High-pressure injector intelligent control system based on 5G technology

Abstract

The invention relates to the technical field of high-pressure injection control, and discloses a high-pressure injector intelligent control system based on the 5G technology. An injection pressure dynamic adjusting module driven by a microcontroller of the system analyzes pressure fluctuation characteristics and generates a primary pressure compensation instruction; the fluid state reconstruction module is used for calculating fluid state reconstruction parameters in combination with the real-time flow velocity data; the multi-channel cooperative control module distributes a pressure compensation weight and generates a channel cooperative instruction set based on reconstruction parameter analysis; the abnormal pressure tracing module monitors abnormal pressure sudden change points of each unit, traces corresponding fluid state parameter change tracks and outputs tracing mark data; the injection path optimization module re-plans a propulsion path trajectory based on the marked data to generate a three-dimensional path optimization parameter table; and the safety threshold self-correction module dynamically adjusts the boundary value of the pressure threshold sequence according to the track offset, and outputs an updated safety threshold parameter set to the microcontroller.

Description

Technical Field

[0001] This invention relates to the field of high-pressure injection control technology, specifically to an intelligent control system for high-pressure injectors based on 5G technology. Background Technology

[0002] High-pressure injectors are key equipment in modern medical imaging and precision industrial coating. Their core function is to inject contrast agents, drugs, or specific fluids into a target area under strict control at high pressure and high speed. The stability and accuracy of the injection process directly affect image quality, treatment efficacy, or coating uniformity. Traditional high-pressure injector control systems often use preset pressure-flow rate curves for open-loop control or rely on local closed-loop feedback for simple adjustments. These control methods have significant shortcomings when facing complex operating conditions. During injection, the viscosity of the fluid medium may change due to temperature variations, drug mixing, or batch differences, causing deviations between the actual injection resistance and the preset model. For multi-channel parallel injection systems, even small differences in the fluid characteristics of each channel can be amplified, leading to pressure imbalances and affecting injection synchronization.

[0003] Existing technologies struggle to perceive and adapt to these dynamic changes in real time and with precision. Limited data acquisition and processing capabilities of local sensors, coupled with information transmission delays, lead to sluggish control system responses and an inability to effectively suppress pressure fluctuations. When abnormal pressure surges occur, the system often only provides emergency shutdowns or simple alarms, lacking the ability to deeply trace the root cause of the anomaly and hindering preventative adjustments. Furthermore, injection path planning is typically based on idealized straight lines or simple curves, failing to consider path deviations caused by mechanical clearances, fluid backlash, and other factors during actual execution, thus affecting injection positioning accuracy. While the high speed and low latency of 5G communication technology enable real-time transmission and centralized processing of massive amounts of sensor data, transforming this data into intelligent control commands for fluid state, equipment coordination, and path accuracy remains a critical technical challenge. Therefore, a control system that deeply integrates the advantages of 5G communication to achieve real-time perception, intelligent decision-making, and adaptive optimization of all parameters in the high-pressure injection process is needed. Summary of the Invention

[0004] The purpose of this invention is to provide an intelligent control system for high-pressure injectors based on 5G technology, so as to solve the problems mentioned in the background art.

[0005] To achieve the above objectives, the present invention provides an intelligent control system for a high-pressure injector based on 5G technology, the system comprising: The microcontroller-driven injection pressure dynamic adjustment module receives pressure sensor data streams from the high-pressure injector in real time via a 5G network, analyzes the pressure fluctuation characteristics of the current injection stage, and generates a primary pressure compensation command by comparing it with a preset pressure threshold sequence. The fluid state reconstruction module identifies the viscosity change trend of the fluid medium in the injection line based on the pressure fluctuation characteristics in the primary pressure compensation command, and calculates the fluid state reconstruction parameters by combining the real-time flow rate data stream transmitted by 5G. Based on the fluid state reconstruction parameters, the multi-channel collaborative control module analyzes the dynamic value of the pressure difference between parallel injection units, allocates pressure compensation weights for each channel, and generates a channel collaborative instruction set. The abnormal pressure tracing module monitors the abnormal pressure mutation points of each injection unit according to the channel coordination instruction set, traces the change trajectory of fluid state reconstruction parameters corresponding to the pressure mutation, and outputs abnormal pressure tracing marker data. Based on the abnormal pressure tracing marker data, the injection path optimization module replans the propulsion path trajectory of the high-pressure injector and generates a three-dimensional path optimization parameter table by combining the real-time location data fed back by the 5G network. The safety threshold self-correction module dynamically adjusts the boundary value range of the pressure threshold sequence based on the trajectory offset in the three-dimensional path optimization parameter table, and outputs the updated safety threshold parameter set to the microcontroller.

[0006] Preferably, the step of obtaining the pressure fluctuation characteristics specifically includes: The microcontroller extracts a sequence of pressure sample values ​​within a continuous time window from the 5G data stream and calculates the rate of change of pressure gradient between adjacent sampling points. Based on the positive and negative polarity distribution of the pressure gradient change rate, the characteristic intervals of the pressure rising phase and the pressure falling phase are divided. By comparing the deviation of the pressure gradient change rate within the characteristic interval from the preset pressure threshold sequence, abnormal fluctuation intervals exceeding the threshold are marked. The duration and amplitude changes of abnormal fluctuation ranges are integrated to generate a pressure fluctuation feature matrix.

[0007] Preferably, the step of generating the fluid state reconstruction parameters specifically includes: Analyze the timestamps of abnormal fluctuation intervals in the pressure fluctuation feature matrix and extract the fluid medium velocity sampling values ​​at the corresponding times. Calculate the percentage deviation between the sampled flow velocity and the reference flow velocity, and establish a two-dimensional parameter mapping table in conjunction with the pressure gradient change rate; Based on the distribution density of parameter points in the two-dimensional parameter mapping table, the boundary thresholds for dividing the fluid state into the stable region and the turbulent region are determined. Based on boundary thresholds, the real-time collected flow velocity data stream is classified into different state levels, generating fluid state reconstruction parameter codes.

[0008] Preferably, the generation step of the channel coordination instruction set specifically includes: Monitor the difference in the rate of change of pressure gradient between parallel injection units at the same time stamp; The priority of the current state level of each unit is determined based on the fluid state reconstruction parameter coding; The execution sequence of pressure compensation instructions is allocated according to the priority of status level, and a channel coordination instruction queue with timestamps is generated. By comparing the time interval between adjacent instructions in the instruction queue with the 5G network latency parameters, the instruction sending time can be optimized.

[0009] Preferably, the steps for generating the abnormal pressure tracing marker data are as follows: Capture pressure mutation events after the execution of the channel cooperative instruction queue, and record the sequence of fluid state reconstruction parameter changes within a continuous time window before the mutation occurs; Analyze the time lag between the inflection point of key indicators and the time of sudden pressure change in the parameter change sequence. Establish a correlation strength marker between pressure abrupt changes and specific fluid state parameters based on the magnitude of the delay; Parameters whose association strength exceeds a threshold are combined into an abnormal pressure source labeling dataset.

[0010] Preferably, the steps for generating the three-dimensional path optimization parameter table are as follows: Analyze the key parameter items in the abnormal pressure tracing marker dataset to locate the real-time position offset of the high-pressure injector propulsion mechanism; Calculate the trajectory correction vector of the propulsion path in the three-dimensional coordinate system based on the offset direction; By combining the data consistency verification results from multiple location sensors fed back by the 5G network, effective trajectory correction vectors are selected; The effective vectors are integrated into a three-dimensional path optimization parameter table according to the time series.

[0011] Preferably, the update step of the security threshold parameter set specifically includes: Extract the magnitude statistics of the trajectory correction vector from the 3D path optimization parameter table; Compare the coverage differences between the statistical values ​​and the original safety thresholds; When the difference exceeds the preset tolerance, the upper and lower boundary values ​​of the pressure threshold sequence are scaled proportionally. The scaled boundary values ​​are compared with the historical security parameters stored in the microcontroller, and conflicting values ​​are removed to generate a new set of security threshold parameters.

[0012] Preferably, the steps for the microcontroller to achieve multi-module collaboration are as follows: Establish a dedicated 5G data transmission channel between the injection pressure dynamic adjustment module and the fluid state reconstruction module; Configure a real-time interrupt response mechanism for the abnormal pressure tracing module in the multi-channel collaborative control module; The parameter transmission compression algorithm from the injection path optimization module to the safety threshold self-correction module is optimized. Coordinate the timing of each module's access to the microcontroller's storage resources to avoid conflicts.

[0013] Preferably, the system further includes: The microcontroller integrates a 5G network status monitoring unit to assess the impact of data transmission latency on stress compensation commands in real time. Dynamically adjust the transmission frequency of data interaction between modules to match changes in network status; When the latency exceeds a critical value, switch to the local cache execution strategy.

[0014] Preferably, the system further includes: The microcontroller manages a historical database of injection parameters, continuously recording the intermediate parameter sets generated by each module; Establish an index linking parameter set version tags with injection events; The strategy of uploading or storing historical data locally is dynamically selected based on the 5G network bandwidth.

[0015] Compared with the prior art, the beneficial effects of the present invention are: This invention's system achieves multi-dimensional, refined control of the high-pressure injection process by constructing a closed-loop intelligent control architecture based on a 5G high-speed data link. The microcontroller-driven pressure dynamic adjustment module, leveraging the low latency of the 5G network, can acquire pressure sensor data streams in near real-time, quickly analyze fluctuation characteristics, and generate preliminary compensation commands, thus improving the system's response speed to pressure disturbances. The innovation of the fluid state reconstruction module lies in its ability not only to respond to pressure changes but also to interpret the underlying physical causes of these changes, namely, the changing trend of the fluid medium's viscosity. By combining real-time flow velocity data for state reconstruction, the system can move from phenomenological perception to state cognition, providing a more fundamental basis for subsequent control.

[0016] The multi-channel collaborative control module, based on reconstructed fluid state parameters, intelligently analyzes the dynamic pressure differences between parallel units and allocates compensation weights to each channel accordingly. This achieves load balancing and synchronous injection across multiple channels, effectively improving the overall stability of the system output. The abnormal pressure tracing module further enhances the system's diagnostic capabilities. When a pressure surge is detected, it can trace its correlation with fluid state parameters, outputting clear tracing markers and upgrading the anomaly from a simple alarm to analyzable diagnostic information. The injection path optimization module utilizes this diagnostic information, combined with precise location data from 5G feedback, to replan the injection path, compensating for deviations in actual execution and ensuring the accuracy of the injection trajectory. The safety threshold self-correction module dynamically adjusts the boundary of the pressure safety threshold based on the trajectory offset after path optimization, enabling the safety strategy to adapt to changes in actual operating conditions and avoiding false alarms or missed alarms under fixed thresholds. The entire system forms a complete intelligent closed loop from data perception, state reconstruction, collaborative control, anomaly diagnosis to path optimization and safety strategy self-adjustment, significantly improving the control accuracy, stability, and safety of the high-pressure injection process. Attached Figure Description

[0017] Figure 1 Monitoring diagram of the overall control flow of the system Figure 2 A flowchart for the steps of obtaining pressure fluctuation characteristics; Figure 3 A flowchart of the steps for generating parameters for fluid state reconstruction; Figure 4 This is a diagram for anomaly tracing and path optimization analysis. Detailed Implementation

[0018] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0019] Please see Figure 1 This invention provides an intelligent control system for a high-pressure injector based on 5G technology. The system includes: a microcontroller-driven dynamic injection pressure adjustment module, a fluid state reconstruction module, a multi-channel collaborative control module, an abnormal pressure tracing module, an injection path optimization module, and a safety threshold self-correction module. Specific implementation details are as follows: A microcontroller-driven injection pressure dynamic adjustment module receives real-time pressure sensor data streams from the high-pressure injector via a 5G network, analyzes the pressure fluctuation characteristics of the current injection phase, and generates a primary pressure compensation command by comparing it with a preset pressure threshold sequence. A fluid state reconstruction module identifies the viscosity change trend of the fluid medium within the injection tubing based on the pressure fluctuation characteristics in the primary pressure compensation command, and calculates fluid state reconstruction parameters by combining this with real-time flow velocity data streams transmitted via 5G. A multi-channel collaborative control module analyzes the dynamic pressure difference between parallel injection units based on the fluid state reconstruction parameters, allocates pressure compensation weights for each channel, and generates a channel collaborative command set. An abnormal pressure tracing module monitors abnormal pressure mutation points in each injection unit based on the channel collaborative command set, traces the fluid state reconstruction parameter change trajectory corresponding to the pressure mutation, and outputs abnormal pressure tracing marker data. An injection path optimization module replans the high-pressure injector's propulsion path trajectory based on the abnormal pressure tracing marker data, and generates a three-dimensional path optimization parameter table by combining real-time position data fed back from the 5G network. A safety threshold self-correction module dynamically adjusts the boundary value range of the pressure threshold sequence based on the trajectory offset in the three-dimensional path optimization parameter table, and outputs an updated safety threshold parameter set to the microcontroller. The microcontroller coordinates the data flow between modules to ensure that the low latency transmission and real-time processing capabilities of the 5G network are matched. The system adopts a modular design to facilitate maintenance and upgrades. The hardware interface of the high-pressure injector is integrated with the microcontroller through a standard protocol to achieve intelligent control closed loop.

[0020] Example 1: See Figure 2The microcontroller extracts a sequence of pressure sampling values ​​within a continuous time window from the 5G data stream. This extraction process relies on the real-time packet parsing function of the 5G network interface chip. The pressure sampling value sequence originates from the pressure sensor array of a high-pressure injector, with sensors distributed across multiple key nodes in the injection tubing. The microcontroller's built-in dedicated algorithm calculates the rate of change of the pressure gradient between adjacent sampling points. The algorithm uses a sliding window mechanism to process time-series data, ensuring that the timestamp of each sampling point is strictly synchronized with the 5G network clock source. The calculation of the rate of change of the pressure gradient employs a first-order difference method. Based on a fixed time interval for aligning sampling points, the microcontroller's arithmetic logic unit performs subtraction and divides by the time difference to obtain the rate of change value. The microprocessor analyzes the positive and negative polarity distribution of the rate of change of the pressure gradient. Polarity distribution identification is achieved through a sign bit detection circuit, which is used to divide the characteristic intervals of the pressure rise and fall phases. The division of characteristic intervals is based on the sign change points of the rate of change of the pressure gradient. The microcontroller's state machine monitors the continuity and jump patterns of the sign, and the boundary determination of the characteristic intervals incorporates the statistical characteristics of historical data. The microcontroller compares the deviation of the pressure gradient change rate within a characteristic interval from a preset pressure threshold sequence. This preset pressure threshold sequence is stored in the microcontroller's non-volatile memory, which uses a paged storage structure to manage threshold parameters for different injection stages. The deviation calculation involves two algorithms: absolute value comparison and percentage deviation. The microcontroller's comparator circuit performs the threshold comparison operation in real time, and the quantified result of the deviation is sent to the decision logic unit. The microcontroller marks abnormal fluctuation intervals exceeding the threshold. This marking operation is completed by setting specific bit fields in the status register. The marking of abnormal fluctuation intervals uses a flag bit array to record the time range and amplitude information. The flag bit array is stored in a circular buffer in the microcontroller's memory, and the buffer's pointers are managed using a circular addressing method. The start and end timestamps of the abnormal fluctuation intervals are associated with the pressure sample value indices. The microcontroller integrates the duration and amplitude change of the abnormal fluctuation intervals. The duration is derived from the timestamp difference obtained from a high-precision timer, and the amplitude change is derived from the range calculation and integration of the pressure sample values. The integration process uses a weighted fusion algorithm, with different weight coefficients corresponding to different types of abnormal fluctuations. The generated pressure fluctuation feature matrix includes time, amplitude, and frequency dimensions.

[0021] The rows of the pressure fluctuation feature matrix represent the sliding time window numbers, and the columns represent various feature dimensions of the pressure parameters. The microcontroller transmits the pressure fluctuation feature matrix to the input buffer of the fluid state reconstruction module via a direct memory access channel. The matrix data structure uses a row-major storage format, with each matrix element containing normalized feature values. The matrix dimensions are dynamically adjusted according to the injection stage. The 5G network ensures the real-time performance of the pressure sampling value sequence. The network interface chip's packet reassembly function handles potentially out-of-order packets, and the microcontroller's clock synchronization mechanism coordinates with the 5G network time protocol. Clock synchronization is achieved through a precision clock synchronization protocol, with timestamp accuracy reaching the microsecond level. The sampling frequency of the pressure sensor data stream is adaptively adjusted to match network bandwidth conditions. The microcontroller drives the actuator of the injection pressure dynamic adjustment module, which includes an electric motor and a hydraulic valve group. The primary pressure compensation command is generated based on the feature extraction results of the pressure fluctuation feature matrix. The command generation algorithm uses fuzzy control rules, and the rule base is stored in the microcontroller's read-only memory. The hydraulic system of the high-pressure injector responds to the primary pressure compensation command. The response characteristics of the hydraulic system are pre-calibrated, and the calibration parameters are stored in the microcontroller's calibration table. The microcontroller's interrupt handling routine manages the priority of the pressure data stream. Interrupt priorities are divided into multiple levels, with high-priority interrupts corresponding to urgent pressure anomalies, ensuring that high-priority data packets are processed first. The update of the pressure fluctuation characteristic matrix is ​​synchronized with the reception of the 5G data stream. The synchronization mechanism adopts a hardware triggering method, and the matrix update period is kept as an integer multiple of the 5G network transmission time interval.

[0022] The microcontroller's digital signal processor (DSP) assists in calculating the pressure gradient change rate. The DSP employs a parallel computing architecture, simultaneously processing pressure sample value sequences from multiple channels. The calculation of the pressure gradient change rate incorporates a smoothing filter using a moving average method to reduce computational errors caused by sensor noise. The microcontroller's branch prediction unit optimizes the feature interval partitioning logic. Branch prediction is based on learning results from historical partitioning patterns, improving the efficiency of feature interval boundary determination. The microcontroller's cache manager optimizes the storage and retrieval mode of the pressure sample value sequences. The cache adopts a multi-level structure: the first-level cache stores the sample values ​​of the current time window, and the second-level cache stores historical sample values.

[0023] The loading of the preset pressure threshold sequence adopts an on-demand loading strategy. The microcontroller's memory management unit monitors the access frequency of the threshold sequence and retains frequently accessed threshold parameters in the cache. Threshold comparison operations use dedicated comparison instructions, which support single-instruction multiple-data operations, simultaneously comparing multiple pressure gradient change rates with threshold parameters. The marking information of abnormal fluctuation intervals is stored in the microcontroller's dedicated register set, which includes an interval start register, an interval end register, and a feature code register. The microcontroller's exception handling unit monitors the marking operation process; when a marking conflict occurs, arbitration logic is initiated, resolving the conflict based on a chronological order. The generation process of the pressure fluctuation feature matrix incorporates matrix compression technology. The compression algorithm uses a lossy compression method to retain the main features of the matrix while reducing storage space usage. The microcontroller's matrix operation coprocessor accelerates the generation of the feature matrix. The coprocessor supports basic operations such as matrix multiplication and transpose, and the operation results are transmitted to the main processor via the bus. The parsing of the 5G network data stream uses a protocol stack accelerator. The protocol stack accelerator hardware parses the header and payload of the 5G data packets, and the extracted pressure sampling values ​​are directly stored in a designated memory area. The microcontroller's power management unit dynamically adjusts the power supply voltage of the computing unit and optimizes power consumption based on load changes in pressure data processing. The adjustment of the power supply voltage follows a predefined power performance curve.

[0024] The communication between the microcontroller and the pressure sensor array uses the Industrial Ethernet protocol, which provides deterministic transmission delay. The acquisition period of the pressure sampling value sequence is configurable. The acquisition period is configured through the microcontroller's configuration register, and the register value is dynamically updated according to the injection stage. The calibration coefficients of the pressure sensor array are stored in the microcontroller's secure storage area. The secure storage area has write protection to prevent accidental modification of the calibration coefficients. The microcontroller's watchdog timer monitors the entire pressure data processing flow. The watchdog timer's timeout is set according to the worst-case processing time; timer overflow will trigger a system reset to ensure system recovery from abnormal states. The transmission of the pressure fluctuation feature matrix uses an error detection mechanism, employing a cyclic redundancy check (CRC) code, which is appended to the end of the matrix data. The receiver verifies the correctness of the CRC code and requests retransmission when an error is detected. The microcontroller's error correction logic handles occasional transmission errors. The microcontroller's temperature sensor monitors the chip's operating temperature. When the temperature exceeds a safe threshold, the processor frequency is automatically reduced, with frequency adjustment following a temperature-frequency curve. The scheduling of stress data processing tasks adopts a fixed priority scheduling algorithm. The scheduler is integrated into the real-time operating system of the microcontroller, and the task priority is allocated according to the criticality of stress control.

[0025] Example 2: See Figure 3The fluid state reconstruction module analyzes the timestamps of abnormal fluctuation intervals in the pressure fluctuation feature matrix. The analysis process uses a timestamp matching algorithm to precisely align pressure and flow velocity data. The timestamps of abnormal fluctuation intervals are derived from the circular buffer records marked by the microcontroller. The fluid state reconstruction module extracts the fluid medium flow velocity sampling values ​​at the corresponding moments. These flow velocity sampling values ​​are obtained from the turbine flow sensor of the high-pressure injector via a 5G network. The sensor data is quantized by an analog-to-digital converter to form a digital signal stream. The extraction operation is based on a time synchronization mechanism; the internal clock of the fluid state reconstruction module is synchronized with the 5G network time server to ensure the time consistency of pressure and flow velocity data. The fluid state reconstruction module calculates the percentage deviation between the flow velocity sampling value and the reference flow velocity. The reference flow velocity is stored in a flash lookup table within the fluid state reconstruction module, which pre-stores multiple sets of reference values ​​according to different injection medium types. The percentage deviation calculation is performed using a floating-point arithmetic unit, and the calculation result retains two decimal places of precision. The fluid state reconstruction module combines this with the pressure gradient change rate to establish a two-dimensional parameter mapping table.

[0026] The horizontal axis of the two-dimensional parameter mapping table represents the normalized value of the pressure gradient change rate, and the vertical axis represents the absolute value of the percentage deviation of the flow velocity. The mapping table is stored in the memory of the fluid state reconstruction module as a two-dimensional array. Array elements contain timestamp indices and parameter state markers. The fluid state reconstruction module divides the fluid state into stable and turbulent regions based on the distribution density of parameter points in the two-dimensional parameter mapping table. The distribution density is calculated using a kernel density estimation algorithm, which runs on the digital signal processor of the fluid state reconstruction module. The boundary thresholds are determined through cluster analysis. The cluster analysis uses the K-means algorithm to iteratively calculate the optimal classification boundary. Based on the boundary thresholds, the fluid state reconstruction module classifies the real-time acquired flow velocity data stream into different state levels.

[0027] The fluid state classification adopts a three-level system, including steady flow, transitional flow, and turbulent flow, with each state level corresponding to a numerical range. The fluid state reconstruction module generates fluid state reconstruction parameter codes, using an 8-bit binary representation. The high 3 bits identify the primary state level, the middle 3 bits record sub-state characteristics, and the low 2 bits are reserved for extended uses. The code generation process undergoes redundancy verification, with a checksum appended to the end of the code to ensure transmission reliability. The multi-channel collaborative control module monitors the difference in pressure gradient change rate between parallel injection units at the same timestamp. Monitoring operations synchronously acquire sensor data from each unit via a multiplexer. The calculation of the pressure gradient change rate difference uses the differential amplifier principle, and the numerical results are stored in a first-in-first-out buffer. The multi-channel collaborative control module determines the priority of each unit's current state level based on the fluid state reconstruction parameter codes. The priority assignment rule is based on the criticality of the fluid state, with turbulent flow receiving the highest priority, followed by transitional flow, and steady flow receiving the lowest priority. The status level priority values ​​are stored in a priority lookup table of the multi-channel collaborative control module. The lookup table can be updated via a configuration interface. The multi-channel collaborative control module allocates the execution sequence of pressure compensation instructions according to the status level priority. The allocation of execution sequence adopts a time-slice round-robin scheduling algorithm, which considers the dynamic value of the pressure difference of each unit and historical execution records to generate a channel collaborative instruction queue with timestamps. The instruction queue is stored in a linked list data structure. Each node contains the instruction opcode, target channel address, execution time point, and parameter list. The multi-channel collaborative control module compares the time interval between adjacent instructions in the instruction queue with the 5G network latency parameters. The network latency parameters are obtained in real time from the 5G network status monitoring unit. The latency parameters include average latency, jitter range, and packet loss rate statistics. The comparison operation is implemented using a window comparator circuit. The multi-channel collaborative control module optimizes the instruction sending time. The optimization strategy includes two modes: look-ahead adjustment and compensation adjustment. Look-ahead adjustment sends instructions in advance based on the network latency prediction model, while compensation adjustment dynamically corrects the sending timing based on the actual latency measurement value.

[0028] The microcontroller coordinates the data exchange process between the fluid state reconstruction module and the multi-channel collaborative control module. Data exchange is achieved through a shared memory area, which is configured with a dual-port access mechanism to avoid conflicts. A reliable transmission protocol is used when encoding fluid state reconstruction parameters via the 5G network. This protocol implements retransmission mechanisms and flow control functions. The parallel channels of the high-pressure injector synchronously execute the channel collaborative instruction queue. The synchronization mechanism uses a hardware synchronization signal; each channel controller executes the current instruction simultaneously upon receiving the synchronization pulse. The microcontroller's arbitration logic resolves conflicts arising from multi-channel access to shared resources. The parameter update cycle of the fluid state reconstruction module maintains a harmonic relationship with the pressure sampling cycle. The update cycle can be configured as an integer multiple of the pressure sampling cycle, reducing contention for computing resources. The instruction sending unit of the multi-channel collaborative control module includes a buffer memory that temporarily stores the queue of instructions to be sent. The instruction sending unit queries the network status flag to determine the sending timing. The maintenance of the two-dimensional parameter mapping table adopts a dynamic update strategy. Newly acquired data points are continuously added to the mapping table, while old data points are aged out based on timestamps. The mapping table update algorithm considers the spatiotemporal locality of data points, giving higher weights to recent data points. The fluid state reconstruction module periodically defragments the mapping table. This defragmentation process reorganizes memory distribution to improve access efficiency. The adaptive adjustment of boundary thresholds follows the gradient descent principle, with the adjustment step size inversely proportional to system stability indicators. The calibration unit of the fluid state reconstruction module periodically verifies the validity of boundary thresholds using cross-validation, and the verification results are used to correct clustering analysis parameters. The priority management unit of the multi-channel collaborative control module implements dynamic priority adjustment, automatically lowering the priority of a channel when it is detected to be in a high-priority state for an extended period, preventing resource starvation.

[0029] The compression algorithm of the channel-coordinated instruction queue uses differential encoding to store the differences between consecutive instructions rather than the complete instruction content. The compressed data is decompressed and restored at the receiving end after transmission via the 5G network. The instruction queue's verification mechanism checks the logical rationality of the instructions before transmission; unreasonable instructions are marked and submitted to the error handling process. The microcontroller's monitoring unit records the success rate statistics of instruction execution. The 5G network latency parameter is measured using round-trip time measurement, with the measurement data packet carrying a high-precision timestamp. The latency calculation subtracts processing time to obtain the pure network transmission latency. The network latency prediction model uses an autoregressive integral moving average model, with model parameters updated online based on historical latency data. The prediction results are sent to the instruction transmission optimization module. The microcontroller's interrupt controller handles interrupt requests generated by the multi-channel coordinated control module. Interrupt priorities are dynamically configured based on the urgency of the fluid state; higher-priority interrupts can preempt the execution of lower-priority interrupts.

[0030] The interface between the fluid state reconstruction module and the multi-channel collaborative control module adopts a standard communication protocol, which includes data format definitions and interaction process specifications. Interface version management ensures backward compatibility. The microcontroller's power management unit allocates independent power domains to the two modules. Each power domain can adjust the supply voltage according to the load, optimizing the overall system power consumption. A fault detection circuit monitors the operating status of the fluid state reconstruction module and the multi-channel collaborative control module. Fault signals trigger the system's safety protection mechanism, which includes module restart and mode degradation. Data consistency checks are performed multiple times during inter-module transmission. The check algorithm combines cyclic redundancy check and parity check, and erroneous data triggers a retransmission process. The microcontroller's performance counters record the processing latency and resource usage of each module. The counter data is used for system performance analysis and optimization.

[0031] Example 3: The abnormal pressure tracing module captures pressure mutation events after the execution of the channel collaborative instruction queue. The definition of pressure mutation events is based on the dynamic threshold comparison principle. The dynamic threshold is calculated based on the historical average and standard deviation of the fluid state reconstruction parameter encoding. The capture mechanism uses a hardware comparator to continuously monitor the pressure sensor readings. When the pressure value exceeds 15% of the current dynamic threshold, an event flag is triggered. The event flag is set simultaneously, and a timestamp accurate to the microsecond level is recorded. The abnormal pressure tracing module records the fluid state reconstruction parameter change sequence within a continuous time window before the mutation occurs. The time window length can be configured from 100 milliseconds to one second before the pressure mutation point. The change sequence is stored in a circular buffer data structure. Buffer management uses a pointer cyclic movement method, with new data overwriting the oldest data to achieve a sliding window effect. The fluid state reconstruction parameter change sequence includes the percentage deviation of flow velocity, the rate of change of pressure gradient, and its first derivative value. The abnormal pressure tracing module analyzes the delay between the inflection point time of key indicators in the parameter change sequence and the pressure mutation time. Key indicators are determined by a feature selection algorithm. The feature selection algorithm calculates the mutual information value between each parameter and the pressure mutation and selects the three parameters with the highest correlation. The inflection point time detection employs a sliding window inflection point identification algorithm. This algorithm calculates local extrema on the parameter sequence and verifies their significance. Delay calculation uses a high-precision timer to measure the time difference between the inflection point and the pressure abrupt change point. The abnormal pressure tracing module establishes a correlation strength marker between pressure abrupt changes and specific fluid state parameters based on the magnitude of the delay. The correlation strength marker is calculated using a weighted scoring model; the smaller the delay, the larger the weight coefficient, and the greater the parameter change amplitude, the higher the contribution value. The scoring model outputs correlation strength values ​​between zero and one hundred. The abnormal pressure tracing module combines parameters with correlation strength markers exceeding a threshold of sixty into an abnormal pressure tracing marker dataset.

[0032] The abnormal pressure tracing marker dataset is stored in a tabular format, with each row recording detailed information for a parameter, including the parameter identifier, correlation strength value, timestamp sequence, and change trajectory feature vector. The injection path optimization module parses the key parameters in the abnormal pressure tracing marker dataset. The parsing process prioritizes parameters with high correlation strength, sorted in descending order of correlation strength. Key parameters are defined as those with a correlation strength value exceeding 75 and appearing consecutively in three or more time windows. The injection path optimization module locates the real-time position offset of the high-pressure injector propulsion mechanism. The positioning method integrates multi-source sensor data, including linear encoder readings, laser rangefinder data, and inertial measurement unit output. The data fusion algorithm uses Kalman filtering to reduce measurement noise, and the real-time position offset is calculated as the Euclidean distance between the actual and expected positions.

[0033] The injection path optimization module calculates the trajectory correction vector of the propulsion path in a three-dimensional coordinate system based on the offset direction. The three-dimensional coordinate system is established with the syringe mounting base as the origin, the X-axis along the injection direction, the Y-axis vertically upward, and the Z-axis forming a right-handed coordinate system. The calculation of the trajectory correction vector incorporates path smoothing constraints, prohibiting sharp turns and movements exceeding mechanical limits. The vector components are obtained by solving a constrained optimization problem. The injection path optimization module also incorporates data consistency verification results from multiple position sensors fed back from the 5G network. The consistency verification uses a combination of majority voting and chi-square tests to filter out abnormal sensor readings that deviate from the group mean by more than two standard deviations. The voting mechanism requires at least two-thirds of the sensor data to be consistent for valid data, while the chi-square test evaluates the normal distribution characteristics of the dataset to eliminate outliers that do not conform to the distribution pattern. The injection path optimization module selects valid trajectory correction vectors based on criteria including vector magnitude limits, directional continuity checks, and compatibility verification with historical paths. The vector length limit stipulates that the vector length must not exceed the maximum allowable correction amount. Directional continuity checks reject vectors whose angle with the recently corrected direction exceeds 45 degrees. Compatibility verification ensures a smooth transition between new vectors and already executed paths. The injection path optimization module integrates effective vectors into a three-dimensional path optimization parameter table based on time series. The parameter table uses a hierarchical storage structure: the top layer is the time index layer, the middle layer is the vector data layer, and the bottom layer is the checksum layer. Each time index corresponds to a set of three-dimensional coordinate correction values ​​and smoothing coefficients. Cyclic redundancy checks are used in the checksum to ensure data integrity. The correlation strength between pressure mutation events and fluid state parameters is quantified using the following mathematical relationship:

[0034] in: Indicates the correlation strength flag value. The moment when the pressure change occurs, This represents the number of sampling points within the time window. The sampling time interval, This is the time decay weighting coefficient. Let t be the percentage change in flow velocity at time t. This represents the maximum permissible deviation in flow velocity. Let be the rate of change of the pressure gradient at time t. This represents the maximum permissible rate of change of the pressure gradient. All terms in the formula are dimensionless to ensure dimensional consistency on both sides of the equation.

[0035] The digital signal processor (DSP) of the abnormal pressure tracing module processes pressure mutation detection tasks from multiple channels in parallel. The processor integrates a dedicated instruction set to accelerate inflection point identification algorithms and correlation strength calculations. The memory controller manages the storage access mode of parameter change sequences, employing burst transmission to improve data throughput. The 5G network interface module initiates data compression when transmitting the abnormal pressure tracing marker dataset. The compression algorithm, based on dictionary encoding principles, reduces network bandwidth usage. The path planning unit of the injection path optimization module includes a geometric operation coprocessor, which hardware-accelerates matrix operations on trajectory correction vectors. The microcontroller's interrupt scheduler allocates high-priority interrupt channels to the abnormal pressure tracing module, ensuring timely responses to pressure mutation events. A system watchdog timer monitors the real-time performance of the entire processing flow. Version management of the abnormal pressure tracing marker dataset employs an incremental storage strategy; only changed data items are written to non-volatile memory, extending memory lifespan. The data encryption engine performs AES encryption on sensitive parameters before storage, and the encryption key is protected by a hardware security module. The 3D path optimization parameter table from the injection path optimization module is output to the motion controller. The motion controller converts vector sequences into motor control commands. Command generation considers the acceleration limitations and motion smoothness requirements of the servo motor. A block transmission mechanism is used when transmitting the 3D path optimization parameter table over the 5G network. The transmission block size is adaptively adjusted based on network latency; small data blocks are used in high-latency environments to improve transmission reliability, while large data blocks are used in low-latency environments to improve transmission efficiency. The calibration procedure of the abnormal pressure tracing module is executed periodically. The calibration procedure injects known pressure mutation modes to verify detection accuracy, and the calibration results are used to adjust the dynamic threshold and correlation strength calculation parameters. The self-test function of the injection path optimization module verifies the normal operation of all computing units upon system startup. The test mode generates a standard test path and compares it with the expected results; if the deviation exceeds the set value, an alarm signal is triggered. The microcontroller's temperature monitoring system monitors the chip temperature of the abnormal pressure tracing module and the injection path optimization module in real time. When the temperature exceeds the safety threshold, the processor frequency is automatically reduced to prevent overheating damage; the frequency adjustment curve is carefully designed according to the chip characteristics. The power management unit provides filtered clean power to the sensitive analog circuits, reducing the impact of power supply noise on the accuracy of pressure mutation detection. The system grounding design follows high-frequency circuit layout principles to minimize electromagnetic interference.

[0036] See Figure 4 The above figure demonstrates the effectiveness of the abnormal pressure tracing module and the injection path optimization module. The top image shows detected pressure surge events using vertical marker lines; the color depth of the marker lines represents the strength of the correlation, and the numerical annotations visually show the degree of correlation between each abnormal event and fluid state parameters. This correlation strength analysis is based on the sequence of fluid state reconstruction parameter changes within a continuous time window before the pressure surge, establishing a quantitative relationship by calculating the delay between the inflection point time of key indicators and the pressure surge time. The bottom image compares the modulus changes of the original injection path and the optimized path, as well as the path improvement calculated by the system. The optimized path fully considers the abnormal pressure tracing marker data, re-plans the propulsion trajectory of the high-pressure injector, and generates a smoother and more stable injection path by combining real-time position data fed back from the 5G network. This optimization ensures the safety and accuracy of the injection process, demonstrating the system's intelligent control level.

[0037] Example 4: The safety threshold self-correction module extracts the amplitude statistics of the trajectory correction vector from the 3D path optimization parameter table. The extraction process involves batch calculation of the vector magnitude. The amplitude statistics include three core indicators: arithmetic mean, standard deviation, and maximum value. The calculation module adopts a streaming processing architecture. Vector data is read directly from the dual-port memory and sent to the vector processing unit, which integrates a hardware-accelerated square root operation circuit. The arithmetic mean calculation uses a recursive algorithm to avoid numerical overflow caused by the accumulation of large arrays. The standard deviation calculation uses the Welford online algorithm for real-time updates, and the maximum value detection is implemented in parallel through a comparator tree circuit. The safety threshold self-correction module compares the coverage difference between the statistical value and the original safety threshold. The coverage difference is defined as the ratio of the overlapping area between the statistical value distribution and the threshold interval. The difference calculation requires normalizing the statistical values ​​to the same dimension. The original safety threshold is stored in the parameter storage area of ​​the microcontroller. The storage area is divided into multiple sectors to store the threshold parameters for different injection stages. The conditions for starting the difference calculation include the number of new statistical samples reaching the window threshold or the detection of path anomalies. The safety threshold self-correction module scales the upper and lower boundary values ​​of the pressure threshold sequence proportionally when the difference exceeds a preset tolerance. The preset tolerance is set to an adjustable range of 5% to 10% based on injection accuracy requirements, and the scaling ratio has a piecewise linear relationship with the difference. The scaling algorithm maintains the symmetry of the threshold interval; the adjustment amount of the upper and lower boundary values ​​is the same but in opposite directions. The update of the pressure threshold sequence uses atomic operations to avoid read-write conflicts. The safety threshold self-correction module compares the scaled boundary values ​​with the historical safety parameters stored in the microcontroller. The historical safety parameters come from the threshold set recorded in past successful injection cycles. The comparison process calculates the Euclidean distance between the new boundary value and the historical value. A weighting coefficient is introduced into the distance calculation, with recent historical values ​​receiving higher weights. After removing conflicting values, the safety threshold self-correction module generates a new safety threshold parameter set. The determination of conflicting values ​​is based on the principle of statistical hypothesis testing. The removal operation uses a two-sided test method with a significance level set to 0.1%. The new safety threshold parameter set is formatted and written to non-volatile memory.

[0038] The microcontroller achieves multi-module collaboration by establishing a dedicated 5G data transmission channel between the injection pressure dynamic adjustment module and the fluid state reconstruction module. This dedicated channel is configured with differentiated quality of service levels to ensure transmission latency is less than ten milliseconds. The channel establishment process includes authentication and key negotiation steps. The authentication protocol employs a two-way challenge-response mechanism, and data transmission uses the AES-GCM encryption algorithm to ensure confidentiality and integrity. The microcontroller is configured with a real-time interrupt response mechanism for the abnormal pressure tracing module. This interrupt response mechanism implements a priority inheritance protocol to prevent priority inversion, and the interrupt service routine is permanently stored in a reserved memory area. The interrupt vector table is set to read-only access via a memory protection unit. The interrupt request signal from the abnormal pressure tracing module is connected to the microcontroller's fast interrupt input pin, configured in edge-triggered mode to reduce response latency. The microcontroller optimizes the parameter transmission compression algorithm from the injection path optimization module to the safety threshold self-correction module. The compression algorithm uses a hybrid scheme combining dictionary encoding and differential encoding. The compression dictionary is dynamically updated based on parameter characteristics. Differential encoding records the changes between consecutive parameter frames rather than absolute values, dictionary encoding replaces frequently occurring parameter patterns with short codes, and frame header markers are added to the compressed data for easy parsing by the receiver.

[0039] The microcontroller coordinates the access timing of various modules to its storage resources to avoid conflicts. Storage resources are divided into multiple bandwidth protection zones, and each module is allocated an independent access time slot. The time slot allocation table is maintained by the storage controller, and its contents are dynamically adjusted based on module load. A conflict detection circuit monitors bus access requests. Conflict resolution employs a round-robin arbitration mechanism, ensuring that each module obtains storage access rights within the longest possible waiting time. The microcontroller's storage manager implements error-correcting coding to enhance data reliability. The amplitude statistics calculation of the security threshold self-correction module uses fixed-point arithmetic to improve processing efficiency. The fixed-point number format uses Q15 notation to balance precision and dynamic range, and the statistics cache is optimized using hardware prefetching. The difference calculation unit integrates a dedicated divider circuit, which uses Newton's iterative algorithm to accelerate division operations. Preset tolerance values ​​are stored in programmable registers for runtime adjustment. Boundary value scaling operations use a saturation operation mode to prevent numerical overflow. The scaling factor is obtained by querying a lookup table, the contents of which are pre-set through a calibration program. The historical security parameter comparison process initiates a similarity matching algorithm. The algorithm calculates the cosine similarity between the new parameter set and the historical template; historical data with similarity below a threshold is marked as expired. The microcontroller's garbage collection mechanism periodically cleans up expired historical records, maintaining storage defragmentation during the cleanup process. A system-wide notification event is triggered upon the generation of a new set of safety threshold parameters. Event notifications are implemented through a message broadcast mechanism; modules subscribing to the event update their local cache parameters. Message delivery uses a publish-subscribe pattern to decouple module dependencies.

[0040] Referring to Table 1, the dedicated 5G data transmission channel of the injection pressure dynamic adjustment module and the fluid state reconstruction module implements flow shaping control. The flow shaper smooths burst data streams to reduce network congestion, and the channel state monitoring unit statistically analyzes bit error rate and latency jitter. The multi-channel collaborative control module is equipped with nested interrupt support for the real-time interrupt response of the abnormal pressure tracing module. High-priority interrupts can preempt low-priority interrupts, and interrupt context switching is optimized to reduce overhead. The parameter transmission compression algorithm of the injection path optimization module to the safety threshold self-correction module enables lossless mode, and forward error correction coding is used to enhance robustness for key parameter transmission. The compression ratio is dynamically adjusted according to network conditions. The microcontroller storage resource access timing coordination introduces a predictive scheduling algorithm. The algorithm analyzes the module access mode to prefetch the storage data that may be needed, and the storage controller implements an arbitration strategy to optimize throughput. The storage protection unit sets access permission boundaries to prevent modules from accessing sensitive data without authorization, and the storage error correction code detects and corrects single-bit errors. The microcontroller's clock synchronization mechanism ensures the time consistency of operation of each module. The synchronization signal is distributed to all functional units, and clock gating technology reduces the power consumption of idle modules.

[0041]

[0042] The version management of the safety threshold self-correction module records the history of each parameter set change. Version numbers are bound to injection event numbers for easy traceability, and a rollback mechanism allows restoration of previous parameter sets upon detection of anomalies. The microcontroller's monitoring subsystem records performance metrics of the safety threshold parameter set update process, including computational latency, storage usage, and communication overhead. This data is used for system optimization analysis. A battery backup circuit maintains parameter set storage during mains power failure, ensuring continuous operation for at least 72 hours. Parameter set integrity verification is performed periodically using cyclic redundancy check (CRC) codes. The system log records all parameter set modification operations, with entries including operation time, module identifier, and modification content. Wear leveling is implemented in the log storage area to extend memory lifespan.

[0043] Example 5: A microcontroller-integrated 5G network status monitoring unit evaluates the impact of data transmission latency on stress compensation commands in real time. The evaluation process is based on continuously measuring the round-trip time of 5G data packets. The 5G network status monitoring unit injects timestamped probe data packets into the network transmission channel. The probe data packets have a fixed length of bytes to maintain measurement consistency. The receiving end parses the timestamps and calculates the end-to-end latency. Latency calculation excludes internal device processing time and focuses on the network transmission link. The 5G network status monitoring unit maintains a latency statistics window recording the most recent 100 measurement results. The statistics window uses a sliding update mechanism to discard the oldest data. Latency thresholds are divided into multiple levels corresponding to different network status codes: green level indicates latency below 20 milliseconds, yellow level indicates latency between 20 and 50 milliseconds, and red level indicates latency exceeding 50 milliseconds. The 5G network status monitoring unit dynamically adjusts the transmission frequency of data interaction between modules to match network status changes. The transmission frequency adjustment follows a predefined mapping table. The mapping table maps network status codes to specific transmission time intervals: green level corresponds to a 10-millisecond transmission interval, yellow level corresponds to a 30-millisecond transmission interval, and red level corresponds to a 100-millisecond transmission interval. The frequency switching process employs a smooth transition algorithm to avoid abrupt changes. The algorithm gradually adjusts to the target frequency over ten cycles. The 5G network status monitoring unit broadcasts network status change event notifications to the subscription module. The event message includes a timestamp, old status code, new status code, and a switching reason explanation. The subscription module adjusts its local transmission strategy based on the event message. When the latency exceeds a critical value, the 5G network status monitoring unit switches to a local cache execution strategy. The critical value is set at 80 milliseconds, corresponding to a severe network congestion state. The switching decision undergoes multi-condition verification to prevent false triggering. Verification conditions include five consecutive probe packet timeouts and latency variance exceeding a threshold. The local cache execution strategy activates the cache manager's prefetch function. The prefetch algorithm loads potentially needed data into the cache based on historical access patterns. A cache consistency protocol ensures data consistency across multiple modules, and the microcontroller's clock synchronization mechanism coordinates the cache switching sequence.

[0044] The microcontroller-managed injection parameter history database continuously records intermediate parameter sets generated by each module. These intermediate parameter sets include data such as pressure fluctuation feature matrices, fluid state reconstruction parameter codes, and channel coordination instruction queues. The recording process uses a structured storage format, with metadata descriptions added to each parameter set, including generation time, module identifier, data version, and checksum. The injection parameter history database establishes an association index between parameter set version markers and injection events. The version marker uses a dual identifier of an incrementing sequence number and a timestamp, while the injection event number comes from the system event counter. The association index is implemented as a hash table structure to support fast lookups, and the hash function optimizes the collision rate based on event number characteristics. Index updates maintain atomicity with parameter records. The injection parameter history database dynamically selects between uploading or local storage strategies for historical data based on 5G network bandwidth. Bandwidth assessment is based on real-time throughput measurements. The uploading strategy is prioritized when bandwidth is sufficient (sufficient bandwidth is defined as available bandwidth exceeding ten megabits per second), and the upload process employs a breakpoint resumption mechanism to enhance reliability. The local storage strategy is activated when bandwidth is insufficient, with data compression algorithms reducing storage space usage and storage media balancing technology extending flash memory lifespan. The strategy selector comprehensively considers network conditions and storage resource status, and the selection algorithm calculates a weighted score to determine the current strategy. The weighting coefficients are configured by the network administrator. The latency measurement module of the 5G network status monitoring unit integrates a hardware timestamp unit with nanosecond-level accuracy, and the measurement results automatically correct for clock drift. The latency statistics window implements an automatic expansion mechanism; when increased network jitter is detected, the window size is expanded to two hundred samples, and the average latency, standard deviation, and percentiles are statistically calculated and output. The network status code mapping table supports runtime updates, and administrators can modify threshold boundaries and corresponding actions through the configuration interface. Mapping table changes are recorded in the audit log. The transmission frequency adjustment mechanism includes security protection logic to prevent frequent frequency oscillations in a short period of time, setting a minimum stability time of one second.

[0045] The local caching strategy divides cache partitions based on data type, storing real-time operation data in high-speed static storage and historical data in large-capacity dynamic storage. The cache replacement algorithm is an improvement on the Least Recently Used (LRU) principle, taking into account data access frequency and priority. Cache consistency maintenance employs a write-back strategy combined with periodic refresh. The microcontroller's power management unit provides adjustable voltage to the cache memory, dynamically optimizing the voltage based on access frequency for energy saving. Cache error detection and correction circuitry ensures data integrity. The injection parameter historical database's recording engine uses a transaction processing mode, guaranteeing atomic records of the parameter set, and a transaction rollback mechanism handles record failure scenarios. Parameter set version markers are generated using a distributed unique identifier algorithm, combining microcontroller identifiers and high-precision timers to avoid conflicts. Injection event-related indexes maintain bidirectional links. Index cleanup tasks periodically remove expired associations, with cleanup conditions based on event completion status and time thresholds. The database reorganization process optimizes storage layout and improves access efficiency. The network bandwidth assessment module implements a combination of active probing and passive monitoring. Active probing sends test packets to measure actual bandwidth, while passive monitoring analyzes daily transmission throughput statistics. The bandwidth assessment results are smoothed to eliminate the impact of instantaneous fluctuations. The smoothing window is dynamically adjusted according to the network type, with a larger smoothing window used for mobile networks and a smaller smoothing window used for wired networks. The data block size of the upload strategy is adaptively adjusted, using large data blocks to reduce protocol overhead in high-bandwidth environments and small data blocks to improve transmission success rate in low-bandwidth environments. The compression algorithm of the local storage strategy selects different dictionaries according to the data type; LZMA algorithm is used for text data, and Delta encoding combined with Huffman coding is used for binary data.

[0046] The collaboration between the 5G network status monitoring unit and the historical injection parameter database is achieved through a message queue. The message queue decouples asynchronous communication, and deep queue monitoring prevents data backlog. A flow control mechanism adjusts the data generation rate to match processing capabilities, a backpressure signal notifies the sender to pause transmission, and the microcontroller's memory protection unit isolates the storage spaces of different modules. Access control is based on role allocation; the network monitoring module has read permissions, and the database module has read and write permissions. A secure startup process verifies the integrity of the module code. The system monitor records the operational metrics of the 5G network status monitoring unit and the historical injection parameter database. Visualization of these metrics helps administrators optimize system parameters, and fault injection testing verifies the system's fault tolerance.

[0047] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

[0048] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. An intelligent control system for a high-pressure injector based on 5G technology, characterized in that, The system includes: The microcontroller-driven injection pressure dynamic adjustment module receives pressure sensor data streams from the high-pressure injector in real time via a 5G network, analyzes the pressure fluctuation characteristics of the current injection stage, and generates a primary pressure compensation command by comparing it with a preset pressure threshold sequence. The fluid state reconstruction module identifies the viscosity change trend of the fluid medium in the injection line based on the pressure fluctuation characteristics in the primary pressure compensation command, and calculates the fluid state reconstruction parameters by combining the real-time flow rate data stream transmitted by 5G. Based on the fluid state reconstruction parameters, the multi-channel collaborative control module analyzes the dynamic value of the pressure difference between parallel injection units, allocates pressure compensation weights for each channel, and generates a channel collaborative instruction set. The abnormal pressure tracing module monitors the abnormal pressure mutation points of each injection unit according to the channel coordination instruction set, traces the change trajectory of fluid state reconstruction parameters corresponding to the pressure mutation, and outputs abnormal pressure tracing marker data. Based on the abnormal pressure tracing marker data, the injection path optimization module replans the propulsion path trajectory of the high-pressure injector and generates a three-dimensional path optimization parameter table by combining the real-time location data fed back by the 5G network. The safety threshold self-correction module dynamically adjusts the boundary value range of the pressure threshold sequence based on the trajectory offset in the three-dimensional path optimization parameter table, and outputs the updated safety threshold parameter set to the microcontroller.

2. The intelligent control system for high-pressure injectors based on 5G technology according to claim 1, characterized in that, The specific steps for obtaining the pressure fluctuation characteristics are as follows: The microcontroller extracts a sequence of pressure sample values ​​within a continuous time window from the 5G data stream and calculates the rate of change of pressure gradient between adjacent sampling points. Based on the positive and negative polarity distribution of the pressure gradient change rate, the characteristic intervals of the pressure rising phase and the pressure falling phase are divided. By comparing the deviation of the pressure gradient change rate within the characteristic interval from the preset pressure threshold sequence, abnormal fluctuation intervals exceeding the threshold are marked. The duration and amplitude changes of abnormal fluctuation ranges are integrated to generate a pressure fluctuation feature matrix.

3. The intelligent control system for high-pressure injectors based on 5G technology according to claim 2, characterized in that, The specific steps for generating the fluid state reconstruction parameters are as follows: Analyze the timestamps of abnormal fluctuation intervals in the pressure fluctuation feature matrix and extract the fluid medium velocity sampling values ​​at the corresponding times. Calculate the percentage deviation between the sampled flow velocity and the reference flow velocity, and establish a two-dimensional parameter mapping table in conjunction with the pressure gradient change rate; Based on the distribution density of parameter points in the two-dimensional parameter mapping table, the boundary thresholds for dividing the fluid state into the stable region and the turbulent region are determined. Based on boundary thresholds, the real-time collected flow velocity data stream is classified into different state levels, generating fluid state reconstruction parameter codes.

4. The intelligent control system for high-pressure injectors based on 5G technology according to claim 3, characterized in that, The specific steps for generating the channel coordination instruction set are as follows: Monitor the difference in the rate of change of pressure gradient between parallel injection units at the same time stamp; The priority of the current state level of each unit is determined based on the fluid state reconstruction parameter coding; The execution sequence of pressure compensation instructions is allocated according to the priority of status level, and a channel coordination instruction queue with timestamps is generated. By comparing the time interval between adjacent instructions in the instruction queue with the 5G network latency parameters, the instruction sending time can be optimized.

5. The intelligent control system for a high-pressure injector based on 5G technology according to claim 4, characterized in that, The specific steps for generating the abnormal pressure tracing marker data are as follows: Capture pressure mutation events after the execution of the channel cooperative instruction queue, and record the sequence of fluid state reconstruction parameter changes within a continuous time window before the mutation occurs; Analyze the time lag between the inflection point of key indicators and the time of sudden pressure change in the parameter change sequence. Establish a correlation strength marker between pressure abrupt changes and specific fluid state parameters based on the magnitude of the delay; Parameters whose association strength exceeds a threshold are combined into an abnormal pressure source labeling dataset.

6. The intelligent control system for a high-pressure injector based on 5G technology according to claim 5, characterized in that, The specific steps for generating the three-dimensional path optimization parameter table are as follows: Analyze the key parameter items in the abnormal pressure tracing marker dataset to locate the real-time position offset of the high-pressure injector propulsion mechanism; Calculate the trajectory correction vector of the propulsion path in the three-dimensional coordinate system based on the offset direction; By combining the data consistency verification results from multiple location sensors fed back by the 5G network, effective trajectory correction vectors are selected; The effective vectors are integrated into a three-dimensional path optimization parameter table according to the time series.

7. The intelligent control system for a high-pressure injector based on 5G technology according to claim 6, characterized in that, The specific steps for updating the security threshold parameter set are as follows: Extract the magnitude statistics of the trajectory correction vector from the 3D path optimization parameter table; Compare the coverage differences between the statistical values ​​and the original safety thresholds; When the difference exceeds the preset tolerance, the upper and lower boundary values ​​of the pressure threshold sequence are scaled proportionally. The scaled boundary values ​​are compared with the historical security parameters stored in the microcontroller, and conflicting values ​​are removed to generate a new set of security threshold parameters.

8. The intelligent control system for a high-pressure injector based on 5G technology according to claim 7, characterized in that, The specific steps for the microcontroller to achieve multi-module collaboration are as follows: Establish a dedicated 5G data transmission channel between the injection pressure dynamic adjustment module and the fluid state reconstruction module; Configure a real-time interrupt response mechanism for the abnormal pressure tracing module in the multi-channel collaborative control module; The parameter transmission compression algorithm from the injection path optimization module to the safety threshold self-correction module is optimized. Coordinate the timing of each module's access to the microcontroller's storage resources to avoid conflicts.

9. The intelligent control system for a high-pressure injector based on 5G technology according to claim 8, characterized in that, The system also includes: The microcontroller integrates a 5G network status monitoring unit to assess the impact of data transmission latency on stress compensation commands in real time. Dynamically adjust the transmission frequency of data interaction between modules to match changes in network status; When the latency exceeds a critical value, switch to the local cache execution strategy.

10. The intelligent control system for a high-pressure injector based on 5G technology according to claim 9, characterized in that, The system further includes: The microcontroller manages a historical database of injection parameters, continuously recording the intermediate parameter sets generated by each module; Establish an index linking parameter set version tags with injection events; The strategy of uploading or storing historical data locally is dynamically selected based on the 5G network bandwidth.

Images