Dynamic environment error compensation system of HART protocol intelligent transmitter

Abstract

The invention relates to the technical field of environmental interference compensation, and discloses a dynamic environmental error compensation system of an HART protocol intelligent transmitter, which comprises a multi-sensor fusion module, a mechanical anti-pulsation structure, a dynamic compensation calculation core, an HART protocol enhancement module and a self-checking unit. The piezoelectric ceramic piece is combined with an LQR algorithm to generate a reverse vibration waveform in real time, and active counteracting is achieved. The dynamic compensation calculation core corrects temperature drift and vibration noise based on a BP neural network and a Kalman filtering model. The HART protocol enhancement module adopts quantum key distribution and Huffman coding compression to ensure safe transmission of data; and the reliability of the self-checking unit is improved through two-way ADC checking and step testing. In the harsh working condition of an oil refinery, the system pressure measurement error is smaller than or equal to 0.03% FS, the communication integrity rate is larger than or equal to 99.99%, and a high-precision and anti-interference solution is provided for industrial intelligent measurement.

Description

Technical Field

[0001] The invention relates to the technical field of environmental interference compensation, in particular to a dynamic environmental error compensation system for a HART protocol intelligent transmitter. Background Art

[0002] In industrial scenarios such as refinery catalytic cracking units and chemical process pumping stations, HART protocol smart transmitters need to operate for a long time in strong vibration (50-500Hz) and high-frequency pressure pulsation (100-500Hz) environments. The pressure pulsation is transmitted to the sensor sensitive diaphragm through the pipeline, causing the following problems: The natural frequency of the sensor of the traditional transmitter is usually 200-300Hz, which is close to the main frequency of pressure pulsation. The main frequency of pressure pulsation is usually 150Hz, which leads to distortion of the resonance amplification signal and the measurement error can reach ±0.2%FS. In addition, high-frequency pulsation is coupled with mechanical vibration, resulting in the superposition of velocity noise in the pressure signal. For example, 10G vibration will introduce ±0.1%FS error. Furthermore, the mechanical friction heat caused by pulsation can cause the local temperature to rise by 20°C, resulting in the drift of the sensor sensitivity coefficient.

[0003] The existing technology mainly alleviates the impact of pressure pulsation by using a homogeneous mass block and a linear spring to form an inertial filter system, but its effective frequency band is narrow at 80-300Hz, and the fixed spring stiffness causes the resonant frequency to drift. When the temperature increases by 10°C, the frequency shifts by 2Hz. The dynamic environment of the HART protocol smart transmitter affected by high-frequency pressure pulsation causes output data errors.

[0004] Therefore, we propose a dynamic environmental error compensation system for HART protocol smart transmitter to solve the problems in the above background. Summary of the invention

[0005] The present invention provides a dynamic environment error compensation system for a HART protocol intelligent transmitter, which can solve the problem that in the prior art, a homogeneous mass block and a linear spring are mainly used to form an inertial filter system to alleviate the influence of pressure pulsation, but the effective frequency band is narrow at 80-300 Hz, and the spring stiffness is fixed, resulting in a resonant frequency drift. When the temperature increases by 10°C, the frequency shifts by 2 Hz. The dynamic environment of the HART protocol intelligent transmitter subjected to high-frequency pressure pulsation causes output data errors.

[0006] In order to solve the above technical problems, the present invention provides the following technical solutions: The dynamic environmental error compensation system of the HART protocol intelligent transmitter includes a multi-sensor fusion module, a mechanical anti-pulsation structure, a dynamic compensation calculation core, a HART protocol enhancement module and a self-checking unit; the mechanical anti-pulsation structure includes a pneumatic damper, a mass and spring inertia filter system and an anti-impact packaging shell; the pneumatic damper is installed at the front end of the multi-sensor fusion module housing, adopts a diaphragm cavity design, and the cavity is filled with high-viscosity silicone oil; The impact-resistant package shell has a micro piezoelectric ceramic unit integrated inside, which detects the pressure pulsation signal in real time, generates a reverse vibration waveform through the inverse piezoelectric effect, and realizes closed-loop cancellation of mechanical vibration in combination with the LQR optimal control algorithm; The dynamic compensation calculation core is composed of a microprocessor (MCU) and an FPGA hardware accelerator. The microprocessor (MCU) and the FPGA hardware accelerator work together to realize dynamic nonlinear compensation based on a hybrid model of BP neural network and Kalman filtering.

[0007] The dynamic compensation calculation core realizes dynamic compensation based on the following nonlinear coupling model: ; in, is the pressure sensitivity coefficient, is the neural network weight, is a nonlinear function of temperature-motion-time, is the dynamic acceleration compensation factor.

[0008] The HART protocol enhancement module supports 1.2Mbps full-duplex communication, sends compensation parameters in real time and transmits verification data back through a custom instruction set.

[0009] Preferably, the pneumatic damper adopts a diaphragm-type cavity, which is installed at the front end of the multi-sensor fusion module housing, and its cavity inlet is connected to the measured medium pipeline flange, and its cavity outlet is connected to the sensitive diaphragm of the pressure sensor through a rigid connecting rod; the cavity is filled with high-viscosity silicone oil to form a viscous shear energy dissipation layer, and the silicone oil circulates between the top and bottom of the cavity through the microchannel, and uses viscous resistance to attenuate high-frequency pressure pulsations; The mass and spring inertia filter system suppresses high-frequency vibration transmission through a silicone spring composite structure. The mass and spring inertia filter system consists of a stainless steel mass block and a disc spring with adjustable preload. A silicone damping layer covers the spring surface, and the vibration transmission rate is ≤5% in the 50-500Hz frequency band. The ceramic base material of the impact-resistant packaging shell is silicon nitride (Si 3 N 4 ), the surface is plated with nickel-copper alloy layer; The impact-resistant packaging shell adopts a ceramic-based composite material and a stainless steel isolation diaphragm to reduce mechanical vibration coupling.

[0010] Compared with the prior art, the beneficial effects achieved by the present invention are: The system of the present invention significantly improves the measurement accuracy and reliability in harsh industrial environments through a multi-level collaborative anti-interference architecture and intelligent dynamic compensation technology. The physical-level anti-interference design of the pneumatic damper and the piezoelectric ceramic sheet is combined with a hybrid compensation model of the BP neural network and the Kalman filter to achieve full-link error elimination from mechanical suppression to digital correction.

[0011] Compared with traditional solutions, traditional transmitters only rely on single hardware filtering or software compensation. This solution solves the coupling problem of high-frequency pressure pulsation and temperature drift by coordinating piezoelectric active cancellation with a response time of less than 1ms and dynamic algorithms.

[0012] The pneumatic damper and the mass and spring system inertial filter system (vibration transmission rate ≤5% in the 50-500Hz frequency band) work together. The pneumatic damper achieves a 100-500Hz pressure pulsation attenuation of ≥20dB. The mass and spring system inertial filter system reduces the vibration transmission rate in the 50-500Hz frequency band to ≤5%, expanding the effective suppression frequency band from 80-300Hz of the traditional solution to 50-500Hz, increasing the vibration energy attenuation efficiency by 66%, and completely eliminating the signal distortion caused by high-frequency resonance. Secondly, based on the hybrid model of BP neural network and Kalman filter, the system corrects temperature drift and vibration coupling noise in real time, reducing the sensitivity drift from ±0.1%FS to ±0.02%FS when the temperature rises by 20℃, and reducing the error introduced by 10G vibration from ±0.1%FS to ±0.03%FS, and improving the comprehensive pressure measurement accuracy to ±0.05%FS. At the same time, the flexible piezoelectric ceramic sheet is combined with the LQR optimal control algorithm to generate a multi-band reverse vibration waveform, with the main frequency randomly changing between 50-150Hz, achieving a stable vibration suppression efficiency of ≥30dB in the inverter interference scenario.

[0013] On the other hand, in terms of data transmission, the combination of quantum key distribution and Huffman coding compression technology has increased the data transmission integrity rate from 85% to 99.99% in a strong electromagnetic interference environment (30V / m), and the communication delay is ≤1ms, meeting the highest safety standards of industrial control systems IEC 62443-3-3. In addition, the self-checking unit passes the dual-channel ADC cross-check and step response test, and is triggered every 5 minutes. The deviation exceeds ±5% to trigger the HART alarm code 0xE2, thereby reducing the sensor failure rate by 90%, extending the system mean time between failures (MTBF) from 50,000 hours to 100,000 hours, and extending the maintenance cycle from 3 months to 2 years. The HART alarm code 0xE2 represents the alarm signal of the sensor signal deviation exceeding ±5% in this system.

[0014] In a refinery catalytic cracking unit with a pressure pulsation of 200Hz, a vibration of 8G, and a temperature of -40~120℃, the system stabilizes the pressure measurement error at ±0.03%FS, the communication packet loss rate is ≤0.01%, and it can adapt to environmental changes without human intervention, providing a high-precision and high-reliability solution for industrial process control. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 This is the working flow chart of the dynamic environmental error compensation system of the HART protocol smart transmitter. DETAILED DESCRIPTION

[0016] The specific embodiments of the present invention are described in detail below, but it should be understood that the protection scope of the present invention is not limited by the specific embodiments.

[0017] Embodiment 1: See also Figure 1 The present invention provides a technical solution: a dynamic environmental error compensation system for a HART protocol intelligent transmitter, including a multi-sensor fusion module, a mechanical anti-pulsation structure, a dynamic compensation calculation core, a HART protocol enhancement module and a self-checking unit; the mechanical anti-pulsation structure includes a pneumatic damper, a mass and spring inertia filter system and an anti-impact packaging shell; the pneumatic damper is installed at the front end of the multi-sensor fusion module housing, adopts a diaphragm cavity design, and the cavity is filled with high-viscosity silicone oil; The pneumatic damper adopts a circulation design of a diaphragm cavity and a high-viscosity silicone oil microchannel, directly connecting the measured pipeline and the sensor diaphragm to form the first-level physical filtering of the pressure pulsation. The silicone oil circulates between the top and bottom of the cavity through the microchannel, generating viscous shear resistance to the 100-500Hz high-frequency pulsation, achieving ≥20dB attenuation; the rigid connecting rod transmits the attenuated pressure to the sensor, avoiding the hysteresis effect of the traditional rubber diaphragm and shortening the response time by 40%.

[0018] The mass and spring inertia filter system achieves broadband vibration suppression through gradient density mass blocks, preload adjustable springs and composite damping layers. The stepped stainless steel cylinder of the gradient density mass block is filled with tungsten powder, forming a gradient distribution with a center density of >15g / cm³ and an edge density of 8g / cm³, so that the system presents multiple resonance points in the 50-500Hz frequency band, effectively suppressing the bandwidth to twice that of the traditional solution. The Nitinol disc spring uses a micro-stepping motor to adjust the preload. At a high temperature of 80°C, the spring stiffness is adaptively increased by 20%-50%, offsetting the resonant frequency shift caused by temperature, and reducing the frequency drift from 2Hz to 0.5Hz per 10°C. Using the composite damping method of silicone and carbon fiber, the alternately stacked silicone and carbon fiber sheets are enhanced in bonding strength through a laser microhole array. The loss factor in the frequency band of 50-500Hz is ≥0.6, and the vibration transmission rate is ≤5%.

[0019] The impact-resistant package shell adopts a combination of silicon nitride ceramic substrate, gradient nickel-copper plating and vacuum brazing isolation diaphragm. The ceramic substrate resists external impact through its own tensile strength greater than ≥800MPa, and the gradient plating controls the thermal expansion coefficient to ≤3×10 -6 / ℃, to avoid temperature stress cracking; the stainless steel isolation diaphragm and the ceramic substrate are brazed by Ag-Cu-Ti alloy vacuum, and the air tightness of the weld is ≤1×10 -9 Pa·m³ / s, thus blocking the vibration transmission path.

[0020] The impact-resistant package shell integrates a micro piezoelectric ceramic unit, which detects the pressure pulsation signal in real time and generates a reverse vibration waveform through the inverse piezoelectric effect. Combined with the LQR optimal control algorithm, it realizes closed-loop cancellation of mechanical vibration. The dynamic compensation calculation core is composed of a microprocessor (MCU) and an FPGA hardware accelerator. The microprocessor (MCU) and the FPGA hardware accelerator work together to achieve dynamic nonlinear compensation based on a hybrid model of BP neural network and Kalman filter.

[0021] The HART protocol enhancement module supports 1.2Mbps full-duplex communication, sends compensation parameters in real time and transmits verification data back through a custom instruction set. The self-checking unit ensures data integrity through dual-channel ADC cross-checking and CRC-16 redundancy check mechanism; The self-checking unit performs a sensor step response test every 5 minutes. If the response time deviation exceeds ±5%, the HART alarm code 0xE2 is triggered. The HART protocol enhancement module uses Huffman coding to compress the transmitted data with a compression rate of ≥50%, and realizes zero-interrupt communication with the microprocessor through the DMA channel, with a communication delay of ≤1ms.

[0022] The self-checking unit ensures system reliability through dual-channel ADC cross-checking, CRC-16 redundant check mechanism and step test. It applies a 0 to 80% range step pressure to the sensor every 5 minutes to detect the response time deviation and triggers a 0xE2 alarm if it exceeds plus or minus 5%. The CRC-16 check combined with dual-channel sampling reduces the data error rate from 1.5% to 0.01%.

[0023] The dynamic compensation calculation core realizes dynamic compensation based on the following nonlinear coupling model: ; in, is the pressure sensitivity coefficient, is the neural network weight, is a nonlinear function of temperature-motion-time, is the dynamic acceleration compensation factor.

[0024] The nonlinear coupling model realizes dynamic environmental error correction through a multi-physical field coupling compensation mechanism. Its specific working principle is as follows: the pressure sensitivity coefficient Kp reflects the basic characteristics of the sensor and produces time-varying deviations due to the influence of temperature drift; the neural network weight W is characterized by the function f(T, t) through the BP neural network to dynamically learn the nonlinear coupling relationship between temperature T and time t, and corrects the sensitivity drift caused by temperature in real time. For example, when the temperature rises by 20°C and causes Kp to drop by 0.5%, W·f(T, t) generates a +0.5% compensation amount; the dynamic acceleration compensation factor α is based on the vibration acceleration a separated by the Kalman filter, which quantifies the coupling interference of mechanical vibration on the pressure signal. For example, the ±0.1% FS error introduced by 10G vibration is offset by the α·a term.

[0025] The calculation process of this formula is: (1) Real-time acquisition of temperature T, vibration acceleration a and raw pressure signal Praw; (2) The BP neural network is trained based on historical data to obtain the weight W and calculate the temperature-time coupling term W·f(T,t); (3) Kalman filter estimates the true pressure value and solves the acceleration compensation factor α; (4) The final compensation pressure Pcomp = Kp × [1 + W·f(T,t)] × Praw + α·a, which eliminates temperature drift and vibration noise simultaneously through nonlinear superposition.

[0026] In the catalytic cracking unit, this model reduces the pressure measurement error from ±0.15%FS of traditional linear compensation to ±0.03%FS, and improves the temperature drift suppression efficiency by 80%.

[0027] In the dynamic compensation algorithm, the hybrid model construction method of BP neural network and Kalman filter includes: 1. Dynamic error modeling, generating temperature-pressure-vibration multi-field coupling data based on the digital twin platform and establishing state space equations: ; in, is the sensor state vector, u is the environmental disturbance input, w and v are the process noise and observation noise; 2. Sliding time window frequency domain decoupling, real-time sampling of the pressure signal at 1kHz, extraction of the main frequency component through FFT, and separation of temperature drift and vibration noise using the time window function; 3. Adaptive threshold detection, dynamically adjust the filter threshold according to environmental parameters, the threshold calculation formula is: Threshold=β·|△T|+γ·|△V|; Among them, β and γ are the sensitivity coefficients of temperature and vibration, which are updated through online learning.

[0028] The pneumatic damper adopts a diaphragm cavity, which is installed at the front end of the multi-sensor fusion module housing. Its cavity inlet is connected to the measured medium pipeline flange, and its cavity outlet is connected to the sensitive diaphragm of the pressure sensor through a rigid connecting rod; the cavity is filled with high-viscosity silicone oil to form a viscous shear energy dissipation layer. The silicone oil circulates between the top and bottom of the cavity through the microchannel, and uses viscous resistance to attenuate high-frequency pressure pulsations; The mass and spring inertia filter system suppresses high-frequency vibration transmission through a silicone spring composite structure. The mass and spring inertia filter system consists of a stainless steel mass block and a disc spring with adjustable preload. The silicone damping layer covers the spring surface. The vibration transmission rate is ≤5% in the 50-500Hz frequency band. The ceramic base material of the impact-resistant packaging shell is silicon nitride (Si 3 N 4 ), the surface is plated with nickel-copper alloy layer; The impact-resistant package shell adopts ceramic matrix composite material and stainless steel isolation diaphragm to reduce mechanical vibration coupling.

[0029] In a specific implementation of the above scheme, the pneumatic damper adopts a diaphragm-type cavity directly installed on the front end of the multi-sensor fusion module housing, the cavity inlet is connected to the measured medium pipeline flange, and the cavity outlet is connected to the sensitive diaphragm of the pressure sensor through a rigid connecting rod; the cavity is filled with high-viscosity silicone oil to form a viscous shear energy dissipation layer, and the silicone oil circulates between the top and bottom of the cavity through the microchannel, using viscous resistance to attenuate high-frequency pressure pulsations; The mass and spring inertial filter system includes an integrated design of a gradient density stainless steel mass block. The gradient density stainless steel mass block adopts a stepped cylindrical structure and is filled with tungsten powder to form a gradient distribution with high density in the center and low density at the edge. It is suspended on the top plate inside the impact-resistant packaging shell through a disc spring with adjustable preload, and the upper flange of the spring is fixed to the top plate of the shell by laser welding; the disc spring group with adjustable preload is made of Nitinol alloy (Nitinol), and the dynamic adjustment of the stiffness of 50-200N / mm is achieved through a micro stepper motor. The mass and spring system fixes the upper flange of the spring to the top plate of the impact-resistant packaging shell by laser welding to form a stable suspension structure. The 120° symmetrical distribution design makes the center of gravity of the mass block coincide with the geometric center of the shell, eliminating the installation eccentric torque; the silicone and carbon fiber composite damping layer covers the outer surface of the spring, the silicone layer directly contacts the metal surface of the spring, and the carbon fiber reinforced polymer sheet covers the outer side of the silicone, and the interface bonding force is enhanced by a laser micropore array; the pressure sensor is embedded in the central cavity at the lower end of the mass block, and the probe is connected through an M3 thread, and the probe is in rigid contact with the stainless steel isolation diaphragm; The structure of the impact-resistant packaging shell is optimized. The shell substrate is made of reaction-sintered silicon nitride ceramic, and the surface is coated with a gradient nickel-copper alloy layer. The nickel layer and the copper layer are deposited alternately to match the thermal expansion coefficient of the ceramic-metal interface; the stainless steel isolation diaphragm is sealed to the ceramic substrate by a vacuum brazing process, and the weld surrounds the sensor interface area to form a double sealing barrier; a quantum channel window is opened on the side wall of the shell, and the window material is anti-reflection coated sapphire, which is arranged orthogonally to the piezoelectric ceramic sheet to prevent the photon transmission path from being disturbed by mechanical vibration.

[0030] Furthermore, the closed-loop cancellation of vibration energy is achieved through piezoelectric ceramics and LQR control: The flexible piezoelectric array uses multi-layer PZT-5H piezoelectric sheets manufactured by MEMS technology, and each layer is 20μm thick and arranged in a matrix on the inner wall of the shell to detect vibration signals in real time and generate reverse waveforms. Multi-band LQR control: First, a state space model is constructed based on the vibration displacement x and velocity v, and the performance index J=∫(x²Q+v²R)dt. The optimal control force F=-Kx is solved by the Riccati equation; Then, independent controllers are designed for the 100 Hz main frequency and its 200 Hz or 300 Hz harmonics, and multi-band cancellation signals are superimposed and output. Finally, the weight matrices Q and R are dynamically adjusted to compensate for temperature drift, so that the Q value is adjusted by 0.5% for every 1°C change in temperature, and the error is further controlled to be ≤0.1%.

[0031] Furthermore, the nonlinear error is corrected by a hybrid model of BP neural network and Kalman filter. The input layer of the BP neural network is pressure, temperature, and vibration acceleration, the hidden layer has 8 nodes, and the output layer generates weights W with a learning rate of 0.01, which is used to fit the nonlinear relationship between pressure and temperature. In conjunction with the Kalman filter scheme, its state equation contains a dynamic acceleration compensation factor α(t), which separates the real pressure signal from the vibration noise in real time. The specific compensation formula is ΔP=W·f(T,t)+α(t)·a, where a is the vibration acceleration, which ultimately reduces the pressure error from ±0.2%FS to ±0.05%FS.

[0032] In another embodiment, signal distortion is reduced by utilizing quantum encrypted HART protocol and closed loop self-checking. The HART protocol enhancement module uses quantum key distribution, Huffman compression, and DMA transmission to ensure data security and real-time performance. The quantum-encrypted polarization-coded photons distribute dynamic keys through a sapphire window, and each frame embeds an 8-bit quantum error correction code. The data integrity rate is ≥99.9% under strong interference of 30V / m. Efficient transmission is achieved, with a Huffman coding compression rate of ≥50%. The DMA channel achieves 1.2Mbps full-duplex communication with a delay of ≤1ms, and supports real-time delivery of LQR parameters.

[0033] The effect of the above scheme is that the full-band pressure pulsation of industrial scenarios is covered by the coordination of the pneumatic damper and the mass and spring system; Compared with traditional solutions, the effective suppression bandwidth is extended from 80-300Hz to 50-500Hz, and the vibration energy attenuation efficiency is increased by 66%. In addition, the dynamic adaptability is enhanced, and the stiffness adjustment range of the preload adjustable spring is 50-200N / mm. It is combined with the temperature adaptive LQR algorithm to improve the resonant frequency stability of the system by 4 times under -40~120℃ working conditions; and the error introduced by the acceleration noise is reduced from ±0.1%FS to ±0.02%FS through the multi-band control of the ceramic piezoelectric array.

[0034] And the security and reliability have been broken through, using quantum encryption to make the key cracking difficulty reach 2¹² 8 The communication anti-interference capability is improved by 100 times, and the bit error rate is reduced from 10 -3 Down to 10 -9 The self-checking mechanism reduces the sensor fault detection response time from 30 minutes to 5 minutes, and increases the mean time between failures from 50,000 hours to 100,000 hours.

[0035] This solution achieves mechanical filtering through pneumatic dampers, mass and spring inertia filtering systems and impact-resistant packaging shells, and realizes active cancellation of vibration energy by combining micro piezoelectric ceramic units with LQR optimal control algorithms. It then uses neural networks and Kalman filter models to achieve intelligent compensation, as well as the four-level collaborative mechanism of quantum encryption transmission. It overcomes the dynamic error problem of industrial transmitters in high-voltage pulsating environments and improves the measurement accuracy to 0.05%FS. It also achieves wideband suppression, anti-interference communication and long-term reliable operation, meeting the intelligent measurement needs of harsh scenarios such as refining and chemical industry.

[0036] The multi-sensor fusion module synchronously collects signals through integrated packaged pressure sensors, temperature sensors and vibration sensors. The natural frequency of the sensor sensitive element is designed to be more than 3 times the pressure pulsation frequency to avoid the resonance effect. The signal conditioning circuit is directly integrated with the sensor to shorten the transmission path.

[0037] This solution achieves multi-sensor collaborative anti-interference through integrated packaging design and natural frequency optimization. The sensitive elements of pressure, temperature and vibration sensors are packaged on the same base in a three-dimensional stacked structure. The resonance risk is completely avoided by the principle of tripling the natural frequency, that is, the natural frequency of the sensor is greater than 3 times the main frequency of the pressure pulsation, for example, a 150Hz pulsation corresponds to 450Hz of the sensor. At the same time, the signal conditioning circuit is directly integrated into the sensor pin, and the transmission path is shortened to ≤5mm, eliminating the electromagnetic interference introduced by long wires, and the signal-to-noise ratio is improved by 12dB to 72dB. In combination with synchronous clock triggering, strict alignment of multi-physical quantity data is achieved, and the time deviation is ≤10ns. Ultimately, in the high-frequency pulsation scenario of the refinery, the pressure measurement error is reduced from ±0.15%FS to ±0.03%FS, the temperature drift suppression efficiency is improved by 80%, and the system failure rate is reduced by 90%.

[0038] In another embodiment, the micro piezoelectric ceramic unit includes a multi-layer flexible piezoelectric array, an adaptive control module and a drive circuit; the multi-layer flexible piezoelectric array is composed of a multi-layer stacked piezoelectric ceramic sheet and a flexible substrate, which are arranged in a matrix on the inner wall of the transmitter housing, and are used to detect mechanical vibration signals in real time and generate reverse vibration waveforms through the inverse piezoelectric effect; wherein the piezoelectric ceramic sheet is a multi-layer PZT-5H piezoelectric ceramic sheet manufactured by MEMS process; The adaptive control module is built based on the LQR optimal control algorithm. It establishes a state space model with vibration displacement and velocity as state variables, calculates the optimal control force by minimizing performance indicators, and dynamically adjusts the weight matrix to compensate for temperature drift. The drive circuit works with the HART protocol enhancement module to convert the control force into a drive signal, driving the piezoelectric array to produce offset vibrations with opposite phases to achieve closed-loop vibration suppression; the state feedback matrix of the LQR optimal control algorithm is dynamically adjusted through the real-time pressure pulsation spectrum characteristics, and the control error is ≤0.1%; The piezoelectric ceramic piece detects the mechanical vibration signal in real time and outputs the charge signal, which is converted into a voltage signal by the charge amplifier and then input into the signal processing module; the signal processing module extracts the spectral characteristics of the main vibration frequency and its integer multiple harmonic components through fast Fourier transform (FFT); the LQR control module constructs a state space model based on the vibration displacement and velocity, calculates the optimal control force with the goal of minimizing the performance index, and compensates for the temperature drift effect of the piezoelectric material by adaptively adjusting the weight matrix; the drive circuit generates a multi-band coordinated PWM drive signal according to the optimal control force, drives the piezoelectric ceramic piece to produce a vibration waveform with opposite phase, and realizes the closed-loop cancellation of vibration energy.

[0039] The adaptive control module designs independent LQR controllers for the main vibration frequency and its harmonic components, and outputs multi-band collaborative cancellation signals after parallel calculation. The HART protocol enhancement module customizes the instruction set, including vibration spectrum data uploading instructions and LQR parameter sending instructions, to support remote optimization and real-time monitoring of vibration suppression strategies.

[0040] This embodiment achieves precise vibration suppression through a flexible piezoelectric array and LQR adaptive control. The micro piezoelectric ceramic unit is integrated with a multi-layer PZT-5H piezoelectric sheet and a flexible substrate, and is arranged in a matrix on the inner wall of the transmitter housing to detect vibration signals in real time and output the charge. The signal processing module extracts the main vibration frequency and its harmonics through FFT, constructs an LQR state space model with displacement and velocity as state variables, and dynamically calculates the optimal control force. The drive circuit generates a multi-band PWM signal to drive the piezoelectric array to generate a reverse vibration waveform, and eliminates vibration energy through phase cancellation and amplitude matching.

[0041] This innovative design implements temperature adaptive compensation and dynamically adjusts the LQR weight matrix according to the temperature sensitivity coefficient of the piezoelectric material to offset the impact of temperature drift. It uses a multi-band collaborative approach to suppress vibrations, independently controls the main frequency and harmonics, and extends the vibration suppression bandwidth to three times the fundamental frequency, from 100Hz to 300Hz. It uses flexible substrate adaptation to allow the piezoelectric array to fit the curved housing, increasing detection sensitivity by 30%.

[0042] Therefore, under 10G vibration shock, the vibration energy attenuation is ≥30dB, the pressure measurement error is ≤0.05%FS, and the control deviation caused by temperature drift during -40~120℃ is reduced from ±0.15% to ±0.03%.

[0043] In another embodiment, a quantum key distribution unit is embedded in the HART protocol enhancement module, and the quantum key distribution unit includes an optical coupling interface, a quantum channel window, and a polarization-coded photon transmitter; The quantum key distribution unit is physically connected to the HART protocol enhancement module through an optical coupling interface; The quantum channel window is set on the side wall of the impact-resistant packaging shell, and the window material is anti-reflection coated sapphire; The polarization-coded photon transmitter is integrated into the multi-sensor fusion module and shares the power supply bus with the pressure sensor and vibration sensor. The photon transmission path is arranged orthogonally to the sensor signal line to avoid crosstalk. The quantum error correction code is embedded in the HART protocol data stream with 8 bits of redundancy per frame, and is transmitted synchronously with the dynamic compensation parameters through time division multiplexing. The communication delay is ≤1ms; the quantum key distribution unit and the HART protocol enhancement module share the DMA channel, the key update cycle is ≤1 minute, and the bit error rate is ≤10 -9 ; In a strong electromagnetic interference environment, the quantum error correction code recovers lost data through redundant photon states to ensure the integrity of the dynamic compensation parameters; the recovered data triggers the LQR control module to re-optimize the vibration cancellation strategy of the piezoelectric array, with a control error of ≤0.1%.

[0044] In the above scheme, this embodiment realizes high-security dynamic compensation through quantum encryption communication and adaptive error correction. The quantum key distribution unit (QKD) uses polarization-coded photons with a wavelength of 1550nm to negotiate keys with the receiving end through an anti-reflection sapphire window, and updates it every 60 seconds to generate a dynamic encryption key; the compensation parameters are encrypted by the quantum error correction code and transmitted through the DMA channel of the HART protocol. When encountering strong electromagnetic interference of 30V / m, the lost data is restored by redundant photon states; the recovered spectrum data triggers the LQR module to recalculate the optimal control force, for example, by adjusting the weight matrix Q by ±5%, driving the piezoelectric array to generate precise reverse vibration, and its phase error is ≤0.1°; the effect of the above scheme is that the difficulty of quantum key cracking is 2¹² 8 The data integrity rate is increased from 85% to 99.99% in the strong electromagnetic environment of the refinery, and the bit error rate is ≤10 -9 ; and the delay introduced by quantum encryption is only 0.8ms, which is 275% higher than the traditional 3ms AES encryption, ensuring the real-time vibration suppression, and the control cycle is within 10ms; after data recovery, the LQR parameters are readjusted to restore the vibration suppression efficiency to the optimal level within 1ms, and the pressure measurement error is stabilized at ±0.03%FS.

[0045] The dynamic environmental error compensation system of the HART protocol intelligent transmitter includes the following steps: S1. The pressure pulsation signal, ambient temperature signal and mechanical vibration signal are collected synchronously through the integrated packaged pressure sensor, temperature sensor and vibration sensor, and the natural frequency of the sensor sensitive element is designed to be more than three times the main frequency of the pressure pulsation. The signal conditioning circuit is directly integrated into the sensor pin to shorten the transmission path; S2. The pneumatic damper initially attenuates high-frequency pressure pulsations through the viscous silicone oil in the diaphragm cavity, while the mass and spring inertia filter system suppresses the transmission of mechanical vibrations to sensitive components through the synergistic effect of the gradient density mass block and the preload adjustable spring; S3. The piezoelectric ceramic piece integrated in the inner wall of the impact-resistant package shell detects the residual vibration signal in real time, converts it into a voltage signal through a charge amplifier, and then uses a fast Fourier transform to extract the spectral characteristics of the main vibration frequency and its harmonic components by the signal processing module; S4. The dynamic compensation calculation core is based on a hybrid model of BP neural network and Kalman filter, combined with the nonlinear coupling relationship between temperature and time and the dynamic acceleration compensation factor, to generate nonlinear compensation parameters to correct the pressure measurement error; The S5.LQR control module constructs a state space model based on vibration displacement and velocity, calculates the optimal control force by minimizing the performance index, and drives the piezoelectric array to generate a reverse waveform opposite to the detected vibration phase, thus achieving closed-loop cancellation of vibration energy; S6. The HART protocol enhancement module compresses the compensation parameters and vibration spectrum data through Huffman coding, embeds the quantum error correction code, encrypts the data through the dynamic key distributed by polarization-coded photons, and transmits the data to the control terminal through the quantum channel of the anti-reflection sapphire window; S7. When data is lost due to strong electromagnetic interference during transmission, the quantum error correction code recovers the lost frame through redundant photon states and feeds the recovered data back to the LQR control module to re-optimize the vibration cancellation strategy; S8. dynamically adjusting the LQR weight matrix and the neural network model parameters according to the recovered spectrum data and the real-time temperature change, and updating the driving signal of the piezoelectric array to achieve adaptive vibration suppression; S9. The self-checking unit periodically performs the sensor step response test and verifies the data integrity through the dual-channel ADC cross-check and CRC-16 redundancy check mechanism. If the response time exceeds the tolerance or the signal is abnormal, the HART protocol alarm code is triggered; S10. The vibration sensor monitors the compensated vibration energy in real time. If the suppression efficiency does not reach the threshold, the compensation process is restarted. The final pressure measurement value is output through the HART protocol after redundancy verification, completing the dynamic environmental error compensation closed loop.

[0046] The above disclosures are only several specific embodiments of the present invention. However, the embodiments of the present invention are not limited thereto. Any changes that can be conceived by those skilled in the art should fall within the protection scope of the present invention.

Claims

1. Dynamic environmental error compensation system for HART protocol intelligent transmitter, characterized in that: include: Multi-sensor fusion module, mechanical anti-pulsation structure, dynamic compensation calculation core, HART protocol enhancement module and self-checking unit; The mechanical anti-pulsation structure includes a pneumatic damper, a mass and spring inertia filtering system and an impact-resistant packaging shell; The pneumatic damper is installed at the front end of the multi-sensor fusion module housing, adopts a diaphragm cavity design, and the cavity is filled with high-viscosity silicone oil; The impact-resistant package shell has a micro piezoelectric ceramic unit integrated inside, which detects the pressure pulsation signal in real time, generates a reverse vibration waveform through the inverse piezoelectric effect, and realizes closed-loop cancellation of mechanical vibration in combination with the LQR optimal control algorithm; The dynamic compensation calculation core is composed of a microprocessor and an FPGA hardware accelerator. The microprocessor and the FPGA hardware accelerator operate in coordination and realize dynamic nonlinear compensation based on a hybrid model of BP neural network and Kalman filtering.

2. The dynamic environmental error compensation system of the HART protocol intelligent transmitter according to claim 1, characterized in that: The HART protocol enhancement module supports 1.2Mbps full-duplex communication, sends compensation parameters in real time and transmits verification data back through a custom instruction set.

3. The dynamic environmental error compensation system of the HART protocol intelligent transmitter according to claim 1, characterized in that: The self-checking unit ensures data integrity through dual-channel ADC cross-checking and CRC-16 redundancy check mechanism; The self-checking unit performs a sensor step response test every 5 minutes. If the response time deviation exceeds ±5%, the HART alarm code 0xE2 is triggered; the HART alarm code 0xE2 represents a sensor signal deviation exceeding the limit in this system; The HART protocol enhancement module adopts Huffman coding to compress transmission data, with a compression rate of ≥50%, and realizes zero-interruption communication of the microprocessor through the DMA channel, with a communication delay of ≤1ms.

4. The dynamic environmental error compensation system of the HART protocol intelligent transmitter according to claim 1, characterized in that: The dynamic compensation calculation core realizes dynamic compensation based on the following nonlinear coupling model: ； in, is the pressure sensitivity coefficient, is the neural network weight, is a nonlinear function of temperature-motion-time, is the dynamic acceleration compensation factor.

5. The dynamic environmental error compensation system of the HART protocol intelligent transmitter according to claim 1, characterized in that: The pneumatic damper adopts a diaphragm-type cavity, which is installed at the front end of the multi-sensor fusion module housing, and its cavity inlet is connected to the measured medium pipeline flange, and its cavity outlet is connected to the sensitive diaphragm of the pressure sensor through a rigid connecting rod; the cavity is filled with high-viscosity silicone oil to form a viscous shear energy dissipation layer, and the silicone oil circulates between the top and bottom of the cavity through the microchannel, using viscous resistance to attenuate high-frequency pressure pulsations; The mass and spring inertia filter system suppresses high-frequency vibration transmission through a silicone spring composite structure. The mass and spring inertia filter system consists of a stainless steel mass block and a disc spring with adjustable preload. A silicone damping layer covers the spring surface, and the vibration transmission rate is ≤5% in the 50-500Hz frequency band. The ceramic base material of the impact-resistant packaging shell is silicon nitride, and the surface is plated with a nickel-copper alloy layer; The impact-resistant packaging shell adopts a ceramic-based composite material and a stainless steel isolation diaphragm to reduce mechanical vibration coupling.

6. The dynamic environmental error compensation system of the HART protocol intelligent transmitter according to claim 1, characterized in that: The multi-sensor fusion module synchronously collects signals through integrated packaged pressure sensors, temperature sensors and vibration sensors. The natural frequency of the sensor sensitive element is designed to be more than 3 times the pressure pulsation frequency to avoid resonance effects. The signal conditioning circuit is directly integrated with the sensor to shorten the transmission path.

7. The dynamic environmental error compensation system of the HART protocol intelligent transmitter according to claim 6, characterized in that: The micro piezoelectric ceramic unit includes a multi-layer flexible piezoelectric array, an adaptive control module and a drive circuit; the multi-layer flexible piezoelectric array is composed of a multi-layer stacked piezoelectric ceramic sheet and a flexible substrate, which are arranged in a matrix on the inner wall of the transmitter housing, and are used to detect mechanical vibration signals in real time and generate reverse vibration waveforms through the inverse piezoelectric effect; the piezoelectric ceramic sheet is a multi-layer PZT-5H piezoelectric ceramic sheet manufactured by MEMS technology; The adaptive control module is constructed based on the LQR optimal control algorithm, establishes a state space model with vibration displacement and velocity as state variables, calculates the optimal control force by minimizing the performance index, and dynamically adjusts the weight matrix to compensate for temperature drift; The drive circuit cooperates with the HART protocol enhancement module to convert the control force into a drive signal, driving the piezoelectric array to generate an opposite-phase offset vibration to achieve closed-loop vibration suppression; the state feedback matrix of the LQR optimal control algorithm is dynamically adjusted through the real-time pressure pulsation spectrum characteristics, and the control error is ≤0.1%; The piezoelectric ceramic piece detects the mechanical vibration signal in real time and outputs a charge signal, which is converted into a voltage signal by a charge amplifier and then input into a signal processing module; The signal processing module extracts the frequency spectrum characteristics of the vibration main frequency and its integer multiple harmonic components through fast Fourier transform; The LQR control module constructs a state space model based on vibration displacement and velocity, calculates the optimal control force with the goal of minimizing performance indicators, and compensates for the temperature drift effect of piezoelectric materials by adaptively adjusting the weight matrix; The driving circuit generates a multi-band coordinated PWM driving signal according to the optimal control force, drives the piezoelectric ceramic piece to generate a vibration waveform with opposite phases, and realizes closed-loop cancellation of vibration energy.

8. The dynamic environmental error compensation system of the HART protocol intelligent transmitter according to claim 7, characterized in that: The adaptive control module designs independent LQR controllers for the vibration main frequency and its harmonic components respectively, and outputs multi-band collaborative cancellation signals after parallel calculation and superposition; the HART protocol enhancement module customizes the instruction set, including vibration spectrum data uploading instructions and LQR parameter sending instructions, to support remote optimization and real-time monitoring of vibration suppression strategies.

9. The dynamic environmental error compensation system of the HART protocol intelligent transmitter according to claim 7, characterized in that: The HART protocol enhancement module is embedded with a quantum key distribution unit, which includes an optical coupling interface, a quantum channel window, and a polarization-encoded photon transmitter; The quantum key distribution unit is physically connected to the HART protocol enhancement module through an optical coupling interface; The quantum channel window is set on the side wall of the impact-resistant packaging shell, and the window material is anti-reflection coated sapphire; The polarization-coded photon transmitter is integrated into the multi-sensor fusion module and shares the power supply bus with the pressure sensor and vibration sensor. The photon transmission path is arranged orthogonally to the sensor signal line to avoid crosstalk. The quantum error correction code is embedded in the HART protocol data stream with 8 bits of redundancy per frame, and is transmitted synchronously with the dynamic compensation parameters through time division multiplexing, with a communication delay of ≤1ms; The quantum key distribution unit and the HART protocol enhancement module share the DMA channel, the key update cycle is ≤ 1 minute, and the bit error rate is ≤ 10 -9 ; In a strong electromagnetic interference environment, quantum error correction codes recover lost data through redundant photon states to ensure the integrity of dynamic compensation parameters; The recovered data triggers the LQR control module to re-optimize the vibration cancellation strategy of the piezoelectric array, with a control error of ≤0.1%.

10. The dynamic environmental error compensation system of the HART protocol intelligent transmitter according to any one of claims 8 or 9, characterized in that: The following steps are involved: S1. The pressure pulsation signal, ambient temperature signal and mechanical vibration signal are collected synchronously through the integrated packaged pressure sensor, temperature sensor and vibration sensor, and the natural frequency of the sensor sensitive element is designed to be more than three times the main frequency of the pressure pulsation. The signal conditioning circuit is directly integrated into the sensor pin to shorten the transmission path; S2. The pneumatic damper initially attenuates high-frequency pressure pulsations through the viscous silicone oil in the diaphragm cavity, while the mass and spring inertia filter system suppresses the transmission of mechanical vibrations to sensitive components through the synergistic effect of the gradient density mass block and the preload adjustable spring; S3. The piezoelectric ceramic piece integrated in the inner wall of the impact-resistant package shell detects the residual vibration signal in real time, converts it into a voltage signal through a charge amplifier, and then uses a fast Fourier transform to extract the spectral characteristics of the main vibration frequency and its harmonic components by the signal processing module; S4. The dynamic compensation calculation core is based on a hybrid model of BP neural network and Kalman filter, combined with the nonlinear coupling relationship between temperature and time and the dynamic acceleration compensation factor, to generate nonlinear compensation parameters to correct the pressure measurement error; The S5.LQR control module constructs a state space model based on vibration displacement and velocity, calculates the optimal control force by minimizing the performance index, and drives the piezoelectric array to generate a reverse waveform opposite to the detected vibration phase, thus achieving closed-loop cancellation of vibration energy; S6. The HART protocol enhancement module compresses the compensation parameters and vibration spectrum data through Huffman coding, embeds the quantum error correction code, encrypts the data through the dynamic key distributed by polarization-coded photons, and transmits the data to the control terminal through the quantum channel of the anti-reflection sapphire window; S7. When data is lost due to strong electromagnetic interference during transmission, the quantum error correction code recovers the lost frame through redundant photon states and feeds the recovered data back to the LQR control module to re-optimize the vibration cancellation strategy; S8. dynamically adjusting the LQR weight matrix and the neural network model parameters according to the recovered spectrum data and the real-time temperature change, and updating the driving signal of the piezoelectric array to achieve adaptive vibration suppression; S9. The self-checking unit periodically performs the sensor step response test and verifies the data integrity through the dual-channel ADC cross-check and CRC-16 redundancy check mechanism. If the response time exceeds the tolerance or the signal is abnormal, the HART protocol alarm code is triggered; S10. The vibration sensor monitors the compensated vibration energy in real time. If the suppression efficiency does not reach the threshold, the compensation process is restarted. The final pressure measurement value is output through the HART protocol after redundancy verification, completing the dynamic environmental error compensation closed loop.

Images