Intelligent long-distance power supply relay voltage regulating device and method for tunnel face

Abstract

The application discloses a kind of intelligent long-distance power supply relay voltage regulating devices and methods in heading face, and the device includes input isolation and state acquisition unit, relay voltage regulating execution unit, terminal measurement unit, linkage breaking unit and controller.Current power supply condition is identified as light load stable, heavy load stable, start transient or abnormal disturbance condition by controller, and recursive parameter identification algorithm is used to update power supply section equivalent impedance parameters online, according to the type of working condition and the available voltage at the end, determine the target level and switching window under the constraints of minimum residence time, level switching frequency limit, phase consistency and voltage change slope, and check after switching, execute degradation power supply strategy according to four priority levels when abnormal.The device also integrates double-link confidence weighted fusion communication fault-tolerant, distributed state diagnosis module and independent hardware safety state machine.The application significantly reduces the mis-switching rate and level oscillation, and improves the voltage quality and safety and reliability of long-distance power supply in heading face.

Description

Technical Field

[0001] This invention relates to the field of power supply and distribution technology in mining tunneling, specifically to an intelligent long-distance power supply relay voltage regulation device and method for tunneling faces. Background Technology

[0002] As the tunneling distance in coal mines and tunnel projects continues to increase, the power supply cables between the upstream substation and the electrical equipment at the tunneling face extend continuously, significantly increasing the line impedance. In a typical underground power supply architecture in a coal mine, 6kV or 10kV is typically stepped down to 1140V or 660V from a surface or underground substation and then transmitted to the tunneling face via long-distance low-voltage cables to power equipment such as roadheaders, belt conveyors, scraper conveyors, local ventilation fans, and drainage pumps. When the cables extend to hundreds or even thousands of meters, the voltage quality at the end of the cable has become one of the main constraints on safe and efficient production at the tunneling face.

[0003] The existing relay voltage regulation scheme has the following five main defects:

[0004] Defect 1: Fixed power supply parameters cannot adapt to tunneling progress. Existing solutions (such as the electromechanical voltage regulating device disclosed in publication number CN117674162A) use pre-set fixed parameters for the voltage drop model. As the cable length continuously increases and the line impedance continuously drifts during tunneling, the prediction error of the fixed parameter model accumulates, leading to a mismatch between the selected voltage regulation level and the actual requirements, and a decrease in voltage regulation accuracy. This is the fundamental reason why this invention introduces a recursive parameter identification algorithm to update the equivalent impedance parameters online.

[0005] Defect 2: Lack of operating condition awareness, resulting in a high false switching rate. Traditional threshold-based voltage regulation schemes trigger switching immediately after a voltage drop at the end, failing to distinguish whether the load is in a startup transient, heavy load stability, or abnormal disturbance state. Switching the stage during the motor startup transient not only fails to effectively suppress the surge voltage drop but also overlaps with the load startup process, exacerbating current surges and contactor arcing. This is precisely why this invention introduces a four-category operating condition identification and constrains voltage regulation decisions based on operating condition type.

[0006] Defect 3: Lack of safety constraints during switching leads to frequent stage oscillations. Existing solutions trigger switching based solely on a single voltage threshold, lacking multi-dimensional constraints such as minimum dwell time, switching frequency limits, and voltage change slope. This results in frequent back-and-forth oscillations between adjacent stages, leading to high contactor operation frequency and significant contact wear. This is precisely why this invention introduces a multi-constraint switching window determination mechanism.

[0007] Defect 4: Lack of post-switching verification and degraded supply guarantee mechanism. Existing devices lack real-time verification of voltage regulation effectiveness after level switching. If contactor malfunction, predictive mismatch, or phase sequence abnormalities occur, the system cannot promptly transition to a safe state, potentially escalating the fault. This is precisely why this invention introduces post-switching verification and graded degraded supply guarantee strategies.

[0008] Defect 5: Lack of safety compliance linkage. The existing pressure regulating device does not achieve hardware-level linkage with the wind power interlock and methane power interlock signals at the tunneling face. When interlock is triggered, there is a risk that the device may not promptly enter a safe power-off state, which does not meet the hardware-level protection requirement of "complete power-off" in coal mine safety regulations. This is precisely why this invention introduces an independent hardware safety state machine module. Summary of the Invention

[0009] Technical Objective: To address the problems in existing long-distance power supply systems for tunneling faces, such as continuous drift of power supply segment parameters during tunneling, frequent switching of load conditions, high error switching rate of traditional threshold-based voltage regulation, lack of post-switching verification and degradation protection capabilities, and lack of safety and compliance linkage, this invention discloses an intelligent long-distance power supply relay voltage regulation device and method for tunneling faces. It deeply integrates four-category working condition identification, online identification of recursive parameters, candidate level prediction, multi-constraint switching windows, post-switching verification, and graded degradation protection.

[0010] Technical solution: To achieve the above technical objectives, the present invention adopts the following technical solution:

[0011] A smart long-distance power supply relay voltage regulation device for a tunnel face, installed between the upstream power supply and the downstream load, includes:

[0012] The input isolation and status acquisition unit is connected to the upstream power supply and is used to acquire the electrical status of the upstream line.

[0013] The relay voltage regulation execution unit has its input terminal connected to the input isolation and status acquisition unit, and its output terminal facing the end load. It is used to regulate the output voltage between multiple voltage regulation stages.

[0014] The terminal measurement unit is located on the output side of the relay voltage regulating unit and is used to collect terminal electrical quantities and load status quantities.

[0015] The linkage interruption unit is connected in series with the output circuit of the relay voltage regulation execution unit and is used to perform interruption and switching operations during the level switching process;

[0016] The controller, connected to the input isolation and status acquisition unit, the relay voltage regulation execution unit, the terminal measurement unit, and the linkage disconnection unit, is used for: identifying the current power supply condition based on the acquired electrical status quantities of the upstream line and the terminal electrical quantities; updating the equivalent impedance parameters of the power supply section online based on the current sampled data using a recursive parameter identification algorithm; predicting the available terminal voltage corresponding to each of the multiple candidate voltage regulation levels based on the current condition and the online updated equivalent impedance parameters; determining the target level and switching window from the multiple candidate voltage regulation levels according to preset switching safety constraints, and controlling the linkage disconnection unit to perform level switching within the switching window; verifying the prediction results based on the measured terminal voltage after switching, and executing a degraded power supply guarantee strategy when the verification is abnormal.

[0017] Preferably, the controller identifies the current power supply condition by comprehensively considering the voltage fluctuation amplitude of the upstream line, the voltage deviation at the terminal, the rate of change of the terminal current, and the load state transition information, and determines the current condition as a light-load stable condition, a heavy-load stable condition, a start-up transient condition, or an abnormal disturbance condition; wherein, the light-load stable condition corresponds to a set proportion in which the terminal current is lower than the rated value and the rate of change of the current is lower than a preset threshold; the heavy-load stable condition corresponds to a set proportion in which the terminal current is continuously higher than the rated value; the start-up transient condition corresponds to a rapid increase in the terminal current within a short period of time exceeding a preset rate; and the abnormal disturbance condition corresponds to scenarios such as abnormal phase sequence, phase loss, contactor failure to operate, or mismatch in switching effect.

[0018] Preferably, the recursive parameter identification algorithm is a recursive least squares algorithm, and the controller updates the equivalent resistance of the power supply section online according to the following formula. : ; ; Where k is the current sampling time, The input voltage difference during the k-th sampling period. The current difference in the kth sampling period. It is the forgetting factor and satisfies 0.95≤λ≤1. For error covariance, The gain coefficient is used for recursion; the controller updates the equivalent inductance of the power supply section using a recursive formula of the same format. After each level switch, the controller uses the deviation between the predicted terminal voltage and the measured terminal voltage as a parameter correction value. and Perform iterative corrections.

[0019] Preferably, the preset switching safety constraints include: the current terminal current is lower than a preset switching current threshold, which is 0.2 to 0.6 times the rated current; the duration from the last level switching to the current moment is not less than the minimum dwell time, which is 5 to 120 seconds; the number of round-trip switching between adjacent levels within the preset statistical window does not exceed a set upper limit; the voltage change slope before and after the candidate level switching meets the set upper limit constraint; the switching window can only be entered when all the above constraints are met simultaneously; the linkage interruption unit sequentially executes the current level exit, target level input, and power supply restoration within the switching window, and enters a short-term switching lockout period of 0.2 to 5 seconds after the switching is completed.

[0020] Preferably, the controller uses model predictive control to determine the target level, constructing a predictive model using the online-updated equivalent impedance parameters, and performing prediction in the time domain. Internally solve the following constrained optimization problems: Where J is the model predictive control objective function value, and i is the prediction step number. The target voltage at the end. , , These are the weighting coefficients for the voltage deviation term, current term, and temperature rise term, respectively. To predict the terminal voltage at step k+i, To predict the maximum phase current in k+i steps, The system predicts the temperature rise of power devices in k+i steps. Constraints include: the compensation amount of the candidate stage does not exceed the rated range of the voltage regulator, the current of each phase does not exceed the rated current of the cable, and the temperature rise of the power devices does not exceed the rated limit. The optimization solution results simultaneously output the target stage command and the step-by-step switching sequence of the linkage interruption unit.

[0021] Preferably, the criteria for verifying anomalies include at least one of the following: the terminal voltage fails to recover to the target range within the set verification time after the switch is completed; more than a threshold number of round-trip oscillations occur between adjacent levels within the set statistical time window; the phase-to-phase voltage imbalance exceeds the set upper limit; the disconnection execution result is inconsistent with the target level instruction; the degradation supply protection strategy is executed in the following priority order: disconnecting non-critical loads and maintaining power supply to critical loads; switching to a pre-set safe fixed level for operation; engaging bypass branches to implement bypass direct supply; and performing safe disconnection when a short circuit, phase loss, or contactor failure to operate is detected.

[0022] Preferably, the system further includes a dual-link redundant communication network, comprising a wired fiber optic link and a wireless redundant link, used to transmit the terminal data collected by the terminal measurement unit to the controller; the controller receives terminal voltage data from the wired fiber optic link and the wireless redundant link respectively within a preset time window, performs anomaly removal on the data from each link, and then performs confidence-weighted fusion based on the confidence coefficient of each link. ;in This is a reference value for the weighted fusion of the two-link terminal voltage. This represents the average terminal voltage of the wired link after anomaly removal. This represents the average terminal voltage of the wireless link after anomaly removal. This represents the reliability coefficient of the fiber optic wired link. Here, is the reliability coefficient of the wireless redundant link, where This represents the historical packet loss rate of the fiber optic wired link. Historical packet loss rate of wireless redundant links. The standard deviation of delay jitter in fiber optic wired links. The standard deviation of delay jitter for wireless redundant links is used; when both links are interrupted, the controller switches to local model degraded operation mode, estimates the terminal voltage with the currently updated equivalent impedance parameters, and maintains voltage regulation control.

[0023] Preferably, the controller further includes a distributed condition diagnostic module, which calculates a node health score based on the housing temperature rise sequence, contact resistance change, and voltage sag events in each diagnostic cycle. : ;in To diagnose the number of times the temperature rise exceeds the limit within the diagnostic time window, The equivalent resistance is a significant jump variable in the short time. This is the reference value for contact resistance drift. This represents the number of voltage sag events. , , For weighting coefficients; when A derating alarm is generated when the derating falls below a first preset threshold. When the value falls below a second preset threshold, a safety bypass is triggered and a maintenance request is initiated; the distributed status diagnostic module will... Feedback is sent in reverse to the voltage regulation control logic, according to The current rise slope constraint value is linearly reduced by the ratio of the drift trigger threshold to prevent thermal runaway at poor contact points due to large current surges.

[0024] Preferably, the device also includes a safety state machine module, which is implemented using independent hardware logic. Its input terminal is directly hardwired to the wind power interlock dry contact signal of the local ventilation fan at the tunneling face and the methane concentration over-limit electrical interlock dry contact signal. When the wind power interlock dry contact signal or the methane concentration over-limit electrical interlock dry contact signal is received, the safety state machine module blocks all control pulses of the relay pressure regulating execution unit in hardware within 10ms, and simultaneously closes and locks the bypass contactor, causing the device to enter a forced power-off safety state. The establishment of the forced power-off safety state does not depend on the software execution path of the controller. The safety state machine module records the triggering reason, triggering time, and duration of each interlocking action and reports it to the controller.

[0025] This invention also provides a method for intelligent long-distance power supply relay voltage regulation at a tunnel face, applied to an intelligent long-distance power supply relay voltage regulation device at a tunnel face as described above, specifically including the following steps:

[0026] S1. Collect the electrical status quantities of the upstream line, the electrical quantities of the terminal line, and the load status quantities;

[0027] S2. Identify the current power supply status based on the collected data;

[0028] S3. The recursive parameter identification algorithm updates the equivalent impedance parameters of the power supply section online based on the current sampled data;

[0029] S4. Based on the current operating conditions and the online updated equivalent impedance parameters, predict the available terminal voltage for each of the multiple candidate voltage regulation stages;

[0030] S5. Determine the target voltage regulation level and switching window from multiple candidate voltage regulation levels according to the preset switching safety constraints, and control the linkage interruption unit to perform level switching within the switching window;

[0031] S6. Verify the prediction results based on the measured terminal voltage after switching, and use the deviation between the predicted terminal voltage and the measured terminal voltage as a parameter correction amount to iteratively correct the equivalent impedance parameter.

[0032] S7. When the verification result meets the preset anomaly criteria, the downgrade supply guarantee strategy is executed.

[0033] Beneficial Effects: The intelligent long-distance power supply relay voltage regulation device and method for tunneling faces provided by this invention have the following beneficial effects:

[0034] (1) High voltage regulation accuracy and strong parameter self-adaptation capability. The recursive parameter identification algorithm tracks the line impedance drift caused by tunneling in real time, and the iterative correction after each switch further eliminates model errors. The steady-state end voltage deviation target can be controlled within ±2% of the rated value.

[0035] (2) High switching safety and fundamental suppression of stage oscillation. The multi-constraint switching window combines the four dimensions of current state, time interval, round trip frequency and slope to constrain the switching timing. Compared with the traditional single threshold scheme, it fundamentally eliminates stage round trip oscillation and significantly reduces the contactor operation frequency and the probability of contact arcing.

[0036] (3) Voltage regulation decision takes into account both voltage quality and equipment safety. The model predicts the control objective function to simultaneously optimize the terminal voltage deviation, phase current and device temperature rise. The constraints include the rated current of the cable and the temperature rise limit of the power device, so as to avoid introducing new equipment damage risks in pursuit of voltage regulation effect.

[0037] (4) Strong power supply continuity and orderly degradation transition. The four-level degradation supply guarantee strategy (cutting off non-critical loads - fixed safety level - bypass direct supply - safe disconnection) forms an orderly abnormal handling echelon, which can still maintain the power supply of critical loads when the voltage regulation link fails locally, and is suitable for engineering scenarios where the downtime cost of tunneling face is high.

[0038] (5) Strong communication fault tolerance. Confidence-weighted fusion utilizes the data quality differences between the two links for adaptive fusion, and does not reduce the accuracy of voltage regulation when a single link fails; when both links are interrupted at the same time, local degradation operation ensures the continuity of voltage regulation.

[0039] (6) Predictive maintenance levels have significantly improved. Health score Quantitatively reflects the combined degradation trend of contact resistance, temperature rise, and sag events; contact resistance drift. Real-time feedback to the current slope constraint proactively reduces the risk of inrush current before a fault occurs, upgrading the maintenance mode from reactive repair to preventative proactive intervention.

[0040] (7) Strong safety compliance. The independent hardware safety state machine does not rely on the controller software to respond to wind power interlock and methane power interlock signals. It completes the establishment of the forced power-off safety state within 10ms, meets the hardware-level mandatory requirement of the coal mine safety regulations to disconnect all power, and retains complete interlock event records to support safety audits. Attached Figure Description

[0041] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the accompanying drawings used in the description of the embodiments or the prior art will be briefly introduced below.

[0042] Figure 1 This is a schematic diagram of the overall architecture of the device of the present invention;

[0043] Figure 2 This is a schematic diagram of the electrical topology of the relay voltage regulation execution unit and the linkage disconnection unit in this invention;

[0044] Figure 3This is a flowchart of the intelligent voltage regulation control logic of the controller of the present invention;

[0045] Figure 4 This is a schematic diagram of the logic for determining the level switching window and verifying the switching process in this invention.

[0046] Figure 5 This is a schematic diagram of the distributed state diagnosis and degradation supply guarantee path of the present invention. Detailed Implementation

[0047] The present invention will now be described more clearly and completely by way of a preferred embodiment in conjunction with the accompanying drawings, but this does not limit the invention to the scope of the described embodiment.

[0048] See Figure 1 The intelligent long-distance power supply relay voltage regulation device for tunneling faces of the present invention comprises five cooperating units: an input isolation and status acquisition unit, a relay voltage regulation execution unit, an end measurement unit, a linkage disconnection unit, and a controller.

[0049] The input isolation and status acquisition unit includes an upstream incoming line interface, a voltage and current transmitter group, a phase sequence detection module, and a switch status acquisition module. The voltage and current transmitter group acquires the three-phase voltage and line current of the upstream line at a sampling rate of not less than 1kHz; the phase sequence detection module determines in real time whether there are phase sequence errors or phase loss; the switch status acquisition module acquires the operating status of the upstream power supply circuit and each contactor inside the device. The above data are sent in real time to the controller's operating condition identification module and recursive parameter identification module.

[0050] The end-point measurement unit includes an end-point voltage acquisition module, an end-point current acquisition module, and a load status acquisition module. The end-point voltage acquisition module acquires the three-phase voltage at the output side of the relay voltage regulating execution unit, with a sampling rate of not less than 1kHz; the end-point current acquisition module acquires the three-phase current at the output side; the load status acquisition module acquires at least one load operating status quantity, which can be one or more of the following: tunneling machine start / stop command, inverter operating status, motor start flag, load priority, or load group switching status. The data acquired by the end-point measurement unit is transmitted to the controller via a dual-link redundant communication network.

[0051] The controller's internal logic is divided into a working condition identification module, a recursive parameter identification module, a voltage prediction and constraint decision module, a switching verification module, a fault degradation module, a distributed state diagnosis module, and a safety state machine module.

[0052] See Figure 2The relay voltage regulation execution unit includes a three-phase voltage regulator, a stage switching component, and a bypass branch. The three-phase voltage regulator adopts a multi-tap autotransformer structure with multiple fixed voltage regulation stages, each stage corresponding to a fixed output voltage boost ratio; the stage switching component consists of stage contactors corresponding to each tap; the bypass branch is connected in parallel with the voltage regulation link and is used to bypass the voltage regulator and directly supply power in the event of a voltage regulation link failure or maintenance.

[0053] The interlocking switching unit comprises a pre-charge contactor, a main contactor, and a set of stage contactors, which work together to achieve a step-by-step, orderly switching process. During stage switching, the controller first disconnects the main contactor to break the main circuit, then switches to the target stage tap via the set of stage contactors, and finally, after buffering the new circuit with the pre-charge contactor, closes the main contactor to restore power supply. The entire switching process is step-by-step and orderly, effectively reducing the amplitude of the switching inrush current and laying the foundation for the verification of the end voltage after switching.

[0054] See Figure 3 The controller executes the voltage regulation control logic according to the following steps:

[0055] S1 (Data Acquisition): Each sampling module samples synchronously according to a set period. The acquired data is aggregated to the controller via a dual-link redundant communication network for fusion processing. The operating condition identification and parameter identification modules use the latest batch of valid fused data as the calculation input.

[0056] S2 (Operating Condition Identification): The operating condition identification module integrates the voltage fluctuation amplitude of the upstream line, the voltage deviation at the end, the rate of change of the current at the end, and the load state transition information to determine the current operating condition into one of the following four categories.

[0057] Light load stable operating condition: The terminal current is lower than a certain percentage of the rated value (e.g., 40%), the terminal voltage deviation is within the target range, the current change rate is lower than the preset threshold, and there is no load start / stop command.

[0058] Heavy load stable operating condition: The terminal current is higher than the rated value by a certain percentage (e.g., 70%) and the duration exceeds the set threshold. The terminal voltage deviation may be at the lower edge of the target range, but the current change rate is lower than the preset threshold.

[0059] Start-up transient condition: The load status acquisition module detects a motor start command or the frequency converter switching from stop to run, and the terminal current rises rapidly within a short period of time, with the rate of rise exceeding the preset threshold.

[0060] Abnormal disturbance conditions: The phase sequence detection module detects an abnormal phase sequence or a missing phase, or the switching verification module reports a verification failure, or the switch status acquisition module detects that the contactor action result is inconsistent with the command.

[0061] The operating condition type serves as a correction factor for subsequent candidate stage predictions in the input voltage prediction and constraint decision module. For example, under the transient operating condition at startup, the lower limit tolerance of the terminal voltage can be appropriately relaxed to avoid triggering unnecessary stage switching during the startup impact.

[0062] S3 (Online Parameter Identification): The recursive parameter identification module is based on the voltage difference in each sampling period. and current difference The equivalent resistance of the power supply section is updated online using a recursive least squares algorithm. and equivalent inductance .

[0063] Equivalent resistance The update formula is as follows:

[0064]

[0065]

[0066]

[0067] Where k is the current sampling time; This is the input-side voltage difference; λ is the current difference; λ is the forgetting factor, which satisfies 0.95≤λ≤1. The smaller λ is, the greater the weighting of recent data, which is suitable for working conditions where parameters change rapidly. To estimate the error covariance, reflecting the uncertainty of the current estimate, the equivalent inductance is used. The same recursive formula is used, and the update is based on the voltage differential component; the formula will not be repeated here.

[0068] Initial values ​​for online identification The parameters can be set based on the cable type and the estimated current tunneling distance to shorten the convergence time. After each level switch, the switching verification module will predict the terminal voltage. Compared with the measured terminal voltage Deviation between As a parameter correction quantity, for and Perform an iterative correction to ensure that the parameter estimates continuously reflect actual working conditions.

[0069] S4 (Candidate Level Prediction): Voltage prediction and constraint decision module calculates voltage from the input side. The equivalent compensation amount corresponding to candidate level k Based on the current working conditions and Estimated line voltage drop Calculate the predicted terminal voltage for each candidate stage:

[0070]

[0071] In a preferred embodiment, the voltage prediction and constraint decision module employs model predictive control. and Build a prediction model in the prediction time domain The following objective function is optimized and solved internally:

[0072]

[0073] In the formula: The target voltage at the end is set by the operator according to the load requirements; , , These are the weighting coefficients for the voltage deviation term, current term, and temperature rise term, respectively. The terminal voltage is predicted for the (k+i)th step, which is obtained by extending the terminal voltage prediction formula to multi-step prediction. To predict the maximum phase current; To predict the device temperature rise, the following constraints are applied:

[0074] (a) The candidate level compensation amount satisfies (Voltage regulator rated range);

[0075] (b) The currents in each phase satisfy (Cable rated current);

[0076] (c) The temperature rise of the power device meets the requirements. (Rated temperature rise limit).

[0077] The optimized solution results are output along with the step-by-step switching timing of the target level bit command and the linkage switching unit, for use in the next step of switching window determination.

[0078] S5 (Window Switching Decision and Level Switching): See [link] Figure 4 Before entering the allowed switching window, the voltage prediction and constraint decision module must simultaneously meet the following five conditions:

[0079] Condition (a): The current terminal current is lower than the switching current threshold. , ,in The coefficient is between 0.2 and 0.6 to prevent excessive arcing during heavy load switching.

[0080] Condition (b): The time interval from the last handover to the current time. , The value range is from 5 seconds to 120 seconds to avoid mechanical wear caused by frequent switching in a short period of time.

[0081] Condition (c): In the preset statistics window Within, the number of round trips between adjacent levels To prevent level oscillation.

[0082] Condition (d): Estimated slope of the end voltage change before and after candidate stage switching This is to prevent voltage surges from impacting end devices.

[0083] Condition (e): Phase consistency is satisfied, that is, the phase difference between the output voltage corresponding to the candidate stage and the current circuit phase does not exceed the allowable range.

[0084] When all five conditions above are met simultaneously, the controller sends a switching command to the linkage interruption unit, which executes in the following order: disconnect the main contactor → switch the stage contactor group to the target stage → close the pre-charge contactor for buffer connection → close the main contactor to restore the main circuit power supply. After the switching is completed, a short switching lockout period of 0.2 seconds to 5 seconds is entered, during which stage switching is prohibited from being triggered again.

[0085] S6 (Post-switch verification and parameter iterative correction): See [link] Figure 4 The switching verification module continuously collects the measured end voltage during the verification period (50 to 500 milliseconds) after the switching is completed. Determine whether any of the following preset anomaly criteria are met:

[0086] (i) The terminal voltage did not recover to the target range within the verification period. ;

[0087] (ii) If the number of round trips between adjacent levels within the set statistical window exceeds the threshold, it is determined to be level oscillation;

[0088] (iii) The phase-to-phase voltage imbalance exceeds the set upper limit;

[0089] (iv) The execution result of the switch (feedback from the switch status acquisition module) is inconsistent with the target level bit instruction.

[0090] If the verification passes (no abnormality criteria triggered), switch the verification module to adjust the prediction deviation. Feedback is sent to the recursive parameter identification module for... and Perform iterative corrections and continue operating at the current level. If any exception criterion is triggered, proceed to step S7.

[0091] S7 (Abnormal Degradation Supply Guarantee): See [link] Figure 5 The fault degradation module attempts degradation paths in the following priority order based on the exception type and the current system state:

[0092] Degradation Path 1: Cut off non-critical loads while maintaining continuous power supply to critical loads (including tunneling machines, safety monitoring equipment, etc.). If the end voltage recovers to an acceptable range for the critical loads after this operation, maintain this state and generate an alarm.

[0093] Degradation Path 2: If the critical load is still undervoltage after cutting off non-critical loads, switch to the pre-set safety fixed level. This level is pre-calibrated according to the maximum expected line length when the equipment is put into operation, ensuring that basic power supply can still be maintained under the worst operating conditions.

[0094] Degradation path 3: If the stage selection link fails (the stage contactor group fails to operate or becomes stuck), the bypass branch is activated to implement bypass direct power supply, bypassing the voltage regulation link to directly supply power to the end.

[0095] Degradation path 4: If a short circuit, phase loss, or upstream power supply failure is detected, which does not meet the conditions for safe operation, a safety disconnection is performed, the main contactor is disconnected to cut off the power supply, and the fault event is reported.

[0096] Throughout the degradation process, the fault degradation module continuously records the anomaly type, trigger time, degradation measures taken, and end voltage recovery status, forming a traceable event log.

[0097] See Figure 1 The dual-link redundant communication network consists of a fiber optic wired link laid along the tunnel and an industrial wireless redundant link. The two links are independent at the physical layer and serve as hot backups for each other.

[0098] The controller employs the following fusion processing procedure for dual-link data:

[0099] First, within a preset time window T (typically 100ms to 500ms), terminal voltage and current data packets are received from both the wired fiber optic link and the wireless redundant link. Outlier removal based on the 3σ criterion is performed independently on the data from each link, and the mean of the valid data is calculated for each link. and .

[0100] Secondly, based on the historical packet loss rate of each link and latency jitter Calculate the credibility coefficient:

[0101]

[0102]

[0103] Next, the two terminal voltages are weighted and fused according to confidence level:

[0104]

[0105] The As the input of the terminal voltage reference value for the switching verification module and the recursive parameter identification module; This represents the average terminal voltage of the wired link after anomaly removal. This represents the average terminal voltage of the wireless link after anomaly removal. This represents the reliability coefficient of the fiber optic wired link. Here, is the reliability coefficient of the wireless redundant link, where This represents the historical packet loss rate of the fiber optic wired link. Historical packet loss rate of wireless redundant links. The standard deviation of delay jitter in fiber optic wired links. The standard deviation of delay jitter for wireless redundant links.

[0106] When the fiber optic wired link is interrupted Setting it to zero relies entirely on data from the wireless redundant link; when the wireless redundant link fails simultaneously... Also set to zero, the controller switches to local model degraded operation mode, using the currently identified recursive parameters maintained by the module. and Estimate the terminal voltage, maintain basic voltage regulation control, and trigger a dual-link interruption alarm to prompt maintenance personnel for handling. After the link is restored, and data consistency verification passes for K consecutive time windows, the system automatically exits the degradation mode and resumes fusion computing.

[0107] The distributed status diagnosis module calculates the node health score by combining the following three types of signals within each diagnosis cycle (typically 1 minute). :

[0108] The first type of signal is the number of times the casing temperature rise exceeds the limit. The first type of signal is obtained by comparing the housing temperature rise sequence collected by the temperature sensor with the upper limit threshold of the temperature rise; the second type of signal is the contact resistance drift. The output of the recursive parameter identification module The short-term significant jump variable is obtained by identifying short-term significant jump variables in the sequence. A short-term significant jump variable is defined as a jump variable within a diagnostic period. The difference between the maximum and minimum values ​​exceeds a preset threshold; the third type of signal is the number of voltage sag events. The number of events where the terminal voltage is lower than 90% of the rated value and the duration exceeds 10ms is counted according to the standard definition of voltage sag.

[0109] Health score calculation formula:

[0110]

[0111] in The reference value for contact resistance drift is 5mΩ (typically). α, β, and γ are weighting coefficients with factory default values ​​of α=2.0, β=30.0, and γ=1.5, respectively. They can be adaptively adjusted based on historical maintenance data from the field. A higher score indicates a better node condition, with a maximum score of 100 corresponding to no degradation events.

[0112] when When the value falls below the first preset threshold (typically 60), the controller generates a derating alarm, prompting maintenance personnel to schedule upcoming repairs; when When the value falls below the second preset threshold (typically 30), the controller triggers the safety bypass (activates the bypass branch) and requests emergency maintenance from the superior system.

[0113] The distributed state diagnosis module will simultaneously include the current The numerical values ​​are fed back to the voltage prediction and constraint decision module, and the maximum allowable current rise slope constraint value is adjusted according to the following formula:

[0114]

[0115] in This is the current maximum allowable current rise rate constraint value. This represents the maximum allowable current rise slope under rated operating conditions. Adjust the trigger threshold for the slope. When When it increases, The linear reduction in current allows for stricter current slope constraints during MPC solving, preventing thermal runaway at poor contact points due to large current surges. The above formulas constitute a closed-loop feedback between the diagnostic results and the control strategy, representing the core embodiment of the deep coupling between predictive maintenance and proactive safety control in this invention.

[0116] The safety state machine module is implemented with independent hardware logic, which is independent of the software execution path of the controller. Its physical input terminal is directly hardwired to the wind power interlock dry contact signal of the local ventilation fan in the tunneling face and the over-limit power interlock dry contact signal of the methane concentration sensor through dedicated terminals.

[0117] Under normal operating conditions, the safety state machine module continuously monitors the electrical status of the two interlocking dry contacts. When a valid wind power interlock signal (triggered by local ventilator power outage) or a valid methane electrical interlock signal (triggered by methane concentration exceeding a specified threshold) is received, the hardware logic of the safety state machine module completes the following sequence of operations within 10ms:

[0118] Operation 1: Apply a hardware blocking signal to the control pulse drive circuit of the relay voltage regulation execution unit to block all drive pulses of the level switching component;

[0119] Operation 2: Apply a closing signal to the bypass contactor drive coil of the bypass branch to engage the bypass branch and maintain the closed state through the hardware self-locking contact;

[0120] Operation 3: Apply a disconnect signal to the main contactor to cut off the power supply to the main circuit;

[0121] Operation 4: Send a hardware interrupt signal to the event logging interface of the controller to report the interlock trigger type (wind power interlock or methane power interlock), trigger time (taken from the hardware clock) and duration.

[0122] Operations 1 to 3 described above are all implemented through hardware logic, without going through any software execution path of the controller. Therefore, even if the controller fails to function properly due to program abnormalities or communication interruptions, the safety state machine module can still independently establish a forced power-off safety state. After the forced power-off safety state is established, the device maintains this state until the interlock signal is released and the maintenance personnel manually reset it through a dedicated confirmation operation.

[0123] The safety state machine module has a built-in non-erasable memory that permanently records the trigger type, time, and duration of each locking action, with a record capacity of no less than 1,000 entries, and supports on-site safety audit queries.

[0124] With a power supply voltage of 1140V (three-phase AC, 50Hz) and a cable length of 1200m (cross-sectional area of ​​35mm²), 2 Using a copper-core rubber-sheathed flexible cable with a single-core resistance of approximately 0.57 Ω / km and a comprehensive power factor of 0.75 as the baseline operating condition, a test scenario was set up with a roadheader (rated power of 300 kW, actual power to be supplemented by the applicant according to on-site parameters) directly starting and impacting the load. The device of this invention was compared with the uncompensated baseline scheme and the traditional threshold-type relay voltage regulation scheme. Simulation verification was performed using the MATLAB / Simulink electromagnetic transient simulation platform, and the cable adopted a segmented π-type equivalent model; the power quality index measurement method referred to GB / T17626.30-2023.

[0125] Table 1. Performance comparison results of the device of the present invention, the uncompensated reference scheme, and the traditional threshold relay voltage regulation scheme.

[0126] steady-state terminal voltage deviation Approximately -12% to -18% Approximately ±5% ≤±2% (target value) Startup transient period handover error probability — Higher (no operating condition classification) Significantly reduced (four categories) Adjacent level oscillation — Yes, the contactor operates frequently. Subject to multi-constraint window suppression Unrecovery rate of terminal voltage after switching — No verification mechanism Automatic downgrade after verification Voltage regulation capability during communication interruption — none Local model downgrade maintenance Early warning of contact resistance degradation none none <![CDATA[H score Real-time assessment Safety interlock response time — Software dependent, >50ms Hardware level ≤10ms

[0127] The above indicators are design target values, and the final results will be based on the test results of the physical prototype under typical tunneling face conditions.

[0128] The above description is only a preferred embodiment of the present invention. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the principle of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A smart long-distance power supply relay voltage regulating device for a tunnel face, installed between the upstream power supply and the end load, characterized in that, include: The input isolation and status acquisition unit is connected to the upstream power supply and is used to acquire the electrical status of the upstream line. The relay voltage regulation execution unit has its input terminal connected to the input isolation and status acquisition unit, and its output terminal facing the end load. It is used to regulate the output voltage between multiple voltage regulation stages. The terminal measurement unit is located on the output side of the relay voltage regulating unit and is used to collect terminal electrical quantities and load status quantities. The linkage interruption unit is connected in series with the output circuit of the relay voltage regulation execution unit and is used to perform interruption and switching operations during the level switching process; The controller is connected to the input isolation and status acquisition unit, the relay voltage regulation execution unit, the terminal measurement unit, and the linkage disconnection unit, respectively, and is used to: identify the current power supply condition based on the acquired electrical status quantities of the upstream line and the electrical quantities of the terminal line; Based on the recursive parameter identification algorithm, the equivalent impedance parameters of the power supply section are updated online according to the current sampled data. Based on the current operating conditions and the online updated equivalent impedance parameters, the available terminal voltages corresponding to multiple candidate voltage regulation levels are predicted. According to the preset switching safety constraints, the target level and switching window are determined from multiple candidate voltage regulation levels, and the linkage interruption unit is controlled to perform level switching within the switching window. The prediction results are checked based on the measured terminal voltage after switching, and a degraded power supply guarantee strategy is executed when the check is abnormal.

2. The intelligent long-distance power supply relay voltage regulation device for a tunnel face according to claim 1, characterized in that, The controller identifies the current power supply condition by comprehensively considering the voltage fluctuation amplitude of the upstream line, the voltage deviation at the terminal, the rate of change of the terminal current, and the load state transition information, and determines the current condition as a light-load stable condition, a heavy-load stable condition, a start-up transient condition, or an abnormal disturbance condition. Among them, the light-load stable condition corresponds to a terminal current lower than the rated value by a set proportion and the current change rate lower than a preset threshold; the heavy-load stable condition corresponds to a terminal current continuously higher than the rated value by a set proportion; the start-up transient condition corresponds to a terminal current rapidly increasing beyond a preset rate within a short period of time; and the abnormal disturbance condition corresponds to scenarios such as abnormal phase sequence, phase loss, contactor failure to operate, or mismatch in switching effect.

3. The intelligent long-distance power supply relay voltage regulation device for a tunnel face according to claim 1, characterized in that, The recursive parameter identification algorithm is a recursive least squares algorithm, and the controller updates the equivalent resistance of the power supply section online according to the following formula. : ; ; Where k is the current sampling time, The input voltage difference during the k-th sampling period. The current difference in the kth sampling period. It is the forgetting factor and satisfies 0.95≤λ≤1. For error covariance, The gain coefficient is used for recursion; the controller updates the equivalent inductance of the power supply section using a recursive formula of the same format. After each level switch, the controller uses the deviation between the predicted terminal voltage and the measured terminal voltage as a parameter correction value. and Perform iterative corrections.

4. The intelligent long-distance power supply relay voltage regulation device for a tunnel face according to claim 1, characterized in that, The preset switching safety constraints include: the current terminal current is lower than the preset switching current threshold, which is 0.2 to 0.6 times the rated current; the duration from the last level switching to the current moment is not less than the minimum dwell time, which is 5 to 120 seconds; the number of round-trip switching between adjacent levels within the preset statistical window does not exceed the set upper limit; the voltage change slope before and after the candidate level switching meets the set upper limit constraint; the switching window can only be entered when all the above constraints are met simultaneously; the linkage interruption unit sequentially executes the current level exit, target level input, and power supply restoration within the switching window, and enters a short-term switching lockout period of 0.2 to 5 seconds after the switching is completed.

5. The intelligent long-distance power supply relay voltage regulation device for a tunnel face according to claim 1, characterized in that, The controller uses model predictive control to determine the target level, constructing a predictive model based on the online-updated equivalent impedance parameters, in the prediction time domain. Internally solve the following constrained optimization problems: Where J is the model predictive control objective function value, and i is the prediction step number. The target voltage at the end. , , These are the weighting coefficients for the voltage deviation term, current term, and temperature rise term, respectively. To predict the terminal voltage at step k+i, To predict the maximum phase current in k+i steps, The system predicts the temperature rise of power devices in k+i steps. Constraints include: the compensation amount of the candidate stage does not exceed the rated range of the voltage regulator, the current of each phase does not exceed the rated current of the cable, and the temperature rise of the power devices does not exceed the rated limit. The optimization solution results simultaneously output the target stage command and the step-by-step switching sequence of the linkage interruption unit.

6. The intelligent long-distance power supply relay voltage regulation device for a tunnel face according to claim 1, characterized in that, The criteria for verifying anomalies include at least one of the following: the terminal voltage fails to recover to the target range within the set verification time after the switch is completed; more than a threshold number of round-trip oscillations occur between adjacent levels within the set statistical time window; the phase-to-phase voltage imbalance exceeds the set upper limit; the disconnection execution result is inconsistent with the target level instruction; the degradation supply protection strategy is executed in the following priority order: disconnecting non-critical loads and maintaining power supply to critical loads; switching to a pre-set safe fixed level for operation; engaging bypass branches to implement bypass direct supply; and performing safe disconnection when a short circuit, phase loss, or contactor failure to operate is detected.

7. The intelligent long-distance power supply relay voltage regulation device for a tunnel face according to claim 1, characterized in that, It also includes a dual-link redundant communication network, which includes a fiber optic wired link and a wireless redundant link, used to transmit the terminal data collected by the terminal measurement unit to the controller. The controller receives end voltage data from the fiber optic wired link and the wireless redundant link respectively within a preset time window, removes anomalies from the data of each link, and then performs a confidence-weighted fusion based on the confidence coefficient of each link. ;in This is a reference value for the weighted fusion of the two-link terminal voltage. This represents the average terminal voltage of the wired link after anomaly removal. This represents the average terminal voltage of the wireless link after anomaly removal. This represents the reliability coefficient of the fiber optic wired link. Here, is the reliability coefficient of the wireless redundant link, where This represents the historical packet loss rate of the fiber optic wired link. Historical packet loss rate of wireless redundant links. The standard deviation of delay jitter in fiber optic wired links. The standard deviation of delay jitter for the wireless redundant links is used; when both links are interrupted, the controller switches to the local model degraded operation mode, estimates the terminal voltage with the currently updated equivalent impedance parameters, and maintains voltage regulation control.

8. The intelligent long-distance power supply relay voltage regulation device for a tunnel face according to claim 1, characterized in that, The controller also includes a distributed condition diagnostic module, which calculates a node health score based on the housing temperature rise sequence, contact resistance change, and voltage sag events in each diagnostic cycle. : ;in To diagnose the number of times the temperature rise exceeds the limit within the diagnostic time window, The equivalent resistance is a significant jump variable in the short time. This is the reference value for contact resistance drift. This represents the number of voltage sag events. , , For weighting coefficients; when A derating alarm is generated when the derating falls below a first preset threshold. When the value falls below a second preset threshold, a safety bypass is triggered and a maintenance request is initiated; the distributed status diagnostic module will... Feedback is sent in reverse to the voltage regulation control logic, according to The current rise slope constraint value is linearly reduced by the ratio of the drift trigger threshold to prevent thermal runaway at poor contact points due to large current surges.

9. The intelligent long-distance power supply relay voltage regulation device for a tunnel face according to claim 1, characterized in that, It also includes a safety state machine module, which is implemented with independent hardware logic. Its input terminal is directly hardwired to the wind power interlock dry contact signal of the local ventilation fan at the tunneling face and the methane concentration over-limit electrical interlock dry contact signal. When the wind power interlock dry contact signal or the methane concentration over-limit electrical interlock dry contact signal is received, the safety state machine module blocks all control pulses of the relay pressure regulating execution unit in hardware within 10ms, and at the same time closes and locks the bypass contactor, so that the device enters the forced power-off safety state. The establishment of the forced power-off safety state does not depend on the software execution path of the controller. The safety state machine module records the triggering reason, triggering time and duration of each interlocking action and reports it to the controller.

10. A method for intelligent long-distance power supply relay voltage regulation at a tunnel face, characterized in that, The intelligent long-distance power supply relay voltage regulation device for tunnel faces, which includes an input isolation and status acquisition unit, a relay voltage regulation execution unit, an end measurement unit, a linkage switching unit, and a controller, specifically includes the following steps: S1. Collect the electrical status quantities of the upstream line, the electrical quantities of the terminal line, and the load status quantities; S2. Identify the current power supply status based on the collected data; S3. The recursive parameter identification algorithm updates the equivalent impedance parameters of the power supply section online based on the current sampled data; S4. Based on the current operating conditions and the online updated equivalent impedance parameters, predict the available terminal voltage for each of the multiple candidate voltage regulation stages; S5. Determine the target voltage regulation level and switching window from multiple candidate voltage regulation levels according to the preset switching safety constraints, and control the linkage interruption unit to perform level switching within the switching window; S6. Verify the prediction results based on the measured terminal voltage after switching, and use the deviation between the predicted terminal voltage and the measured terminal voltage as a parameter correction amount to iteratively correct the equivalent impedance parameter. S7. When the verification result meets the preset anomaly criteria, the downgrade supply guarantee strategy is executed.

Images