Remote control system and control method for high-pressure air blasting device

Abstract

The invention relates to the technical field of non-electrical variable control or regulation systems, and particularly discloses a remote control system and a remote control method for a high-pressure air blasting device. The system comprises a remote control terminal, a system main controller, a nonlinear dynamic modeling unit, a self-adaptive prediction control unit, a multi-modal execution driving unit and a high-precision sensing acquisition network. By constructing a real-time updated nonlinear dynamic model and combining adaptive predictive control and multi-mode valve cooperative driving, millisecond-level high-precision tracking and stable output of bursting pressure are achieved, and operation safety and control reliability are effectively improved.

Description

Technical Field

[0001] This invention belongs to the technical field of control or regulation systems for non-electrical variables, and specifically relates to a remote control system and control method for a high-pressure air blasting device. Background Technology

[0002] High-pressure air blasting technology, as a non-explosive rock breaking and demolition method, has significant application value in mining, urban demolition, and special engineering fields. This technology generates shock waves and stress waves by instantaneously releasing high-pressure gas, thereby breaking up media such as rock or concrete. Its core lies in the precise control of gas pressure and the release process.

[0003] The control system of the high-pressure air blasting device directly determines the accuracy and reliability of the blasting operation. This system must manage the entire process of high-pressure gas energy storage, release, and pressure regulation within milliseconds to ensure the stability and predictability of the blasting energy output.

[0004] In existing technologies, high-pressure air blasting devices mostly employ a set-execute open-loop control mode or a closed-loop control strategy based on traditional PID algorithms. However, high-pressure gas exhibits strong nonlinear dynamic characteristics during rapid release, with factors such as pressure decay, temperature changes, and flow field disturbances coupling with each other, making it difficult for traditional control methods to achieve high-precision tracking and stable output of blasting pressure. This lack of control precision directly leads to fluctuations in blasting power, resulting in uneven fragment size distribution and inconsistent operational effects, while also increasing the safety risks associated with overshoot or undershoot. Therefore, there is an urgent need for a remote control solution that can adapt to the nonlinear dynamic process of high-pressure gas and achieve precise pressure regulation. Summary of the Invention

[0005] The technical problem to be solved by the present invention is to overcome the shortcomings of the existing high-pressure air blasting device control methods in terms of insufficient accuracy when dealing with the nonlinear dynamic characteristics of high-pressure gas, and to provide a remote control system and control method for high-pressure air blasting devices, which aims to achieve millisecond-level high-precision tracking and stable output of blasting pressure.

[0006] To achieve the above objectives, the present invention provides a remote control system for a high-pressure air blasting device, the core architecture of which consists of a remote control terminal, a system main controller, a nonlinear dynamic modeling unit, an adaptive predictive control unit, a multimodal execution drive unit, and a high-precision sensing and acquisition network.

[0007] The remote control terminal establishes a bidirectional data connection with the system's main controller via an encrypted communication link. This connection is used to receive operational commands and send them to the main controller, while simultaneously displaying system status and blasting process data in real time. The main controller, as the core of the system, is responsible for coordinating the work of each unit, parsing commands from the remote control terminal, and overseeing the execution of control logic.

[0008] The nonlinear dynamic modeling unit is connected to the system's main controller and is used to construct and update in real time a dynamic mathematical model that characterizes the nonlinear coupling relationship between pressure, temperature, and flow rate during the high-pressure gas release process.

[0009] The adaptive predictive control unit is connected to the nonlinear dynamic modeling unit and the system main controller, respectively. Based on the dynamic mathematical model, it performs forward rolling optimization calculations to generate the optimal control sequence for the multimodal execution drive unit.

[0010] The multimodal execution drive unit receives control commands output by the adaptive predictive control unit and drives the electro-pneumatic servo proportional valve and the fast relief valve in the high-pressure gas circuit to work together to precisely regulate the release process of high-pressure gas in the accumulator.

[0011] A high-precision sensor acquisition network is distributed and deployed at key nodes of the high-pressure gas path to monitor accumulator pressure, pipeline flow and valve inlet and outlet pressure difference in real time, and feeds the acquired data back to the system main controller and nonlinear dynamic modeling unit.

[0012] Furthermore, the internal workflow of the nonlinear dynamic modeling unit is as follows: First, the unit receives real-time pressure, flow, and temperature data from a high-precision sensing network; second, based on the received data stream, it uses a recursive least squares method with a forgetting factor to identify online the model parameters describing the gas equation of state and flow resistance characteristics; the model parameters include at least the equivalent value of the gas adiabatic index, the equivalent friction coefficient of the flow channel, and the volume effect time constant; finally, using the updated model parameters, a nonlinear dynamic model expressed in the form of a differential-algebraic equation system is constructed, which can predict the system pressure response within a specific future time window.

[0013] Furthermore, the adaptive predictive control unit's operation mechanism comprises three core components: a prediction component, an optimization component, and a feedback correction component. The prediction component invokes the current nonlinear dynamic model output by the nonlinear dynamic modeling unit, using the current system state as initial conditions, to predict the system pressure trajectory over the next 20ms to 50ms time domain. The optimization component sets an objective function that aims to minimize the cumulative deviation between the predicted pressure trajectory and the set pressure command, while constraining the rate of change of the control variable to prevent actuator saturation; by solving the constrained optimization problem, a series of optimal control variables are obtained within the future control time domain. At the end of each control cycle, the feedback correction component compares the system pressure measured by the high-precision sensor network with the predicted value, generating a prediction error. This error is then used to compensate for subsequent predictions, enhancing the control robustness under model mismatch or external disturbances.

[0014] Furthermore, the multimodal actuation drive unit includes an electro-pneumatic servo proportional valve and a parallel fast-release valve. The electro-pneumatic servo proportional valve receives an analog voltage control signal output from the adaptive predictive control unit and precisely adjusts the main valve core opening through its internal pilot stage and power amplifier stage, thereby achieving continuous proportional control of the high-pressure gas flow. The fast-release valve receives a digital switch control signal and fully opens within 5ms when the system main controller determines that rapid pressure relief or an emergency situation is required, ensuring rapid system unloading. The coordination strategy between the electro-pneumatic servo proportional valve and the fast-release valve is jointly determined by the system main controller based on the output of the adaptive predictive control unit and safety logic.

[0015] Furthermore, the high-precision sensing network specifically includes an absolute pressure sensor, a mass flow meter, and a differential pressure sensor. The absolute pressure sensor is installed at the accumulator outlet, with a measurement range covering 0 to 100 MPa, and is used to directly monitor the pressure within the accumulator. The mass flow meter is installed in the main gas path, with a response time of less than 1 ms, and is used to accurately measure the instantaneous gas mass flow rate. The differential pressure sensor is connected across the inlet and outlet ports of the electro-pneumatic servo proportional valve to monitor the pressure difference across the valve core, providing data support for valve status diagnosis and flow estimation. All sensor data is sampled and quantized by a 24-bit analog-to-digital converter and transmitted to the system main controller via the controller area network bus.

[0016] Furthermore, the system's main controller integrates a safety monitoring and fault diagnosis module. This module continuously analyzes data from the high-precision sensor acquisition network and the status feedback from the multimodal actuators. By comparing preset safety thresholds and logical rules, it diagnoses in real time whether the system has overpressure, leakage, or actuator jamming faults. Once a fault is detected, it immediately triggers a tiered safety response, including but not limited to switching to a safe mode, shutting off the main gas path, or activating a rapid relief valve.

[0017] Furthermore, the present invention also provides a control method for the remote control system of the above-mentioned high-pressure air blasting device, the method comprising the following steps:

[0018] Step 1, System Initialization and Self-Test: The remote control terminal is powered on and establishes a secure connection with the system main controller; the system main controller sequentially performs power-on self-tests and communication tests on the high-precision sensor acquisition network, nonlinear dynamic modeling unit, adaptive predictive control unit, and multimodal execution drive unit. After confirming that each unit is in normal condition, it enters standby mode.

[0019] Step 2, Blasting Parameter Setting and Model Construction: The operator inputs the target pressure curve, blasting duration, and safety parameters for this blasting operation through a remote control terminal; after receiving the parameters, the system main controller instructs the nonlinear dynamic modeling unit to start initial modeling; the nonlinear dynamic modeling unit constructs the initial nonlinear dynamic model of the high-pressure gas system based on the initial steady-state data provided by the high-precision sensor acquisition network.

[0020] Step 3, Adaptive Predictive Control Start-up: The system main controller issues an energy storage command, and the electro-pneumatic servo proportional valve in the multimodal execution drive unit slowly opens to charge the energy storage device; at the same time, the adaptive predictive control unit starts up and begins rolling optimization calculation based on the initial nonlinear dynamic model to generate a preliminary control sequence that acts on the electro-pneumatic servo proportional valve, so that the energy storage device pressure rises steadily to the target initial pressure.

[0021] Step 4, Precise Control of the Blasting Process: After the system's main controller receives the blasting execution command from the remote control terminal, the adaptive predictive control unit enters a high-frequency control cycle. Within each control cycle, the unit executes prediction, optimization, and feedback correction processes, adjusting the control signal output to the electro-pneumatic servo proportional valve in real time to drive the valve's action. This aims to ensure that the pressure trajectory measured by the high-precision sensor acquisition network closely tracks the preset target pressure curve. Simultaneously, the nonlinear dynamic modeling unit continuously updates the model parameters online using the latest sensor data to ensure that the dynamic model accurately reflects the current gas release characteristics.

[0022] Step 5, End of blasting and system reset: When the blasting duration reaches the set value or the pressure drops below the safety threshold, the system main controller determines that the blasting process has ended; then, the main controller instructs the multi-modal execution drive unit to close the electro-pneumatic servo proportional valve and activates the quick relief valve to release residual pressure as needed; the system status data is recorded and uploaded to the remote control terminal, and each control unit is reset, waiting for the next operation instruction.

[0023] Furthermore, in the precise control of the blasting process in step 4, if the safety monitoring and fault diagnosis module detects that the pressure overshoot is greater than the preset safety limit and continues for more than 3 control cycles, the system main controller will immediately override the output of the adaptive predictive control unit, forcibly intervene in the control, and prioritize the activation of the fast relief valve for rapid depressurization until the pressure is restored to the safe range.

[0024] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0025] 1. This invention constructs a control system capable of accurately describing and actively compensating for the nonlinear and coupled dynamic characteristics during high-pressure gas release by introducing a nonlinear dynamic modeling unit and an adaptive predictive control unit. The adaptive predictive control unit performs forward rolling optimization based on a real-time updated nonlinear model, enabling it to anticipate future system dynamics and apply control actions in advance. This effectively overcomes the inherent defects of traditional PID control, such as lag and large overshoot, in rapidly changing and non-minimum phase systems. Thus, it achieves high-precision tracking and stable output of blasting pressure at the millisecond level, improving the uniformity of blasting power and the consistency of operational results.

[0026] 2. This invention employs a multi-modal actuator unit, combining the continuous and precise control of an electro-pneumatic servo proportional valve with the rapid and safe response of a quick-release valve, and achieving intelligent coordination through the system's main controller. This design not only ensures precise pressure regulation under normal operating conditions but also provides the system with rapid and safe redundancy to cope with abnormal conditions such as overpressure, effectively reducing the risk of overshoot or undershoot caused by control failure or slow response, and enhancing the safety of the entire blasting operation process.

[0027] 3. The high-precision sensor acquisition network and safety monitoring and fault diagnosis module integrated in this invention enable comprehensive, high-frequency monitoring and intelligent diagnosis of key system status parameters. This system can sense subtle changes in parameters such as pressure and flow rate in real time, and quickly identify potential faults based on preset logic, executing graded safety strategies. This deep state awareness and proactive safety protection mechanism further enhances the reliability and risk management capabilities of remote control of high-pressure air blasting devices, providing a solid technical guarantee for safe operation in complex engineering environments. Attached Figure Description

[0028] Figure 1 This is a schematic diagram of the overall technical architecture of the remote control system for the high-pressure air blasting device proposed in this invention;

[0029] Figure 2 This is a schematic diagram of the core principle framework of nonlinear dynamic modeling and adaptive predictive control in this invention;

[0030] Figure 3This is a logical flowchart of the operation mechanism of the adaptive predictive control unit in this invention;

[0031] Figure 4 This is a schematic diagram of the interaction relationship and data flow between the multimodal execution driving unit and the high-precision sensing acquisition network in this invention;

[0032] Figure 5 This is a hierarchical response logic framework diagram for system safety monitoring and fault diagnosis in this invention; Detailed Implementation

[0033] This embodiment details a specific implementation scheme of a remote control system for a high-pressure air blasting device. Please refer to the appendix. Figure 1 The overall architecture of the system consists of a remote control terminal, a main system controller, a nonlinear dynamic modeling unit, an adaptive predictive control unit, a multimodal execution drive unit, and a high-precision sensing and acquisition network.

[0034] The remote control terminal establishes a bidirectional full-duplex data connection with the system's main controller via a wireless communication link employing advanced encryption standards. The terminal is equipped with a ten-point capacitive touchscreen display and a physical emergency stop button. It receives blasting parameter commands input by the operator and sends them to the main controller. Simultaneously, it dynamically displays key system status parameters and pressure-flow curves during the blasting process in real time at a refresh rate of 60fps. The main controller uses a multi-core processor with a clock speed of 2.4GHz as its core computing unit and incorporates a real-time operating system.

[0035] The system's main controller is responsible for coordinating the working sequence of all units within the system, parsing instruction data packets from remote control terminals, and coordinating the cyclical execution of all control logic and security policies.

[0036] The nonlinear dynamic modeling unit is connected to the system main controller via a high-speed parallel bus. This unit contains a dedicated digital signal processor for constructing and updating in real time a dynamic mathematical model characterizing the nonlinear coupling relationship between pressure, temperature, and flow during the release of high-pressure gas.

[0037] The adaptive predictive control unit interacts with the nonlinear dynamic modeling unit and the system main controller through a dual-port memory. This unit performs forward rolling optimization calculations based on the dynamic mathematical model to generate the optimal control sequence for the multimodal execution drive unit.

[0038] The multimodal execution drive unit receives control commands from the adaptive predictive control unit through an isolated digital-to-analog converter interface and an analog output module. It drives the electro-pneumatic servo proportional valve and the fast relief valve in the high-pressure gas circuit to work together to precisely regulate the release process and rate of high-pressure gas in the accumulator.

[0039] The high-precision sensor acquisition network adopts a distributed architecture and is deployed at key physical nodes of the high-pressure gas path, including the accumulator outlet, the straight section of the main gas path, and both ends of the valve. It monitors the accumulator pressure, the instantaneous gas mass flow rate in the pipeline, and the pressure difference before and after the valve in real time, and synchronously feeds back the collected data to the system main controller and the nonlinear dynamic modeling unit at a rate of 1000fps through the controller local area network bus.

[0040] The specific hardware components of the remote control terminal include an industrial-grade embedded computer module, a wireless communication module, and a human-machine interface.

[0041] The embedded computer module uses a 64-bit processor and is equipped with 8GB of RAM and 128GB of solid-state storage. The wireless communication module supports 5G mobile communication technology and wireless protocols, and uses Transport Layer Security (TLS) version 1.3 for end-to-end encryption at the communication link layer to ensure the confidentiality and integrity of control commands and status data during transmission.

[0042] The human-machine interface uses a 12.1-inch LCD screen with a resolution of 1920×1080 pixels. The interface layout is divided into a parameter setting area, a real-time data display area, a historical curve playback area, and an alarm information area. Operators input parameters such as the target pressure curve, total blast duration, and maximum allowable pressure overshoot in the parameter setting area. These parameters are then validated by the interface logic and encapsulated into a specific format instruction frame, which is sent to the system's main controller via the wireless communication module.

[0043] Meanwhile, the remote control terminal continuously receives system status data packets from the system's main controller. After parsing, it displays the current accumulator pressure, instantaneous flow rate, valve opening, system operating mode, and remaining burst time in the real-time data display area in digital and instrument panel format. It also plots the trend of pressure and flow rate over time in the historical curve playback area.

[0044] The system's main controller serves as the command center of the entire control system. Its hardware platform is built on a multi-core processor, which contains four processing cores, each with a main frequency of 2.4 GHz, and is equipped with a real-time operating system to ensure the timely scheduling of control tasks.

[0045] The system's main controller interconnects with other units within the system through multiple communication interfaces: it communicates with remote control terminals via an Ethernet interface; it communicates with high-precision sensor acquisition networks via a controller area network bus interface; it exchanges data with nonlinear dynamic modeling units and adaptive predictive control units via high-speed serial peripheral interfaces; and it connects to multimodal execution drive units via digital-to-analog converters and digital input / output interfaces.

[0046] The system's main controller internally runs a main control program, a communication management program, a safety monitoring program, and a data logging program. The main control program is responsible for parsing instructions from the remote control terminal and calling the corresponding functional modules based on the instruction type, such as the initialization module, inflation control module, blasting execution module, and system reset module. The communication management program is responsible for sending, receiving, verifying, encapsulating, and parsing data from each interface, ensuring the reliability and real-time performance of the data stream.

[0047] The safety monitoring program runs continuously, and its detailed functions will be explained later in conjunction with the safety monitoring and fault diagnosis modules. The data logging program stores key operational events, control parameters, sensor readings, and alarm information in a timestamp format in non-volatile memory, with a storage capacity that can support continuous recording of more than 30 days of operational data.

[0048] Nonlinear dynamic modeling units are the theoretical foundation for achieving high-precision control. Please refer to the appendix. Figure 2 At the heart of this unit is a high-performance digital signal processor whose workflow begins with receiving real-time data streams from a high-precision sensor acquisition network.

[0049] These data streams include pressure sequences from absolute pressure sensors, flow sequences from mass flow meters, and temperature data from temperature sensors, updated 1000 times per second. The digital signal processor internally runs online parameter identification and model building algorithms. The online parameter identification algorithm employs recursive least squares with a forgetting factor; its purpose is to dynamically identify key model parameters describing the dynamic characteristics of the high-pressure gas system using continuously flowing real-time sensor data.

[0050] These model parameters include at least three core parameters: the equivalent value of the gas adiabatic index, which reflects the approximate characteristics of the thermodynamic behavior of high-pressure gas during rapid expansion; the equivalent friction coefficient of the flow channel, which characterizes the energy loss characteristics of gas flowing through valves and pipes; and the volume effect time constant, which reflects the inertial delay effect of the system volume on pressure changes.

[0051] A typical value for the forgetting factor is 0.98. Its purpose is to assign higher weights to new data, enabling the model to track the slow drift of system characteristics. The model building algorithm then uses the newly identified model parameters to construct a nonlinear dynamic model expressed as a system of differential-algebraic equations. This model is essentially a set of mathematical equations describing the dynamic relationships between state variables such as pressure, flow rate, and temperature, and can predict the system pressure response trajectory within a time window of 20ms to 50ms, using the current system state as initial conditions.

[0052] The adaptive predictive control unit is the core decision-making unit for performing precise control. Please refer to the appendix. Figure 3The unit's operating mechanism comprises three closely linked core components: prediction, optimization, and feedback correction.

[0053] The prediction phase is triggered at the beginning of each control cycle, and it calls the latest nonlinear dynamic model output by the nonlinear dynamic modeling unit.

[0054] The prediction phase uses the measured system state values ​​at the current moment provided by a high-precision sensor network as initial conditions. It then uses this nonlinear dynamic model for numerical integration to predict the system's pressure change trajectory over the next 20ms to 50ms. The optimization phase follows immediately after the prediction phase, and this phase sets a clear objective function.

[0055] The objective function aims to minimize the cumulative quadratic deviation between the predicted future pressure trajectory and the target pressure command set by the remote control terminal.

[0056] Meanwhile, constraints on the rate of change of the control quantity are introduced into the optimization problem to prevent the electro-pneumatic servo proportional valve in the multimodal execution drive unit from entering the saturation region and losing its regulating ability due to the rapid change of the command.

[0057] The optimization process involves solving a constrained quadratic programming problem to calculate a series of optimal control variables in the future control time domain. Typically, only the first control variable is applied to the controlled object.

[0058] The feedback correction stage is executed at the end of each control cycle. This stage compares the current system pressure value measured by the high-precision sensor network with the prediction value for that moment in the previous cycle, and calculates the prediction error. This prediction error is used to compensate for the prediction output of the next control cycle. Typically, a weighted method is used to incorporate the error into the new initial prediction value, thereby effectively enhancing the system's control robustness when facing model parameter mismatch or unknown external disturbances.

[0059] The multimodal execution drive unit is the final execution mechanism for control commands. Please refer to the appendix. Figure 4 Specifically, the unit includes an electro-pneumatic servo proportional valve and a fast relief valve connected in parallel with it in the pneumatic circuit.

[0060] The electro-pneumatic servo proportional valve receives an analog voltage control signal from the adaptive predictive control unit, with a signal range of 0 to 10V. The electro-pneumatic servo proportional valve internally includes a pilot stage and a power amplifier stage.

[0061] The pilot stage is a low-power electromagnetic converter that proportionally converts the input voltage signal into pilot gas pressure. The power amplifier stage uses this pilot gas pressure to drive the main valve spool to move against spring force and hydrodynamic force, thereby precisely and continuously adjusting the flow area of ​​the main valve port to achieve proportional control of the high-pressure gas mass flow rate. The valve spool displacement is detected by a linear variable differential transformer and forms a local closed loop to ensure the accuracy of valve position control.

[0062] The quick-release valve receives digital switch control signals from the digital output module of the system's main controller, typically 24V DC level signals.

[0063] The quick-release valve adopts an electromagnet-driven valve core structure, designed for extremely fast response speed. Upon receiving an opening signal, it can switch from fully closed to fully open within 5ms, ensuring that the main air path can be quickly bypassed to achieve rapid unloading when the system needs rapid pressure relief or an emergency occurs.

[0064] The coordinated action strategy of the electro-pneumatic servo proportional valve and the quick-release valve is jointly determined by the safety logic in the system's main controller and the output state of the adaptive predictive control unit.

[0065] During normal, precise control, only the electro-pneumatic servo proportional valve operates; when the system's main controller determines that rapid pressure relief is required, it will activate the rapid relief valve first or simultaneously.

[0066] A high-precision sensor network provides accurate status awareness for the system. This network comprises three key sensors: an absolute pressure sensor, a mass flow meter, and a differential pressure sensor. The absolute pressure sensor is directly mounted on the gas outlet interface of the accumulator, with a measurement range covering 0 to 100 MPa. Internally, this sensor employs silicon micro-fusion technology to convert the pressure signal into a millivolt-level voltage signal output. The mass flow meter is installed in the straight section of the main gas path. Operating based on the Coriolis force principle, it directly measures instantaneous gas mass flow rate with a response time of less than 1 ms and an accuracy of 0.5% of the reading.

[0067] A differential pressure sensor is connected across the inlet and outlet of an electro-pneumatic servo proportional valve to monitor the pressure difference across the valve. Its range is typically 0 to 5 MPa, with an accuracy of 0.25% of the full range. The pressure difference data before and after the valve plays a crucial role in estimating the real-time flow through the valve and diagnosing whether the valve is blocked or stuck.

[0068] The analog signals output by all sensors are connected to a centralized signal conditioning and acquisition module, which contains a multi-channel 24-bit analog-to-digital converter to perform anti-aliasing filtering, amplification, and digital sampling of the signals.

[0069] The digitized sensor data is transmitted to the system's main controller via the Controller Area Network (CLAN) bus. The CLAN bus uses a communication rate of 500 kbit per second, the data frame format conforms to the standard data frame format, and each sensor data is assigned a specific identifier.

[0070] The system's main controller integrates a safety monitoring and fault diagnosis module, which ensures the system's safe operation. Please refer to the appendix. Figure 5 This module runs as an independent software task on the real-time operating system, and its execution frequency is the same as that of the main control loop.

[0071] The safety monitoring and fault diagnosis module continuously analyzes real-time data from the high-precision sensor acquisition network and status feedback information returned by the multi-modal execution drive unit, such as valve drive current and valve position feedback. This module has multiple preset safety thresholds and logical diagnostic rules.

[0072] Safety thresholds include upper limits for absolute pressure, upper limits for pressure rise rate, and abnormal flow rate thresholds. Logical diagnostic rules are used to determine system status. For example, analyzing the correlation between the changing trend of absolute pressure sensor readings and mass flow meter readings can determine if there is a gas leak; comparing the deviation between the control command and the valve position feedback signal of an electro-pneumatic servo proportional valve can determine if there is actuator jamming; and monitoring whether the pressure response under specific control actions exceeds the expected range can diagnose the degree of model mismatch.

[0073] Once the safety monitoring and fault diagnosis module detects that any parameter exceeds the safety threshold or meets the fault diagnosis rules, it will immediately send an interrupt request to the decision core of the system's main controller, triggering a graded safety response.

[0074] The graded safety response is divided into different levels according to the severity of the fault: for minor overshoot or transient anomalies, only a warning signal may be issued and the event recorded; for persistent overshoot or clear actuator faults, it will switch to a safety mode, such as locking the current valve opening; for severe overpressure or leakage faults, an emergency command will be issued immediately to close the front-end shut-off valve of the main gas line and forcibly activate the quick relief valve to release the high-pressure gas in the accumulator and pipeline as quickly as possible to ensure the safety of equipment and personnel.

[0075] Data interaction between the system's main controller and the nonlinear dynamic modeling unit is achieved through a shared memory region. The system's main controller writes control mode commands from the remote control terminal, target pressure curve parameters, and status flags from the safety monitoring module to a specific address in the shared memory.

[0076] The nonlinear dynamic modeling unit periodically reads these instructions and status information from the shared memory to determine its own operating mode, such as initial modeling, online updates, or model persistence.

[0077] Meanwhile, the nonlinear dynamic modeling unit writes its completed model parameters, model confidence index, and model readiness status flags into another area of ​​shared memory, which can then be read by the system's main controller and adaptive predictive control unit. This shared memory communication mechanism ensures low-latency transmission of critical data between core units.

[0078] The interface between the adaptive predictive control unit and the multimodal execution drive unit involves the conversion and output of control signals.

[0079] The optimal control quantity calculated by the adaptive predictive control unit is a digital quantity, which is first sent to the system's main controller.

[0080] The dedicated output management module within the system's main controller performs final arbitration on the control quantity based on the current system mode and safety status. In normal control mode, this digital control quantity is converted into an analog voltage signal of 0 to 10V via a 16-bit digital-to-analog converter. This analog voltage signal then passes through a voltage-to-current conversion circuit, transforming it into a current signal of 4 to 20mA to improve anti-interference capability, ultimately driving the pilot magnet of the electro-pneumatic servo proportional valve.

[0081] For the quick-release valve, the digital output port of the system main controller directly outputs a 24V DC switching signal. This signal is amplified by a solid-state relay to drive the electromagnet of the quick-release valve. All signal lines output to the actuator unit use shielded twisted-pair cables and are opto-isolated at the system main controller to suppress the negative impact of electromagnetic interference on the control system.

[0082] Power supply and signal integrity are crucial for high-precision sensor acquisition networks. All sensors in the network are powered by isolated DC regulated power supplies, typically 24V DC, with a power supply ripple factor of less than 1%. The analog signal output lines of the sensors use shielded cables, with the shield grounded at a single point on the system's main controller. The 24-bit analog-to-digital converter within the signal conditioning module features a programmable gain amplifier that automatically adjusts the amplification factor according to the sensor's range to fully utilize the converter's dynamic range.

[0083] The controller area network (CLAN) uses a bus topology with 120Ω terminating resistors at both ends to ensure signal transmission integrity and avoid reflections.

[0084] The system's main controller, acting as a node on the controller area network bus, polls the data frames of each sensor at a preset cycle through its built-in controller area network controller and performs cyclic redundancy checks to ensure correct data reception.

[0085] During the system initialization phase, after the remote control terminal is powered on, it first performs its own hardware and software self-test, including checking the memory, display screen, communication module, etc. After the self-test passes, the remote control terminal attempts to establish a secure connection with the system's main controller.

[0086] The connection establishment process includes identity authentication and key exchange to ensure the uniqueness and security of the connection. After the system main controller is powered on, it first loads the bootloader, and then starts the real-time operating system and main control program. The system main controller sequentially sends self-test commands to the high-precision sensor acquisition network, the nonlinear dynamic modeling unit, the adaptive predictive control unit, and the multimodal execution drive unit.

[0087] In the high-precision sensing acquisition network, each sensor performs zero-point and full-scale self-tests; the nonlinear dynamic modeling unit and the adaptive predictive control unit perform memory tests and algorithm initializations; the electro-pneumatic servo proportional valve in the multimodal execution drive unit performs a test stroke from fully closed to fully open and back to fully closed, while the quick-release valve performs a brief opening and closing action test.

[0088] Each unit returns a self-test result status word to the system master controller. Only when all units return a normal status word will the system master controller send a system ready signal to the remote control terminal, and the entire system enters standby mode.

[0089] After the operator sets and confirms the blasting parameters via the remote control terminal, the system enters the accumulator charging and initial modeling stage. The system's main controller parses the received parameters and sends key information such as the target pressure curve and blasting duration to the nonlinear dynamic modeling unit and the adaptive predictive control unit.

[0090] Subsequently, the system's main controller issues an energy storage command. The electro-pneumatic servo proportional valve in the multi-modal execution drive unit slowly opens to a preset small initial opening degree, such as five percent, and the high-pressure gas source begins to charge the accumulator.

[0091] Simultaneously, the nonlinear dynamic modeling unit is activated, and it begins receiving pressure and flow data provided by the high-precision sensor acquisition network during inflation. Based on this initial dynamic data, the nonlinear dynamic modeling unit uses its recursive least squares method to identify the initial model parameters of the system and construct an initial nonlinear dynamic model.

[0092] The adaptive predictive control unit also starts simultaneously. It first loads this initial model and then begins rolling optimization calculations.

[0093] During the inflation phase, the goal of adaptive predictive control is to ensure that the accumulator pressure rises smoothly and without overshoot to the starting pressure point of the target pressure curve. It finely regulates the inflation flow rate by adjusting the control signal output to the electro-pneumatic servo proportional valve, thus avoiding pressure surges.

[0094] Upon receiving the blasting execution command, the system enters the most critical stage of precise control of the blasting process. At this point, the adaptive predictive control unit enters a high-frequency control loop, typically set to a 1ms control period. Within each 1ms control cycle, the adaptive predictive control unit strictly follows a process of prediction, optimization, and feedback correction. The prediction stage utilizes the latest model provided by the nonlinear dynamic modeling unit to predict the pressure trajectory for the next 20ms. The optimization stage solves the optimization problem, calculating the optimal valve control quantity that makes the predicted pressure closest to the target pressure.

[0095] The feedback correction stage uses the current measured pressure to correct the prediction for the next cycle. The generated optimized control quantity is output to the electro-pneumatic servo proportional valve in real time, driving its valve core to precisely control the flow rate of high-pressure gas released from the accumulator, thereby ensuring that the measured pressure closely tracks the preset target pressure curve.

[0096] During this process, the nonlinear dynamic modeling unit does not stop working. It continues to run in the background, using the latest pressure, flow, and temperature data continuously provided by the high-precision sensor acquisition network to update its model parameters online once per second. This ensures that the dynamic model can accurately reflect the nonlinear time-varying characteristics of the current gas release process, such as the change in the adiabatic index caused by the decrease in gas temperature.

[0097] During precise blasting control, the safety monitoring and fault diagnosis module remains on high alert. It continuously analyzes the absolute pressure sensor readings at 1ms intervals, calculating the deviation between the instantaneous pressure value and the target value, i.e., the pressure overshoot. This module has a preset safety limit for pressure overshoot, such as 5% of the target pressure. The safety monitoring logic stipulates that if the detected pressure overshoot exceeds this safety limit, and this overshoot persists for more than three consecutive control cycles (3ms), the system's main controller's safety monitoring program will determine that the system faces an overpressure risk.

[0098] At this point, the safety monitoring program will immediately issue a highest-priority interrupt, forcibly overriding the current control signal output by the adaptive predictive control unit and directly taking over the control of the multimodal execution drive unit. The system main controller will first completely shut down the electro-pneumatic servo proportional valve, cutting off the main pneumatic circuit, and simultaneously forcibly activate the rapid relief valve, opening it fully to rapidly depressurize.

[0099] Opening the quick-release valve establishes a low-resistance release path, allowing high-pressure gas in the accumulator and pipelines to be rapidly discharged into the atmosphere, causing a rapid drop in system pressure. The safety monitoring module continuously monitors the pressure value until it returns to a preset safe pressure range, such as below 2% of the target pressure, at which point the emergency status is lifted and a fault event is reported to the host computer.

[0100] When the blasting duration reaches the time value set by the remote control terminal, or when the system pressure drops below a preset safety threshold due to gas release, the system main controller determines that the blasting process has officially ended. The system main controller then issues a serialized command to initiate the system reset procedure. First, the multi-modal execution drive unit is instructed to completely close the electro-pneumatic servo proportional valve. Next, based on the residual pressure in the accumulator, the system main controller decides whether to activate the rapid relief valve for complete depressurization.

[0101] 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.

[0102] 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. A remote control system for a high-pressure air blasting device, characterized in that, include: The remote control terminal establishes a two-way data connection with the system's main controller through an encrypted communication link. It is used to receive operation commands and send them to the system's main controller, while simultaneously displaying the system status and blasting process data in real time. The system's main controller, as the core of the system, is responsible for coordinating the work of each unit, parsing instructions from the remote control terminal, and overseeing the execution of control logic. The nonlinear dynamic modeling unit, connected to the system's main controller, is used to construct and update in real time a dynamic mathematical model characterizing the nonlinear coupling relationship between pressure, temperature, and flow rate during the high-pressure gas release process. The adaptive predictive control unit is connected to the nonlinear dynamic modeling unit and the system main controller, respectively. It performs forward rolling optimization calculations based on the dynamic mathematical model to generate the optimal control sequence for the multimodal execution drive unit. The multimodal execution drive unit receives control commands output by the adaptive predictive control unit and drives the electro-pneumatic servo proportional valve and the fast relief valve in the high-pressure gas circuit to work together to precisely regulate the release process of high-pressure gas in the accumulator. A high-precision sensor acquisition network is distributed and deployed at key nodes of the high-pressure gas path to monitor accumulator pressure, pipeline flow and valve inlet and outlet pressure difference in real time, and feed the acquired data back to the system main controller and nonlinear dynamic modeling unit.

2. The remote control system for a high-pressure air blasting device according to claim 1, characterized in that, The internal workflow of the nonlinear dynamic modeling unit is as follows: First, the unit receives real-time pressure sequence, flow sequence and temperature data from a high-precision sensor acquisition network; Secondly, based on the received data stream, the recursive least squares method with a forgetting factor is used to identify the model parameters describing the gas equation of state and flow resistance characteristics online. The model parameters include at least the equivalent value of the gas adiabatic index, the equivalent friction coefficient of the flow channel, and the volume effect time constant. Finally, using the updated model parameters, a nonlinear dynamic model expressed in the form of a system of differential-algebraic equations is constructed, which can predict the system pressure response within a specific future time window.

3. The remote control system for a high-pressure air blasting device according to claim 1, characterized in that, The operating mechanism of the adaptive predictive control unit comprises three core components: In the prediction phase, the current nonlinear dynamic model output by the nonlinear dynamic modeling unit is invoked, and the current system state is used as the initial condition to predict the system pressure trajectory in the future time domain. In the optimization phase, an objective function is set to minimize the cumulative deviation between the predicted pressure trajectory and the set pressure command, while constraining the rate of change of the control quantity to prevent actuator saturation. By solving the constrained optimization problem, a series of optimal control variables in the future control time domain are obtained; In the feedback correction stage, at the end of each control cycle, the measured system pressure from the high-precision sensor acquisition network is compared with the predicted value to generate a prediction error. This error is then used to compensate for subsequent predictions, thereby enhancing the control robustness under model mismatch or external disturbances.

4. The remote control system for a high-pressure air blasting device according to claim 1, characterized in that, The multimodal execution drive unit includes an electro-pneumatic servo proportional valve and a parallel fast relief valve; The electro-pneumatic servo proportional valve receives the analog voltage control signal output by the adaptive predictive control unit, and through its internal pilot stage and power amplifier stage, precisely adjusts the opening of the main valve core, thereby achieving continuous proportional control of the high-pressure gas flow. The coordination strategy between the electro-pneumatic servo proportional valve and the fast relief valve is determined by the system main controller based on the output of the adaptive predictive control unit and the safety logic.

5. The remote control system for a high-pressure air blasting device according to claim 1, characterized in that, The high-precision sensing and acquisition network specifically includes an absolute pressure sensor, a mass flow meter, and a differential pressure sensor; An absolute pressure sensor is installed at the accumulator outlet to directly monitor the pressure inside the accumulator; The mass flow meter is installed in the main gas line to accurately measure the instantaneous gas mass flow rate; The differential pressure sensor is connected across the inlet and outlet ports of the electro-pneumatic servo proportional valve to monitor the pressure difference across the valve core, providing data support for valve condition diagnosis and flow estimation. All sensor data is sampled and quantized by a 24-bit analog-to-digital converter and transmitted to the system main controller via the controller area network bus.

6. The remote control system for a high-pressure air blasting device according to claim 1, characterized in that, The system's main controller integrates a security monitoring and fault diagnosis module; This module continuously analyzes the data from the high-precision sensor acquisition network and the status feedback from the multimodal execution drive unit. By comparing the preset safety thresholds and logic rules, it diagnoses in real time whether the system has overpressure, leakage, or actuator jamming faults. Once a fault is detected, a tiered safety response is immediately triggered, including but not limited to switching to a safe mode, shutting off the main gas line, or activating the quick relief valve.

7. The remote control system for a high-pressure air blasting device according to claim 2, characterized in that, In the recursive least squares method with a forgetting factor, the forgetting factor is set to 0.

98. The purpose of this forgetting factor is to assign higher weights to new data, enabling the model to track the slow drift of system characteristics.

8. The remote control system for a high-pressure air blasting device according to claim 3, characterized in that, The optimization process involves solving a constrained optimization problem, specifically a constrained quadratic programming problem. The solution process for this quadratic programming problem is performed in real time within each control cycle, and typically only the first optimal control variable calculated for the future control time domain is applied to the controlled object.

9. A remote control system for a high-pressure air blasting device according to claim 4, characterized in that, The electro-pneumatic servo proportional valve also contains a linear variable differential transformer, which is used to detect the valve core displacement and form a local closed loop to ensure the accuracy of valve position control.

10. A remote control method for a high-pressure air blasting device, characterized in that, Control is achieved using the remote control system for the high-pressure air blasting device according to any one of claims 1-9, wherein the remote control method for the high-pressure air blasting device comprises the following steps: Step 1, System Initialization and Self-Test: The remote control terminal is powered on and establishes a secure connection with the system main controller; the system main controller sequentially performs power-on self-tests and communication tests on the high-precision sensor acquisition network, nonlinear dynamic modeling unit, adaptive predictive control unit, and multimodal execution drive unit. After confirming that each unit is in normal condition, it enters standby mode. Step 2, Blasting Parameter Setting and Model Construction: The operator inputs the target pressure curve, blasting duration, and safety parameters for this blasting operation through a remote control terminal; after receiving the parameters, the system main controller instructs the nonlinear dynamic modeling unit to begin initial modeling; The nonlinear dynamic modeling unit constructs an initial nonlinear dynamic model of the high-pressure gas system based on the initial steady-state data provided by the high-precision sensor acquisition network. Step 3, Adaptive Predictive Control Start-up: The system main controller issues an energy storage command, and the electro-pneumatic servo proportional valve in the multimodal execution drive unit slowly opens to charge the energy storage device; at the same time, the adaptive predictive control unit starts up and begins rolling optimization calculation based on the initial nonlinear dynamic model to generate a preliminary control sequence that acts on the electro-pneumatic servo proportional valve, so that the energy storage device pressure rises steadily to the target initial pressure. Step 4, Precise Control of the Blasting Process: After the system's main controller receives the blasting execution command from the remote control terminal, the adaptive predictive control unit enters a high-frequency control cycle. Within each control cycle, the unit executes prediction, optimization, and feedback correction processes, adjusting the control signal output to the electro-pneumatic servo proportional valve in real time to drive the valve's action. This aims to ensure that the pressure trajectory measured by the high-precision sensor acquisition network closely tracks the preset target pressure curve. Simultaneously, the nonlinear dynamic modeling unit continuously updates the model parameters online using the latest sensor data to ensure that the dynamic model accurately reflects the current gas release characteristics. Step 5, End of blasting and system reset: When the blasting duration reaches the set value or the pressure drops below the safety threshold, the system main controller determines that the blasting process has ended; then, the main controller instructs the multi-modal execution drive unit to close the electro-pneumatic servo proportional valve and activates the quick relief valve to release residual pressure as needed; the system status data is recorded and uploaded to the remote control terminal, and each control unit is reset, waiting for the next operation instruction.

Images