Control or regulation system for response characteristic calibration of logging equipment

Abstract

The invention relates to the technical field of control or regulation systems, and discloses a control or regulation system for response characteristic calibration of logging equipment, which comprises a frequency spectrum characteristic extraction module, a topological form locking controller, a thermodynamic benchmark evolution module and an amplitude response calibration controller, a frequency spectrum feature extraction module is used for extracting power values of intrinsic background noise in a passband core frequency band and a stop band edge frequency band, a dimensionless frequency spectrum shape factor is constructed, a topological form locking controller adjusts bandwidth control parameters according to deviation between the factor and a topological constant reference, a physical bandwidth is locked, and the bandwidth is locked. On the premise of bandwidth locking, the amplitude response calibration controller adjusts gain in combination with the deviation between the theoretical target power value calculated by the thermodynamic reference evolution module and the actually measured total power, and the bandwidth is locked by using the intrinsic noise spectrum shape invariant through a double-closed-loop time sequence decoupling mechanism. And the problems of model mismatch and multivariable coupling control caused by no external standard source and aging of devices are solved.

Description

Technical Field

[0001] This invention relates to a control or adjustment system for calibrating the response characteristics of well logging equipment, belonging to the technical field of control or adjustment systems. Background Technology

[0002] In current extreme operating scenarios of deep geological exploration, the analog signal acquisition link of logging equipment needs to maintain high response consistency under high temperature, high pressure and long-term continuous operation conditions to ensure the accuracy of geological parameter inversion. For the control of such non-stationary thermodynamic environment systems, existing technologies often adopt a static model lookup table compensation strategy. The strategy pre-establishes a static mapping relationship between temperature and key system parameters. When the controller is running, it reads the ambient temperature, retrieves the corresponding control quantity based on the stored factory calibration data, and performs open-loop adjustment of the circuit gain or cutoff frequency.

[0003] However, the above control strategies are based on the time-invariant ideal assumption of the physical model of the controlled object. In actual engineering applications, long-term thermal stress loads cause irreversible aging and drift of passive components such as resistors and capacitors, resulting in the current physical transfer function of the system deviating from the factory-preset static model. Model mismatch causes cumulative errors in the compensation mechanism based on fixed lookup tables. To address model mismatch, existing technologies attempt to introduce soft measurement or model algorithms to improve calibration accuracy. For example, Chinese invention patent application CN119620228A discloses a method, device, equipment, and storage medium for calibrating the response characteristics of well logging equipment. This method establishes a pressure-fluidity coupled response model and introduces a dual-channel deep neural network to extract features and optimize multi-objective parameters. Although the scheme uses algorithms to approximate and solve the nonlinear fitting of multi-parameter coupling, the core calibration logic is still based on model-data statistical mapping and does not deviate from the preset operating conditions. In deep well closed systems, where a large number of training samples are required, the underlying hardware may experience drift in physical parameters such as the RC constant due to long-term thermal stress. Such algorithm models, lacking an absolute physical reference, face the dilemma of reference failure. They cannot distinguish between device aging and environmental interference from a physical perspective, making it difficult to accurately anchor the system's physical bandwidth and absolute gain. If a conventional closed-loop feedback mechanism is introduced to correct the error, it faces the dual dilemma of missing reference source and variable coupling. Since a long-term stable external physical standard source cannot be implanted in the closed space downhole, the controller cannot obtain an absolute reference. The gain drift and bandwidth drift of the simulation front end are strongly coupled. Relying solely on the observation of the total energy of the output signal, it is mathematically impossible to distinguish whether the energy fluctuation originates from changes in passband gain or cutoff frequency drift. In this case, if multiple feedback loops are used for parallel adjustment, the strong correlation between control variables will cause competitive oscillations, preventing the system from converging to a steady state.

[0004] Therefore, how to utilize the intrinsic characteristics of the system to construct an absolute control reference independent of external standard sources and aging parameters, and achieve decoupling and closed-loop locking of gain and bandwidth timing, is the technical problem to be solved by this invention. Summary of the Invention

[0005] To address the problems mentioned in the background art, the technical solution of the present invention is as follows: A control or adjustment system for calibrating the response characteristics of well logging equipment, the system comprising: The spectrum feature extraction module is used to access the signal output terminal of the controlled object, collect the intrinsic background noise output by the controlled object during non-signal excitation, perform multi-channel parallel digital filtering on the intrinsic background noise, extract the first power value of the intrinsic background noise in the core frequency band of the passband and the second power value in the edge frequency band of the stopband, and calculate the ratio of the second power value to the first power value to construct a dimensionless spectrum shape factor. The topology shape locking controller, coupled to the spectrum feature extraction module, is used to execute the first-level closed-loop adjustment logic. It compares the spectrum shape factor with the preset topology constant benchmark in real time, generates a bandwidth adjustment command based on the shape deviation generated by the comparison, and adjusts the bandwidth control parameters of the passive filter network inside the controlled object through a negative feedback mechanism until the spectrum shape factor converges and locks to the topology constant benchmark, thereby establishing a frequency domain steady state premise in which the physical bandwidth is fixed at the preset value inside the controlled object. The thermodynamic reference evolution module is used to store the thermal noise physical model and, in response to the locking status signal of the topology locking controller, collects real-time ambient temperature data. Combining the real-time ambient temperature data and the locked physical bandwidth, it uses the thermal noise physical model to calculate the theoretical target power value under the current temperature and bandwidth conditions. The amplitude response calibration controller, coupled to the spectrum feature extraction module and the thermodynamic reference evolution module, is used to execute the second-stage closed-loop regulation logic under the constraint of the physical bandwidth being locked. It calculates the power deviation between the measured total power of the intrinsic background noise and the theoretical target power value, and generates a gain adjustment command based on the power deviation. The gain control parameters of the programmable gain amplifier inside the controlled object are adjusted through a negative feedback mechanism until the measured total power converges to the theoretical target power value.

[0006] Preferably, the spectral shape factor constructed by the spectral feature extraction module is a physical invariant independent of the absolute gain of the controlled object and the ambient temperature; the topology constant reference is a numerical constant uniquely determined by the filter transfer function order and cutoff frequency design value of the controlled object; the topology shape locking controller is used to change the pole distribution position of the transfer function of the controlled object by adjusting the input value of the variable capacitor array or the bias current of the transconductance amplifier in the controlled object.

[0007] Preferably, the system further includes a timing logic arbitration unit, which is used to execute hierarchical control logic: only when the absolute value of the shape deviation output by the topology shape locking controller is continuously less than a preset dead zone threshold for N consecutive clock cycles, an activation enable signal is sent to the amplitude response calibration controller; and during the operation of the amplitude response calibration controller, if the absolute value of the shape deviation exceeds the dead zone threshold, the timing logic arbitration unit immediately suspends the adjustment action of the amplitude response calibration controller and reactivates the topology shape locking controller.

[0008] Preferably, the thermodynamic reference evolution module further includes an aging state observer, which is used to monitor in real time the control effort output by the topology shape-locking controller to maintain spectral shape factor locking, and when it detects that the monotonic drift of the control effort at a constant temperature exceeds a preset threshold, it corrects the equivalent input impedance parameter in the thermal noise physical model according to the following drift compensation relationship: ,in, This is the corrected equivalent input impedance. This is the factory-specified impedance. The preset aging sensitivity coefficient, To control the real-time deviation of the effort amount relative to the system's factory calibration reference value. This refers to the system's cumulative service time.

[0009] Preferably, the system further includes a non-Gaussian interference fuse unit, which is used to calculate the fourth-order statistical moments of the intrinsic background noise in real time and compare the calculated fourth-order statistical moments with the theoretical kurtosis value of the Gaussian distribution. When the comparison result shows that the current signal exceeds the Gaussian confidence interval, the non-Gaussian interference fuse unit immediately cuts off the parameter update path of the topology morphology locking controller and the amplitude response calibration controller, and maintains the control state of the controlled object at the previous moment.

[0010] Preferably, the amplitude response calibration controller adopts an integral adjustment strategy, and its integral gain coefficient is controlled by the convergence state of the topology shape lock controller. When the rate of change of the deviation between the spectral shape factor and the topology constant reference exceeds the preset rate of change threshold, the amplitude response calibration controller automatically reduces its integral gain coefficient to suppress the overshoot of the gain adjustment loop.

[0011] Preferably, the spectral feature extraction module includes a multi-stage cascaded digital decimation filter bank. The digital decimation filter bank is used to perform multi-rate undersampling processing on the intrinsic background noise before extracting the first power value and the second power value, so as to obtain the high-frequency noise spectral components aliased into the linear operating region of the baseband processing unit, thereby improving the quantization resolution for high-frequency cutoff characteristics.

[0012] Preferably, the system further includes a temperature gradient compensation unit, which monitors the time rate of change of ambient temperature and applies a feedforward compensation amount based on the time rate of change to the topology locking controller when the time rate of change exceeds a preset thermal shock threshold. The feedforward compensation amount is used to offset the transient thermal drift of passive device parameters caused by temperature changes.

[0013] Preferably, the controlled object is the analog signal conditioning front-end circuit of a deep geological exploration instrument, and the intrinsic background noise is the thermal noise generated by the internal resistive elements of the analog signal conditioning front-end circuit when there is no external excitation signal input; the system is embedded in the digital logic core of the analog signal conditioning front-end circuit and performs self-calibration with the intrinsic background noise as the sole reference.

[0014] Preferably, the system is equipped with non-volatile memory to record historical convergence parameters of the topology morphology locking controller and the amplitude response calibration controller; when the system is powered on again, the system reads the historical convergence parameters of the most recently successfully calibrated system as the initial control value.

[0015] Compared with the prior art, the beneficial effects of the present invention are: 1. In the control or regulation of well logging equipment response characteristic calibration, the intrinsic thermal noise spectrum shape characteristics of the controlled object are extracted, and a topological invariant independent of absolute gain and ambient temperature is constructed as the first-level control feedback source. The ratio of the power of the passband core component to the stopband edge component depends on the physical characteristics of the pole distribution of the transfer function. An absolute bandwidth reference is established under closed working conditions without external standard signal sources. This solves the technical problem that traditional control strategies rely too much on external high-precision reference sources and cannot maintain the consistency of system frequency response in extreme thermodynamic environments.

[0016] 2. A time-series decoupling control based on physical causal chains is adopted. The physical bandwidth is locked by utilizing the immunity of the spectrum shape factor to amplitude changes, which establishes the necessary boundary conditions for subsequent gain control. The hierarchical adjustment logic decomposes the complex gain-bandwidth coupling disturbance into two orthogonal single-variable control processes, ensuring that the gain adjustment loop has a unique physical convergence solution in the thermal noise power equation. This eliminates the competitive oscillations common in multivariable parallel feedback systems and guarantees the steady-state response accuracy of precision electronic systems in a wide temperature range environment.

[0017] 3. An integrated inverse model identification mechanism based on control effort monitoring is used to analyze the long-term drift trend of control parameters required to maintain frequency response shape locking, quantitatively invert the aging state of passive device materials, and transform the controller adjustment action into a basis for dynamic correction of system physical model parameters. This enables the reference benchmark to adaptively evolve with the characteristics of the device throughout its entire life cycle, solving the engineering problem of control accuracy degradation caused by the mismatch between theoretical models and actual physical objects due to component aging. A non-Gaussian interference fuse mechanism is constructed using higher-order statistical moments to calculate the background noise kurtosis characteristics in real time to identify transient pulse disturbances. When non-stationary statistical characteristics are detected, the control loop state is immediately frozen to cut off the transmission path of external sudden noise to control parameter errors, maintain the system's adjustment stability under non-ideal electromagnetic environments, and prevent parameter divergence due to misjudgment of interference signals. Attached Figure Description

[0018] Figure 1 This is a block diagram of the dual closed-loop timing decoupling control system based on intrinsic noise of the present invention; Figure 2 This is a curve comparing the bandwidth locking effect and uncalibrated drift under wide temperature range conditions of the present invention; Figure 3 This is a flowchart of the self-calibration control logic that integrates interference fuse and aging compensation in this invention. Detailed Implementation

[0019] This specific embodiment is only used to illustrate the present invention and is not intended to limit the claims of the present invention. Unless otherwise specified, the embodiments and features in the embodiments of this application can be combined with each other.

[0020] This invention provides a control or adjustment system for calibrating the response characteristics of well logging equipment. It comprises a spectrum feature extraction module, a topology locking controller, a thermodynamic reference evolution module, and an amplitude response calibration controller. These modules work collaboratively through a timing decoupling mechanism. Utilizing the statistical characteristics of the intrinsic background noise of the controlled object, closed-loop locking of the system's physical bandwidth and absolute gain is achieved under conditions of no external standard source and device aging drift. The spectrum feature extraction module is connected to the signal output of the controlled object and is used to collect intrinsic background noise during non-signal excitation periods. This module uses a high-precision analog-to-digital converter with a sampling rate 10 times the system's rated bandwidth to digitize the analog noise. The module internally contains two sets of parallel digital decimation filter banks, which perform frequency domain segmentation processing on the digitized noise stream. The first low-pass filter extracts the noise component within the core passband frequency band, i.e., the component ranging from DC to 0.1 times the cutoff frequency, and calculates the root mean square power of this component to obtain a first power value. The second set of bandpass filters extracts the noise components within the stopband edge frequency band, specifically the components ranging from 0.8 to 1.2 times the cutoff frequency, and calculates the root mean square power of these components to obtain the second power value. The module performs further calculations. and The ratio of these ratios is used to construct a dimensionless spectral shape factor. The spectral shape factor It depends only on the location of the poles of the circuit's transfer function, i.e., the topological shape of the physical bandwidth, and is independent of the circuit's absolute gain. and ambient temperature This provides orthogonalized feedback variables for bandwidth control.

[0021] The topology-locked controller is coupled to the spectral feature extraction module to execute the first-stage closed-loop regulation logic. The system has a pre-stored topology constant reference. This constant is uniquely determined by the order of the ideal filter transfer function of the controlled object and the design value of the cutoff frequency. The controller calculates the real-time spectrum shape factor in real time. With topological constant benchmark The shape deviation is measured and input into the digital proportional-integral controller. Greater than When this occurs, it indicates that the actual bandwidth exceeds the design value, and the regulator output command increases the input value of the variable capacitor array in the passive filter network or decreases the bias current of the transconductance amplifier; when Less than At this time, reverse adjustment is performed. The controller has built-in convergence judgment logic. Only when the absolute value of the morphological deviation is continuously less than the dead zone threshold of 0.5% for 100 consecutive clock cycles will a bandwidth lock-in state signal be output to establish the premise of frequency domain steady state. The thermodynamic reference evolution module responds to the bandwidth lock-in state signal, providing a theoretical reference for gain calibration. The module stores a thermal noise physical model based on the Johnson-Nyquist principle, i.e. The module collects ambient temperature data in real time. Read the locked physical bandwidth And call the currently stored equivalent input impedance. Calculate the theoretical target power value under the current conditions. This module includes an aging state observer for real-time monitoring of the control effort of the topology-locked controller. If it is observed that the capacitor array control word or bias voltage required to maintain bandwidth locking at a constant temperature increases with the cumulative service time of the system... If a monotonic drift is observed and the drift exceeds a preset threshold, the observer uses the drift compensation formula to inversely calculate the impedance change. And update accordingly. ,Right now ,in This is the aging sensitivity coefficient. To control the relative offset of effort.

[0022] The amplitude response calibration controller is activated under the constraint that the physical bandwidth is locked, and executes the second-stage closed-loop regulation logic. The controller calculates the measured total power of the intrinsic background noise. Compared with the theoretical target power value Based on the power deviation between the two, the controller uses an integral adjustment strategy to drive the gain control interface of the programmable gain amplifier inside the controlled object, thereby adjusting the system gain. Until the actual measured total power Converging to the theoretical target power value When the rate of change of the spectral shape factor from the topological constant reference exceeds a preset threshold, the controller automatically reduces the integral gain coefficient to suppress overshoot of the gain adjustment loop. The system also includes a non-Gaussian interference fuse unit to maintain the robustness of the control loop. This unit calculates the fourth-order statistical moment of the intrinsic background noise in real time. Since thermal noise follows a Gaussian distribution, its theoretical kurtosis value is close to 3. When the calculated kurtosis value exceeds the preset Gaussian confidence interval, i.e., the range of 2.5 to 3.5, the fuse unit determines that the current signal is mixed with non-Gaussian pulse interference, immediately cuts off the parameter update path of the topology shape lock controller and the amplitude response calibration controller, and forcibly maintains the control state of the previous moment until the signal statistical characteristics are restored to the confidence interval.

[0023] Example 1: This example describes the specific application of the above-mentioned control system based on spectral shape invariant in deep high-temperature and high-pressure logging operations. In a typical deep well logging scenario, the logging instrument is lowered to a depth of 5000 meters underground. At this time, the ambient temperature rises to 175℃ and the ambient pressure exceeds 100MPa. As the continuous operation time increases, the resistive components in the analog front-end circuit undergo irreversible aging drift due to long-term high-temperature stress. At this time, if the traditional static compensation strategy based on factory calibration data is adopted, the measurement gain error will accumulate over time because the actual physical parameters have deviated from the factory model, which cannot meet the stringent requirements of signal consistency for precise geological parameter inversion. For the above working conditions, this system automatically starts the calibration process during the non-signal excitation period of each logging cycle. The spectral feature extraction module is connected to the output of the front-end circuit to collect the intrinsic background noise at a sampling rate of 200kHz. The low-pass filter inside the module extracts the first power value within the range of DC to 0.1 times the cutoff frequency. The bandpass filter extracts the second power value within the range of 0.8 to 1.2 times the cutoff frequency. The system calculates the ratio of the two values ​​to obtain the real-time spectral shape factor. Even if the absolute gain of the circuit drifts due to high temperature, because and Both are affected by gain. As a dimensionless ratio, it remains insensitive to changes in gain, reflecting the change in physical bandwidth caused by circuit pole shift.

[0024] The topology-locking controller will calculate Compared with the preset topological constant benchmark A comparison was performed, and the detection... Deviation At this time, the controller adjusts the input value of the variable capacitor array through a negative feedback mechanism. For example, if As the indicated bandwidth narrows, the controller gradually reduces the capacitance value, forcing the spectrum shape to return to its original state. The defined theoretical form of this process utilizes the spectral shape factor as an intermediate control variable independent of the gain to lock the physical bandwidth under conditions of unknown gain. Once the physical bandwidth is locked, the thermodynamic reference evolution module reads the current downhole temperature of 175°C and the locked bandwidth value. Regarding the resistor aging issue, the aging state observer within the module analyzes historical data on the control effort required to maintain bandwidth locking. If a monotonic offset is observed in the capacitor array setpoint relative to the factory reference, the observer uses a preset model to inversely calculate the actual aging amount of the resistor. And correct the thermal noise equation In The parameter, amplitude response calibration controller uses a modified physical model to calculate the theoretical target power value. The programmable gain amplifier is adjusted to bring the measured background noise power to the reference, thereby completing the calibration of the absolute gain. Through the above timing decoupling and model adaptation mechanism, the system can autonomously correct the circuit aging drift by utilizing intrinsic thermal noise in the closed environment of a deep well without external standard sources. The logging instrument maintains the consistency of response characteristics during long-term high-temperature operation.

[0025] Example 2: This example describes an accelerated aging test used to verify the calibration performance of the above control system under wide temperature range and long-term service conditions. It aims to demonstrate the effectiveness of the invention in achieving closed-loop locking without an external standard source and its adaptive correction capability for model mismatch. The system under test is constructed on a hardware-in-the-loop (HIL) simulation platform equipped with a precision programmable temperature chamber and a low-noise electromagnetic shielding room. This system includes a prototype analog front-end circuit based on CMOS technology. The transfer function is designed as a fourth-order Butterworth low-pass filter with a cutoff frequency of 10kHz to realistically simulate the non-ideal characteristics under deep well conditions. A programmable impedance network is connected in series in the signal link to simulate the aging drift of passive components with temperature and time. A 24-bit high-precision dynamic signal analyzer is used as the data acquisition end, acting as a third-party arbitrator to independently monitor the intrinsic noise of the system output. Its measurement results are only used for post-verification and do not participate in closed-loop control.

[0026] The core of the experiment lies in verifying whether the system can recover the preset bandwidth and gain solely based on intrinsic thermal noise when device parameters experience unknown drift. To this end, a two-stage test procedure was established. The first stage was a temperature lock-in test, where the chamber temperature was linearly increased from room temperature (25°C) to 175°C at a rate of 2°C / min and then kept constant. During this process, the programmable impedance network was manually controlled to induce an initial -20% drift in the filter's cutoff frequency and a -3dB gain attenuation. The second stage was an accelerated aging test, running continuously for 1000 hours at a constant temperature of 175°C. During this period, the impedance network simulated a monotonically drifting resistance value at a rate of 1% per year. After the test started, the spectral feature extraction module collected background noise in real time. During the first stage of temperature increase, although temperature changes caused fluctuations in the total thermal noise power, monitoring data showed that the spectral shape factor... Before closed-loop control intervenes, the frequency drifts and deviates from the topological constant reference. When the closed-loop control is activated, the topology locking controller begins to adjust the capacitor array. As shown in the table below, at different temperature points, although the bandwidth error before calibration increases with the temperature, the measured bandwidth is always locked near the design value after closed-loop adjustment.

[0027] Table 1: Example Table of Bandwidth Lockout Data During Temperature Variation Process

[0028] Referring to Table 1, the data shows that regardless of changes in ambient temperature, as long as Locked in The system's physical bandwidth was thus forcibly anchored to the design target of 10kHz, verifying that the spectral shape factor, as a temperature- and gain-independent invariant, can effectively decouple multivariate interference. In the second stage of accelerated aging testing, with the introduction of simulated resistor aging drift, the control voltage required to maintain bandwidth lockout exhibited a monotonically increasing trend. The aging state observer captured this trend and calculated the resistor correction amount accordingly. At this point, the thermodynamic baseline evolution module is dynamically updated. Parameters, recalculate the theoretical target power The amplitude response calibration controller then adjusts the PGA gain. At the end of the 1000-hour test, compared with the control group without the aging observer enabled, the gain error of the sample group of this invention is reduced. The control group, due to its fixed model parameters, results in a higher calculated gain error. The mismatch with the actual physical state ultimately resulted in a gain accumulation error of about 1.2dB; however, the sample of this invention controlled the gain error within 0.1dB through model self-evolution.

[0029] Example 3: This example combines Figures 1 to 3A description of the control or regulation system for calibrating the response characteristics of a well logging device, such as... Figure 1 As shown, the control or adjustment system for calibrating the response characteristics of this logging equipment mainly consists of a spectrum feature extraction module containing multi-channel parallel digital filtering processing logic. It receives the intrinsic background noise from the controlled object, i.e., the analog signal conditioning front-end circuit. It extracts the passband to stopband power ratio to construct a dimensionless spectral shape factor. The measured total power and shape factor are transmitted to the amplitude response calibration controller and the topology shape locking controller, respectively. The topology shape locking controller, as the first-level closed loop, compares the shape factor with a topological constant reference to generate a bandwidth adjustment command to adjust the passive filter network parameters in the controlled object. It also outputs aging monitoring data to the thermodynamic reference evolution module. The information control effort module integrates an aging state observer with a thermodynamic reference evolution module. It calculates the theoretical target power value by combining ambient temperature and aging correction and provides it to the amplitude response calibration controller. The amplitude response calibration controller acts as a second-level closed loop. It adjusts the programmable gain amplifier in the controlled object based on the thermal noise power deviation to achieve gain calibration. The system also includes a timing logic arbitration unit that connects each unit. It is used to execute hierarchical control logic and monitor the morphological deviation dead zone threshold to ensure that the bandwidth is locked before the gain is calibrated. It also includes a non-Gaussian interference fuse unit, which is used to calculate the fourth-order statistical moment, i.e., kurtosis, to identify non-Gaussian impulse interference and perform fuse or hold operations in case of anomalies.

[0030] like Figure 2 As shown, the horizontal axis represents temperature, with a scale range from 25℃ to 175℃. The left vertical axis corresponds to ambient temperature, and the right vertical axis corresponds to the percentage change in bandwidth. The graph contains three data curves: the first is a solid line with hollow circles, representing the linear increase in ambient temperature from 25℃ to 175℃ during the test; the second is a dashed line with hollow circles, representing the percentage of uncalibrated bandwidth drift. This curve shows that the bandwidth in the uncalibrated state exhibits a non-linear negative drift with temperature changes, with the drift fluctuating around -20%; the third is a dotted line with hollow circles, representing the percentage of bandwidth error after calibration. This curve shows that the bandwidth error after system adjustment remains stably maintained at the 0% baseline across the entire temperature range. Figure 3As shown, the system's workflow begins with the logging operation master control system being triggered during non-excitation periods, entering the full-process master control stage of self-calibrating the execution response characteristics. This stage collects intrinsic background noise and extracts the spectral shape factor to calculate the passband to stopband power ratio. The process then enters branch processing: on the one hand, it performs fourth-order moment detection protection through a non-Gaussian interference fuse unit, blocking control if an anomaly is detected; on the other hand, it enters two-level closed-loop control paths, namely, the first-level control closed loop locks the physical bandwidth and controls the controlled object, the analog front-end circuit, by adjusting parameters, and the second-level control closed loop calibrates the absolute gain and controls the analog front-end circuit by adjusting the gain. During the control process, the system reads the temperature and pressure data from the environmental sensing sensors, calculates the thermodynamic target power in conjunction with the thermodynamic benchmark evolution module, and compensates for device aging drift and corrects the model impedance parameters in the process, ultimately making the gain calibration dependent on the corrected theoretical target value.

[0031] Example 4: This example addresses the black box issues in the algorithm path and parameter determination of the aging state observer and thermodynamic benchmark evolution module in the original document, providing targeted repair and refinement. The aim is to make the calculation process of the aging correction coefficient transparent. In industrial scenarios where long-term monitoring of resistive element aging is required, the dynamic update mechanism distinguishes between normal ambient temperature drift and irreversible material aging drift. To address this, the thermodynamic reference evolution module incorporates a parameter identification engine based on recursive least squares (RLS). This engine is activated each time the system powers on or performs a periodic self-test, allowing the system to read the current absolute temperature. and control effort to maintain bandwidth locking Controlling effort Defined as the normalized input value of the variable capacitor array in the passive filter network, with a value ranging from 0 to 1; the system performs aging trend separation and determination, and establishes an impedance model that evolves over time: ,in, The nominal impedance at the reference temperature, The preset temperature coefficient of resistance, This represents the deviation of the current temperature from the reference temperature. For the aging factors to be identified, in the known and Under the premise that the observer monitors To estimate the rate of change Specifically, if at a constant temperature or after temperature compensation In continuous Within a single observation period (e.g.) The aging process exhibits monotonous changes, and the rate of change exceeds the preset aging threshold. (For example If the value is less than or equal to 1 / hour, then aging drift is determined to have occurred.

[0032] Once aging is detected, the system starts. The update procedure utilizes identified aging factors. Calculate the current equivalent impedance correction amount. Update the equivalent impedance parameters in the thermal noise physical model: The updated It was then used to calculate the new theoretical target power value. To ensure that the reference of the gain calibration loop is always anchored to the actual physical state, and to prevent erroneous corrections due to measurement noise, the system introduces a smoothing filter based on a sliding time window, which only applies the calculated value when the reference value is true. Updates are only allowed when the variance within the sliding window is less than a preset stability threshold. .

[0033] Example 5: This example describes an offline calibration procedure for pre-deployment in the field. It aims to ensure the initial accuracy of core model parameters in the thermodynamic baseline evolution module and establish the system's baseline adaptability to hardware differences of specific controlled objects. Before the system is officially put into logging operations, impedance-temperature baseline calibration must be performed on each assembled circuit board in a controlled laboratory. The circuit board to be calibrated is placed in a high-precision constant temperature chamber, and the output of the topology-locked controller is connected to a standard impedance analyzer. The system is configured to enter calibration mode. In this mode, the automatic calibration loop is bypassed, and the control logic only performs open-loop data acquisition. The chamber temperature is gradually increased from -20℃ to 180℃ in 5℃ increments, and held constant at each temperature point for 30 minutes to reach thermal equilibrium. The system records the actual physical temperature at each temperature point. And the corresponding measured values ​​of the passive filter network port impedance, using this set The initial temperature coefficient of resistance for this specific circuit board was obtained by fitting the corresponding impedance data using the least squares method. and nominal impedance at reference temperature These two parameters are then burned into the non-volatile memory of the thermodynamic reference evolution module, serving as the calculation basis for subsequent online aging monitoring.

[0034] In addition, to ensure that the false trigger rate of the non-Gaussian interference fuse unit meets the industry standard under different electromagnetic environments, the field noise baseline calibration procedure must be performed. Before the logging instrument is lowered into the well, it is placed in the wellhead standby state and the noise learning mode is activated. In this mode, the system continuously collects intrinsic background noise data for no less than 10 minutes and calculates the probability density distribution of its fourth-order statistical moments (kurtosis). The system automatically adjusts the upper and lower limit thresholds of the Gaussian confidence interval according to the statistical characteristics of the measured distribution. For example, if a steady-state non-Gaussian colored noise component is detected in the environment itself, the system will adaptively relax the kurtosis threshold.

[0035] Example 6: This example describes a standardized engineering procedure for systematically calibrating and solidifying key parameters in a topology morphology locking controller before formally performing online calibration tasks at the simulated front end of a controlled object logging equipment. In a temperature-controlled and electromagnetically controlled laboratory, the physical parameters of the controlled object in its initial state are confirmed. A high-precision vector network analyzer is used to measure the actual transfer function of the controlled object under room temperature conditions, and the reference curves of the passband cutoff frequency and stopband attenuation characteristics are recorded. The topology constant reference is calibrated, and the system is placed in calibration mode. An external signal generator inputs a spectral flatness better than [previous value]. The broadband white noise signal is processed by the spectral feature extraction module, which calculates the spectral shape factor and defines this measured value as the system's topological constant benchmark. And write it into the controller's non-volatile register.

[0036] Next, the dead-zone threshold is calibrated. While keeping the input signal constant, a series of small-amplitude bandwidth disturbances are artificially introduced. By gradually increasing the disturbance amplitude, the response of the topology-locked controller is monitored. The minimum bandwidth deviation at which the controller begins to generate effective regulation is recorded, and the rate of change of the shape factor corresponding to this deviation is set as the dead-zone threshold. This procedure establishes the controller's sensitivity boundary to bandwidth drift, avoiding frequent malfunctions caused by measurement noise while ensuring timely response to substantial drift. Finally, the integral gain coefficient is optimized. Under closed-loop system operation, a step temperature disturbance is applied to the controlled object through a temperature control chamber. This is done to induce rapid parameter drift, monitor the overshoot and settling time of the system during the recovery to steady state, and iteratively adjust the integral gain coefficient until the system response curve meets preset engineering parameters, such as an overshoot of less than 1%. And the adjustment time is less than The integral gain coefficient at this point is then fixed as the system's default control parameter. Through the standardized calibration process described above, the system establishes a set of optimal control parameters that are adapted to the current hardware characteristics, providing a definite engineering foundation for stable operation under complex conditions.

[0037] Example 7: The aging state observer uses a recursive least squares algorithm with a forgetting factor, with a preset system clock cycle as the step size, to synchronously acquire the normalized control effort output of the topology locking controller. and environmental sensing sensor outputs real-time temperature Based on the preset impedance evolution model Iterative calculation of real-time data stream separation aging drift factor ,in To store the factory-specified impedance, The temperature coefficient of resistance. For calibration reference temperature; only when the calculated temperature is obtained. In length The variance statistic within the sliding time window is less than the preset convergence threshold. At that time, the thermodynamic baseline evolution module performs parameter updates according to the update formula. Calculate the corrected equivalent input impedance Then, by substituting the thermal noise power equation, a theoretical target power value matching the current physical state of the device is generated.

[0038] Topological constant benchmark and temperature coefficient of resistance Derived from the offline thermodynamic calibration procedure before equipment operation: the controlled object is placed in a temperature control environment with an accuracy better than... In a constant temperature environment, at multiple discrete temperature points covering the rated operating temperature range of the equipment, the real part of the port impedance of the passive filter network was measured using a standard impedance analyzer. The slope of the measured impedance versus temperature characteristic curve was extracted using linear regression fitting. Injecting spectral flatness into the controlled object is better than The arithmetic mean of the ratio of passband power to stopband power output by the white noise excitation signal at each temperature point spectral feature extraction module is defined as the topological constant benchmark. It is stored in non-volatile memory and used as the basis for establishing frequency domain steady-state addressing during system power-on initialization.

[0039] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0040] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. A control or regulation system for calibration of response characteristics of a well logging device, characterized in that, The system comprises: a spectrum feature extraction module, configured to access a signal output end of a controlled object, collect intrinsic background noise output by the controlled object during a non-signal excitation period, perform multi-channel parallel digital filtering processing on the intrinsic background noise, extract a first power value of the intrinsic background noise in a passband core frequency band and a second power value of the intrinsic background noise in a stopband edge frequency band, and calculate a ratio of the second power value to the first power value to construct a dimensionless spectrum shape factor; a topological morphology locking controller, coupled to the spectrum feature extraction module, configured to perform a first-stage closed-loop adjustment logic, compare the spectrum shape factor with a preset topological constant reference in real time, generate a bandwidth adjustment instruction according to a morphology deviation generated by the comparison, and adjust a bandwidth control parameter of a passive filter network inside the controlled object through a negative feedback mechanism until the spectrum shape factor converges and is locked to the topological constant reference, thereby establishing a frequency domain steady-state premise with a physical bandwidth fixed at a preset value inside the controlled object; a thermodynamic reference evolution module, configured to store a thermal noise physical model, and in response to a locking state signal of the topological morphology locking controller, collect real-time ambient temperature data, combine the real-time ambient temperature data and the locked physical bandwidth, and calculate a theoretical target power value under a current temperature and bandwidth condition by using the thermal noise physical model; an amplitude response calibration controller, coupled to the spectrum feature extraction module and the thermodynamic reference evolution module, configured to perform a second-stage closed-loop adjustment logic under the constraint condition that the physical bandwidth is locked, calculate a power deviation between a measured total power of the intrinsic background noise and the theoretical target power value, and generate a gain adjustment instruction according to the power deviation, and adjust a gain control parameter of a programmable gain amplifier inside the controlled object through a negative feedback mechanism until the measured total power converges to the theoretical target power value.

2. The control or regulation system for calibrating the response characteristics of a well logging device according to claim 1, characterized in that The spectrum shape factor constructed by the spectrum feature extraction module is a physical invariant independent of the absolute gain of the controlled object and the ambient temperature; the topological constant reference is a numerical constant uniquely determined by the filter transfer function order and the cutoff frequency design value of the controlled object; The topological morphology locking controller is configured to change the pole distribution position of the transfer function of the controlled object by adjusting the access value of a variable capacitance array or the bias current of a transconductance amplifier in the controlled object.

3. The control or regulation system for calibrating the response characteristics of a well logging device according to claim 1, characterized in that The system further comprises a timing logic arbitration unit, which is configured to perform hierarchical control logic: only when it is monitored that the absolute value of the morphology deviation output by the topological morphology locking controller continuously remains less than a preset deadband threshold for consecutive N clock periods, an activation enable signal is sent to the amplitude response calibration controller; and during the operation of the amplitude response calibration controller, if the absolute value of the morphology deviation exceeds the deadband threshold, the timing logic arbitration unit immediately suspends the adjustment action of the amplitude response calibration controller and reactivates the topological morphology locking controller.

4. The control or regulation system for calibrating the response characteristics of a well logging device according to claim 1, characterized in that The thermodynamic benchmark evolution module further comprises an aging state observer configured to monitor in real time the control effort output by the topology morphological locking controller to maintain the spectral shape factor lock, and to correct the equivalent input impedance parameter in the thermal noise physical model according to the following drift compensation relationship when detecting that the monotonic drift amount of the control effort at constant temperature exceeds a preset threshold value: wherein, is the corrected equivalent input impedance, is the factory nominal impedance, is a preset aging sensitivity coefficient, is the real-time offset of the control effort relative to the system factory calibration benchmark value, is the cumulative service time of the system.

5. The control or regulation system for calibrating the response characteristics of a well logging device according to claim 1, characterized in that The system further comprises a non-Gaussian interference fusing unit, which is configured to calculate the fourth-order statistical moment of the intrinsic background noise in real time, and compare the calculated fourth-order statistical moment with a theoretical kurtosis value of Gaussian distribution; when the comparison result shows that the current signal exceeds the Gaussian confidence interval, the non-Gaussian interference fusing unit immediately cuts off the parameter updating path of the topological morphology locking controller and the amplitude response calibration controller, and maintains the control state of the controlled object at the previous moment.

6. The control or regulation system for calibrating the response characteristics of a well logging device according to claim 1, characterized in that The amplitude response calibration controller adopts an integral regulation strategy, and the gain coefficient of the integral action is controlled by the convergence state of the topological morphology locking controller; when the change rate of the deviation of the spectral shape factor from the topological constant reference exceeds a preset change rate threshold, the amplitude response calibration controller automatically reduces the integral gain coefficient.

7. The control or regulation system for calibrating the response characteristics of a well logging device according to claim 1, characterized in that, The spectral feature extraction module comprises a multi-stage cascaded digital decimation filter bank, which is configured to perform multi-rate under-sampling processing on the intrinsic background noise before extracting the first power value and the second power value, so as to obtain high-frequency noise spectral components that are aliased into the linear working area of the baseband processing unit.

8. The control or regulation system for calibrating the response characteristics of a well logging device according to claim 1, characterized in that, The system further comprises a temperature gradient compensation unit, which is configured to monitor the time change rate of the ambient temperature, and when the time change rate exceeds a preset thermal shock threshold, apply a feedforward compensation amount based on the time change rate to the topological morphology locking controller, the feedforward compensation amount being used to offset the transient thermal drift of the passive device parameters caused by temperature change.

9. The control or regulation system for calibrating the response characteristics of a well logging device according to claim 1, characterized in that The controlled object is an analog signal conditioning front-end circuit of a deep geological exploration instrument, and the intrinsic background noise is thermal noise generated by internal resistive elements of the analog signal conditioning front-end circuit during no external excitation signal input; the system is embedded in the digital logic core of the analog signal conditioning front-end circuit, and performs self-calibration with the intrinsic background noise as the only reference.

10. The control or regulation system for calibrating the response characteristics of a well logging device according to claim 1, characterized in that, The system is configured with a non-volatile memory for recording the historical convergence parameters of the topological morphology locking controller and the amplitude response calibration controller; when the system is restarted, the system reads the historical convergence parameters of the last successful calibration as the initial control value.

Images