculation of the closed-loop calibration and compensation module, achieving state-level coupling. The compensation parameters output by the closed-loop calibration and compensation module are aggregated to the collaborative synchronization optimization module, achieving collaborative control between channels through parameter sharing, thus achieving logic-level coupling. The collaborative synchronization optimization module feeds back the optimization results to the multi-parameter model driving module for model updates, forming a complete closed loop of forward transmission → performance evaluation → reverse feedback → parameter adjustment.

[0011] This system achieves several synergistic effects: through parameter exchange between channels, the optimized parameters of high-quality channels can guide the parameter adjustments of low-quality channels, achieving mutual promotion; through multi-channel collaborative compensation, the overall synchronization accuracy of the system is higher than the sum of the accuracies of independent compensation by a single channel, achieving synergistic effects; through the generation of collaborative strategies, parameter conflicts and resource competition between different channels are resolved, achieving conflict resolution; through global state updates and adaptive adjustment, the system can dynamically adjust the parameters of each module according to overall performance, achieving adaptive optimization. These synergistic effects result in a non-linear growth characteristic of the system's technical performance, where 1+1>2.

[0012] Compared with the prior art, the present invention has the following beneficial effects:

[0013] First, this invention achieves comprehensive optimization of the acquisition process by establishing a multi-parameter driven model that integrates sampling rate, channel load, and error trend. The multi-parameter model comprehensively considers various influencing factors, resulting in more accurate and reasonable optimized parameters. Compared to the fixed parameter configurations of existing technologies, this invention can dynamically adjust parameters according to actual acquisition needs, significantly improving the system's adaptability and acquisition quality. Experiments show that with a 20% change in signal frequency, the synchronization error of this invention increases by only 5%, while the synchronization error of existing technologies increases by more than 30%.

[0014] Secondly, this invention achieves real-time quality assessment and dynamic resource allocation for each channel through an adaptive channel scheduling mechanism. The system prioritizes channels based on their signal-to-noise ratio (SNR) and stability metrics, allocating more sampling frequency and buffer space to high-quality channels while selectively optimizing low-quality channels. This adaptive scheduling mechanism significantly improves overall system performance compared to the equal resource allocation of existing technologies. Test results show that even with a 10dB SNR drop in some channels, this invention maintains an overall data quality degradation of no more than 3%, while existing technologies experience an overall data quality degradation exceeding 15%.

[0015] Third, this invention achieves real-time monitoring and compensation of dynamic errors during the acquisition process through a closed-loop calibration compensation mechanism. The error monitoring unit continuously detects the synchronization error and drift trend between channels, while the compensation calculation unit calculates the compensation amount based on the error characteristics and selects an appropriate compensation strategy. The compensation effect influences the update of the parameter model through a feedback loop. Compared to existing technologies that lack dynamic error compensation, this closed-loop compensation mechanism effectively suppresses error accumulation and maintains long-term high-precision synchronization. Long-term operation tests show that the synchronization error of this invention remains within ±50ns after 24 hours of continuous operation, while the synchronization error of existing technologies accumulates to over ±500ns.

[0016] Fourth, this invention achieves parameter sharing and mutual promotion among multiple channels through a collaborative synchronization optimization mechanism. Optimized parameters for high-quality channels can guide parameter adjustments for low-quality channels. The accuracy of multi-channel collaborative compensation is higher than that of single-channel independent compensation. Parameter conflicts between channels are resolved through a negotiation mechanism, and the system can dynamically adjust the collaborative strategy based on the global state. These synergistic effects result in a non-linear increase in overall system performance, where 1+1>2. Experimental data show that the average synchronization accuracy after 16-channel collaborative optimization is 40% higher than that after single-channel independent optimization, and the overall data quality is improved by 35%.

[0017] Fifth, this invention achieves self-learning and self-optimization of the system by establishing a complete closed-loop feedback loop. The collaborative synchronization optimization module feeds the optimization results back to the multi-parameter model-driven module for model parameter updates, enabling the system to continuously adjust its optimization strategy based on actual operating results and gradually adapt to the characteristics of specific application scenarios. Compared to the fixed strategies of existing technologies, this self-learning capability significantly improves the long-term stability and adaptability of the system. Long-term application tests show that after 3 days of self-learning optimization in different application scenarios, the synchronization accuracy and data quality of this invention have improved by 25% and 30%, respectively.

[0018] Sixth, compared to existing hardware synchronization solutions, this invention offers lower costs and greater flexibility. This invention primarily achieves synchronization optimization through software algorithms and parameter models, eliminating the need for dedicated PXIe backplanes and complex FPGA circuits. It can be implemented based on general-purpose data acquisition hardware modules, significantly reducing hardware costs. Furthermore, the software algorithms of this invention are easy to modify and upgrade, quickly adapting to different application requirements. Compared to the fixed architecture of hardware solutions, it offers greater flexibility and scalability. Cost analysis shows that the hardware cost of this invention is more than 60% lower than existing PXIe solutions, while maintaining comparable or better performance. Attached Figure Description

[0019] Figure 1 This is a schematic diagram of the overall architecture of the 16-channel high-precision synchronous data acquisition system based on a multi-parameter model according to the present invention.

[0020] Figure 2 This is a detailed structural diagram of the multi-parameter model driving module 1 of the present invention.

[0021] Figure 3 This is a detailed structural diagram of the adaptive channel scheduling module 2 of the present invention.

[0022] Figure 4 This is a detailed structural diagram of the closed-loop calibration compensation module 3 of the present invention.

[0023] Figure 5 This is a detailed structural diagram of the collaborative synchronization optimization module 4 of the present invention.

[0024] Figure 6 This is a flowchart of the data acquisition process of this invention. Detailed Implementation

[0025] Please refer to the attached document. Figures 1-6 The technical solution of the present invention will be described in detail below with reference to the accompanying drawings and specific embodiments. It should be noted that the following embodiments are for illustrative purposes only and are not intended to limit the invention. Unless otherwise specified, the embodiments and features described in the present invention can be combined with each other.

[0026] Reference Figure 1 The 16-channel high-precision synchronous data acquisition system based on a multi-parameter model provided in this embodiment includes a multi-parameter model driving module 1, an adaptive channel scheduling module 2, a closed-loop calibration and compensation module 3, and a collaborative synchronization optimization module 4.

[0027] The multi-parameter model-driven module 1 receives historical acquisition data as input and establishes a multi-dimensional optimization model that integrates sampling rate parameters, channel load parameters, and error trend parameters through its internal parameter optimization unit and model update unit. It then outputs the optimized parameters to the adaptive channel scheduling module 2. The core function of this module is to provide optimized parameter configurations for the entire system, enabling subsequent acquisition processes to adaptively adjust based on the characteristics of historical data.

[0028] The adaptive channel scheduling module 2 receives the optimized parameters output by the multi-parameter model driving module 1. Through the quality assessment unit, it performs real-time quality assessment on the 16 acquisition channels, calculating the signal-to-noise ratio and stability indicators for each channel. Then, through the resource allocation unit, it prioritizes the channels based on the quality scores, dynamically allocating sampling frequencies and buffer space, and generating a scheduling instruction containing sampling timing and buffering strategies. This scheduling instruction is sent to the closed-loop calibration and compensation module 3, which also receives acquisition data fed back from the data acquisition hardware interface for the next round of quality assessment.

[0029] The closed-loop calibration compensation module 3 receives the scheduling instructions generated by the adaptive channel scheduling module 2. During data acquisition, the error monitoring unit continuously monitors the synchronization error between channels, analyzes the drift trend of the error over time, and identifies the main sources of error. Then, the compensation calculation unit calculates the compensation amount for each channel based on the detected error characteristics, selects an appropriate compensation strategy (including time offset compensation, phase correction compensation, and amplitude compensation strategies), and outputs the compensation parameters to the collaborative synchronization optimization module 4.

[0030] The collaborative synchronization optimization module 4 receives the compensation parameters output by the closed-loop calibration compensation module 3, and aggregates the compensation parameters of each channel through the parameter sharing unit to achieve parameter exchange and global state update between channels. Then, the collaborative control unit generates a multi-channel collaborative control strategy based on the global state, generating 16 channels of synchronization control signals and sending them to the data acquisition hardware interface. This module also includes a feedback optimization subunit, which feeds back the optimization results to the multi-parameter model drive module 1 for model parameter update, forming a complete closed-loop feedback loop.

[0031] The system's workflow is as follows: First, the multi-parameter model-driven module 1 establishes an optimization model based on historical data and outputs optimization parameters. Second, the adaptive channel scheduling module 2 performs quality assessment and resource allocation for 16 channels based on the optimization parameters, generating scheduling instructions. Then, the hardware interface executes synchronous acquisition of the 16 channels according to the scheduling instructions and synchronization control signals. Next, the closed-loop calibration and compensation module 3 monitors and compensates for errors during the acquisition process. Finally, the collaborative synchronization optimization module 4 implements collaborative control of multiple channels and feeds the optimization results back to the multi-parameter model-driven module 1 for model updates. This process continuously cycles, resulting in continuous optimization and improvement.

[0032] The system in this embodiment achieves comprehensive optimization and adaptive control of the multi-channel data acquisition process through deep coupling and closed-loop feedback among four modules. Compared with existing hardware synchronization solutions, this embodiment can significantly improve synchronization accuracy and data quality through intelligent software algorithms while reducing hardware costs, and also has stronger environmental adaptability and long-term stability.

[0033] Reference Figure 2 The multi-parameter model-driven module 1 includes a parameter optimization unit and a model update unit. The parameter optimization unit further includes a sampling rate optimization subunit, a channel load assessment subunit, and an error trend prediction subunit; the model update unit further includes a parameter fusion subunit, a weight allocation subunit, and a model output subunit.

[0034] The sampling rate optimization subunit receives historically acquired data and analyzes the frequency characteristics and data throughput of each channel. For each channel, this subunit uses spectral analysis to determine the main frequency components of the signal, calculates the theoretical minimum sampling rate based on the Nyquist sampling theorem, and then, considering the requirements for spectral aliasing and reconstruction accuracy, multiplies the theoretical minimum sampling rate by a safety factor to obtain the actual recommended sampling rate. Preferably, for signals with a frequency range of 0-1kHz, the sampling rate is set to 5kHz; for signals with a frequency range of 0-10kHz, the sampling rate is set to 50kHz. The output of this subunit is the recommended sampling rate vector for each channel.

[0035] The channel load assessment subunit evaluates channel load based on the data generation rate of each channel and the system's processing capacity. This subunit calculates the data traffic of each channel using the data generation rate (equal to the sampling rate multiplied by the data bit width), and then compares the data traffic of each channel with the total system data processing bandwidth to obtain the channel load percentage. Simultaneously, this subunit also considers cache space occupancy, assessing cache load by monitoring the cache queue length. Preferably, when the data traffic percentage of a channel exceeds 15% of the total system bandwidth, or the cache queue length exceeds 80% of the total cache capacity, this subunit marks that channel as a high-load channel and outputs a load warning signal. The output of this subunit is the load assessment vector and warning signal for each channel.

[0036] The error trend prediction subunit analyzes historical error data to predict future error trends. This subunit maintains a historical error database, recording the synchronization error values ​​of each channel over a past period (preferably the most recent hour). Then, time series analysis is used to fit the error data to a trend. Preferably, this subunit uses the Exponentially Weighted Moving Average (EWMA) method for trend prediction, calculated using the following formula: ,in The predicted error trend value, This represents the measured error at the current moment. This is the predicted value from the previous moment. The smoothing coefficient is preferably set to 0.3. The output of this sub-unit is the error trend prediction vector for each channel.

[0037] The parameter fusion subunit receives parameter vectors output from the sampling rate optimization subunit, channel load assessment subunit, and error trend prediction subunit, and performs weighted fusion. This subunit first normalizes each parameter vector, mapping parameters of different dimensions to the 0-1 range, and then assigns fusion weights based on the degree of influence of each parameter on the acquisition accuracy. The innovative parameter fusion algorithm proposed in this embodiment adopts an adaptive weighting mechanism, adjusting the weight allocation according to the dynamic changes of each parameter. Specifically, the mathematical expression of the algorithm is:

[0038] ,

[0039] in, Let be the fusion parameters for the i-th channel. The normalized sampling rate parameter, These are the normalized load parameters. The normalized error trend parameter, , and These are the fusion weights for the three parameters. The weights are calculated using a variance-based adaptive allocation strategy:

[0040] ,

[0041] ,

[0042] ,

[0043] in, , and These represent the variances of the sampling rate parameter, load parameter, and error trend parameter across all channels. A larger variance indicates a greater difference in the parameter across different channels, and a more significant impact on acquisition accuracy; therefore, it is assigned a higher weight. In a preferred embodiment, when... , , At that time, the calculation yielded , , The innovation of this algorithm lies in the dynamic adjustment of weights according to the parameter variance, which can automatically adapt to the characteristics of different application scenarios. Compared with the fusion method with fixed weights, the fusion accuracy of this algorithm is improved by more than 20%.

[0044] The weight allocation subunit receives the channel fusion parameters output by the parameter fusion subunit and allocates resource weights to each channel based on the magnitude of the fusion parameters. This subunit uses a priority queue data structure, assigning higher priority to channels with larger fusion parameters, thus prioritizing the needs of high-priority channels during resource allocation. Preferably, this subunit divides the 16 channels into three priority levels: high, medium, and low. The top 5 channels in terms of fusion parameters are high priority, channels ranked 6-11 are medium priority, and the bottom 5 channels are low priority. The resource weight allocation ratio is 50% for high-priority channels, 35% for medium-priority channels, and 15% for low-priority channels. The output of this subunit is the resource weight vector for each channel.

[0045] The model output subunit integrates sampling rate parameters, load parameters, error trend parameters, and resource weights, outputting a complete optimized parameter vector to the adaptive channel scheduling module 2. This optimized parameter vector includes the recommended sampling rate, load status, error trend, and resource weights for each channel, providing comprehensive parameter basis for subsequent adaptive scheduling.

[0046] The core innovation of the multi-parameter model-driven module 1 lies in the establishment of an optimization model that integrates multi-dimensional parameters. Through an adaptive weighting mechanism, parameter fusion is achieved. Compared with the fixed parameter configuration of existing technologies, it can automatically adjust the optimization strategy according to the characteristics of historical data and the dynamic changes of each parameter, which significantly improves the accuracy of parameter configuration and the adaptability of the system.

[0047] Reference Figure 3The adaptive channel scheduling module 2 includes a quality assessment unit and a resource allocation unit. The quality assessment unit further includes a signal-to-noise ratio calculation subunit, a stability evaluation subunit, and a channel priority ranking subunit; the resource allocation unit further includes a dynamic sampling allocation subunit, a buffer space scheduling subunit, and a scheduling instruction generation subunit.

[0048] The signal-to-noise ratio (SNR) calculation subunit receives real-time acquired data from the data acquisition hardware interface and calculates the SNR for each channel. This subunit employs a classic SNR calculation method, evaluating signal quality by analyzing the ratio of signal power to noise power. Specifically, the subunit first segments the acquired data, with each segment containing 256 sampling points. Then, it calculates the root mean square (RMS) value for each segment as an estimate of the signal power, and simultaneously calculates the RMS value of the difference between adjacent sampling points as an estimate of the noise power. The SNR is calculated as the ratio of signal power to noise power, expressed in decibels (dB). Preferably, a SNR higher than 30 dB is considered good signal quality; a SNR lower than 20 dB is considered poor signal quality, requiring targeted optimization. The output of this subunit is the SNR vector for each channel.

[0049] The stability evaluation subunit assesses the volatility and stability of the data from each channel. This subunit evaluates stability by calculating the standard deviation and coefficient of variation of the data. Specifically, this subunit maintains a sliding window, preferably with a window size of 512 sampling points, and calculates the standard deviation of the data within the window. Then calculate the coefficient of variation. ,in This represents the mean of the data within the window. A smaller coefficient of variation indicates more stable data. Preferably, a coefficient of variation less than 0.05 indicates good channel stability, while a coefficient of variation greater than 0.15 indicates poor channel stability. The output of this sub-unit is the stability evaluation vector for each channel.

[0050] The channel prioritization subunit prioritizes the 16 channels based on a combination of signal-to-noise ratio and stability metrics. This subunit employs a weighted scoring method to calculate the overall quality score. ,in Let i be the overall quality score for the i-th channel. This represents the signal-to-noise ratio (linear value). The coefficient of variation is 1. The weighting coefficient is 0.6, with a preferred value of 0.6. This formula considers the combined impact of signal-to-noise ratio and stability on data quality. A comprehensive score is obtained through weighted summation, and then the channels are sorted in descending order based on the comprehensive score, with higher-scoring channels having higher priority. The output of this sub-unit is the channel priority ranking result and the quality score vector for each channel.

[0051] The dynamic sampling allocation subunit dynamically allocates the sampling frequency of each channel based on channel priority and the optimization parameters output by the multi-parameter model driving module 1. The basic principle of this subunit is: high-priority channels receive higher sampling frequencies to ensure data quality, while the sampling frequencies of low-priority channels can be appropriately reduced to conserve system resources. Specifically, this subunit first determines the baseline sampling frequency for each channel based on the recommended sampling rate in the optimization parameters, and then dynamically adjusts it based on quality scores and resource weights. The adjustment strategy is as follows: for high-priority channels, the actual sampling frequency is set to 1.2 times the recommended sampling rate; for medium-priority channels, the actual sampling frequency is equal to the recommended sampling rate; and for low-priority channels, the actual sampling frequency is set to 0.8 times the recommended sampling rate. Simultaneously, this subunit also sets upper and lower limits for the sampling frequency: the highest sampling frequency does not exceed 100kHz, and the lowest sampling frequency is not lower than 1kHz, to ensure stable system operation. The output of this subunit is the actual sampling frequency vector for each channel.

[0052] The cache space scheduling subunit allocates cache space based on the data generation rate and processing capacity of each channel. This subunit maintains a cache pool, preferably with a total capacity of 16MB, and dynamically allocates cache space according to the data traffic and priority of each channel. Specifically, channels with high data traffic and high priority are allocated larger cache spaces to avoid buffer overflow and data loss; channels with low data traffic or low priority are allocated smaller cache spaces to improve cache utilization. The cache space allocation adopts a proportional allocation strategy, with the cache space for the i-th channel... The calculation is as follows:

[0053] ,

[0054] in, This represents the total capacity of the cache pool. The data generation rate for the i-th channel is equal to the sampling frequency multiplied by the data bit width. Let i be the resource weight of the i-th channel. This is the weighted sum of data traffic across all channels. This algorithm ensures that cache space allocation considers both the actual data traffic demand and channel priority, significantly reducing buffer overflow and data loss rates compared to an equal allocation strategy. In a preferred embodiment, using this algorithm reduces the buffer overflow rate by 70% and improves data integrity by 25%.

[0055] The scheduling instruction generation subunit integrates the sampling frequency and buffer space allocation results of each channel to generate a scheduling instruction that includes sampling timing and buffering strategy. This scheduling instruction contains detailed parameters for each channel, such as sampling start time, sampling interval, number of sampling points, buffer start address, and buffer length, and is encoded in binary format before being sent to the closed-loop calibration and compensation module 3 and the data acquisition hardware interface. This subunit is also responsible for the synchronous transmission of the scheduling instruction, ensuring precise alignment of the sampling timing of the 16 channels, with the synchronization time deviation controlled within ±10ns.

[0056] The core innovation of the adaptive channel scheduling module 2 lies in the establishment of an adaptive scheduling mechanism based on real-time quality assessment. It can dynamically allocate sampling frequency and buffer space according to the signal-to-noise ratio and stability of each channel. Compared with the fixed resource allocation of existing technologies, it can significantly improve the utilization efficiency of system resources and the overall data quality.

[0057] Reference Figure 4 The closed-loop calibration compensation module 3 includes an error monitoring unit and a compensation calculation unit. The error monitoring unit further includes a synchronous error detection subunit, a drift trend analysis subunit, and an error source identification subunit; the compensation calculation unit further includes a compensation amount calculation subunit, a compensation strategy selection subunit, and a compensation parameter output subunit.

[0058] The synchronization error detection subunit continuously monitors the time synchronization error between channels during data acquisition. The basic principle of this subunit is to assess the synchronization error by comparing the differences in sampling timestamps between channels. Specifically, this subunit records a precise timestamp for each sampling point of each channel, preferably with an accuracy of 10 ns. Then, a reference channel (preferably the one with the highest signal-to-noise ratio) is selected, and the time deviation of each other channel relative to the reference channel is calculated. The time deviation is calculated by aligning the sampling sequences of each channel and finding the time delay between each channel and the reference channel; this time delay is the synchronization error. Preferably, this subunit uses a cross-correlation algorithm to determine the time delay. The time corresponding to the peak position of the cross-correlation function between each channel signal and the reference channel signal is the time delay. The output of this subunit is the synchronization error vector of each channel relative to the reference channel.

[0059] The drift trend analysis subunit analyzes the drift characteristics of synchronization error over time. This subunit maintains an error history queue, recording the synchronization error values ​​of each channel over a past period (preferably the most recent 10 minutes). Then, linear regression is performed on the error sequence to calculate the error drift rate. The formula for calculating the drift rate is as follows: ,in This represents the change in error over a given time period. The time period is defined as the drift rate. A positive drift rate indicates that the error increases with time, while a negative drift rate indicates that the error decreases with time. Preferably, when the absolute value of the drift rate exceeds 5 ns / min, it is considered that there is significant drift and compensation is required. The output of this sub-unit is the error drift rate vector for each channel.

[0060] The error source identification subunit analyzes the primary sources of error. This subunit determines whether the error is caused by clock phase shift, signal transmission delay, or ADC sampling jitter by analyzing the spectral and temporal characteristics of the error. Specifically, if the error is mainly concentrated in the low-frequency band (less than 1Hz), it is considered to be caused by clock phase shift or signal transmission delay; if the error is distributed across a wider frequency band, it is considered to be caused by ADC sampling jitter. Simultaneously, this subunit also analyzes the correlation between error and channel load. If the error increases with increasing load, it is considered to be caused by system resource contention. The output of this subunit is an error source type identifier for each channel.

[0061] The compensation calculation subunit calculates the compensation amount for each channel based on the detected synchronization error and drift trend. For clock phase offset type errors, the compensation amount is equal to the detected synchronization error value; for drift type errors, the compensation amount needs to consider the drift trend and adopt a predictive compensation strategy. The innovative compensation calculation algorithm proposed in this embodiment combines the current error and drift prediction, and its mathematical expression is as follows:

[0062] ,

[0063] in, Let be the compensation amount for the i-th channel at time t. The current synchronization error is... For the error drift rate, The preferred prediction time window is 60 seconds. The innovation of this algorithm lies in the introduction of a drift prediction term. It can compensate for error drift in the future in advance. Compared with the method of only compensating for the current error, it can significantly reduce the accumulation rate of error and improve the long-term synchronization accuracy by 35%.

[0064] The compensation strategy selection subunit selects a suitable compensation strategy based on the error characteristics. This subunit supports three compensation strategies: time offset compensation, phase correction compensation, and amplitude compensation. Time offset compensation is achieved by adjusting the sampling timestamps of each channel and is suitable for errors caused by clock phase offset. Phase correction compensation is achieved by adjusting the sampling clock phase of each channel and is suitable for periodic phase errors. Amplitude compensation is achieved by adjusting the gain coefficients of each channel and is suitable for errors caused by amplitude inconsistencies between channels. This subunit automatically selects the most suitable compensation strategy based on the error source identification results and error characteristics, and can apply multiple strategies simultaneously for combined compensation. Preferably, for channels with synchronization errors less than 50ns, time offset compensation is sufficient; for channels with synchronization errors between 50-200ns, a combination of time offset compensation and phase correction compensation is used; for channels with synchronization errors exceeding 200ns, all three compensation strategies need to be applied simultaneously. The output of this subunit is the compensation strategy identifier for each channel.

[0065] The compensation parameter output subunit integrates the compensation amount and compensation strategy of each channel, and generates a compensation parameter vector which is output to the collaborative synchronization optimization module 4. The compensation parameter vector contains information such as the compensation amount, compensation strategy type, and compensation accuracy of each channel, providing an accurate compensation basis for collaborative synchronization optimization.

[0066] The core innovation of the closed-loop calibration compensation module 3 lies in establishing a complete closed-loop compensation mechanism. Through the tight coupling of error monitoring, drift analysis, and compensation calculation, it achieves real-time compensation for dynamic errors. In particular, the introduction of a drift prediction compensation algorithm can compensate for future errors in advance. Compared with existing technologies that lack dynamic error compensation solutions, this significantly improves long-term synchronization accuracy and system stability. Simultaneously, the compensation effect of this module is fed back to the multi-parameter model driving module 1 through the collaborative synchronization optimization module 4, forming a complete closed-loop circuit and realizing continuous improvement from error monitoring → compensation calculation → performance evaluation → parameter optimization.

[0067] Reference Figure 5 The collaborative synchronization optimization module 4 includes a parameter sharing unit and a collaborative control unit. The parameter sharing unit further includes a parameter aggregation subunit, an inter-channel parameter exchange subunit, and a global state update subunit; the collaborative control unit further includes a collaborative strategy generation subunit, a synchronization signal generation subunit, and a feedback optimization subunit.

[0068] The parameter aggregation subunit receives the compensation parameters for each channel output by the closed-loop calibration compensation module 3 and aggregates the compensation parameters for all 16 channels into a centralized parameter library. This parameter library maintains information such as the compensation amount, compensation strategy, error status, and historical compensation records for each channel, providing a data foundation for parameter exchange between channels and global status updates.

[0069] The inter-channel parameter exchange subunit transfers parameters from high-quality channels to low-quality channels. The basic principle of this subunit is to leverage successful compensation experience from high-quality channels to guide parameter optimization in low-quality channels. Specifically, this subunit first identifies high-quality channels (top 5 in quality score) and low-quality channels (bottom 5 in quality score) based on their quality ratings. Then, it analyzes the compensation parameter characteristics of high-quality channels, extracting effective compensation strategies and parameter settings. Finally, these effective parameters are transferred to low-quality channels as a reference for parameter optimization in low-quality channels. Preferably, this subunit uses a weighted average method for parameter transfer, with the adjustment parameters for low-quality channels being... ,in These are the original parameters for the low-quality channel. This is a weighted average of the parameters for high-quality channels. This mechanism enables mutual improvement between channels, allowing optimization experience from high-quality channels to help low-quality channels quickly improve their performance.

[0070] The global state update subunit updates the system's global synchronization state based on the compensation parameters and quality scores of each channel. This subunit maintains a global state vector containing key indicators such as overall system synchronization accuracy, average signal-to-noise ratio (SNR), and channel load balancing. Global synchronization accuracy is calculated by weighting the root mean square (RMS) values ​​of the synchronization errors of each channel, with the weights proportional to the channel quality scores. Average SNR is calculated by arithmetically averaging the SNRs of each channel. Channel load balancing is assessed by calculating the standard deviation of the load for each channel; a smaller standard deviation indicates a more balanced load. The output of this subunit is the global state vector, used to guide the generation of collaborative strategies.

[0071] The cooperative strategy generation subunit generates multi-channel cooperative control strategies based on the global state vector. The core of this subunit is a cooperative optimization algorithm that comprehensively considers the current state, historical performance, and interrelationships of each channel to generate a cooperative strategy that optimizes the overall system performance. The innovative cooperative optimization algorithm proposed in this embodiment employs a multi-objective optimization method, simultaneously optimizing synchronization accuracy, data quality, and resource utilization. The mathematical expression of this algorithm is:

[0072] ,

[0073] ,

[0074] in, For the overall system synchronization error, The average signal-to-noise ratio. For load imbalance, , , The weighting coefficients for the three objectives are preferably set to 0.5, 0.3, and 0.2. The data generation rate for the i-th channel. This represents the upper limit of the system's total data processing capacity. Let i be the buffer space for the i-th channel. To ensure the minimum buffer space required for normal channel operation, this optimization problem is solved using the Particle Swarm Optimization (PSO) algorithm. The algorithm iteratively searches for the parameter configuration that minimizes the objective function. The innovation of this algorithm lies in transforming the multi-channel collaborative optimization problem into a constrained multi-objective optimization problem. It can balance synchronization accuracy, data quality, and resource utilization while satisfying system resource constraints, achieving optimal overall system performance. Experimental results show that after adopting this collaborative optimization algorithm, the overall system synchronization accuracy is improved by 30%, the average signal-to-noise ratio is improved by 25%, and the load balancing is improved by 40%.

[0075] The synchronization signal generation subunit generates 16 channels of synchronization control signals according to a cooperative strategy. These synchronization control signals include parameters such as the sampling trigger time, sampling clock frequency, and sampling phase for each channel, used to control the synchronous acquisition process of the data acquisition hardware interface. The key technology of this subunit is precise timing control, generating nanosecond-level precision synchronization signals through a high-precision clock source. Preferably, this subunit uses digital delay line (DDL) technology to achieve programmable time delays with a delay accuracy of up to 5ns, thereby achieving precise alignment of the sampling timing of each channel. Simultaneously, this subunit also considers compensation parameters for each channel, pre-adding compensation delays when generating the synchronization signals, ensuring precise synchronization of the compensated sampling timing. The output of this subunit is the 16-channel synchronization control signal, which is sent to the data acquisition hardware interface.

[0076] The feedback optimization subunit is crucial for forming a closed-loop feedback between the collaborative synchronization optimization module 4 and the multi-parameter model-driven module 1. This subunit continuously monitors the system's operational performance, including actual synchronization accuracy, data quality, and resource utilization, comparing these against expected targets to evaluate the optimization effect. If the actual performance exceeds expectations, it indicates that the current parameter configuration and collaborative strategy are effective. The subunit then feeds these successful experiences back to the multi-parameter model-driven module 1 to update the parameters and weights of the optimization model. If the actual performance falls short of expectations, it indicates that the current configuration needs adjustment. The subunit generates parameter adjustment suggestions and feeds them back to the multi-parameter model-driven module 1, triggering a new round of parameter optimization. This subunit's feedback mechanism employs an incremental update strategy, adjusting only a portion of the parameters each time to avoid excessive parameter changes that could lead to system instability. Preferably, the parameter adjustment range is controlled within ±10% of the current value. This feedback optimization mechanism forms a complete closed-loop circuit, enabling the system to continuously adjust and optimize itself based on actual operational performance, gradually adapting to the characteristics of specific application scenarios and achieving self-learning and self-optimization.

[0077] The core innovation of the collaborative synchronization optimization module 4 lies in establishing a deep collaborative mechanism among multiple channels and a complete closed-loop feedback loop. Through parameter sharing and inter-channel parameter exchange, high-quality channels assist low-quality channels; through collaborative optimization algorithms, the overall system performance is optimized; and through feedback optimization, the system achieves self-learning and self-adaptation. These innovations result in a non-linear growth characteristic of overall system performance (1+1>2), demonstrating significant performance advantages compared to the channel-independent operating mode of existing technologies.

[0078] In this embodiment, the data acquisition hardware interface is implemented using the ADM-4586 analog-to-digital signal acquisition and control module. This module includes two acquisition cards, each providing eight analog current signal acquisition channels, for a total of 16 channels. The analog current signal types support two standard industrial signal types: 0-20mA and 4-20mA.

[0079] The data acquisition hardware interface receives the synchronization control signal generated by the collaborative synchronization optimization module 4, and synchronously acquires data from 16 channels according to the sampling trigger time and sampling frequency specified in the signal. Each acquisition card includes an FPGA and multiple ADCs. The FPGA receives the synchronization control signal and generates the sampling clock and trigger signal for each channel. The ADC performs analog-to-digital conversion based on the sampling clock and trigger signal, converting the analog current signal into a digital signal. The converted digital data is transmitted to the host computer via the data bus. The host computer feeds back the acquired data to the adaptive channel scheduling module 2 for quality evaluation, and simultaneously stores the data in the designated buffer space.

[0080] The key performance parameters of the data acquisition hardware interface are as follows: 16-bit sampling accuracy, adjustable sampling frequency range of 1-100kHz, inter-channel synchronization accuracy better than ±50ns, input impedance of 250Ω, measurement range of 0-20mA or 4-20mA, and measurement accuracy of ±0.1%. These performance parameters can meet the needs of applications such as industrial monitoring, environmental detection, and scientific experiments.

[0081] Reference Figure 6 This embodiment describes in detail the complete workflow of a 16-channel high-precision synchronous data acquisition system based on a multi-parameter model.

[0082] Step 1: The system begins the data acquisition task. The user initiates data acquisition through the host computer software, and the system initializes the parameters and status of each module.

[0083] Step 2: The multi-parameter model-driven module 1 establishes a multi-dimensional optimization model based on historical data. The parameter optimization unit of this module analyzes historical data, calculates the recommended sampling rate, load status, and error trend for each channel; the model update unit fuses these parameters and allocates resource weights according to an adaptive weighting mechanism, ultimately outputting an optimized parameter vector.

[0084] Step 3: Adaptive channel scheduling module 2 executes adaptive scheduling based on optimization parameters. The quality assessment unit of this module calculates the signal-to-noise ratio and stability index of each channel and prioritizes them; the resource allocation unit dynamically allocates sampling frequency and buffer space according to priority and generates scheduling instructions.

[0085] Step 4: The data acquisition hardware interface executes synchronous acquisition of 16 channels according to the scheduling instructions and synchronization control signals. Each channel samples simultaneously according to the specified sampling sequence, the ADC performs analog-to-digital conversion, and the converted digital data is transmitted to the host computer and fed back to the adaptive channel scheduling module 2.

[0086] Step 5: Closed-loop calibration compensation module 3 monitors and compensates for errors during the acquisition process. The error monitoring unit of this module detects the synchronization error between channels, analyzes the drift trend, and identifies the error source; the compensation calculation unit calculates the compensation amount based on the error characteristics, selects the compensation strategy, and outputs the compensation parameters.

[0087] Step 6: The collaborative synchronization optimization module 4 implements multi-channel collaborative control. The parameter sharing unit of this module summarizes the compensation parameters of each channel, realizing parameter exchange and global state update between channels; the collaborative control unit generates collaborative strategies and synchronization control signals to control the next round of data acquisition.

[0088] Step 7: Determine if the data acquisition task is complete. If the user has not stopped the data acquisition task, return to Step 2. The feedback optimization subunit will feed back the optimization results to the multi-parameter model-driven module 1 for model updates, starting a new round of optimization and data acquisition. If the data acquisition task is complete, proceed to Step 8.

[0089] Step 8: End data acquisition, save the acquired data and close all modules.

[0090] The core feature of this workflow is the formation of a complete closed-loop feedback loop. The results of each round of data collection are fed back to the multi-parameter model-driven module 1 to update and optimize the model parameters, guiding the next round of collection. Through continuous optimization and improvement, the system can gradually adapt to the characteristics of specific application scenarios, achieving self-learning and self-optimization, and continuously improving the synchronization accuracy and data quality over long-term operation.

[0091] This embodiment presents a comprehensive performance test of a 16-channel high-precision synchronous data acquisition system based on a multi-parameter model, and compares it with an existing hardware synchronization solution (the system in reference document CN110488718A).

[0092] Test environment: A standard industrial environment simulation platform was used, including 16 temperature sensors (measurement range 0-100℃, output signal 4-20mA) and 16 pressure sensors (measurement range 0-10MPa, output signal 4-20mA), all connected to a data acquisition hardware interface. The test lasted continuously for 72 hours, with a sampling frequency of 10kHz, and the ambient temperature varied between 20-30℃ to simulate temperature fluctuations in a real industrial environment.

[0093] The test results are as follows:

[0094] Synchronization accuracy test: The inter-channel synchronization error of the system of this invention is ±35ns, while the synchronization error of the comparative system is ±250ns. After 72 hours of continuous operation, the synchronization error of the system of this invention accumulated to ±48ns, while that of the comparative system accumulated to ±520ns. The long-term synchronization accuracy of the system of this invention is approximately 10 times higher than that of the comparative system.

[0095] Data quality testing: The average signal-to-noise ratio (SNR) of the 16 channels in this invention's system is 42 dB, and the average stability (1 / CV) is 35, while the average SNR of the comparative system is 38 dB, and the average stability is 28. The data quality indicators of this invention's system are comprehensively superior to those of the comparative system.

[0096] Resource utilization test: The CPU utilization of the system of this invention was 45%, the memory utilization was 60%, and the number of cache overflows was 0. In contrast, the CPU utilization of the comparison system was 55%, the memory utilization was 70%, and the number of cache overflows was 15 times per hour. The system of this invention significantly improves resource utilization efficiency and avoids data loss caused by cache overflows through adaptive resource allocation.

[0097] Environmental adaptability test: As the ambient temperature increased from 20℃ to 30℃, the synchronization error of the system of the present invention increased by 12%, while the synchronization error of the comparative system increased by 45%. The system of the present invention effectively suppressed the impact of temperature changes on synchronization accuracy through a closed-loop calibration compensation mechanism, and its environmental adaptability is significantly better than that of the comparative system.

[0098] Collaborative optimization effect test: Before and after enabling 16-channel collaborative optimization, the average synchronization accuracy of the system of this invention improved by 38%, and the overall data quality improved by 32%, fully verifying the effectiveness of the collaborative synchronization optimization mechanism. The comparison system, lacking a collaborative mechanism, could not achieve mutual promotion and overall optimization between channels.

[0099] Cost Comparison: The system of this invention uses the general-purpose ADM-4586 data acquisition module, and the hardware cost is approximately 35% of that of the comparative system (dedicated PXIe backplane and data acquisition card). Furthermore, the system of this invention achieves synchronous optimization through software algorithms, resulting in significantly lower development and maintenance costs compared to the hardware solution.

[0100] In summary, the system of the present invention is superior to existing hardware synchronization solutions in terms of synchronization accuracy, data quality, resource utilization, environmental adaptability, and cost, which fully demonstrates the advanced nature and practicality of the technical solution of the present invention.

[0101] The above description is merely a preferred embodiment of the present invention and is not intended to limit the 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 16-channel high-precision synchronous data acquisition system based on a multi-parameter model, characterized in that, include: The multi-parameter model-driven module is used to build a multi-dimensional optimization model based on historical data, which integrates sampling rate parameters, channel load parameters, and error trend parameters, and outputs the optimized parameters. The multi-parameter model driving module includes a parameter optimization unit and a model update unit; An adaptive channel scheduling module, connected to the multi-parameter model driving module, is used to perform real-time quality assessment of the 16 acquisition channels according to the optimization parameters, prioritize the channels based on the quality scores, dynamically allocate sampling resources and buffer space, and generate scheduling instructions; the adaptive channel scheduling module includes a quality assessment unit and a resource allocation unit. The closed-loop calibration compensation module is connected to the adaptive channel scheduling module. It is used to perform synchronization error detection and drift trend analysis during the data acquisition process according to the scheduling instructions, calculate the compensation amount after identifying the error source, select the compensation strategy, and output the compensation parameters. The closed-loop calibration compensation module includes an error monitoring unit and a compensation calculation unit; The collaborative synchronization optimization module is connected to both the closed-loop calibration compensation module and the multi-parameter model driving module. It is used to summarize the compensation parameters of each channel to realize parameter exchange and global state update between channels, and generate collaborative strategies and synchronization control signals. The synchronization control signals are sent to the data acquisition hardware interface to control the synchronous acquisition of 16 channels. The acquired data is fed back to the adaptive channel scheduling module for quality assessment. The feedback optimization subunit of the collaborative synchronization optimization module feeds back the optimization results to the multi-parameter model driving module for model update, forming a closed-loop feedback loop.

2. The 16-channel high-precision synchronous data acquisition system based on a multi-parameter model according to claim 1, characterized in that, The parameter optimization unit includes a sampling rate optimization subunit, a channel load assessment subunit, and an error trend prediction subunit. The sampling rate optimization subunit is used to calculate the optimal sampling rate based on the signal frequency characteristics and data throughput of each channel. The channel load assessment subunit is used to assess the processing load of each channel and predict resource usage. The error trend prediction subunit is used to analyze historical error data and predict future error change trends.

3. The 16-channel high-precision synchronous data acquisition system based on a multi-parameter model according to claim 1, characterized in that, The model update unit includes a parameter fusion subunit, a weight allocation subunit, and a model output subunit. The parameter fusion subunit is used to perform weighted fusion of sampling rate parameters, channel load parameters, and error trend parameters. The weight allocation subunit is used to dynamically allocate fusion weights based on the degree of influence of each parameter on the acquisition accuracy. The model output subunit is used to output the optimized multidimensional parameter vector.

4. The 16-channel high-precision synchronous data acquisition system based on a multi-parameter model according to claim 1, characterized in that, The quality assessment unit includes a signal-to-noise ratio (SNR) calculation subunit, a stability evaluation subunit, and a channel priority ranking subunit. The SNR calculation subunit is used to calculate the real-time SNR of each channel. The stability evaluation subunit is used to evaluate the fluctuation and stability of the data in each channel. The channel priority ranking subunit is used to prioritize the channels by combining the SNR and stability indicators.

5. The 16-channel high-precision synchronous data acquisition system based on a multi-parameter model according to claim 1, characterized in that, The resource allocation unit includes a dynamic sampling allocation subunit, a buffer space scheduling subunit, and a scheduling instruction generation subunit; the dynamic sampling allocation subunit is used to dynamically allocate sampling frequencies according to channel priorities. The cache space scheduling subunit is used to allocate cache space according to the data generation rate and processing capability; The scheduling instruction generation subunit is used to generate scheduling instructions that include sampling timing and caching strategies.

6. The 16-channel high-precision synchronous data acquisition system based on a multi-parameter model according to claim 1, characterized in that, The error monitoring unit includes a synchronization error detection subunit, a drift trend analysis subunit, and an error source identification subunit; the synchronization error detection subunit is used to detect the time synchronization error between channels; the drift trend analysis subunit is used to analyze the drift characteristics of the error over time; and the error source identification subunit is used to identify the main sources of error generation.

7. The 16-channel high-precision synchronous data acquisition system based on a multi-parameter model according to claim 1, characterized in that, The compensation calculation unit includes a compensation amount calculation subunit, a compensation strategy selection subunit, and a compensation parameter output subunit; the compensation amount calculation subunit is used to calculate the compensation amount of each channel based on the detected synchronization error; The compensation strategy selection subunit is used to select a time offset compensation, phase correction compensation, or amplitude compensation strategy based on the error characteristics. The compensation parameter output subunit is used to output compensation parameters to the collaborative synchronization optimization module.

8. The 16-channel high-precision synchronous data acquisition system based on a multi-parameter model according to claim 1, characterized in that, The parameter sharing unit of the collaborative synchronization optimization module includes a parameter aggregation subunit, an inter-channel parameter exchange subunit, and a global state update subunit; the parameter aggregation subunit is used to aggregate the compensation parameters of each channel; the inter-channel parameter exchange subunit is used to transfer high-quality channel parameters to low-quality channels; and the global state update subunit is used to update the global synchronization state of the system.

9. The 16-channel high-precision synchronous data acquisition system based on a multi-parameter model according to claim 1, characterized in that, The collaborative control unit of the collaborative synchronization optimization module includes a collaborative strategy generation subunit, a synchronization signal generation subunit, and a feedback optimization subunit; the collaborative strategy generation subunit is used to generate a multi-channel collaborative control strategy based on the global state; the synchronization signal generation subunit is used to generate 16-channel synchronization control signals; and the feedback optimization subunit is used to feed the optimization results back to the multi-parameter model driving module for model parameter updates.

10. The 16-channel high-precision synchronous data acquisition system based on a multi-parameter model according to claim 1, characterized in that, The data acquisition hardware interface is connected to the ADM-4586 analog-to-digital acquisition and control module. The ADM-4586 analog-to-digital acquisition and control module includes two acquisition cards, each of which provides eight channels for acquiring analog current signals. The analog current signal types of the 16 acquisition channels are 0-20mA or 4-20mA.

Images