Aluminum electrolysis cell series ground fault detection method and system based on multi-data fusion

Abstract

The application discloses a kind of aluminum electrolytic cell series ground fault detection method and system based on multi-data fusion, the method of the present application includes to aluminum electrolytic cell series Load AC excitation signal;Synchronous acquisition aluminum electrolytic cell series DC voltage data, AC voltage data and bus current data;Respectively based on DC voltage data, AC voltage data and bus current data carry out aluminum electrolytic cell series ground fault diagnosis to obtain corresponding diagnostic results;Three kinds of diagnostic results are fused using Dempster-Shafer data fusion method to obtain the final ground fault detection result of aluminum electrolytic cell series.The present application aims to improve the reliability of electrolytic cell series insulation diagnosis, reduce the uncertainty of monitoring results, correctly warn electrolytic cell series ground fault, and accurately locate the electrolytic cell with ground fault.

Description

Technical Field

[0001] This invention relates to grounding fault detection technology in the field of aluminum electrolysis process, specifically to a method and system for detecting grounding faults in aluminum electrolysis cells based on multi-data fusion. Background Technology

[0002] Aluminum electrolytic cells utilize an ungrounded DC power supply system. Due to their harsh operating environment, grounding faults are unavoidable in the electrolytic cell series, such as insulation damage at the busbar and cell support, or contact between the cell shell and the ground. These faults are typically multi-point, causing severe zero-point drift, increased losses, decreased production efficiency, and threats to equipment and personnel, resulting in significant losses for aluminum electrolysis enterprises. Therefore, conducting grounding fault detection in aluminum electrolytic cells and providing real-time early warning of grounding faults is crucial for ensuring safe production and improving the level of intelligent operation in aluminum electrolysis production.

[0003] Existing methods for detecting the insulation resistance of electrolytic cells to ground mainly include: (1) insulation resistance detection method, which uses a multimeter to directly detect the insulation resistance of the electrolytic cell to ground, thereby determining whether each cell has a fault; (2) cell voltage detection method, which uses a PLC to monitor the voltage at both ends of each electrolytic cell and calculates and analyzes the approximate location of the fault point; (3) zero-point drift detection method, which finds the series of zero-point positions after a fault occurs, thereby determining the location of the fault. However, the insulation resistance detection method can only measure a small range of grounding resistance, cannot measure in the case of metallic grounding, and has a high labor intensity; both the cell voltage detection method and the zero-point drift detection method can accurately measure the location of a single-point grounding fault, but cannot accurately measure in the case of two or more points of grounding. In terms of online monitoring of the insulation resistance of electrolytic cell series to ground, existing technologies use the principle of bridge balance to form multiple sets of bridges in the electrolytic cell series circuits, and determine the fault location of the aluminum electrolytic cell by measuring the amplitude and direction change characteristics of the bridge current. This method involves too many wirings and requires high monitoring sensitivity, making it difficult to implement in practice. Existing technologies inject AC signals into a series of electrolytic cells, install voltage sensors on the cells, and measure the response voltage of the series of electrolytic cells under AC excitation. By comparing and analyzing the cell voltages of adjacent electrolytic cells, grounding faults in the series of electrolytic cells can be detected and warned. However, the electromagnetic environment at the monitoring site is harsh, and the monitoring system units, including sensors and data communication systems, are subjected to severe tests of strong magnetic fields, high dust, and high temperatures. The operating conditions of the electrolytic cells are complex and variable, and the dynamic changes and dispersion of the cell voltage detection data under AC excitation are large. The algorithm model constructed from the series of cell voltages is difficult to handle the impact of uncertain changes in the detection data, which easily leads to missed detections and false alarms. Existing technologies propose a grounding fault detection method for electrolytic cells based on bus AC current detection. This method installs current sensors at the same location on the bus of each series of electrolytic cells, monitors the bus current under AC power excitation, and then compares and analyzes the difference in bus AC current between adjacent electrolytic cells to determine whether a grounding fault exists in the electrolytic cells. This method is not affected by the operating conditions of the electrolytic cells, i.e., the cell resistance, but it is affected by the sensor installation location and on-site electromagnetic interference, and the diagnosis may be misjudged.

[0004] In summary, the existing methods for detecting grounding faults in aluminum electrolytic cells mainly have the following problems: (1) The aluminum electrolytic cell series is exposed to strong magnetic fields, high temperatures, and high dust environments, and the electrolytic cell operating conditions are complex and changeable. The existing online insulation monitoring methods all have large uncertainties in the monitoring results. Using data from a single type of sensor for insulation diagnosis is prone to false judgments and false alarms, resulting in low reliability of the monitoring results. (2) The electrolytic cell voltage detection method and the zero-point drift detection method are relatively accurate in locating fault points when there is a single-point grounding fault, but they cannot accurately determine the situation of multi-point grounding faults. The ground insulation strength detection method requires manual use of a megohmmeter to detect all voltages in the series one by one, and the voltage of each electrolytic cell itself already has errors, which cannot be eliminated by this method. Due to the special nature of the DC power supply system of the aluminum electrolytic cell series, online monitoring of the aluminum electrolytic cell series is still in its initial stage. How to improve the accuracy of insulation fault diagnosis of the electrolytic cell series has become a key technical problem that urgently needs to be solved. Summary of the Invention

[0005] The technical problem to be solved by the present invention is to provide a method and system for detecting grounding faults in aluminum electrolytic cells based on multi-data fusion, in order to address the above-mentioned problems in the prior art. The present invention aims to improve the reliability of insulation diagnosis of electrolytic cells, reduce the uncertainty of monitoring results, correctly warn of grounding faults in electrolytic cells, and accurately locate electrolytic cells with grounding faults.

[0006] To solve the above-mentioned technical problems, the technical solution adopted by the present invention is as follows: A method for detecting grounding faults in aluminum electrolytic cells based on multi-data fusion includes the following steps: S101, Apply an AC excitation signal to the aluminum electrolysis cell series; S102, synchronously collects DC voltage data, AC voltage data and bus current data of aluminum electrolysis cell series; S103, based on DC voltage data, AC voltage data and bus current data, performs a series of grounding fault diagnosis for aluminum electrolytic cells to obtain corresponding diagnostic results; S104 uses the Dempster-Shafer data fusion method to fuse the three diagnostic results to obtain the final grounding fault detection result for the aluminum electrolytic cell series.

[0007] Optionally, in step S102, the synchronously acquired DC voltage data of the aluminum electrolytic cell series includes the DC voltage of each electrolytic cell in the aluminum electrolytic cell series. , , The number of electrolytic cells; in step S103, the series grounding fault diagnosis of aluminum electrolytic cells based on DC voltage data includes: S201, obtain DC voltage data for multiple recent batches of aluminum electrolytic cells; S202, Calculate the average DC voltage of each electrolytic cell based on DC voltage data from multiple batches of aluminum electrolytic cells. and DC voltage standard deviation ; S203, according to Calculate the mean deviation of DC voltage in each electrolytic cell. ,in and These are the average DC voltage values ​​of the m-th and m-1-th electrolytic cells, respectively. S204, according to Calculate the uncertainty of each electrolytic cell ; S205, based on the average deviation of DC voltage in each electrolytic cell The maximum and minimum DC deviations are preset to determine the probability of grounding faults in each electrolytic cell. Probability of no grounding fault and uncertainty If the average DC voltage deviation If the DC deviation exceeds the preset maximum, the probability of a ground fault is... for probability of no grounding fault The uncertainty is 0. for If the average DC voltage deviation If the DC deviation is less than the preset minimum, the probability of a ground fault is... for Uncertainty for ,in Maximum DC deviation preset; no grounding fault probability. for .

[0008] Optionally, the maximum DC deviation is preset to 100 in step S205. The preset minimum DC deviation is 5. In step S205, the probability of a grounding fault in each electrolytic cell is determined. Probability of no grounding fault and uncertainty This also includes: if If true, then: ; ; ; if If true, then: ; ; ; if If true, then: ; ; .

[0009] Optionally, in step S102, the AC voltage data of the aluminum electrolytic cell series collected synchronously includes the AC voltage phasor of each electrolytic cell in the aluminum electrolytic cell series. , , The number of electrolytic cells; in step S103, the series grounding fault diagnosis of aluminum electrolytic cells based on AC voltage data includes: S301, obtain AC voltage data for multiple recent batches of aluminum electrolytic cells; S302, Calculate the average AC voltage of each electrolytic cell based on AC voltage data from multiple batches of aluminum electrolytic cells. and AC voltage standard deviation ; S303, according to Calculate the mean deviation of AC voltage for each electrolytic cell. ,in and These are the average AC voltages of the m-th and m-1-th electrolytic cells, respectively. S304, according to Calculate the uncertainty of each electrolytic cell ; S305, based on the average deviation of AC voltage in each electrolytic cell The maximum and minimum AC deviations are preset to determine the probability of grounding faults in each electrolytic cell. Probability of no grounding fault and uncertainty If the average AC voltage deviation If the AC deviation exceeds the preset maximum, the probability of a ground fault is... for probability of no grounding fault The uncertainty is 0. for If the average AC voltage deviation If the AC deviation is less than the preset minimum, the probability of a ground fault is... for Uncertainty for ,in Maximum AC deviation preset; no grounding fault probability. for .

[0010] Optionally, the maximum AC deviation is preset to 150 in step S305. The preset minimum AC deviation is 15. In step S305, the probability of a grounding fault in each electrolytic cell is determined. Probability of no grounding fault and uncertainty This also includes: if If true, then: ; ; ; if If true, then: ; ; ; if If true, then: ; ; .

[0011] Optionally, in step S102, the bus current data of the aluminum electrolytic cell series collected synchronously includes the AC current phasor of the bus of each electrolytic cell in the aluminum electrolytic cell series. , , The number of electrolytic cells; in step S103, the series grounding fault diagnosis of aluminum electrolytic cells based on bus current data includes: S401, obtain bus current data for multiple recent batches of aluminum electrolytic cells; S402, Calculate the average bus current of each electrolytic cell based on bus current data from multiple batches of aluminum electrolytic cells. and AC voltage standard deviation ; S403, according to Calculate the mean deviation of the bus current for each electrolytic cell. ,in and These are the average bus currents of the m-th and m-1-th electrolytic cells, respectively. S404, according to Calculate the uncertainty of each electrolytic cell ; S405, based on the average deviation of the bus current of each electrolytic cell The maximum and minimum current deviations are preset to determine the probability of grounding faults in each electrolytic cell. Probability of no grounding fault and uncertainty If the average value of the bus current deviates If the current deviation exceeds the preset maximum, the probability of a ground fault is... for probability of no grounding fault The uncertainty is 0. for If the average value of the bus current deviates If the current deviation is less than the preset minimum, the probability of a ground fault is... for Uncertainty for ,in Maximum current deviation is preset; probability of no grounding fault. for .

[0012] Optionally, the maximum current deviation is preset to 720 in step S405. The preset minimum current deviation is 36. In step S405, the probability of a grounding fault in each electrolytic cell is determined. Probability of no grounding fault and uncertainty This also includes: if If true, then: ; ; ; if If true, then: ; ; ; if If true, then: ; ; .

[0013] Optionally, step S104 includes: S501, the diagnostic results of the aluminum electrolytic cell series grounding fault diagnosis based on DC voltage data, AC voltage data and bus current data for each electrolytic cell are respectively used as three evidence sources: evidence source 1 to evidence source 3. Each diagnostic result includes the probability of grounding fault, the probability of no grounding fault and uncertainty. S502, the probability of grounding fault, probability of no grounding fault, and uncertainty in evidence source 1 and evidence source 2 are fused using the Dempster-Shafer data fusion method to obtain the diagnosis result after the first fusion; the diagnosis result after the first fusion, the probability of grounding fault, probability of no grounding fault, and uncertainty in evidence source 3 are fused using the Dempster-Shafer data fusion method to obtain the final fused diagnosis result. S503, the probability of grounding fault, the probability of no grounding fault, and the uncertainty in the final fused diagnostic results are normalized to obtain the normalized probability of grounding fault, the probability of no grounding fault, and the uncertainty; and the higher probability between the normalized probability of grounding fault and the probability of no grounding fault is taken as the grounding fault detection result of the electrolytic cell.

[0014] Furthermore, the present invention also provides a series grounding fault detection system for aluminum electrolytic cells based on multi-data fusion, including a microprocessor and a memory interconnected thereto, wherein the microprocessor is programmed or configured to execute the series grounding fault detection method for aluminum electrolytic cells based on multi-data fusion.

[0015] In addition, the present invention also provides a computer program product, including a computer program or instructions, which are programmed or configured to execute the aforementioned multi-data fusion-based aluminum electrolytic cell series grounding fault detection method by a processor.

[0016] Compared with existing technologies, this invention mainly achieves the following beneficial effects: To improve the reliability of grounding fault diagnosis in aluminum electrolytic cells, multi-sensor data fusion technology is introduced into the grounding fault detection algorithm of electrolytic cells. The method of this invention includes detecting the AC and DC voltages of the electrolytic cells and the AC current on the electrolytic cell bus under AC signal excitation, acquiring the cell DC voltage data, AC voltage data, and bus current data of the electrolytic cell series. Grounding fault diagnosis algorithms are then constructed based on the cell DC voltage data, the cell AC voltage data, and the cell bus current data, respectively, and corresponding diagnostic results are given. Dempster-Shafer data fusion technology is then applied to fuse the diagnostic results of the three types of sensor data to obtain the final diagnostic result. This significantly improves the reliability of monitoring, reduces the uncertainty of diagnostic results, achieves accurate detection and location of grounding faults in the electrolytic cell series, prevents and warns of the occurrence of grounding faults, and provides a guarantee for the safe and efficient production of aluminum electrolysis series. Attached Figure Description

[0017] Figure 1 This is a schematic diagram of the basic process of the method in an embodiment of the present invention.

[0018] Figure 2 This is a schematic diagram of the equivalent circuit of the aluminum electrolytic cell series in the embodiments of the present invention.

[0019] Figure 3 This is a schematic diagram of the equivalent circuit when the aluminum electrolytic cell series is grounded at multiple points in an embodiment of the present invention.

[0020] Figure 4 The image shows the DC voltage distribution curves of the electrolytic cell series in this embodiment of the invention.

[0021] Figure 5 This is a flowchart of ground fault diagnosis based on DC voltage data in an embodiment of the present invention.

[0022] Figure 6 This is a schematic diagram of the AC / DC equivalent circuit of the aluminum electrolytic cell series in an embodiment of the present invention.

[0023] Figure 7 This is a schematic diagram of the equivalent circuit for the current relationship between adjacent slots in an embodiment of the present invention.

[0024] Figure 8 This is a flowchart of ground fault diagnosis based on AC voltage data in an embodiment of the present invention.

[0025] Figure 9 This is a diagram showing the AC voltage distribution of a series of slots under AC excitation during a two-point grounding fault in an embodiment of the present invention.

[0026] Figure 10This is a stepped distribution diagram fitted after processing for a two-point grounding fault in an embodiment of the present invention.

[0027] Figure 11 This is the first set of data on the voltage distribution of slot #52 when there is a grounding fault in the embodiment of the present invention, where (a) is the AC voltage of the slot and (b) is the AC voltage of the slot.

[0028] Figure 12 This is the second set of data on the voltage distribution of slot #52 when there is a grounding fault in the slot in this embodiment of the invention, where (a) is the AC voltage of the slot and (b) is the AC voltage of the slot.

[0029] Figure 13 This is a schematic diagram of the current sensor arrangement in an embodiment of the present invention.

[0030] Figure 14 This is a flowchart of ground fault diagnosis based on bus current data in an embodiment of the present invention.

[0031] Figure 15 This is a schematic diagram of confidence interval division in an embodiment of the present invention. Detailed Implementation

[0032] To enable those skilled in the art to better understand the technical solutions of the present invention, the technical solutions of the present invention will be further described in detail below with reference to the accompanying drawings of the embodiments of the present invention. Figure 1 As shown, this embodiment of the aluminum electrolytic cell series grounding fault detection method based on multi-data fusion includes the following steps: S101, Apply an AC excitation signal to the aluminum electrolysis cell series; S102, synchronously collects DC voltage data, AC voltage data and bus current data of aluminum electrolysis cell series; S103, based on DC voltage data, AC voltage data and bus current data, performs a series of grounding fault diagnosis for aluminum electrolytic cells to obtain corresponding diagnostic results; S104 uses the Dempster-Shafer data fusion method to fuse the three diagnostic results to obtain the final grounding fault detection result for the aluminum electrolytic cell series.

[0033] In step S101, when applying an AC excitation signal to the aluminum electrolytic cell series, in this embodiment, a low-frequency AC signal voltage of 36V / 60Hz is applied to the first end of the electrolytic cell series. Alternatively, the required voltage and frequency can be used as needed.

[0034] In step S102, when simultaneously collecting DC voltage data, AC voltage data, and bus current data for the aluminum electrolytic cell series, a comprehensive voltage sensor and a bus current sensor can be installed on each electrolytic cell in the series to simultaneously detect the cell's DC voltage, AC voltage, and bus AC current. Based on this, grounding fault detection algorithms for electrolytic cells can be constructed using DC voltage data, AC voltage data, and bus current data respectively. DS multi-sensor data fusion technology is then applied to fuse the three types of sensor data at the decision level, accurately detecting grounding faults in the electrolytic cell's DC system and indicating the location of the faulty electrolytic cell. Compared to diagnostic results from single-sensor data, the results from multi-sensor data fusion offer higher reliability and detection accuracy.

[0035] In this embodiment, the synchronously acquired DC voltage data of the aluminum electrolytic cell series includes the DC voltage of each electrolytic cell in the aluminum electrolytic cell series. , , This refers to the number of electrolytic cells. In this embodiment, the DC system operating voltage of the electrolytic cell series is 1160V, and the current is 300kA. During normal operation, the cell resistance... =14uΩ, the tank voltage is approximately 4.2V, and the insulation resistance of the tank to ground is R1, R2, ..., R n When insulation is good, the capacitance is greater than 5MΩ, so let R = 5MΩ; under DC conditions, the distributed capacitance to ground can be ignored, and its equivalent circuit is as follows: Figure 2 Where V is the voltmeter and 01 to 62 are the electrolytic cells. The voltage between the cell and ground decreases sequentially from the beginning to zero. In practice, zero-point drift often occurs, mainly caused by grounding due to insulation damage of the series busbars and electrolytic cells. Usually, there are multiple grounding points, and the drifted zero-point position is the result of the combined effect of multiple grounding points.

[0036] The following is an analysis of the grounding fault types in the aluminum electrolytic cell series: (1) Single-point grounding fault. If the DC voltage distribution curve of the electrolytic cell series... y ( k (1) If there are no staggered singularities and the zero point is not at the geometric center of the series, then there is a grounding fault in the series, and the zero potential slot is the grounding fault slot. The zero point position can be obtained by using the DC voltage of the series head to ground and the DC voltage data of the comprehensive voltage sensor of each electrolytic cell, and the grounding fault slot can be determined. (2) Detection of multi-point grounding faults. Grounding faults in the DC system of the electrolytic cell series are usually multi-point grounding faults. There is a significant change in the voltage of the slot before and after the fault slot, and the grounding slot # m The DC voltage is greater than the voltage of the subsequent slots. The magnitude of the change depends on the grounding current in the grounding trench. The grounding current is related to the grounding resistance and the number of electrolytic cells between the grounding fault points. For metallic grounding faults, a larger grounding current results in a larger voltage change in the faulty cell, leading to a more obvious fault detection effect. Based on the DC voltage difference between adjacent electrolytic cells, the location of the grounding faulty cell can be accurately detected. The equivalent circuit is as follows: Figure 3 As shown, when there are N slots between the two grounding slots #m and #k, For slot resistors, and The grounding resistances of the two grounding slots #m and #k are... and These are the bus currents for grounding slots #m and #m+1, respectively. The slot voltage is approximately 4.2V, and the slot resistance is approximately... If the grounding resistance is Then the grounding current is Voltage change at grounding trench #m In the extreme case, N=1, at which point... =6 For example, if a grounding fault is set in cells #20 and #50 of the electrolytic cell series, the voltage sensor will measure the DC voltage data of the cells as follows: Figure 4 It is obvious that there is a significant voltage change between the front and rear cells of #20 and #50. In this embodiment, the DC voltage detection sensitivity of the integrated voltage sensor for the electrolytic cell is 5. It can meet the grounding resistance requirement. Metal grounding fault detection. Due to the slight variations in cell resistance caused by the electrolytic cell's operating conditions, severe electromagnetic interference in the field, and the inherent variability in sensor performance, the DC voltage detection data for the electrolytic cell contains a certain degree of error. Therefore, the detection and diagnostic results are subject to uncertainty, expressed as relative uncertainty: Relative uncertainty = (standard deviation / average) × 100%. Therefore, if... Figure 5 As shown, step S103 of this embodiment, which involves diagnosing grounding faults in aluminum electrolytic cells based on DC voltage data, includes: S201, obtain DC voltage data for multiple recent batches of aluminum electrolytic cells; S202, Calculate the average DC voltage of each electrolytic cell based on DC voltage data from multiple batches of aluminum electrolytic cells. and DC voltage standard deviation ; S203, according to Calculate the mean deviation of DC voltage in each electrolytic cell. ,in and These are the average DC voltage values ​​of the m-th and m-1-th electrolytic cells, respectively. S204, according to Calculate the uncertainty of each electrolytic cell ; S205, based on the average deviation of DC voltage in each electrolytic cell The maximum and minimum DC deviations are preset to determine the probability of grounding faults in each electrolytic cell. Probability of no grounding fault and uncertainty If the average DC voltage deviation If the DC deviation exceeds the preset maximum, the probability of a ground fault is... for probability of no grounding fault The uncertainty is 0. for If the average DC voltage deviation If the DC deviation is less than the preset minimum, the probability of a ground fault is... for Uncertainty for ,in Maximum DC deviation preset; no grounding fault probability. for .

[0037] Specifically, in step S205 of this embodiment, the preset maximum DC deviation is 100. The preset minimum DC deviation is 5. In step S205, the probability of a grounding fault in each electrolytic cell is determined. Probability of no grounding fault and uncertainty This also includes: if If true, then: ; ; ; if If true, then: ; ; ; if If true, then: ; ; .

[0038] Therefore, the criterion for determining whether there is a grounding fault in #m slot is: #m slot has a grounding fault, with very little uncertainty; #m slot has a grounding fault, but the uncertainty is small; #m slot has a grounding fault, and the uncertainty is relatively large; #m slot has a grounding fault, resulting in high uncertainty; The #m slot has no grounding fault and has low uncertainty.

[0039] In step S102, the synchronously acquired AC voltage data of the aluminum electrolytic cell series includes the AC voltage phasor of each electrolytic cell in the aluminum electrolytic cell series. , , This refers to the number of electrolytic cells. A 36V / 60Hz AC excitation voltage is applied at the beginning of the series. The AC component of the electrolytic cell voltage is related not only to the cell resistance but also to the distributed inductance of the busbar and the electrolytic cell itself. Based on field tests and calculations, the cell resistance R... m Approximately The tank impedance under 60Hz AC excitation is approximately Electrolytic cell to ground capacitance Approximately 1 nF, capacitive reactance 2.65 Insulation resistance to ground Normally should It was estimated that the leakage current of a properly insulated electrolytic cell is negligible. The equivalent circuit of the aluminum electrolytic cell series under AC source excitation is as follows: Figure 6 There are 276 electrolytic cells, from #1 to #276. Each electrolytic cell is equivalent to an inductor and a circuit consisting of a resistor and a capacitor in parallel in the main circuit. The electrodes are equivalent to a circuit consisting of a resistor and a capacitor in parallel. The equivalent circuits for any cell #m and #m+1 are as follows: Figure 7 As shown, where and Let C be the grounding resistance of slots #m and #m+1, and C be the capacitance of the electrode components. The requirement is to be able to detect the grounding resistance as... The grounding fault corresponds to a grounding current of approximately 0.036A, and the difference in AC current between the two adjacent tanks is... A, therefore the AC voltage difference between the two adjacent slots is A. Considering the influence of electrolytic cell operating conditions, on-site electromagnetic environment, and sensor dispersion on the test results, the AC voltage measurement data of the electrolytic cell contains a certain error. Uncertainty is introduced to characterize the impact of various factors on the test results, such as… Figure 8 As shown, step S103 of this embodiment, which involves diagnosing grounding faults in aluminum electrolytic cells based on AC voltage data, includes: S301, obtain AC voltage data for multiple recent batches of aluminum electrolytic cells; S302, Calculate the average AC voltage of each electrolytic cell based on AC voltage data from multiple batches of aluminum electrolytic cells. and AC voltage standard deviation ; S303, according to Calculate the mean deviation of AC voltage for each electrolytic cell. ,in and These are the average AC voltages of the m-th and m-1-th electrolytic cells, respectively. S304, according to Calculate the uncertainty of each electrolytic cell ; S305, based on the average deviation of AC voltage in each electrolytic cell The maximum and minimum AC deviations are preset to determine the probability of grounding faults in each electrolytic cell. Probability of no grounding fault and uncertainty If the average AC voltage deviation If the AC deviation exceeds the preset maximum, the probability of a ground fault is... for probability of no grounding fault The uncertainty is 0. for If the average AC voltage deviation If the AC deviation is less than the preset minimum, the probability of a ground fault is... for Uncertainty for ,in Maximum AC deviation preset; no grounding fault probability. for .

[0040] Furthermore, in step S305 of this embodiment, the maximum AC deviation is preset to be 150. The preset minimum AC deviation is 15. In step S305, the probability of a grounding fault in each electrolytic cell is determined. Probability of no grounding fault and uncertainty This also includes: if If true, then: ; ; ; if If true, then: ; ; ; If 1 If true, then: ; ; .

[0041] Therefore, the criterion for determining whether there is a grounding fault in #m slot is: The corresponding grounding resistance is <20 ohms, so #m slot definitely has a grounding fault with low uncertainty; The corresponding grounding resistance is 20-50 ohms. There is a grounding fault in slot #m, but the uncertainty is small. The corresponding grounding resistance is 50-100 ohms. A grounding fault exists in slot #m, with significant uncertainty. The corresponding grounding resistance is 100-200 ohms. There is a grounding fault in slot #m, and the uncertainty is large. The corresponding grounding resistance is >200 ohms, #m slot has no grounding fault, and the uncertainty is small.

[0042] To verify the above method, in this embodiment, a monitoring system was installed on 62 electrolytic cells in an aluminum electrolysis plant, and a grounding fault was set in cells #30 and #51. Under excitation from a 36V / 60Hz excitation source, the AC voltage detected by the voltage sensors in each cell was as follows: Figure 9 As shown, after denoising and algorithm processing, the AC voltage distribution of the stepped slot is obtained as follows: Figure 10 As shown. It can be seen that the AC voltage of the series of cells changes randomly due to multiple factors such as the electrolytic cell operating conditions, electromagnetic environment, and sensor installation position, but the voltage distribution of each cell in the series is related to the grounding fault. (1) Since the grounding current flows through the upstream cell but not the downstream cell, the voltage of the upstream cell of the faulty cell is generally higher than that of the downstream cell, forming a step-like jump at the faulty cell. Combining the grounding fault criterion of the electrolytic cell, the step-like distribution of the series of cell voltages is fitted by the time-domain waveform processing algorithm as shown. Figure 10 As shown, the step point in the distribution diagram is the ground fault slot. The diagnostic results of the example are slots #30 and #51, which are consistent with the set ground fault location. (2) The AC voltage of the ground fault slot is related to the sensor measuring point and the grounding point location. Since the electrolytic cell adopts an asymmetrical six-point power input bus configuration, the cathode output terminals on the A and B sides are not connected to a unified conductive bus, but are connected to the 6 bus in sections; the slot voltage sensor measuring point is on the first cathode bus on the A side. If the ground fault occurs in the connection area of ​​this bus, most of the ground current flows through the sensor branch, and the current diversion in other branches is small, so the measured slot voltage is low. It's bigger. In the formula For grounding current, The equivalent resistance of the #m electrolytic cell. (3) The measured AC voltage of the ground fault cell is significantly higher than that of the upstream cell. For the upstream cell, regardless of the location of the grounding point of the fault cell, the current flowing through the sensor branch can be approximated as Its slot voltage In the formula This is the equivalent resistance of electrolytic cell #m-1. Therefore, if the grounding fault location of cell #m is in the bus connection area of ​​the sensor measuring point, Due to the complex and variable operating conditions of electrolytic cells, the AC voltage data of electrolytic cells fluctuates significantly, such as... Figure 11 and Figure 12 As shown, the curves of the two sets of monitoring data are not entirely the same, and the diagnostic results are also inconsistent. The diagnostic result of the first set of monitoring data is #52, which is consistent with the setting; however, the diagnostic result of the second set of monitoring data is #55, which is an incorrect diagnosis. Therefore, in this embodiment, a confidence level (uncertainty) of the diagnostic result is set in the fault diagnosis to characterize the above-mentioned diagnostic uncertainty, which can provide support for multi-sensor data fusion.

[0043] In step S102 of this embodiment, the bus current data of the aluminum electrolytic cell series collected synchronously includes the AC current phasor of the bus of each electrolytic cell in the aluminum electrolytic cell series. , , This represents the number of electrolytic cells.

[0044] A current sensor based on the principle of electromagnetic induction is installed on the busbar of the electrolytic cell to detect the AC current on the busbar under 36V / 60Hz AC power excitation. The AC current phasors between adjacent cells are calculated and compared to determine if there is a grounding fault in the electrolytic cell. The measurement principle of the busbar current sensor is as follows: Figure 13 As shown. A detection coil is installed on the busbar of the electrolytic cell, with the coil plane perpendicular to the busbar surface, to maximize the magnetic flux passing through the coil plane, as shown. Figure 13 As shown in the box on the right. Let DC be... I The alternating current is , For amplitude, Angular frequency, Given time, a flat coil with N turns and an area of ​​S, the magnetic flux linkage passing through the coil is: : ; in, This is the proportionality coefficient. This refers to the number of turns of the small coil. Let the area of ​​the small coil be... This refers to the current in the busbar of the electrolytic cell. It is direct current. The current is alternating current. The induced electromotive force in the coil. for: ; in, For coefficients, For amplitude, Angular frequency, For time. Induced electromotive force. There is no DC magnetic field component in the sensor. If the sensor output load is R, then the sensor output current will be... for: ; in, A proportionality coefficient related to the number of turns, area, and frequency. Clearly, by adjusting the number of coil turns, area, and excitation signal frequency, and through data preprocessing, the detection data can meet the requirements for detecting grounding faults in electrolytic cells, and the sensor detection accuracy can reach 1mA. The detection data obtained by this method is independent of cell resistance and reactance, and therefore unaffected by changes in the electrolytic cell's operating conditions. Compared to diagnostic methods based on cell voltage monitoring data, the test results have higher reliability; however, the measurement data is closely related to the sensor's installation location and orientation, and the diagnostic results based on this method still have some uncertainty. Figure 14 As shown, in step S103 of this embodiment, the diagnosis of grounding faults in the aluminum electrolytic cell series based on bus current data includes: S401, obtain bus current data for multiple recent batches of aluminum electrolytic cells; S402, Calculate the average bus current of each electrolytic cell based on bus current data from multiple batches of aluminum electrolytic cells. and AC voltage standard deviation ; S403, according to Calculate the mean deviation of the bus current for each electrolytic cell. ,in and These are the average bus currents of the m-th and m-1-th electrolytic cells, respectively. S404, according to Calculate the uncertainty of each electrolytic cell ; S405, based on the average deviation of the bus current of each electrolytic cell The maximum and minimum current deviations are preset to determine the probability of grounding faults in each electrolytic cell. Probability of no grounding fault and uncertainty If the average value of the bus current deviates If the current deviation exceeds the preset maximum, the probability of a ground fault is... for probability of no grounding fault The uncertainty is 0. for If the average value of the bus current deviates If the current deviation is less than the preset minimum, the probability of a ground fault is... for Uncertainty for ,in Maximum current deviation is preset; probability of no grounding fault. for .

[0045] Furthermore, in step S405 of this embodiment, the preset maximum current deviation is 720. mA The preset minimum current deviation is 36. mA In step S405, the probability of a grounding fault in each electrolytic cell is determined. Probability of no grounding fault and uncertainty This also includes: if If true, then: ; ; ; if If true, then: ; ; ; if If true, then: ; ; .

[0046] Since the applied AC signal voltage is 36V / 60Hz and the tank impedance is very small, it can be assumed that the AC voltage of each tank to ground is approximately 36V. When a ground fault occurs in tank #m, if It is determined that the tank has a grounding fault with a grounding resistance of less than 500 ohms. Therefore, the criterion for determining whether tank #m has a grounding fault is: #m slot has a grounding fault, corresponding to the grounding resistance. The uncertainty is very small, almost zero; #m slot has a grounding fault, corresponding to the grounding resistance. The uncertainty is small; #m slot has a grounding fault, corresponding to the grounding resistance. The uncertainty is relatively high; #m slot has a grounding fault, corresponding to the grounding resistance. High uncertainty; #m slot does not have a grounding fault, corresponding grounding resistance The uncertainty is 0.

[0047] The diagnostic results from single-type sensor monitoring data are subject to considerable uncertainty, leading to false alarms and missed detections. Therefore, this embodiment utilizes the Dempster-Shafer data fusion method to fuse three types of data—tank AC voltage data, DC voltage data, and bus current data—at the decision-level. This avoids the limitations and uncertainties of single-type sensor perception and significantly improves the reliability of the detection results. The Dempster-Shafer data fusion method is a mathematical tool specifically designed to handle uncertainty and incomplete information. It provides an effective method for reasoning and decision-making under imprecise or uncertain information and has wide applications in equipment fault diagnosis and online monitoring. Generally, using single-sensor monitoring data for equipment fault diagnosis results in considerable uncertainty, easily leading to false alarms and missed detections. Equipment faults typically generate multiple types of externally perceptible information from the same source, such as voltage, current, sound, and light signals. By monitoring these signals with appropriate sensors, multi-sensor data fusion technology can be used to fuse feature-level or decision-level information, avoiding the limitations and uncertainties of single-type sensor perception. This achieves comprehensive perception of internal equipment faults, significantly improving the reliability of the detection results and reducing their uncertainty. The identification framework of the Dempster-Shafer data fusion method represents the set of all possible hypotheses, denoted as the identification framework. For example, in target recognition, It can include all possible target types. The basic probability assignment, i.e., the quality function, is denoted as... For the recognition framework each subset , Indicates to The level of trust satisfies: And satisfy: , ; Then it is called To identify the subset of frames The basic probability assignment represents the evidence for a subset. The level of support. Among them... Let be an empty set. The formulas that need to be satisfied above represent that the basic trust assignment for the empty proposition (empty set) is 0, and the sum of the basic probability assignments for all subsets is 1. Given a recognition framework and any subset thereof Reliability function It is all support The sum of the basic probability assignments (BPAs) of a subset of , i.e.: ; plausibility function It indicates the degree of uncertainty regarding a certain hypothesis or set of hypotheses, that is: ; Belief function and similarity function These are the upper and lower bound probabilities defined by Dempster, and their confidence intervals are divided as follows: Figure 15 As shown. When there are multiple sources of evidence, the Dempster-Shafer evidence theory uses... Dempster The combination rule fuses information from different sources of evidence to derive the final basic probability allocation and confidence function. For the sources of evidence... and The combined mass function is defined as: ; in, It is the source of evidence and The level of trust after the combination For combinational operations, subsets The intersection of subsets B and C. As a source of evidence The degree of trust in subset B, As a source of evidence The degree of trust in subset C, This represents the summation of the trust levels for all subsets whose intersection with subset C equals subset A. This indicates that the intersection of all subsets B and C is empty. The trust levels are summed. The denominator is used for normalization to ensure that the combined quality function still satisfies the probability assignment conditions. In the analysis and decision-making process, the maximum basic probability assignment function rule is used to evaluate the diagnostic target, requiring: ① the diagnostic level determination result... m ( A max1 ① It should have the maximum basic probability allocation value; ② Diagnostic grade determination result m ( A max1 The difference between the base probability assignment value and any other level is not less than the set threshold. ③ The basic probability distribution value of uncertainty must be less than a certain set threshold. Using decision rules, the allocation probability with the highest overall evidence is selected as the final judgment result. In this embodiment, combined with the Dempster-Shafer data fusion method, step S104 includes: S501, the diagnostic results of the aluminum electrolytic cell series grounding fault diagnosis based on DC voltage data, AC voltage data, and bus current data for each electrolytic cell are used as three evidence sources: evidence source 1 to evidence source 3. Each diagnostic result includes the probability of grounding fault, the probability of no grounding fault, and uncertainty, which respectively include: There is a probability of grounding fault. Probability of no grounding fault and uncertainty ; There is a probability of grounding fault. Probability of no grounding fault and uncertainty ; There is a probability of grounding fault. Probability of no grounding fault and uncertainty ; Among them, A, B and C refer to three possible situations: grounding fault, no grounding fault and uncertain, respectively; S502, the probability of grounding fault, probability of no grounding fault, and uncertainty in evidence source 1 and evidence source 2 are fused using the Dempster-Shafer data fusion method to obtain the diagnosis result after the first fusion; the diagnosis result after the first fusion, the probability of grounding fault, probability of no grounding fault, and uncertainty in evidence source 3 are fused using the Dempster-Shafer data fusion method to obtain the final fused diagnosis result. For example, in this embodiment, evidence source 1 and evidence source 2 are first fused to obtain the diagnosis result after the first fusion: There is a probability of grounding fault. Probability of no grounding fault and uncertainty ; Then, the fusion results of evidence sources 1 and 2 are fused with evidence source 3 to calculate the final fused diagnostic result: There is a probability of grounding fault. Probability of no grounding fault and uncertainty ; S503, the probability of grounding fault, probability of no grounding fault, and uncertainty in the final fused diagnostic results are normalized to obtain normalized grounding fault probability, probability of no grounding fault, and uncertainty; and the higher of the normalized grounding fault probability and probability of no grounding fault is taken as the grounding fault detection result of the electrolytic cell. The normalization function expression is as follows: ; ; .

[0048] Ultimately, comparison , The result with the highest probability is taken as the diagnosis result for the grounding fault in that trench. Typically... The uncertainty after fusion is greatly reduced, thus improving the reliability of the diagnostic results.

[0049] To verify the multi-data fusion-based grounding fault detection method for aluminum electrolytic cells in this embodiment, this embodiment examines the DC voltage difference between two adjacent cells at cell #k of a certain electrolytic cell series. The DC voltage difference is 6μV, resulting in a diagnosis of no grounding fault; the AC voltage difference is 110μV, resulting in a diagnosis of a grounding fault; and the measured bus current difference is 0.1A, also resulting in a diagnosis of a grounding fault. These three results are inconsistent. The diagnostic results of each electrolytic cell based on DC voltage data, AC voltage data, and bus current data for grounding fault detection in the aluminum electrolytic cell series are used as evidence sources 1 to 3. The probability of a grounding fault, the probability of no grounding fault, and the uncertainty of each diagnostic result are as follows: Probability of a grounding fault for evidence source 1. Probability of no grounding fault and uncertainty The values ​​are (0.086, 0.581, 0.333); the probability of a grounding fault in evidence source 2. Probability of no grounding fault and uncertainty (0.817, 0.133, 0.05); Probability of grounding fault in evidence source 3 Probability of no grounding fault and uncertainty (0.482, 0.418, 0.10); By fusing and normalizing evidence source 1 and evidence source 2, the result is: =0.428; =0.470; =0.101. This is then fused with evidence source 3 to obtain the probability of a grounding fault. Probability of no grounding fault and uncertainty Then, after normalization, the final fusion results (normalized grounding fault probability, no-grounding fault probability, and uncertainty) are as follows: =0.506, =0.4817, =0.012, as shown in Table 1.

[0050] Table 1: Three sources of evidence and their fusion results in this embodiment

[0051] Table 1 shows that the fused results indicate a grounding fault in the cell, with a significantly reduced uncertainty of only 0.012, indicating improved reliability. After data fusion, the system's uncertainty is significantly reduced, and the fused results eliminate misjudgments from DC voltage testing, improving the system's identification capability. Compared to the electrolytic cell voltage detection method and the zero-point drift detection method, the multi-data fusion-based grounding fault detection method for aluminum electrolytic cells in this embodiment can more effectively identify the location of multiple faults. Compared to single DC voltage, AC voltage, and bus current methods, the multi-data fusion-based grounding fault detection method for aluminum electrolytic cells in this embodiment can improve the accuracy and distinguishability of diagnostic results and reduce diagnostic uncertainty.

[0052] Furthermore, this embodiment also provides a multi-data fusion-based aluminum electrolytic cell series grounding fault detection system, including a microprocessor and a memory interconnected, wherein the microprocessor is programmed or configured to execute the multi-data fusion-based aluminum electrolytic cell series grounding fault detection method. Additionally, this embodiment also provides a computer program product, including a computer program or instructions, which is programmed or configured to execute the multi-data fusion-based aluminum electrolytic cell series grounding fault detection method via a processor.

[0053] Those skilled in the art will understand that the technical solutions provided by this invention may take the form of a method, system, or computer program product. Therefore, this invention may take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this invention may take the form of a computer program product embodied on one or more computer-readable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code. This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, produce an implementation of the flowchart... Figure 1 One or more processes and / or boxes Figure 1 The computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to operate in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The functions specified in one or more boxes. These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable apparatus for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0054] The above description is merely a preferred embodiment of the present invention. The scope of protection of the present invention is not limited to the above embodiments. All technical solutions falling within the scope of the present invention's concept are within the scope of protection of the present invention. It should be noted that for those skilled in the art, any improvements and modifications made without departing from the principles of the present invention should also be considered within the scope of protection of the present invention.

Claims

1. A method for detecting grounding faults in aluminum electrolytic cells based on multi-data fusion, characterized in that... Includes the following steps: S101, Apply an AC excitation signal to the aluminum electrolysis cell series; S102, synchronously collects DC voltage data, AC voltage data and bus current data of each electrolytic cell in the aluminum electrolytic cell series; S103, based on DC voltage data, AC voltage data and bus current data, performs a series of grounding fault diagnosis for aluminum electrolytic cells to obtain corresponding diagnostic results; Grounding fault diagnosis of aluminum electrolytic cells based on DC voltage data includes: S201, obtaining DC voltage data from multiple recent batches of aluminum electrolytic cells; S202, calculating the average DC voltage of each electrolytic cell based on the DC voltage data from multiple batches of aluminum electrolytic cells. and DC voltage standard deviation S203, according to Calculate the mean deviation of DC voltage in each electrolytic cell. ,in and These are the average DC voltages of the m-th and m-1-th electrolytic cells, respectively; S204, according to Calculate the uncertainty of each electrolytic cell S205, based on the average deviation of DC voltage in each electrolytic cell The maximum and minimum DC deviations are preset to determine the probability of grounding faults in each electrolytic cell. Probability of no grounding fault and uncertainty If the average DC voltage deviation If the DC deviation exceeds the preset maximum, the probability of a ground fault is... for probability of no grounding fault The uncertainty is 0. for If the average DC voltage deviation If the DC deviation is less than the preset minimum, the probability of a ground fault is... for Uncertainty for ,in Maximum DC deviation preset; no grounding fault probability. for ;in, , This refers to the number of electrolytic cells; Grounding fault diagnosis of aluminum electrolytic cells based on AC voltage data includes: S301, obtaining AC voltage data from multiple recent batches of aluminum electrolytic cells; S302, calculating the average AC voltage of each electrolytic cell based on the AC voltage data from multiple batches of aluminum electrolytic cells. and AC voltage standard deviation S303, according to Calculate the mean deviation of AC voltage for each electrolytic cell. ,in and These are the average AC voltages of the m-th and m-1-th electrolytic cells, respectively; S304, according to Calculate the uncertainty of each electrolytic cell S305, based on the average deviation of AC voltage in each electrolytic cell The maximum and minimum AC deviations are preset to determine the probability of grounding faults in each electrolytic cell. Probability of no grounding fault and uncertainty If the average AC voltage deviation If the AC deviation exceeds the preset maximum, the probability of a ground fault is... for probability of no grounding fault The uncertainty is 0. for If the average AC voltage deviation If the AC deviation is less than the preset minimum, the probability of a ground fault is... for Uncertainty for ,in Maximum AC deviation preset; no grounding fault probability. for ; The grounding fault diagnosis of aluminum electrolytic cells based on bus current data includes: S401, obtaining bus current data from multiple recent batches of aluminum electrolytic cells; S402, calculating the average bus current of each electrolytic cell based on the bus current data from multiple batches of aluminum electrolytic cells. and AC voltage standard deviation S403, according to Calculate the mean deviation of the bus current for each electrolytic cell. ,in and These are the average bus currents of the m-th and m-1-th electrolytic cells, respectively; S404, according to Calculate the uncertainty of each electrolytic cell S405, based on the average deviation of the bus current of each electrolytic cell. The maximum and minimum current deviations are preset to determine the probability of grounding faults in each electrolytic cell. Probability of no grounding fault and uncertainty If the average value of the bus current deviates If the current deviation exceeds the preset maximum, the probability of a ground fault is... for probability of no grounding fault The uncertainty is 0. for If the average value of the bus current deviates If the current deviation is less than the preset minimum, the probability of a ground fault is... for Uncertainty for ,in Maximum current deviation is preset; probability of no grounding fault. for ; S104 uses the Dempster-Shafer data fusion method to fuse the three diagnostic results to obtain the final grounding fault detection result for the aluminum electrolytic cell series.

2. The method for detecting grounding faults in aluminum electrolytic cells based on multi-data fusion according to claim 1, characterized in that, In step S102, the synchronously acquired DC voltage data of the aluminum electrolytic cell series includes the DC voltage of each electrolytic cell in the aluminum electrolytic cell series. .

3. The method for detecting grounding faults in aluminum electrolytic cells based on multi-data fusion according to claim 2, characterized in that, In step S205, the preset maximum DC deviation is 100. The preset minimum DC deviation is 5. In step S205, the probability of a grounding fault in each electrolytic cell is determined. Probability of no grounding fault and uncertainty This also includes: if If true, then: ； ； ； if If true, then: ； ； ； if If true, then: ； ； 。 4. The method for detecting grounding faults in aluminum electrolytic cells based on multi-data fusion according to claim 1, characterized in that, In step S102, the synchronously acquired AC voltage data of the aluminum electrolytic cell series includes the AC voltage phasor of each electrolytic cell in the aluminum electrolytic cell series. .

5. The method for detecting grounding faults in aluminum electrolytic cells based on multi-data fusion according to claim 4, characterized in that, In step S305, the maximum AC deviation is preset to 150. The preset minimum AC deviation is 15. In step S305, the probability of a grounding fault in each electrolytic cell is determined. Probability of no grounding fault and uncertainty This also includes: if If true, then: ； ； ； if If true, then: ； ； ； if If true, then: ； ； 。 6. The method for detecting grounding faults in aluminum electrolytic cells based on multi-data fusion according to claim 1, characterized in that, In step S102, the bus current data of the aluminum electrolytic cell series collected synchronously includes the AC current phasor of the bus of each electrolytic cell in the aluminum electrolytic cell series. .

7. The method for detecting grounding faults in aluminum electrolytic cells based on multi-data fusion according to claim 6, characterized in that, In step S405, the preset maximum current deviation is 720. The preset minimum current deviation is 36. In step S405, the probability of a grounding fault in each electrolytic cell is determined. Probability of no grounding fault and uncertainty This also includes: if If true, then: ； ； ； if If true, then: ； ； ； if If true, then: ； ； 。 8. The method for detecting grounding faults in aluminum electrolytic cells based on multi-data fusion according to claim 1, characterized in that, Step S104 includes: S501, the diagnostic results of the aluminum electrolytic cell series grounding fault diagnosis based on DC voltage data, AC voltage data and bus current data for each electrolytic cell are respectively used as evidence source 1 to evidence source 3. S502, the probability of grounding fault, probability of no grounding fault, and uncertainty in evidence source 1 and evidence source 2 are fused using the Dempster-Shafer data fusion method to obtain the diagnosis result after the first fusion; the diagnosis result after the first fusion, the probability of grounding fault, probability of no grounding fault, and uncertainty in evidence source 3 are fused using the Dempster-Shafer data fusion method to obtain the final fused diagnosis result. S503, the probability of grounding fault, the probability of no grounding fault, and the uncertainty in the final fused diagnostic results are normalized to obtain the normalized probability of grounding fault, the probability of no grounding fault, and the uncertainty; and the higher probability between the normalized probability of grounding fault and the probability of no grounding fault is taken as the grounding fault detection result of the electrolytic cell.

9. A series grounding fault detection system for aluminum electrolytic cells based on multi-data fusion, comprising a microprocessor and a memory interconnected, characterized in that, The microprocessor is programmed or configured to execute the aluminum electrolytic cell series grounding fault detection method based on multi-data fusion as described in any one of claims 1 to 8.

10. A computer program product, comprising a computer program or instructions, characterized in that, The computer program or instructions are programmed or configured to execute, via a processor, the aluminum electrolytic cell series grounding fault detection method based on multi-data fusion as described in any one of claims 1 to 8.

Images