Data acquisition adaptive control method and system of multi-channel intelligent instrument

Abstract

The application discloses a data acquisition adaptive control method and system of a multi-channel intelligent instrument, relates to the technical field of data acquisition and processing, and comprises the following steps: synchronously acquiring real-time data streams of a plurality of sensor channels to generate an original data set, and integrating and processing the original data set into a standardized data matrix; dynamically converting the standardized data matrix into protocol format data and transmitting the protocol format data to an industrial control network to analyze and generate control decision instructions; and driving a plurality of relays to perform reverse control, verifying an adaptive control strategy formulated after optimization, and comprehensively managing and controlling. The application solves the technical problems of poor synchronism, insufficient data processing adaptability, inflexible protocol conversion and low reverse control precision of a traditional data acquisition control system in multi-channel data acquisition, achieves the technical effects of real-time and synchronous multi-channel data acquisition and adaptive processing, dynamic adaptation of industrial protocol conversion, optimization of control precision through a reverse control closed loop, and meeting the requirements of an industrial scene for precise multi-channel data acquisition and comprehensive management.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of data acquisition and processing, in particular to a data acquisition adaptive control method and system for multi-channel intelligent instruments. BACKGROUND

[0002] In industrial control processes, data acquisition control systems are widely used in various scenarios. Multi-channel intelligent instruments, as key components of these systems, are crucial for accurate data acquisition and control. However, traditional data acquisition control systems have significant shortcomings when it comes to multi-channel data acquisition. The single acquisition architecture used by these systems struggles to achieve real-time synchronous acquisition of data from multiple channels, and the data processing methods lack adaptability, making it difficult to effectively handle data diversity and complexity. Additionally, traditional systems cannot dynamically adjust to the state of industrial control networks during protocol conversion, resulting in low data transmission efficiency. Furthermore, there is a lack of real-time verification and backtracking optimization mechanisms for control results during reverse control, making it difficult to ensure control accuracy. These issues make it difficult for traditional data acquisition control systems to meet the demands of industrial scenarios for data accuracy, transmission efficiency, and control precision in multi-channel data acquisition and comprehensive management. SUMMARY

[0003] The present application provides a data acquisition adaptive control method and system for multi-channel intelligent instruments to address the technical problems of poor synchronization, insufficient data processing adaptability, inflexible protocol conversion, and low reverse control accuracy in traditional data acquisition control systems during multi-channel data acquisition.

[0004] In a first aspect, the present application provides a data acquisition adaptive control method for multi-channel intelligent instruments. The method includes synchronously acquiring real-time data streams from multiple sensor channels, generating a multi-channel raw data set, performing adaptive data integration processing on the multi-channel raw data set to generate a standardized data matrix, dynamically converting the standardized data matrix to obtain protocol format data transmission to an industrial control network for multi-channel flow rate analysis, generating control decision instructions, executing the control decision instructions to drive multi-channel relays for reverse control, generating control results, real-time verification of the control results for backtracking optimization, and developing adaptive control strategies for comprehensive management and control of data from multi-channel intelligent instruments.

[0005] A second aspect of this application provides a data acquisition adaptive control system for multi-channel intelligent instruments. The system includes: a standardized data matrix generation module, used to synchronously acquire real-time data streams from multiple sensor channels, generate a multi-channel raw dataset, and perform adaptive data integration processing on the multi-channel raw dataset to generate a standardized data matrix; a control decision command generation module, used to dynamically convert the standardized data matrix to obtain protocol-formatted data transmission to an industrial control network for multi-channel flow rate analysis, generating control decision commands; and a comprehensive management and control execution module, used to execute the control decision commands to drive multiple relays for reverse control, generate control results, verify the control results in real time for backtracking optimization, and formulate adaptive control strategies for comprehensive management and control of the data from the multi-channel intelligent instruments.

[0006] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0007] This application achieves comprehensive management and control of multi-channel data by synchronously acquiring real-time data streams from multiple sensor channels, generating a standardized matrix through adaptive data integration, dynamically converting it into protocol data for flow rate analysis, driving relays for reverse control and backtracking optimization, thereby accurately realizing comprehensive management and control of multi-channel data. This makes the data acquisition and control system more precise and reliable in acquiring, processing, and controlling multi-channel data, achieving real-time synchronous acquisition and adaptive processing of multi-channel data, dynamic adaptation to industrial protocol conversion, and improved control accuracy through reverse control closed-loop optimization, thus meeting the technical requirements of accurate acquisition and comprehensive management of multi-channel data in industrial scenarios. Attached Figure Description

[0008] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0009] Figure 1 This is a flowchart illustrating the data acquisition adaptive control method for multi-channel intelligent instruments provided in this application embodiment.

[0010] Figure 2 This is a schematic diagram of the data acquisition adaptive control system for a multi-channel intelligent instrument provided in an embodiment of this application.

[0011] Figure labeling: Standardized data matrix generation module 1, control decision instruction generation module 2, comprehensive management control execution module 3. Detailed Implementation

[0012] This application provides a data acquisition adaptive control method and system for multi-channel intelligent instruments, which solves the technical problems of poor synchronization, insufficient data processing adaptability, inflexible protocol conversion, and low reverse control accuracy in traditional data acquisition and control systems in multi-channel data acquisition.

[0013] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application.

[0014] It should be noted that the terms "first," "second," etc., in the specification and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or server that includes a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or modules not explicitly listed or inherent to such processes, methods, products, or devices.

[0015] Example 1, as Figure 1 As shown, a data acquisition adaptive control method for multi-channel intelligent instruments includes:

[0016] Step A100: Synchronously acquire real-time data streams from multiple sensor channels to generate a multi-channel raw dataset, perform adaptive data integration processing on the multi-channel raw dataset, and generate a standardized data matrix.

[0017] Specifically, the process involves synchronously acquiring real-time data streams from multiple sensor channels to generate a raw dataset, including configuring channels, building a priority table, activating acquisition, clock synchronization and noise reduction, and dynamically allocating data streams. The specific steps are detailed in A110-A140.

[0018] Adaptive integration processing is performed on the multi-channel raw dataset to generate a standardized matrix, covering validity and integrity detection, anomaly identification compensation, dataset update and local normalization reconstruction. Specific steps are explained in detail in A150-A190.

[0019] Step A200: Dynamically convert the standardized data matrix to obtain protocol format data, transmit it to the industrial control network for multi-channel flow velocity analysis, and generate control decision commands.

[0020] Optionally, the standardized matrix can be dynamically converted into a protocol format and transmitted to the industrial control network for flow rate analysis to generate control decision commands. This includes steps such as detecting network status, converting protocols, formatting, and analyzing flow rates. Specific steps are detailed in A210-A240.

[0021] Step A300: Execute the control decision command to drive multiple relays for reverse control, generate control results, verify the control results in real time for backtracking optimization, and formulate an adaptive control strategy to comprehensively manage and control the data of the multi-channel intelligent instrument.

[0022] In one embodiment of this application, the execution of control decision instructions drives the relay to perform reverse control and backtracks to optimize and formulate an adaptive control strategy, including parsing the action sequence, calculating the conflict matrix, verifying the control results, diagnosing the root cause and optimizing the strategy. The specific steps are described in detail in A310-A350.

[0023] When using a well-defined adaptive control strategy to comprehensively manage and control data from multi-channel smart instruments, for control decisions generated by flow velocity analysis, the mutual exclusion relationship between K1 and K2 is predefined through the relay action conflict matrix to eliminate execution conflicts. When the turbulence suppression mode is triggered (i.e., I is greater than 0.35 and the pulsation frequency f is less than 2.0Hz), K1 is executed first to close the valve to 70% and K2 is delayed. At the same time, the target flow velocity is compared with the actual feedback through a real-time verification mechanism. If the deviation exceeds 10%, backtracking is initiated to calculate the dynamic deviation value, such as ΔV = 0.8m / s. Based on this, the three-dimensional diagnostic map is updated, and the relay drive current is adjusted, such as increasing K3 from 50mA to 80mA to shorten the response time to 40ms. Finally, a closed-loop management is formed, which includes adaptive acquisition parameters, dynamic optimization of protocol conversion, elimination of control execution conflicts, and iterative backtracking of results. This improves the efficiency of multi-channel data acquisition and achieves a relay control success rate of 99%, meeting the real-time data and control accuracy requirements in industrial scenarios.

[0024] Furthermore, step A100 in the method provided in this application embodiment includes:

[0025] A110: Configure data acquisition for multiple sensor channels and build a channel priority mapping table.

[0026] A120: Activate multiple sensor channels according to the channel priority mapping table to acquire data in real time and obtain a real-time data stream.

[0027] A130: The real-time data stream is clock-synchronized according to the timestamp, and noise reduction and filtering are performed based on the synchronization result to generate a cached data stream.

[0028] A140: Dynamically allocate the cached data stream according to the channel priority mapping table to determine the original dataset for multiple channels.

[0029] In this embodiment, the sensor channel is an independent data acquisition path in a multi-channel smart meter used to access sensors or instrument devices. Its hardware is based on a high-end ARM processor, and each channel is equipped with an independent acquisition chip, such as the channel corresponding to the RS232 instrument communication interface terminal block, which can realize real-time synchronous acquisition of multi-source data.

[0030] Specifically, in industrial data acquisition scenarios, when a multi-channel smart instrument needs to construct a channel priority mapping table for four RS232 instrument channels, priority evaluation dimensions are first set based on the criticality of the process parameters and the frequency of data changes of the sensors connected to each channel. For example, temperature and pressure monitoring channels, which reflect the core operating status of the equipment, are classified as high priority, while auxiliary parameter monitoring channels are classified as low priority. In practice, the four RS232 channels are numbered CH1 to CH4 by the instrument's built-in ARM processor, and priority labels are assigned to each channel according to the MODBUS industrial control protocol: CH1 and CH2 are set to P1 level, i.e., the highest priority, corresponding to the acquisition of critical process parameters, with an acquisition cycle of 50ms; CH3 and CH4 are set to P2 level, i.e., the second highest priority, corresponding to the acquisition of auxiliary parameters, with an acquisition cycle of 100ms. The priority mapping table is stored in the instrument's memory as a two-dimensional array. The array fields include the channel number, protocol type, acquisition cycle, and priority level, such as [CH1,RS232,50ms,P1], [CH2,RS232,50ms,P1], [CH3,RS232,100ms,P2], and [CH4,RS232,100ms,P2]. This mapping table can be dynamically modified via instrument buttons or host computer software. When industrial processes are adjusted, the channel priority and acquisition cycle can be adjusted in real time to ensure that the key data acquisition frequency reaches 20 times per second with a 50ms cycle, and the auxiliary data acquisition frequency reaches 10 times per second with a 100ms cycle, thereby achieving differentiated management of data acquisition from the four RS232 channels.

[0031] Next, based on the constructed channel priority mapping table, the high-end ARM processor reads the priority configuration data stored in EEPROM (Electrically Erasable Programmable Read-Only Memory, used to store configuration data such as the channel priority mapping table) via the SPI bus, and parses the acquisition cycle and priority tag for each RS232 channel, namely CH1 to CH4. For P1-level channels CH1 and CH2, the processor sets a 50ms interrupt cycle through the timer module to trigger the corresponding independent ADC acquisition chip, such as ADS1115, to sample the sensor signal; for P2-level channels CH3 and CH4, a 100ms interrupt cycle is set. Each time an interrupt is triggered, the ARM processor activates the analog switch of the corresponding channel in priority order, and the independent acquisition chip converts the sensor analog signal into a digital data stream at 24-bit resolution, i.e., acquires the real-time data stream.

[0032] Then, when synchronizing the acquired real-time data stream by timestamp, the ARM processor's built-in RTC real-time clock module generates timestamps accurate to 0.1ms, such as 2025-07-03-10:30:22.543, and bundles them with the data from each channel, for example, 0x3A2F of CH1, to form a time-series data stream. The hardware clock synchronization mechanism triggers the synchronization clock pins of each independent acquisition chip via the SPI bus, keeping the timestamp deviation of the four channels within 1ms. Noise reduction uses a median filtering algorithm: using five consecutive sampling points as a window, such as Dn-2 to Dn+2, the data within the window is sorted by value, and the median value is used as the filtering result; simultaneously, the mean of the window data is calculated. If the amplitude of a data point exceeds ±20% of the mean (e.g., if the mean is 100, exceeding 120 or falling below 80), it is determined to be impulse noise and replaced with the median. After this processing, the signal-to-noise ratio of the original data stream is improved, and the proportion of noise points in the generated cached data stream is reduced to less than 5%, ensuring that the data integrity reaches more than 99.5%.

[0033] Finally, the acquired data is transmitted to the processor's internal FIFO buffer via a high-speed SPI (Serial Peripheral Interface) bus. Cache space is allocated according to a priority mapping table: P1 level data is stored in the dual-port RAM high-speed area, and P2 level data is stored in the ordinary SRAM area (Static Random Access Memory, used for temporary storage of the acquired data stream; cache space is dynamically allocated according to channel priority). This ultimately forms a three-dimensional data stream structure containing channel number, timestamp, and raw data, such as [CH1, 2025-07-03 10:30:22.543, 0x3A2F]. After integration, a multi-channel raw dataset is obtained. During this process, the P1 level channels achieve a sampling frequency of 20 times / second, and the P2 level achieves 10 times / second, resulting in an overall four-channel parallel sampling frequency exceeding 60 times / second. Combined with the anti-aliasing filter circuit of the independent acquisition chip, this ensures that the time synchronization error of the multi-channel raw dataset is less than 1ms, and the signal-to-noise ratio is better than 80dB.

[0034] By constructing a channel priority mapping table, real-time acquisition, clock synchronization noise reduction, and dynamic data stream allocation, real-time synchronous acquisition and orderly management of multi-channel data are achieved, providing an accurate and complete raw data foundation for subsequent data integration and processing. This effectively solves the problems of poor data synchronization and untimely acquisition of key data in traditional technologies.

[0035] Furthermore, step A100 in the method provided in this application embodiment includes:

[0036] A150: Traverse the multi-channel original dataset to perform data validity checks and generate the first detection result.

[0037] A160: Traverse the multi-channel original dataset to perform data integrity checks and generate a second detection result.

[0038] A170: Based on the overlap analysis of the first detection result and the second detection result, anomaly identification is performed on the multi-channel raw dataset based on the overlap data to determine abnormal data points.

[0039] A180: Based on the abnormal data points, the multi-channel original dataset is dynamically compensated, and the multi-channel original dataset is updated according to the compensation results to obtain the multi-channel updated dataset.

[0040] A190: The multi-channel updated dataset is locally normalized to the continuous data stream, and the data processing results are reconstructed according to the multi-channel dimension to generate the standardized data matrix.

[0041] Optionally, firstly, a comprehensive scan of the multi-channel raw dataset is performed to check data validity. Taking four RS232 channels as an example, based on the sensor range of each channel (e.g., a temperature sensor range of -20℃ to 100℃), it is determined whether the data exceeds the physically reasonable range. If the value collected by channel CH1 at a certain moment is 150℃, exceeding the upper limit of 100℃, it is marked as invalid data; otherwise, it is marked as valid data, generating the first detection result. This step can detect data that exceeds the range limit.

[0042] Next, a data integrity check is performed. The time series data in the original dataset is traversed to check for missing values ​​exceeding two consecutive sampling periods. For example, in a P1-level channel with a 50ms acquisition period, there might be no data for 100ms consecutively. Taking four-channel parallel acquisition as an example, if channel CH1 has no data records between 10:30:22.543 and 10:30:22.643, it is determined to be an integrity missing value; otherwise, it is determined to be normal integrity, and a second detection result is generated. This step can pinpoint the data loss situation.

[0043] An overlap analysis is performed on the first and second detection results to construct an anomaly data matrix: the coordinates of invalid data from the validity test are cross-compared with the coordinates of missing data from the integrity test, and data points in the overlapping areas are marked as anomaly data points. For example, if a data point is both out of range and has adjacent missing data, it is preferentially marked as a high-risk anomaly point. This step can accurately locate compound anomaly data.

[0044] Based on outlier data points, a cubic spline interpolation method is used for dynamic compensation: for a single outlier point, an interpolation curve is constructed using three valid data points before and after it; for continuously missing data, cross-channel compensation is performed by considering the correlation between adjacent channels, such as the Pearson correlation coefficient r > 0.9 between CH1 and CH2 temperature data. After compensation, the original dataset is updated to obtain a multi-channel updated dataset, thereby improving data integrity.

[0045] Finally, continuous data flow is extracted from the multi-channel updated dataset through continuous data detection. Based on this, data mutation detection is performed to divide the window boundary and set a dynamic window. Local normalization processing is performed on the continuous data flow to generate data processing results containing channel normalization sequences. Channel coupling compensation is performed according to this sequence to determine data compensation terms. The spatial weight matrix defined by the physical arrangement of sensor channels is retrieved and topological mapping is performed to generate multi-channel dimensions. Finally, the standardized data matrix is ​​reconstructed by combining the data compensation terms. The specific steps are explained in detail in A191-A196.

[0046] Through steps such as validity and integrity collaborative detection, overlap analysis to locate anomalies, dynamic interpolation compensation, and local normalization reconstruction, adaptive integration processing of multi-channel data is achieved. The generated standardized data matrix improves the data signal-to-noise ratio, providing a high-quality data foundation for subsequent protocol conversion and multi-channel flow velocity analysis. It effectively solves the problems of missed data anomaly detection, insufficient compensation accuracy, and inconsistent formats in traditional technologies.

[0047] Furthermore, step A190 in the method provided in this application embodiment includes:

[0048] A191: Perform continuous data detection based on the multi-channel updated dataset and extract continuous data streams.

[0049] A192: Perform data mutation detection based on the continuous data stream, divide the window boundary according to the mutation detection result, and set a dynamic window.

[0050] A193: Based on the dynamic window, perform local normalization processing on the continuous data stream to generate data processing results, the data processing results including channel normalization sequences.

[0051] A194: Perform channel coupling compensation according to the channel normalization sequence to determine the data compensation term.

[0052] A195: Retrieve the physical arrangement definition spatial weight matrix of multiple sensor channels, perform topological structure mapping based on the spatial weight matrix, and generate multi-channel dimensions.

[0053] A196: Based on the multi-channel dimensions, the data processing results are reconstructed in conjunction with the data compensation terms to generate the standardized data matrix.

[0054] In this embodiment of the application, channel coupling compensation refers to the process of correcting the deviation of the correlation between adjacent channel data after local normalization processing.

[0055] Specifically, firstly, continuous data detection is performed on the multi-channel update dataset: taking the temperature data of 4 RS232 channels as an example, the continuous sampling interval threshold is set to twice the acquisition period, that is, 100ms for P1 channel and 200ms for P2 channel. If there is no data interruption in CH1 channel from 10:30:22.543 to 10:30:22.643, it is determined to be a continuous data stream, and a continuous sequence containing 10 timestamps is extracted.

[0056] When detecting data mutations based on continuous data streams, a third-order differential gradient algorithm is used: the gradient value |ΔDn|=|Dn-Dn-1| between adjacent data points is calculated. When the gradient value exceeds 1.5 times the mean gradient (e.g., the mean gradient is 5℃ / s, and the mutation threshold is set to 7.5℃ / s), it is determined to be a mutation point. Taking the CH2 channel continuous data stream as an example, if the gradient value reaches 8℃ / s at the 5th sampling point, exceeding the threshold, a window boundary is defined at the next sampling point after that point, and a dynamic window [Dn-4 to Dn] containing the previous 5 points is set to accurately identify the data mutation boundary.

[0057] Next, local normalization is performed on the continuous data stream within the dynamic window, using the minimum-maximum normalization formula: Dn norm =(Dn-Dwindow) min ) / (Dwindow max -Dwindow min ), where Dwindow min and Dwindow max These represent the minimum and maximum values ​​of the data within the window. Taking the data [25,30,35,40,45] within a dynamic window of the CH3 channel as an example, after normalization, a channel normalization sequence of [0,0.25,0.5,0.75,1] is generated, which uniformly maps the data range to the [0,1] interval and reduces the local standard deviation.

[0058] Then, when performing channel coupling compensation according to the channel normalized sequence, the Pearson correlation coefficient between adjacent channels is first calculated. For example, if the correlation coefficient between CH1 and CH2 is r = 0.92, and r > 0.8, then the compensation term is determined based on the difference in the normalized sequence: ΔD = |DCH1 norm -DCH2 norm The offset is calculated by multiplying |×0.1 to compensate for the offset of the normalized CH2 sequence, ensuring data consistency between channels. Taking a normalized CH1 value of 0.8 and a normalized CH2 value of 0.6 as an example, the compensation term ΔD = 0.02, and the updated CH2 value is 0.62, reducing the channel coupling error to below 5%.

[0059] Next, the physical arrangement of the four sensor channels is retrieved to define the spatial weight matrix: Assuming CH1-CH4 are distributed in a rectangular array with a spacing of 20cm between adjacent channels, a 4×4 weight matrix W is constructed, where the diagonal element Wii = 1 and the off-diagonal element Wij = 1 / (dij+1), where dij is the distance between channels i and j in cm. For example, the spacing between CH1 and CH2 is 20cm, W12 = 1 / (20+1) = 0.0476, generating the spatial weight matrix: W = [1, 0.0476, 0.0244, 0.0169; 0.0476, 1, 0.0476, 0.0244; 0.0244, 0.0476, 1, 0.0476; 0.0169, 0.0244, 0.0476, 1]. Based on this matrix, a topological structure mapping is performed, transforming the one-dimensional channel sequence into a two-dimensional spatial matrix, generating a multi-channel dimension containing spatial weights.

[0060] Finally, based on multi-channel dimensions, the data processing results and data compensation terms are reconstructed using spatial weights: Standardized matrix element Mij = Wij × (Dij) norm +ΔDij), where Dij norm Let W13 be the j-th normalized value of channel i, and ΔDij be the corresponding compensation term. Taking the 3rd data point of CH1 as an example, W13 = 0.0244, D13 norm =0.5, ΔD13=0.01, then M13=0.0244×(0.5+0.01)=0.0124, generating a 4×10 standardized data matrix, improving the uniformity of matrix elements, and controlling the spatial correlation error within 8%.

[0061] By continuously detecting and extracting effective sequences, dynamically dividing windows to adapt to data mutations, locally normalizing to unify the numerical range, channel coupling compensation to eliminate dimensional bias, and mapping the spatial weight matrix to the physical topology, a standardized data matrix is ​​finally reconstructed and generated. This improves the consistency of multi-channel data in the spatiotemporal dimensions and provides a standardized data foundation for subsequent protocol conversion and multi-channel flow rate analysis. It effectively solves the problems of data dimensional distortion and insufficient spatial correlation processing in traditional methods.

[0062] Furthermore, step A200 in the method provided in this application embodiment includes:

[0063] A210: Detect the industrial control network, obtain network status parameters, dynamically transform the standardized data matrix based on the network status parameters, and determine the target industrial protocol data.

[0064] A220: Based on the target industrial protocol data, perform protocol format layout to generate protocol format data.

[0065] A230: Transmit the protocol format data to the industrial control network for protocol analysis to determine the conversion protocol.

[0066] A240: Perform multi-channel flow rate analysis on the protocol format data according to the conversion protocol, and generate the control decision command.

[0067] In this embodiment, the industrial control network refers to a communication link used to connect multi-channel smart instruments and external control systems, supporting data transmission of industrial protocols such as HART and MODBUS. Industrial protocol data refers to data formats conforming to industrial control standard protocols, such as HART and MODBUS.

[0068] Specifically, firstly, the network monitoring module built into the ARM processor is used to detect the real-time status parameters of the industrial control network, including key indicators such as bandwidth, latency, packet loss rate, and network load rate. Taking the detected bandwidth parameter as an example, when the network bandwidth is below 20Mbps, it is determined to be a low-bandwidth environment, triggering an efficient protocol switching strategy; if the bandwidth is above 80Mbps, a high-speed protocol transmission mode is adopted. This step can achieve real-time and accurate monitoring of network status parameters.

[0069] Based on the acquired network status parameters, the standardized data matrix is ​​dynamically converted to determine the target industrial protocol data: When a HART protocol request instruction is received from an external control system, the ARM processor encapsulates the 4×10 standardized data matrix according to the HART protocol frame structure, adds an 8-bit device identifier code, such as 0xAB, and a 16-bit CRC check bit, forming a complete HART frame containing a preamble, address field, data field, and check field, with a data transmission rate adapted to the 1.2kbps HART protocol standard; if the request instruction is for the MODBUS protocol, the standardized data matrix is ​​mapped to a MODBUS register address sequence, such as the starting address 40001, with each matrix element corresponding to a 16-bit register value, and the protocol data packet is output at a rate of 100Mbps through the industrial Ethernet interface. This conversion process has high protocol adaptation accuracy and a conversion delay of less than 1ms.

[0070] Next, when formatting the protocol based on the target industrial protocol data, if it is the HA RT protocol, it should be formatted according to the three-layer structure of physical layer FSK signal, data link layer frame structure, and application layer command set, ensuring that the data field length does not exceed 25 bytes; if it is the MODBUS protocol, it should be formatted in the order of function code (such as 03H read holding register), start address, data length, and register value, with CRC-16 check code added to generate formatted data that conforms to the protocol specification, and the formatting error rate should be controlled below 0.01%.

[0071] Then, after transmitting the protocol format data to the industrial control network via RS485 or HART communication interface, the protocol parsing module is activated: it parses the preamble, address field, and checksum of the data frame and matches them with the built-in protocol library containing protocol standards such as HART 7.0 and MODBUS RTU to determine the actual converted protocol type. Taking the parsing of HART frames as an example, by recognizing the preamble 55H and address field 0-15, the protocol type can be determined within 5ms, with high protocol analysis accuracy.

[0072] Finally, after executing the parsing conversion protocol to determine the first and second conversion protocols, the flow rate data of the former is sorted according to the channel priority mapping table and divided into N data segments. A frame header is added to construct the first data packet. The latter is reassembled based on the flow rate data to construct the second data packet. The two data packets are weighted and fused. Based on the fusion result, multi-channel control analysis is performed to generate control decision instructions. The specific steps are explained in detail in A241-A248.

[0073] By real-time monitoring of network status parameters, dynamic adaptation of HART / MODBUS protocol conversion, standardized formatting, and precise protocol analysis, the efficient transmission of standardized data matrices in industrial control networks is achieved. This improves protocol conversion power and reduces data transmission latency, providing a stable and reliable protocol data foundation for multi-channel flow rate analysis. It effectively solves the problems of poor network adaptability and low protocol conversion efficiency in traditional technologies.

[0074] Furthermore, step A240 in the method provided in this application embodiment includes:

[0075] A241: Parse the conversion protocol to determine the first conversion protocol and the second conversion protocol.

[0076] A242: When the conversion protocol is the first conversion protocol, the flow rate data is sorted according to the channel priority mapping table to generate a data chain.

[0077] A243: Divide the data chain into N data segments, where N is an integer greater than 1.

[0078] A244: Add frame headers to the N data segments and reassemble them to construct the first data packet.

[0079] A245: When the conversion protocol is the second conversion protocol, the register address is defined based on the flow rate data.

[0080] A246: Reconstruct the flow rate data according to the registered address to build a second data packet.

[0081] A247: The first data packet and the second data packet are weighted and calculated, and the first data packet and the second data packet are fused according to the weight coefficient to generate a data fusion result.

[0082] A248: Perform multi-channel control analysis based on the data fusion results to generate the control decision command.

[0083] In this embodiment, the first conversion protocol is the HART protocol. The second conversion protocol is the MODBUS protocol.

[0084] Specifically, firstly, the conversion protocol is identified through the protocol parsing module: when the FSK modulation signal of the HART protocol with a center frequency of 1200Hz / 2200Hz or the RTU frame start flag of the MODBUS protocol is detected, and the time interval is 3.5 consecutive bytes, the first conversion protocol is determined to be HART and the second conversion protocol is determined to be MODBUS.

[0085] When the conversion protocol is HART, the flow rate data of the four channels are sorted in descending order based on the channel priority mapping table, i.e., the P1 / P2 level configuration of CH1-CH4, to generate a one-dimensional data chain containing priority labels. Taking P1 level channels CH1 (flow rate 10 m / s), CH2 (flow rate 8 m / s), and P2 level channels CH3 (flow rate 5 m / s), CH4 (flow rate 3 m / s) as an example, the sorted data chain is [10, 8, 5, 3], corresponding to the channel priority sequence [P1, P1, P2, P2]. This data chain is divided into segments according to the maximum length of 25 bytes of HART protocol data field, with each segment consisting of 38 bytes (including frame header / checksum) and N=2 segments. A HART frame header, a preamble of 55H, and an address field of 0x01 are added to each segment, and they are reassembled into the first data packet conforming to the HART 7.0 protocol.

[0086] When the conversion protocol is MODBUS, the register addresses are defined based on the flow rate data: the four channels of flow rate data are mapped to MODBUS holding register addresses 40001-40004, with each 16-bit register storing one channel flow rate value in units of 0.1 m / s. For example, a CH1 flow rate of 10 m / s is stored as register value 0x0064. The flow rate data is reassembled according to the register address order to generate a MODBUS RTU frame containing function code 03H, start address 40001, data length 04H, and register values. This forms the second data packet, which is transmitted at a rate of 9600 bps via the RS485 interface.

[0087] Next, when weighting the first and second data packets, the weight coefficients are dynamically allocated based on network status parameters: when the industrial control network bandwidth is ≥80Mbps and the latency is ≤10ms, the MODBUS protocol weight is set to 0.6 and the HART protocol weight is 0.4; when the bandwidth is <20Mbps, the weights are adjusted to 0.3 and 0.7. Taking a certain operating condition with a MODBUS weight of 0.6 and a HART weight of 0.4 as an example, the flow rate data of the two data packets are weighted and fused: the fused flow rate = 0.6 × MODBUS flow rate + 0.4 × HART flow rate, generating the data fusion result, with the fusion error controlled within ±5%.

[0088] Finally, flow analysis is performed based on the data fusion results to obtain turbulence intensity parameters and pulsation frequency parameters. Based on these parameters, flow velocity feature vectors are extracted. By mapping these vectors with multi-channel intelligent instruments, a flow velocity-control mapping rule base is constructed. The rule base is then traversed to filter and determine control decision instructions. The specific steps are explained in detail in A248-1-A248-3.

[0089] By accurately identifying protocol types, processing data by protocol, and dynamically weighting and fusing data, adaptive analysis of multi-channel flow rate data under HART and MODBUS protocols is achieved, improving the data fusion accuracy to within ±5%. This provides highly reliable flow rate data support for the generation of control decision commands and effectively solves the problems of poor protocol compatibility and large data fusion errors in traditional technologies.

[0090] Furthermore, step A248 in the method provided in this application embodiment includes:

[0091] A248-1: Based on the data fusion results, flow analysis is performed to obtain turbulence intensity parameters and pulsation frequency parameters.

[0092] A248-2: Extract flow velocity features based on the turbulence intensity parameters and the pulsation frequency parameters to generate a flow velocity feature vector.

[0093] A248-3: Based on the flow velocity feature vector, control mapping is performed on the multi-channel intelligent instrument to construct a flow velocity-control mapping rule base, and the flow velocity-control mapping rule base is traversed to perform control filtering to determine the control decision command.

[0094] In this embodiment, turbulence intensity is a dimensionless parameter used to characterize the degree of fluid flow turbulence, obtained by calculating the ratio of the standard deviation of multi-channel fused velocity data to the average velocity. Pulsation frequency refers to the characteristic frequency of the periodic fluctuation of fluid velocity over time, obtained by performing Fast Fourier Transform (FFT) analysis on continuous velocity data.

[0095] In one embodiment, based on the aforementioned data fusion results, such as the velocity data fused from HART and MODBUS protocols, a multi-channel velocity analysis algorithm is used to analyze the flow state: First, the average velocity V is calculated from the fused velocity data of the four channels, such as Va = 10 m / s, Vb = 8 m / s, Vc = 5 m / s, and Vd = 3 m / s. avg Then, the turbulence intensity parameter I = σ / V is characterized by calculating the standard deviation σ of the flow velocity data. avg Meanwhile, the Fast Fourier Transform (FFT) was used to perform spectral analysis on 100 sets of continuous flow velocity data, and the frequency f = 2.5 Hz corresponding to the peak power spectral density was extracted as the pulsating frequency parameter. This step can control the error of turbulence intensity calculation within ±5% and achieve a frequency resolution of 0.1 Hz.

[0096] Next, when extracting flow velocity features based on turbulence intensity parameters and fluctuation frequency parameters, a three-dimensional feature vector V = [I, f, V] is constructed. avg Taking a certain operating condition as an example, when I = 0.42 exceeds the threshold of 0.35, and f = 1.8 Hz is lower than the threshold of 2.0 Hz, it is determined to be a turbulent state; if V avg If the flow rate is 4.5 m / s, which is lower than the minimum flow rate threshold of 5 m / s, it is considered an insufficient flow rate. During feature vector generation, parameters are updated in real time through a sliding window with a window length of 10 seconds to ensure that the feature vector matches the current flow state with a degree of over 95%.

[0097] Finally, when controlling and mapping the multi-channel smart instrument based on the flow velocity feature vector, a turbulence suppression mode is triggered when the turbulence intensity parameter is greater than the maximum intensity threshold and the pulsation frequency parameter is less than the minimum frequency threshold. Control decision instructions are generated by matching the flow velocity-control mapping rule base according to this mode. At the same time, the average flow velocity data is calculated based on the flow velocity feature vector. When the average flow velocity is less than the minimum flow velocity threshold, a flow velocity compensation mode is triggered, and control decision instructions are generated by matching the rule base according to this mode. The specific steps are explained in detail in A248-3A-A248-3D.

[0098] By integrating turbulence intensity and pulsation frequency calculations from flow velocity data, constructing three-dimensional feature vectors, and matching with decision tree rule bases, the system achieves accurate identification of multi-channel flow velocity states and generation of control commands. This improves the accuracy of matching control decisions with actual flow conditions, provides a scientific basis for relay reverse control, and effectively solves the problems of disconnect between control strategies and flow state changes and insufficient control precision in traditional technologies.

[0099] Furthermore, step A248-3 in the method provided in this application embodiment includes:

[0100] A248-3A: When the turbulence intensity parameter is greater than the maximum intensity threshold and the pulsation frequency parameter is less than the minimum frequency threshold, the turbulence suppression mode is triggered.

[0101] A248-3B: Match the flow velocity-control mapping rule base according to the turbulence suppression mode to generate the control decision command.

[0102] A248-3C: Calculate average flow velocity data based on the flow velocity feature vector. When the average flow velocity data is less than the minimum flow velocity threshold, trigger the flow velocity compensation mode.

[0103] A248-3D: Match the flow rate-control mapping rule base according to the flow rate compensation mode to generate the control decision command.

[0104] In this embodiment, the flow rate-control mapping rule base is a database used in multi-channel smart meters to store the correspondence between flow rate characteristics and control strategies.

[0105] Optionally, a maximum intensity threshold I is set based on the turbulence intensity parameter I and the fluctuation frequency parameter f in the flow velocity feature vector. max =0.35, minimum frequency threshold f min =2.0Hz. When I>0.35 and f<2.0Hz, the fluid is determined to be in a turbulent state, triggering the turbulence suppression mode. For example, under a certain operating condition, I=0.42 and f=1.8Hz are calculated, which meets the triggering conditions. At this time, the turbulence suppression rule is executed: relay K1 is closed and K2 is opened, correspondingly adjusting the valve opening to 70% to reduce fluid disturbance. This judgment logic, together with the aforementioned multi-channel flow velocity analysis algorithm, such as eliminating abnormal flow velocities and calculating average flow velocity, forms a data closed loop, improving the accuracy of turbulence state identification.

[0106] When the turbulence suppression mode is triggered, the system traverses the flow velocity-control mapping rule base according to this mode. The rule base is built based on the output capability of the instrument's four relays and stores more than 20 sets of operating condition-control mapping relationships. Taking relays K1-K4 as an example, the corresponding rule for turbulence suppression in the rule base is K1 closed (valve opening 70%) + K2 open (bypass prohibited). The matching process uses the Euclidean distance algorithm to calculate the matching degree between the current feature vector and the rule conditions, selects the rule with the smallest distance to generate the control decision command, and the command generation delay is less than 200ms, and the relay action response time is ≤50ms.

[0107] Meanwhile, based on the average flow velocity data V in the flow velocity feature vector avg Set minimum flow rate threshold V min = 5.0 m / s, when V avg A flow rate < 5.0 m / s is considered insufficient, triggering the flow rate compensation mode. For example, if the fused flow rate of the 4 channels is [4.2, 4.5, 4.8, 4.3] m / s, the calculated V avg =4.45m / s < 5.0m / s, triggering the flow rate compensation mode and executing the flow rate compensation rule: relay K3 closes to start the auxiliary pump and increase the pipeline flow rate. This threshold setting is consistent with the logic of triggering compensation when the flow rate is less than (1-10%) of the reference value for 5 consecutive times. The flow rate anomaly detection sensitivity reaches ±0.5m / s.

[0108] After the flow rate compensation mode is triggered, the system matches the corresponding rule in the rule base according to the mode, such as K3 closing (starting the auxiliary pump) + K4 opening (closing the relief valve). The optimal control strategy is quickly selected through the decision tree structure to ensure that the control decision command is generated within 100ms, so that the flow rate rises back to above the threshold within 2s.

[0109] By setting turbulence intensity and frequency thresholds to trigger suppression modes and average flow velocity thresholds to trigger compensation modes, and combining precise matching with the relay control rule base, adaptive identification and control of multi-channel flow velocity anomalies are achieved. This improves the success rate of turbulence suppression and shortens the flow velocity compensation response time, effectively solving the problems of lagging flow anomaly identification and coarse control strategies in traditional technologies, and improving the accuracy and reliability of industrial fluid control.

[0110] Furthermore, step A300 in the method provided in this application embodiment includes:

[0111] A310: Execute the control decision command to drive multiple relays to perform analysis, extract the relay action sequence, perform conflict calculation based on the relay action sequence, and determine the action conflict matrix.

[0112] A320: Perform reverse control on the multiple relays according to the aforementioned action conflict matrix to generate control results.

[0113] A330: The control results are verified in real time. If the verification fails, the control results are backtracked to calculate the dynamic deviation value.

[0114] A340: Based on the dynamic deviation value, perform root cause diagnosis on the control results and generate a three-dimensional diagnostic map.

[0115] A350: Optimize the control results according to the three-dimensional diagnostic map and formulate the adaptive control strategy.

[0116] In one embodiment, the control decision command is first parsed to extract the action sequence of four relays (K1-K4), such as K1 closing and K3 opening. A 4×4 action conflict matrix C is constructed, where C[i][j] = 1 indicates that relays i and j have conflicting actions. For example, if K1 and K2 close simultaneously, it will cause mechanical interference in the valve, so C[1][2] = 1. This matrix is ​​based on the predefined physical connection relationship of the relays, and the conflict detection accuracy reaches 100%, which can identify conflicting items in the action sequence in real time.

[0117] Next, when implementing reverse control of multiple relays based on the action conflict matrix, if a conflict is detected, such as K1 and K2 being closed simultaneously by command, the system automatically adjusts the action sequence, prioritizing the execution of the relay action corresponding to the higher priority channel. For example, K1 associated with the P1 level channel takes precedence over K2, generating a conflict-free control result. Taking the command sequence of K1 and K2 closing under a certain operating condition as an example, after the conflict matrix detects a conflict, it retains the K1 action and delays the K2 action, ensuring that the control execution success rate is increased to over 99%.

[0118] When verifying the control results in real time, the target parameter (e.g., target flow velocity of 5 m / s) is compared with the actual feedback (e.g., actual flow velocity of 4.2 m / s). If the deviation exceeds 10%, the verification is deemed unsuccessful. At this point, a backtracking mechanism is activated to calculate the dynamic deviation value ΔV = |5 - 4.2| = 0.8 m / s, and a deviation time series is established based on historical data to pinpoint the time node and parameter dimension of the deviation occurrence.

[0119] When performing root cause diagnosis based on dynamic deviation values, a three-dimensional diagnostic atlas is invoked: the X-axis represents time, such as 0-10s; the Y-axis represents deviation parameters, such as flow velocity and turbulence intensity; and the Z-axis represents possible causes, such as sensor malfunction and relay delay. Taking flow velocity deviation as an example, the atlas shows a sudden drop in flow velocity at t=5s. Combined with sensor data, this is determined to be a delay in the K3 action, with a response time of 200ms exceeding the standard of 50ms, thus improving diagnostic accuracy.

[0120] Finally, when optimizing the control results based on the 3D diagnostic map, if the diagnosis is a relay response delay, the drive current of the relay is adjusted, for example, by increasing it from 50mA to 80mA, shortening the response time to 40ms, and the control strategy is updated to K3 priority triggering + current compensation. After optimization, the control deviation can be reduced to within ±5%, and the adaptive strategy formulation cycle is shortened to 1s.

[0121] By eliminating relay control conflicts through action conflict matrix, real-time verification and backtracking to locate deviations, three-dimensional graph root cause diagnosis and strategy optimization, adaptive adjustment of the control process is achieved, which improves the success rate of relay control, shortens the deviation response time, and effectively solves the problems of frequent control conflicts, difficult fault location and lagging strategy adjustment in traditional technologies.

[0122] In summary, the data acquisition adaptive control method for multi-channel intelligent instruments provided in this application has the following technical effects:

[0123] This application achieves the following technical effects: real-time acquisition of data streams from each channel of a multi-channel sensor; generation of buffered data streams through clock synchronization and noise reduction filtering; dynamic allocation of data by constructing a channel priority mapping table; and integration of abnormal data identification and dynamic compensation results to accurately generate a standardized data matrix. This makes the data acquisition and adaptive control of multi-channel smart instruments more precise and reliable, achieving real-time synchronous acquisition and adaptive processing of multi-channel data, dynamic adaptation to industrial protocol conversion, and improved control accuracy through reverse control closed-loop optimization. This meets the technical requirements of accurate acquisition and comprehensive management of multi-channel data in industrial scenarios.

[0124] Example 2, as Figure 2 As shown, based on the same inventive concept as in Embodiment 1 above, this application provides a data acquisition adaptive control system for a multi-channel intelligent instrument, the system comprising:

[0125] The standardized data matrix generation module 1 is used to synchronously collect real-time data streams from multiple sensor channels, generate multi-channel raw datasets, and perform adaptive data integration processing on the multi-channel raw datasets to generate a standardized data matrix.

[0126] The control decision command generation module 2 is used to dynamically convert the standardized data matrix, obtain the protocol format data, transmit it to the industrial control network for multi-channel flow velocity analysis, and generate control decision commands.

[0127] The integrated management and control execution module 3 is used to execute the control decision command to drive multiple relays to perform reverse control, generate control results, verify the control results in real time for backtracking optimization, and formulate adaptive control strategies to perform integrated management and control of the data of multi-channel smart instruments.

[0128] Furthermore, the standardized data matrix generation module 1 is used to perform the following steps:

[0129] Data acquisition configuration is performed for multiple sensor channels, and a channel priority mapping table is constructed. Multiple sensor channels are activated for real-time acquisition according to the channel priority mapping table to obtain a real-time data stream. The real-time data stream is clock-synchronized according to timestamps, and noise reduction and filtering are performed based on the synchronization results to generate a cached data stream. The cached data stream is dynamically allocated according to the channel priority mapping table to determine the original dataset for multiple channels.

[0130] Furthermore, the standardized data matrix generation module 1 is used to perform the following steps:

[0131] The process involves: traversing the multi-channel original dataset to perform data validity checks and generate a first detection result; traversing the multi-channel original dataset to perform data integrity checks and generate a second detection result; performing overlap analysis based on the first and second detection results, identifying anomalies in the multi-channel original dataset based on the overlapping data, and determining abnormal data points; dynamically compensating the multi-channel original dataset based on the abnormal data points, updating the multi-channel original dataset based on the compensation result, and obtaining a multi-channel updated dataset; performing local normalization processing on the continuous data stream using the multi-channel updated dataset, and reconstructing the data processing result according to the multi-channel dimension to generate the standardized data matrix.

[0132] Furthermore, the standardized data matrix generation module 1 is used to perform the following steps:

[0133] Continuous data detection is performed on the multi-channel updated dataset to extract a continuous data stream; data mutation detection is performed on the continuous data stream, and a window boundary is defined based on the mutation detection results to set a dynamic window; local normalization processing is performed on the continuous data stream based on the dynamic window to generate a data processing result, which includes a channel normalization sequence; channel coupling compensation is performed according to the channel normalization sequence to determine the data compensation term; the physical arrangement of multiple sensor channels is retrieved to define a spatial weight matrix, and topological mapping is performed based on the spatial weight matrix to generate a multi-channel dimension; the data processing result is reconstructed based on the multi-channel dimension and combined with the data compensation term to generate the standardized data matrix.

[0134] Furthermore, the control decision instruction generation module 2 is used to perform the following steps:

[0135] The industrial control network is monitored to obtain network status parameters. Based on the network status parameters, the standardized data matrix is ​​dynamically transformed to determine the target industrial protocol data. Based on the target industrial protocol data, the protocol format is arranged to generate protocol format data. The protocol format data is transmitted to the industrial control network for protocol analysis to determine the conversion protocol. According to the conversion protocol, multi-channel flow rate analysis is performed on the protocol format data to generate the control decision command.

[0136] Furthermore, the control decision instruction generation module 2 is used to perform the following steps:

[0137] The conversion protocol is parsed to determine a first conversion protocol and a second conversion protocol. When the conversion protocol is the first conversion protocol, the flow rate data is sorted according to the channel priority mapping table to generate a data chain. The data chain is segmented to obtain N data segments, where N is an integer greater than 1. Frame headers are added to the N data segments and reassembled to construct a first data packet. When the conversion protocol is the second conversion protocol, a register address is defined based on the flow rate data. The flow rate data is reassembled according to the register address to construct a second data packet. The first data packet and the second data packet are weighted and calculated, and the first data packet and the second data packet are fused according to the weight coefficient to generate a data fusion result. Multi-channel control analysis is performed based on the data fusion result to generate the control decision command.

[0138] Furthermore, the control decision instruction generation module 2 is used to perform the following steps:

[0139] Flow analysis is performed based on the data fusion results to obtain turbulence intensity parameters and pulsation frequency parameters; flow velocity features are extracted based on the turbulence intensity parameters and pulsation frequency parameters to generate flow velocity feature vectors; control mapping is performed on multi-channel intelligent instruments based on the flow velocity feature vectors to construct a flow velocity-control mapping rule base; control filtering is performed by traversing the flow velocity-control mapping rule base to determine the control decision instructions.

[0140] Furthermore, the control decision instruction generation module 2 is used to perform the following steps:

[0141] When the turbulence intensity parameter is greater than the maximum intensity threshold and the pulsation frequency parameter is less than the minimum frequency threshold, a turbulence suppression mode is triggered; the flow velocity-control mapping rule base is matched according to the turbulence suppression mode to generate the control decision command; average flow velocity data is calculated based on the flow velocity feature vector, and when the average flow velocity data is less than the minimum flow velocity threshold, a flow velocity compensation mode is triggered; the flow velocity-control mapping rule base is matched according to the flow velocity compensation mode to generate the control decision command.

[0142] Furthermore, the integrated management and control execution module 3 is used to perform the following steps:

[0143] The control decision command is executed to drive multiple relays for parsing, extracting relay action sequences, and performing conflict calculations based on these sequences to determine an action conflict matrix. The multiple relays are then controlled in reverse according to the action conflict matrix to generate control results. These control results are verified in real time; if verification fails, the control results are backtracked to calculate dynamic deviation values. Based on these dynamic deviation values, root cause diagnosis is performed on the control results to generate a three-dimensional diagnostic map. Finally, the control results are optimized according to the three-dimensional diagnostic map to formulate the adaptive control strategy.

[0144] The data acquisition adaptive control system for multi-channel intelligent instruments provided in this embodiment of the invention can execute the data acquisition adaptive control method for multi-channel intelligent instruments provided in any embodiment of the invention, and has the corresponding functional modules and beneficial effects of the execution method.

[0145] Although this application makes various references to certain modules in the system according to the embodiments of this application, any number of different modules can be used and run on user terminals and / or servers. The various units and modules included are only divided according to functional logic, but are not limited to the above division, as long as the corresponding functions can be achieved; in addition, the specific names of each functional unit are only for easy distinction between each other and are not used to limit the scope of protection of this invention.

[0146] The specific embodiments described above do not constitute a limitation on the scope of protection of this application. Those skilled in the art should understand that various modifications, combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this application should be included within the scope of protection of this application. In some cases, the actions or steps described in this application can be performed in a different order than that shown in the embodiments and still achieve the desired results. Furthermore, the processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

Claims

1. A data acquisition and adaptive control method for multi-channel intelligent instruments, characterized in that, The method includes: Real-time data streams from multiple sensor channels are acquired synchronously to generate a multi-channel raw dataset. Adaptive data integration processing is then performed on the multi-channel raw dataset to generate a standardized data matrix. The standardized data matrix is ​​dynamically transformed to obtain protocol format data which is then transmitted to the industrial control network for multi-channel flow velocity analysis and to generate control decision commands. The control decision command is executed to drive multiple relays to perform reverse control, generate control results, verify the control results in real time for backtracking optimization, and formulate an adaptive control strategy to comprehensively manage and control the data of multi-channel smart instruments. The method involves dynamically converting the standardized data matrix to obtain protocol-formatted data, transmitting it to an industrial control network for multi-channel flow velocity analysis, and generating control decision commands. The industrial control network is detected to obtain network status parameters. Based on the network status parameters, the standardized data matrix is ​​dynamically transformed to determine the target industrial protocol data. Based on the target industrial protocol data, the protocol format is formatted to generate protocol format data; The protocol format data is transmitted to the industrial control network for protocol analysis to determine the conversion protocol. Perform multi-channel flow rate analysis on the protocol format data according to the conversion protocol, and generate the control decision command; The method for generating the control decision command by performing multi-channel flow rate analysis on protocol format data according to the conversion protocol includes: The conversion protocol is parsed to determine the first conversion protocol and the second conversion protocol; When the conversion protocol is the first conversion protocol, the flow rate data is sorted according to the channel priority mapping table to generate a data chain; The data chain is divided into N data segments, where N is an integer greater than 1; The N data segments are reassembled by adding frame headers to construct the first data packet; When the conversion protocol is the second conversion protocol, the register address is defined based on the flow rate data; The flow rate data is reassembled according to the registered address to construct a second data packet; The first data packet and the second data packet are weighted and calculated, and then the first data packet and the second data packet are fused according to the weight coefficient to generate a data fusion result; Based on the data fusion results, multi-channel control analysis is performed to generate the control decision instructions.

2. The data acquisition adaptive control method for multi-channel intelligent instruments as described in claim 1, characterized in that, The method for simultaneously acquiring real-time data streams from multiple sensor channels to generate a multi-channel raw dataset includes: Configure data acquisition for multiple sensor channels and construct a channel priority mapping table; Activate multiple sensor channels according to the channel priority mapping table to collect data in real time and obtain a real-time data stream. The real-time data stream is clock-synchronized according to timestamps, and noise reduction and filtering are performed based on the synchronization results to generate a cached data stream. The cached data stream is dynamically allocated according to the channel priority mapping table to determine the original dataset for multiple channels.

3. The data acquisition adaptive control method for multi-channel intelligent instruments as described in claim 2, characterized in that, The method involves adaptive data integration processing of the multi-channel raw dataset to generate a standardized data matrix, including: The multi-channel raw dataset is traversed to perform data validity checks, and a first check result is generated. The multi-channel raw dataset is traversed to perform data integrity checks, and a second check result is generated. Based on the overlap analysis of the first detection result and the second detection result, anomalies are identified in the multi-channel raw dataset based on the overlap data to determine abnormal data points. Based on the abnormal data points, the multi-channel original dataset is dynamically compensated, and the multi-channel original dataset is updated according to the compensation results to obtain a multi-channel updated dataset. The multi-channel updated dataset is locally normalized to the continuous data stream, and the data processing results are reconstructed according to the multi-channel dimensions to generate the standardized data matrix.

4. The data acquisition adaptive control method for multi-channel intelligent instruments as described in claim 3, characterized in that, The method involves performing local normalization on the continuous data stream using the multi-channel updated dataset, reconstructing the data processing results according to the multi-channel dimensions, and generating the standardized data matrix. Perform continuous data detection based on the multi-channel updated dataset to extract continuous data streams; Data mutation detection is performed based on the continuous data stream, and a window boundary is defined according to the mutation detection result, and a dynamic window is set. Based on the dynamic window, a continuous data stream is locally normalized to generate a data processing result, which includes a channel normalization sequence. Perform channel coupling compensation according to the channel normalization sequence to determine the data compensation term; The physical arrangement of multiple sensor channels is retrieved to define a spatial weight matrix. Based on the spatial weight matrix, a topological structure mapping is performed to generate a multi-channel dimension. Based on the multi-channel dimensions, the data processing results are reconstructed in conjunction with the data compensation terms to generate the standardized data matrix.

5. The data acquisition adaptive control method for multi-channel intelligent instruments as described in claim 1, characterized in that, Based on the data fusion results, multi-channel control analysis is performed to generate the control decision command. The method includes: Flow analysis is performed based on the data fusion results to obtain turbulence intensity parameters and fluctuation frequency parameters. Based on the turbulence intensity parameter and the fluctuation frequency parameter, flow velocity features are extracted to generate a flow velocity feature vector; Based on the flow velocity feature vector, control mapping is performed on the multi-channel smart instrument to construct a flow velocity-control mapping rule base. The flow velocity-control mapping rule base is traversed to perform control filtering and determine the control decision command.

6. The data acquisition adaptive control method for multi-channel intelligent instruments as described in claim 5, characterized in that, Based on the flow velocity feature vector, control mapping is performed on a multi-channel intelligent instrument to construct a flow velocity-control mapping rule base. The flow velocity-control mapping rule base is then traversed to perform control filtering, and the control decision command is determined. The method includes: When the turbulence intensity parameter is greater than the maximum intensity threshold and the pulsation frequency parameter is less than the minimum frequency threshold, the turbulence suppression mode is triggered. The flow velocity-control mapping rule base is matched according to the turbulence suppression mode to generate the control decision command; The average flow velocity data is calculated based on the flow velocity feature vector. When the average flow velocity data is less than the minimum flow velocity threshold, the flow velocity compensation mode is triggered. The flow rate-control mapping rule base is matched according to the flow rate compensation mode to generate the control decision command.

7. The data acquisition adaptive control method for multi-channel intelligent instruments as described in claim 1, characterized in that, The method includes executing the control decision command to drive multiple relays for reverse control, generating control results, verifying the control results in real time for backtracking optimization, and formulating an adaptive control strategy. The control decision command is executed to drive multiple relays to be parsed, the relay action sequence is extracted, and conflict calculation is performed based on the relay action sequence to determine the action conflict matrix. The multi-channel relays are reverse-controlled according to the aforementioned action conflict matrix to generate control results; The control results are verified in real time. If the verification fails, the control results are backtracked to calculate the dynamic deviation value. Based on the dynamic deviation value, root cause diagnosis is performed on the control results to generate a three-dimensional diagnostic map; The control results are optimized based on the three-dimensional diagnostic map to formulate the adaptive control strategy.

8. A data acquisition and adaptive control system for multi-channel intelligent instruments, characterized in that, The system is used to implement the data acquisition adaptive control method for a multi-channel intelligent instrument according to any one of claims 1-7, the system comprising: The standardized data matrix generation module is used to synchronously acquire real-time data streams from multiple sensor channels, generate multi-channel raw datasets, and perform adaptive data integration processing on the multi-channel raw datasets to generate a standardized data matrix. The control decision command generation module is used to dynamically convert the standardized data matrix, obtain protocol format data transmission, transmit it to the industrial control network for multi-channel flow velocity analysis, and generate control decision commands. The integrated management and control execution module is used to execute the control decision instructions to drive multiple relays for reverse control, generate control results, verify the control results in real time for backtracking optimization, and formulate adaptive control strategies to comprehensively manage and control the data of multi-channel smart instruments.

Images