CN113049985A - Aluminum cell series earth leakage fault positioning and early warning detection system and method - Google Patents
Aluminum cell series earth leakage fault positioning and early warning detection system and method Download PDFInfo
- Publication number
- CN113049985A CN113049985A CN202110460764.9A CN202110460764A CN113049985A CN 113049985 A CN113049985 A CN 113049985A CN 202110460764 A CN202110460764 A CN 202110460764A CN 113049985 A CN113049985 A CN 113049985A
- Authority
- CN
- China
- Prior art keywords
- electrolytic cell
- aluminum electrolytic
- insulation
- aluminum
- main control
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
- G01R31/50—Testing of electric apparatus, lines, cables or components for short-circuits, continuity, leakage current or incorrect line connections
- G01R31/52—Testing for short-circuits, leakage current or ground faults
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Testing Of Short-Circuits, Discontinuities, Leakage, Or Incorrect Line Connections (AREA)
Abstract
The invention discloses a system and a method for positioning, early warning and detecting an earth leakage fault of an aluminum electrolytic cell series, wherein each aluminum electrolytic cell is connected with an earth leakage fault detection loop, and each earth leakage fault detection loop comprises a signal sending lead, a voltage receiving lead, an insulation resistor and a main control computer; one end of each signal transmission wire is connected with the main control machine, and the other end of each signal transmission wire is connected with the grounding point of the corresponding aluminum electrolytic cell; the grounding point is also connected with the corresponding aluminum electrolytic cell through an insulation resistor; the grounding point is also connected with one end of a voltage receiving lead, the other end of the voltage receiving lead is connected with a main control computer, the main control computer calculates the resistance value of the insulation resistance, judges whether the corresponding aluminum electrolytic cell has the earth leakage fault or not, and accurately positions the aluminum electrolytic cell with the earth leakage fault; meanwhile, the LSTM network is used for predicting the change of the insulation resistance of the aluminum electrolytic cell and each point to the ground, so that the early warning of the earth leakage fault is realized. The invention can accurately position and early warn the earth leakage fault of the aluminum electrolytic cell and each point in time.
Description
Technical Field
The invention relates to the technical field of detection of an aluminum electrolytic cell on an earth leakage fault, in particular to a system and a method for positioning, early warning and detecting the earth leakage fault of an aluminum electrolytic cell series.
Background
When an aluminum electrolysis enterprise produces electrolytic aluminum, the aluminum electrolysis cell of an electrolytic aluminum plant works by supplying power to a plurality of aluminum electrolysis cells connected in series through a direct current high-voltage bus (the voltage is from hundreds of volts to thousands of volts), and the direct current is utilized to electrolyze the aluminum to generate aluminum water for output. The power supply voltage of each aluminum electrolytic cell is about 4V, but the power supply current is very large, from tens of thousands of amperes to hundreds of thousands of amperes, so the whole direct current high-voltage bus passage is equivalent to a pure short-circuit conductor, in order to ensure the safe operation of the aluminum electrolytic cell, the aluminum electrolytic cell and the ground are supported by an insulator component, along with the long-term operation of the electrolytic aluminum, the supported insulator can gradually deteriorate, the ground insulation of the aluminum electrolytic cell is poor, the insulation resistance can be reduced, and the leakage current of the aluminum electrolytic cell to the ground can be caused in serious cases. The electricity leakage of any point of the aluminum electrolytic cell to the ground can also cause electric shock of operators or damage of related electrical equipment of the aluminum electrolytic cell, if the faults are not discovered and cleared in time, the aluminum electrolytic cell can even be scrapped, and meanwhile, the aluminum electrolytic cell causes electric energy to flow into the ground due to the electricity leakage to the ground, thereby causing energy waste and causing great economic loss and serious safety accidents to enterprises.
At present, the judgment of the ground insulation fault of the aluminum electrolytic cell series at home and abroad mostly adopts a bus midpoint voltage drift detection method and a single-frequency signal injection detection method. However, the mid-point voltage drift detection method can only infer the approximate aluminum cell position of the aluminum cell in which the ground leakage accident occurs, and is difficult to accurately locate at a certain aluminum cell, while the single-frequency signal injection detection method can only detect the ground leakage fault of the aluminum cell series, and cannot accurately locate the specific fault cell and fault point. The two methods can not detect the resistance value of each point-to-ground insulation resistance of each cell and each electrolytic cell of the aluminum electrolytic cell series on line in real time, and can not monitor the resistance value change of each point-to-ground insulation resistance of each cell and each electrolytic cell of the aluminum electrolytic cell series on line in real time due to the failure of detecting the resistance value of each cell and each point-to-ground insulation resistance on line in real time, thereby failing to realize the accurate positioning of the earth leakage fault of each cell and each electrolytic cell of the aluminum electrolytic cell series and the early warning detection of the earth leakage fault of each cell and each electrolytic cell of the aluminum electrolytic cell series.
For example, the patent with publication number CN1948975B entitled "a device and method for detecting insulation resistance to ground of electrolytic cell series" in 2011, and the patent application with publication number CN112034283A entitled "device for detecting and locating ground fault of aluminum electrolytic cell and system and process thereof" in 2020. Both the above two patents belong to single-frequency signal injection detection methods, which can only detect that the aluminum electrolytic cell series has earth leakage fault, but can not accurately locate specific fault cells and fault points, can not detect the resistance value of each point-to-ground insulation resistance of each electrolytic cell and each electrolytic cell in real time on line, and can not realize early warning detection of earth leakage fault of each cell and each point of each electrolytic cell of the aluminum electrolytic cell series.
In summary, the reported techniques and methods in the prior art can not measure the insulation resistance value of the aluminum electrolytic cell to the ground, can not know the insulation performance of the aluminum electrolytic cell and accurately determine and position the ground leakage fault point of the aluminum electrolytic cell, and can not predict the performance change trend of the insulation performance deterioration of the aluminum electrolytic cell.
Disclosure of Invention
The invention provides a system and a method for positioning, early warning and detecting the earth leakage fault of an aluminum electrolysis cell series, aiming at the problem that the earth leakage fault point of the aluminum electrolysis cell series cannot be accurately positioned in the prior art.
In order to achieve the purpose, the invention provides the following technical scheme:
the aluminum cell series earth leakage fault positioning and early warning detection system comprises a voltage source, a direct current bus and i aluminum cells connected in series through the direct current bus, and is characterized in that each aluminum cell is connected with an earth leakage fault detection loop, and each earth leakage fault detection loop comprises a signal sending lead, a voltage receiving lead, an insulation resistor and a main control computer processing unit;
one end of each signal transmission conducting wire is connected with the main control computer processing unit, and the other end of each signal transmission conducting wire is connected with the grounding point of the corresponding aluminum electrolytic cell; the grounding point is also connected with the corresponding aluminum electrolytic cell through an insulation resistor; the grounding point is also connected with one end of the voltage receiving lead, the other end of the voltage receiving lead is connected with the main control computer processing unit, and the main control computer processing unit calculates the resistance value of the insulation resistance, so that the aluminum electrolytic cell with earth leakage fault is positioned.
Preferably, each aluminum reduction cell has a unique number, and each number corresponds to a unique installation location.
Preferably, the signal transmission lead is connected with the main control computer processing unit through a current limiting resistor.
Preferably, the main control computer processing unit comprises a signal generating module, a sending module, a receiving module, a calculating module, a judging module, a predicting alarm module and a controller, wherein the signal generating module, the sending module, the receiving module, the calculating module, the judging module and the predicting alarm module are respectively connected with the controller;
the signal generating module is used for generating a measuring signal; the sending module is used for sending the measuring signal to the grounding point of the aluminum electrolytic cell to be measured; the receiving module is used for receiving the voltage of the grounding point; the calculation module is used for calculating the resistance value of the corresponding insulation resistance of the aluminum electrolytic cell to be tested according to the voltage of the grounding point; the judging module is used for comparing the resistance values calculated by the calculating module; and the prediction alarm module is used for sending out an alarm prompt according to the result of the comparison module.
Preferably, the measuring signal generated by the signal generating module is a sinusoidal multi-carrier FSK signal, and the frequency range of the measuring signal is 500 Hz-5 KHz.
Preferably, the main control computer processing unit further comprises an early warning module for predicting the change of the insulation resistance according to the LSTM neural network and the historical insulation resistance value, so as to realize the early warning of the earth leakage fault by the aluminum electrolytic cell.
The invention also provides a method for positioning, early warning and detecting the earth leakage fault of the aluminum electrolytic cell series, which is characterized by comprising the system for positioning, early warning and detecting the earth leakage fault of the aluminum electrolytic cell series according to any one of claims 1 to 6, and the method specifically comprises the following steps:
s1: when the aluminum electrolytic cell Ci-1 to be tested is detected, the processing unit of the main control computer generates a measuring signal and transmits the measuring signal to the grounding point i-1 through the i-1 th signal transmission lead, and voltage Vi-1 is generated at the grounding point i-1;
s2: the measured signal returns to the processing unit of the main control computer after passing through the i-1 insulation resistor and the aluminum electrolytic cell Ci-1 to be measured; the voltage Vi-1 is fed back to the main control computer processing unit by the i-1 th voltage receiving lead, and the main control computer processing unit simultaneously receives the voltages of grounding points of other aluminum electrolytic cells;
s3: and the main control computer processing unit calculates the resistance value of the i-1 insulation resistance according to the received voltage, compares the resistance value with a threshold value, and if the insulation resistance value is smaller than the threshold value, the aluminum electrolytic cell is indicated to have a fault and needs to give an alarm immediately, and simultaneously displays the number of the aluminum electrolytic cell, thereby positioning the aluminum electrolytic cell with the earth leakage fault.
Preferably, the resistance value of the i-1 th insulation resistance is calculated as:
s3-1: respectively calculating the resistance RG between the grounding point i-1 of the i-1 th insulation resistor and the adjacent grounding point i-2aAnd a resistance RG between the grounding point i-1 and the adjacent grounding point ib：
In formula (1), RGaRepresenting the resistance between ground point i-1 and ground point i-2; RG (route group)bRepresenting the resistance between ground point i-1 and ground point i; vi-1Represents the voltage at ground point i-1; vi-2Represents the voltage at ground point i-2; vinRepresenting the measurement signal voltage; viRepresents the voltage of ground point i; r0Representing the resistance value of the current limiting resistor;
s3-2: respectively calculating the resistance values Z of the adjacent insulation resistors Zi-2 and Zi of the i-1 insulation resistori-2、Zi：
In the formula (2), Zi-2Denotes the resistance value, Z, of the i-2 th insulation resistor Zi-2 adjacent to the i-1 th insulation resistoriA resistance value of an i-th insulation resistor Zi adjacent to the i-1 th insulation resistor is represented;
s3-3: calculating the resistance value Z of the i-1 th insulation resistance Zi-1i-1:
First calculate
Ra＝RGa+Zi-2，Rb＝RGb+Zi，
Then let Rab＝Ra//RbDue to Rz＝Rab//Zi-1，
Then one can get:
the resistance value Z of the i-1 th insulation resistor Zi-1 is obtainedi-1：
In equation (3),// denotes the resistor parallel operator, RzIntermediate variables introduced for ease of calculation are shown for the total equivalent resistances Ra and Rb of the i-1 th ground points.
Preferably, the method comprises the following steps of S4: and respectively calculating the historical insulation resistance value of each aluminum electrolytic cell collected by the processing unit of the main control computer by adopting an LSTM neural network, and outputting the predicted value of the insulation resistance.
Preferably, the calculation of the predicted value of the insulation resistance includes the steps of:
s4-1: constructing an LSTM-based aluminum electrolytic cell earth insulation resistance prediction network;
s4-2: acquiring a historical insulation resistance value of the aluminum electrolysis cell stored by a processing unit of a main control computer, and dividing the historical insulation resistance value into a training set, a verification set and a test set;
s4-3: initializing LSTM-based aluminum electrolytic cell earth insulation resistance prediction network model parameters:
output y of the network and an initial value c<t-1>Initialization is 0; weight matrix w of network hiding layercAnd bias bcWeight matrix w of input gatesiAnd bias biUpdating the weight matrix w of the gateuAnd bias buWeight matrix w of output gatesoAnd bias boAnd weight matrix w of forgetting gatefAnd bias bfInitializing the random number to be 0-1; setting the number of neurons of a network input layer as i +1, and setting the number of neurons of a network hiding layer, an input gate, a forgetting gate and an output gate as i; c. C<t>Memory information indicating time t; a is<t>Representing the output at time t of the model; x is the number of<t>Representing the input of the model at time t;
s4-4: training an LSTM-based aluminum electrolysis cell earth insulation resistance prediction network model:
training the model by using the divided sample training set, and optimizing the model weight and bias by using a reverse tuning algorithm J according to the difference between the predicted value and the true value;
in a single LSTM cell architecture, with a forgetting gate Γf <t>Output a according to time t-1<t-1>And input x at time t<t>Calculating the information quantity inheriting the last time step:
Γf <t>＝σ(Wf[a<t-1>,x<t>]+bf)；
at the same time according to a<t-1>And x<t>Calculating the memory information needing to be updated
By updating the gate gammau <t>Computing
The ratio of the node memory information to the final node memory information c<t>Jointly determining by a forgetting gate and an updating gate:
Γu <t>＝σ(Wu[a<t-1>,x<t>]+bu)，
the model output is then passed through an output gate Γo <t>And the node memory information is calculated
Γo <t>＝σ(Wo[a<t-1>,x<t>]+bo)，a<t>＝Γo <t>·tanh(c<t>) σ denotes sigmoid function;
s4-5: and inputting the collected insulation resistance value of the aluminum electrolytic cell into the trained model to obtain the predicted value of the insulation resistance.
In summary, due to the adoption of the technical scheme, compared with the prior art, the invention at least has the following beneficial effects:
the invention can monitor the earth insulation resistance value of the aluminum electrolytic cell series on line in real time, and can judge whether the insulation performance of the aluminum electrolytic cell is good or bad and whether the insulation failure fault occurs or not according to the insulation resistance value and accurately position the aluminum electrolytic cell with the earth leakage fault, thereby informing maintainers of removing the fault in time, avoiding the occurrence of accidents and ensuring the safety of operators and the efficient normal operation of the aluminum electrolytic cell. Meanwhile, the resistance value detected by positioning is utilized, and the change of insulation resistance of each point of the aluminum electrolytic cell and the aluminum electrolytic cell to the ground is predicted through an LSTM neural network, so that the early warning of the earth leakage fault is realized. The invention can timely and accurately detect, position and early warn the earth leakage fault of a single aluminum electrolysis cell or a plurality of aluminum electrolysis cells and all points of the aluminum electrolysis cells in the aluminum electrolysis cell series; the current efficiency, the maintenance efficiency and the energy utilization rate are improved, and the production and maintenance cost is reduced.
Description of the drawings:
fig. 1 is a schematic diagram of a system for positioning and early warning ground leakage faults of an aluminum electrolysis cell series according to an exemplary embodiment of the invention.
Fig. 2 is a schematic diagram of a method for detecting the location of an earth leakage fault by an aluminum electrolysis cell series according to an exemplary embodiment of the invention.
Fig. 3 is a schematic diagram of an equivalent circuit of an insulation resistance according to an exemplary embodiment of the present invention.
Fig. 4 is a schematic diagram of an early warning method for an earth leakage fault of an aluminum electrolysis cell series according to an exemplary embodiment of the invention.
FIG. 5 is a schematic diagram of an LSTM-based aluminum electrolysis cell series insulation resistance to ground prediction network according to an exemplary embodiment of the present invention.
Detailed Description
The present invention will be described in further detail with reference to examples and embodiments. It should be understood that the scope of the above-described subject matter is not limited to the following examples, and any techniques implemented based on the disclosure of the present invention are within the scope of the present invention.
In the description of the present invention, it is to be understood that the terms "longitudinal", "lateral", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", and the like, indicate orientations or positional relationships based on those shown in the drawings, and are used merely for convenience of description and for simplicity of description, and do not indicate or imply that the referenced devices or elements must have a particular orientation, be constructed in a particular orientation, and be operated, and thus, are not to be construed as limiting the present invention.
As shown in figure 1, the invention provides a system for positioning and detecting earth leakage faults of an aluminum electrolysis cell series, which comprises a voltage source and a direct current bus, wherein m aluminum electrolysis cells are connected in series through the direct current bus to form the aluminum electrolysis cell series (the aluminum electrolysis cell is numbered as C)1,C2,...,Ci,Ci+1,...,Cm-2,Cm-1,Cm) And each aluminum electrolytic cell is respectively connected with a ground leakage fault detection loop.
The earth leakage fault detection loop comprises a signal sending lead, a voltage receiving lead, an insulation resistor, a grounding resistor, a current limiting resistor and a main control computer processing unit U1 (which can be called as a main control computer for short). One end of each signal transmission conducting wire is connected with the main control computer processing unit, and the other end of each signal transmission conducting wire is connected with the grounding point of the corresponding aluminum electrolytic cell; the grounding point is also connected with the corresponding aluminum electrolytic cell through an insulation resistor; the grounding point is also connected with one end of the voltage receiving lead, the other end of the voltage receiving lead is connected with the main control computer processing unit, and the main control computer processing unit calculates the resistance value of the insulation resistance, so as to position the aluminum electrolytic cell with earth leakage fault and the specific fault points (such as air doors, inlets, outlets and the like) of the aluminum electrolytic cell; the grounding point is also grounded through a grounding resistor.
I.e. the system comprises m signal transmission wires (T)1,T2,...,Ti,Ti+1,...,Tm-2,Tm-1,Tm) M voltage receiving conductors (R)1,R2,...,Ri,Ri+1,...,Rm-2,Rm-1,Rm) M insulation resistances (Z)1,Z2,...,Zi,Zi+1,...,Zm-2,Zm-1,Zm) M ground Resistances (RG)1,RG2,...,RGi,RGi+1,...,RGm-2,RGm-1,RGm) M current limiting resistors (Ro)1,Ro2,...,Rom) And a master computer processing unit U1.
In this embodiment, each aluminum electrolysis cell has a unique number, and each number corresponds to a unique position. The signal transmitting wire and the voltage receiving wire are both high-temperature fluorine-proof twisted pair wires.
m-th signal transmission wire T of m signal transmission wiresmOne end of (3) and the m current limiting resistor ROmIs connected to the mth current limiting resistor ROmThe other end of the main control computer is connected with a sending module of the main control computer processing unit; mth signal transmission wire TmAnd the other end of (2) and the corresponding mth aluminum electrolytic cell CmIs connected with the m-th voltage receiving wire RmIs connected to the mth voltage receiving conductor RmThe other end of the main control computer is connected with a receiving module of the main control computer processing unit. The grounding point m is also respectively connected with the m insulation resistor ZmAnd m-th ground resistance RGmIs connected to the m-th ground resistance RGmThe other end of (a) is grounded, and an m-th insulation resistance ZmThe other end of the aluminum electrolysis cell C and the mth aluminum electrolysis cell CmAnd (4) connecting.
For example, the 1 st signal transmission conductor T1And a first current limiting resistor Ro1Is connected to a first current limiting resistor Ro1The other end of the master control meterThe sending module of the computer processing unit is connected; no. 1 signal transmission wire T1The other end of the aluminum electrolysis cell C is connected with the corresponding 1 st aluminum electrolysis cell C1The grounding point 1 is connected with the 1 st voltage receiving conducting wire R1Is connected to the 1 st voltage receiving conductor R1The other end of the main control computer is connected with a receiving module of the main control computer processing unit. The grounding point 1 is also respectively connected with the 1 st insulation resistor Z1And 1 st ground resistance RG1Is connected to the 1 st ground resistance RG1The other end of (1) is grounded, and the 1 st insulation resistance Z1The other end of the aluminum electrolysis cell C is connected with the 1 st aluminum electrolysis cell C1And (4) connecting.
The signal generating module of the processing unit of the main control computer generates a measuring signal, the signal transmitting module transmits the measuring signal to the grounding point of the corresponding aluminum electrolytic cell through a signal transmitting lead, and the measuring signal passes through the insulation resistor, passes through the aluminum electrolytic cell to be measured and the direct current BUS and then returns to the Vs end of the processing unit of the main control computer through the S-BUS. And the measuring signal generates voltage (V) at the grounding point corresponding to the aluminum electrolytic cell1,V2,...,Vi,Vi+1,...,Vm-2,Vm-1,Vm) The voltage returns to the main control computer processing unit through the voltage receiving lead, the main control computer processing unit calculates the insulation resistance value according to the voltage and compares the insulation resistance value with the threshold value required by the electrolytic aluminum industry for the insulation resistance of the electrolytic cell to the ground to judge whether the aluminum electrolytic cell has poor insulation or insulation damage, if the insulation resistance value is smaller than the threshold value, the aluminum electrolytic cell is indicated that the leakage fault occurs, the alarm is immediately given, and the number of the aluminum electrolytic cell is displayed at the same time, so that the aluminum electrolytic cell positioning of the leakage fault is realized; if the insulation resistance value is larger than or equal to the threshold value, the aluminum electrolytic cell does not have earth leakage fault.
The bus current of the aluminum electrolysis field is very large, which can reach hundreds of thousands of amperes, so the electromagnetic interference generated by the bus current is very serious, the bus current is anti-interference, the transmitted measuring signal adopts a sine multi-carrier FSK signal format, generally adopts four carrier frequencies (such as 500Hz, 2.4KHz, 4KHz and 5KHz), four resistance values can be measured by the four frequencies, the characteristic of the insulation resistance can be determined to be capacitive or inductive according to the change of the resistance value along with the frequency, and the resistance value measured by which frequency is more suitable for being used as the final insulation resistance value can be determined according to the characteristic. Meanwhile, the multi-frequency point is also beneficial to the anti-interference of field measurement, and frequencies outside the interference frequency band range can be selected as the measurement signal frequency, the frequency range is from 500Hz to 5KHz, and the specific frequency is determined according to a field interference spectrogram. The carrier wave change of the measuring signal is switched according to the frequency of a fixed period (4Hz), the initial phases of the carrier waves are the same, the amplitude of the signal is not more than 15V, and the driving current is less than 1A.
The 1 st aluminum electrolysis cell is taken as an example for explanation:
the signal generating module of the main control computer processing unit generates a measuring signal, the signal sending module sends the measuring signal to the grounding point 1 of the aluminum electrolytic cell C1 through a signal sending lead T1, and the measuring signal passes through the insulation resistor Z1, then passes through the aluminum electrolytic cell C1 and the direct current bus and then returns to the main control computer processing unit. Meanwhile, the measurement signal generates a voltage V1 at the grounding point 1 of the aluminum electrolysis cell C1, the voltage V1 returns to the main control computer processing unit through a voltage receiving lead, and the main control computer processing unit calculates the resistance value of the insulation resistance Z1 according to the voltage V1 and compares the resistance value with a threshold value (each aluminum electrolysis cell has a threshold value requirement). If the resistance value of the insulation resistance Z1 is smaller than the threshold value, the aluminum cell is indicated to have a fault and an alarm is given immediately, and the number of the aluminum cell is displayed at the same time, so that the aluminum cell with the earth leakage fault is positioned.
The main control computer processing unit comprises a signal generating module, a sending module, a receiving module, a calculating module, a judging module, a prediction alarm module and a controller, wherein the signal generating module, the sending module, the receiving module, the calculating module, the judging module and the prediction alarm module are respectively connected with the controller.
The signal generating module is used for generating a measuring signal; the sending module is used for sending the measuring signal to the grounding point of the aluminum electrolytic cell to be measured; the receiving module is used for receiving the voltage of the grounding point; the calculation module is used for calculating the resistance value of the corresponding insulation resistance of the aluminum electrolytic cell to be tested according to the voltage of the grounding point; the judging module is used for comparing the resistance values calculated by the calculating module; and the prediction alarm module is used for sending an alarm prompt according to the result of the comparison module, if the insulation resistance value is smaller than a standard value, the aluminum electrolytic cell is indicated to have a fault and an alarm is given immediately, and meanwhile, the number of the aluminum electrolytic cell is displayed, so that the aluminum electrolytic cell with the fault of the ground leakage is positioned.
The main control computer processing unit also comprises an early warning module which is used for predicting the change of the insulation resistance according to the LSTM neural network and the historical earth insulation resistance value so as to realize the early warning of the earth leakage fault of each point of the aluminum electrolytic cell and the aluminum electrolytic cell.
Based on the system, as shown in fig. 2, the invention also provides a method for positioning and detecting the earth leakage fault of the aluminum electrolytic cell series, which comprises a system for positioning and early warning and detecting the earth leakage fault of the aluminum electrolytic cell series, and specifically comprises the following steps:
s1: when the aluminum electrolytic cell Ci-1 to be tested is detected, the processing unit of the main control computer generates a measuring signal and transmits the measuring signal to the grounding point i-1 through the i-1 th signal transmission lead, and voltage Vi-1 is generated at the grounding point i-1;
s2: the measured signal returns to the processing unit of the main control computer after passing through the i-1 insulation resistor and the aluminum electrolytic cell Ci-1 to be measured; the voltage Vi-1 is fed back to the main control computer processing unit by the i-1 th voltage receiving lead, and the main control computer processing unit simultaneously receives the voltage values of grounding points of other aluminum electrolysis cells.
In the embodiment, the grounding points of the aluminum electrolysis cells are connected through the grounding network, so when the grounding point i-1 of the aluminum electrolysis cell to be tested generates the voltage Vi-1, the grounding points of other aluminum electrolysis cells also generate voltages correspondingly and transmit the voltages to the processing unit of the main control computer through the corresponding voltage receiving wires for storage.
S3: and the main control computer processing unit calculates the resistance value of the i-1 insulation resistance according to the received voltage, compares the resistance value with a threshold value, and if the insulation resistance value is smaller than the threshold value, indicates that the aluminum electrolytic cell has the earth leakage fault and immediately alarms, and simultaneously displays the number of the aluminum electrolytic cell, thereby realizing the positioning of the aluminum electrolytic cell having the earth leakage fault.
In this embodiment, the resistance value of the i-1 th insulation resistor is calculated as shown in fig. 3, which is an equivalent circuit diagram of the resistance value of the insulation resistor:
the main control computer processing unit sends a frequency point FSK signal, and generates a voltage Vi-1 at an earthing point i-1 through a current-limiting resistor ROi-1 and a signal sending lead, and other earthing points also generate corresponding voltages through an earthing network.
S3-1: i.e. calculating the resistance RG between the grounding point i-1 of the i-1 th insulation resistance and the adjacent grounding point i-2aAnd a resistance RG between the grounding point i-1 and the adjacent grounding point ib：
In formula (1), RGaRepresenting the resistance between ground point i-1 and ground point i-2; RG (route group)bRepresenting the resistance between ground point i-1 and ground point i; vi-1Represents the voltage at ground point i-1; vi-2Represents the voltage at ground point i-2; vinRepresenting the measurement signal voltage; viRepresents the voltage of ground point i; r0Representing the resistance value of the current limiting resistor ROi-1.
S3-2: respectively calculating the resistance values Z of the adjacent insulation resistors Zi-2 and Zi of the i-1 insulation resistori-2、Zi：
In the formula (2), Zi-2Denotes the resistance value, Z, of the i-2 th insulation resistor Zi-2 adjacent to the i-1 th insulation resistoriA resistance value of an i-th insulation resistor Zi adjacent to the i-1 th insulation resistor is represented;
s3-3: calculating the resistance value Z of the i-1 th insulation resistance Zi-1i-1:
First calculate
Ra＝RGa+Zi-2，Rb＝RGb+Zi，
Then let Rab＝Ra//RbDue to Rz＝Rab//Zi-1
Then one can get:
the resistance value Z of the i-1 th insulation resistor Zi-1 is obtainedi-1：
In equation (3),// denotes the resistor parallel operator, RzRepresenting the total equivalent resistance of the i-1 th ground point, Ra and Rb are intermediate variables introduced for ease of calculation.
In this example, when the first insulation resistance is calculated as the first aluminum electrolytic cell is measured, the previous value is considered to be infinite, and Ra is infinite in the corresponding equation (3), and R isab＝Rb(ii) a When the last aluminum electrolytic cell is measured, namely the ith insulation resistance is calculated, the latter value is considered to be infinite and corresponds to R in the formula (3)bIs infinite, Rab＝Ra。
In this embodiment, for those aluminum electrolysis cells in which no earth leakage fault occurs, the processing unit of the main control computer may predict the variation trend of the insulation resistance value by using a neural network method according to the historical insulation resistance value. The neural network method is realized by adopting a long-short period memory neural network (LSTM). LSTM neural networks are suitable for processing and predicting significant events of very long intervals and delays in time series. Because the change of the insulation resistance of the aluminum electrolytic cell to the ground is a process which gradually deteriorates for a long time and is matched with an object suitable for the LSTM method, the change trend of the corresponding insulation resistance of the aluminum electrolytic cell can be well predicted by adopting the LSTM method, and the early warning success rate is high.
Further comprising S4: and respectively calculating the historical insulation resistance value of each aluminum electrolytic cell collected by the processing unit of the main control computer by adopting an LSTM neural network, and outputting the predicted value of the insulation resistance value. As shown in figure 4 of the drawings,
s4-1: constructing an LSTM-based aluminum electrolytic cell earth insulation resistance prediction network;
in this embodiment, because the LSTM network has long-term memory for data, fully considers the time dependency of data, and can automatically adjust the contribution of history information to the current prediction according to the current state, a prediction network based on the ground insulation resistance of the LSTM aluminum electrolysis cell is constructed, as shown in fig. 5.
S4-2: acquiring a historical insulation resistance value of the aluminum electrolysis cell stored by a processing unit of a main control computer, and dividing the historical insulation resistance value into a training set, a verification set and a test set;
in this embodiment, the historical insulation resistance value of the aluminum electrolytic cell is obtained from the storage module of the processing unit of the main control computer, the historical insulation resistance value is normalized, and the normalized insulation resistance value is obtained according to the ratio of 8: 1: the scale of 1 is divided into a training set, a validation set, and a test set.
In the embodiment, the historical insulation resistance value of any aluminum electrolytic cell is collected once every time period T, collected for n times in total and used as model input, the resistance value of the corresponding time T is used as a label, and whether the aluminum electrolytic cell has the ground leakage or not is judged according to the size relation between the predicted value and the set threshold value.
S4-3: initializing LSTM-based aluminum electrolytic cell earth insulation resistance prediction network model parameters:
output y of the network and an initial value c<t-1>Initialization is 0; weight matrix w of network hiding layercAnd bias bcWeight matrix w of input gatesiAnd bias biUpdating the weight matrix w of the gateuAnd bias buWeight matrix w of output gatesoAnd bias boAnd weight matrix w of forgetting gatefAnd bias bfInitializing the random number to be 0-1; setting the number of neurons of a network input layer as i +1, and setting the number of neurons of a network hiding layer, an input gate, a forgetting gate and an output gate as i; c. C<t>Memory information indicating time t; a is<t>Representing the model at time tAn output of (d); x is the number of<t>Representing the input at time t of the model.
S4-4: training an LSTM-based aluminum electrolysis cell earth insulation resistance prediction network model:
setting a hidden layer neuron activation function A, an output unit activation function B, a model loss function L, a reverse optimization algorithm J and a target error value E; and training the model by using the divided sample training set, and optimizing the model weight and the bias by using a reverse tuning algorithm J according to the difference between the predicted value and the true value.
In a single LSTM cell architecture, with a forgetting gate Γf <t>Output a according to time t-1<t-1>And input x at time t<t>Calculating the information quantity inheriting the last time step:
Γf <t>＝σ(Wf[a<t-1>,x<t>]+bf)；
at the same time according to a<t-1>And x<t>Calculating the memory information needing to be updated
By updating the gate gammau <t>Computing
The ratio of the node memory information to the final node memory information c<t>Jointly determining by a forgetting gate and an updating gate:
Γu <t>＝σ(Wu[a<t-1>,x<t>]+bu)，
the model output is then passed through an output gate Γo <t>And the node memory information is calculated
Γo <t>＝σ(Wo[a<t-1>,x<t>]+bo)
a<t>＝Γo <t>·tanh(c<t>) (ii) a Sigma represents a sigmoid function;
s4-5: optimizing an LSTM-based aluminum electrolytic cell earth insulation resistance prediction network model:
bringing the verification set into a trained model for performance verification, and finely adjusting the hyper-parameters of the model according to the deviation and variance of the model; and finally evaluating the model performance by using a test set, and taking an average absolute error evaluation method as an LSTM performance evaluation index.
S4-6: and inputting the collected insulation resistance value of the aluminum electrolytic cell into the trained model to obtain an insulation resistance prediction result.
It will be understood by those of ordinary skill in the art that the foregoing embodiments are specific examples for carrying out the invention, and that various changes in form and details may be made therein without departing from the spirit and scope of the invention in practice.
Claims (10)
1. The aluminum cell series earth leakage fault positioning and early warning detection system comprises a voltage source, a direct current bus and aluminum cell series connected in series through the direct current bus, and is characterized in that each aluminum cell is connected with an earth leakage fault detection loop, and each earth leakage fault detection loop comprises a signal sending lead, a voltage receiving lead, an insulation resistor and a main control computer processing unit;
one end of each signal transmission conducting wire is connected with the main control computer processing unit, and the other end of each signal transmission conducting wire is connected with the grounding point of the corresponding aluminum electrolytic cell; the grounding point is also connected with the corresponding aluminum electrolytic cell through an insulation resistor; the grounding point is also connected with one end of the voltage receiving lead, the other end of the voltage receiving lead is connected with the main control computer processing unit, and the main control computer processing unit calculates the resistance value of the insulation resistance, so that the aluminum electrolytic cell with earth leakage fault is positioned.
2. The aluminum electrolysis cell series earth leakage fault positioning and early warning detection system according to claim 1, wherein each aluminum electrolysis cell has a unique number, and each number corresponds to a unique installation position.
3. The aluminum electrolysis cell series earth leakage fault location and early warning detection system of claim 1, wherein the signal transmission lead is connected with the main control computer processing unit through a current limiting resistor.
4. The aluminum electrolysis cell series earth leakage fault positioning and early warning detection system of claim 1, wherein the main control computer processing unit comprises a signal generating module, a transmitting module, a receiving module, a calculating module, a judging module, a prediction alarm module and a controller, and the signal generating module, the transmitting module, the receiving module, the calculating module, the judging module and the prediction alarm module are respectively connected with the controller;
the signal generating module is used for generating a measuring signal; the sending module is used for sending the measuring signal to the grounding point of the aluminum electrolytic cell to be measured; the receiving module is used for receiving the voltage of the grounding point; the calculation module is used for calculating the resistance value of the corresponding insulation resistance of the aluminum electrolytic cell to be tested according to the voltage of the grounding point; the judging module is used for comparing the resistance values calculated by the calculating module; and the prediction alarm module is used for sending out an alarm prompt according to the result of the comparison module.
5. The aluminum electrolysis cell series earth leakage fault positioning and early warning detection system as claimed in claim 4, wherein the measurement signal generated by the signal generation module is a sinusoidal multi-carrier FSK signal, and the frequency range of the measurement signal is 500 Hz-5 KHz.
6. The aluminum reduction cell series earth leakage fault positioning and early warning detection system as claimed in claim 4, wherein the main control computer processing unit further comprises an early warning module for predicting the change of insulation resistance according to the LSTM neural network and the historical insulation resistance value to realize the early warning of the aluminum reduction cell to the earth leakage fault.
7. The method for positioning and early warning and detecting the earth leakage fault of the aluminum electrolytic cell series is characterized by comprising the system for positioning and early warning and detecting the earth leakage fault of the aluminum electrolytic cell series according to any one of claims 1 to 6, and specifically comprises the following steps:
s1: when the aluminum electrolytic cell Ci-1 to be tested is detected, the processing unit of the main control computer generates a measuring signal and transmits the measuring signal to the grounding point i-1 through the i-1 th signal transmission lead, and voltage Vi-1 is generated at the grounding point i-1;
s2: the measured signal returns to the processing unit of the main control computer after passing through the i-1 insulation resistor and the aluminum electrolytic cell Ci-1 to be measured; the voltage Vi-1 is fed back to the main control computer processing unit by the i-1 th voltage receiving lead, and the main control computer processing unit simultaneously receives the voltages of grounding points of other aluminum electrolytic cells;
s3: and the main control computer processing unit calculates the resistance value of the i-1 insulation resistance according to the received voltage, compares the resistance value with a threshold value, and if the insulation resistance value is smaller than the threshold value, the aluminum electrolytic cell is indicated to have a fault and needs to give an alarm immediately, and simultaneously displays the number of the aluminum electrolytic cell, thereby positioning the aluminum electrolytic cell with the earth leakage fault.
8. The method for positioning and early warning of ground leakage fault of aluminum electrolysis cell series according to claim 7, wherein the resistance value of the i-1 insulation resistor is calculated as:
s3-1: respectively calculating the resistance RG between the grounding point i-1 of the i-1 th insulation resistor and the adjacent grounding point i-2aAnd a resistance RG between the grounding point i-1 and the adjacent grounding point ib：
In formula (1), RGaRepresenting the resistance between ground point i-1 and ground point i-2; RG (route group)bRepresenting the resistance between ground point i-1 and ground point i; vi-1Represents the voltage at ground point i-1; vi-2Represents the voltage at ground point i-2; vinRepresenting the measurement signal voltage; viRepresents the voltage of ground point i; r0Representing the resistance value of the current limiting resistor;
s3-2: respectively calculating the resistance values Z of the adjacent insulation resistors Zi-2 and Zi of the i-1 insulation resistori-2、Zi：
In the formula (2), Zi-2Denotes the resistance value, Z, of the i-2 th insulation resistor Zi-2 adjacent to the i-1 th insulation resistoriA resistance value of an i-th insulation resistor Zi adjacent to the i-1 th insulation resistor is represented;
s3-3: calculating the resistance value Z of the i-1 th insulation resistance Zi-1i-1:
First calculate
Ra＝RGa+Zi-2，Rb＝RGb+Zi，
Then let Rab＝Ra//RbDue to Rz＝Rab//Zi-1，
Then one can get:
the resistance value Z of the i-1 th insulation resistor Zi-1 is obtainedi-1：
In equation (3),// denotes the resistor parallel operator, RzIntermediate variables introduced for ease of calculation are shown for the total equivalent resistances Ra and Rb of the i-1 th ground points.
9. The method for positioning and early warning of ground leakage fault of aluminum electrolysis cell series according to claim 7, comprising the steps of S4: and respectively calculating the historical insulation resistance value of each aluminum electrolytic cell collected by the processing unit of the main control computer by adopting an LSTM neural network, and outputting the predicted value of the insulation resistance.
10. The method for positioning and early warning of ground leakage fault of aluminum electrolysis cell series according to claim 9, wherein the calculation of the predicted value of the insulation resistance comprises the following steps:
s4-1: constructing an LSTM-based aluminum electrolytic cell earth insulation resistance prediction network;
s4-2: acquiring a historical insulation resistance value of the aluminum electrolysis cell stored by a processing unit of a main control computer, and dividing the historical insulation resistance value into a training set, a verification set and a test set;
s4-3: initializing LSTM-based aluminum electrolytic cell earth insulation resistance prediction network model parameters:
output y of the network and an initial value c<t-1>Initialization is 0; weight matrix w of network hiding layercAnd bias bcWeight matrix w of input gatesiAnd bias biUpdating the weight matrix w of the gateuAnd bias buWeight matrix w of output gatesoAnd bias boAnd weight matrix w of forgetting gatefAnd bias bfInitializing the random number to be 0-1; setting the number of neurons of a network input layer as i +1, and setting the number of neurons of a network hiding layer, an input gate, a forgetting gate and an output gate as i; c. C<t>Memory information indicating time t; a is<t>Representing the output at time t of the model; x is the number of<t>Representing the input of the model at time t;
s4-4: training an LSTM-based aluminum electrolysis cell earth insulation resistance prediction network model:
training the model by using the divided sample training set, and optimizing the model weight and bias by using a reverse tuning algorithm J according to the difference between the predicted value and the true value;
in a single LSTM cell architecture, with a forgetting gate Γf <t>Output a according to time t-1<t-1>And input x at time t<t>Calculating the information quantity inheriting the last time step:
Γf <t>＝σ(Wf[a<t-1>,x<t>]+bf)；
at the same time according to a<t-1>And x<t>Calculating the memory information needing to be updated
By updating the gate gammau <t>Computing
The ratio of the node memory information to the final node memory information c<t>Jointly determining by a forgetting gate and an updating gate:
Γu <t>＝σ(Wu[a<t-1>,x<t>]+bu)，
the model output is then passed through an output gate Γo <t>And the node memory information is calculated
Γo <t>＝σ(Wo[a<t-1>,x<t>]+bo)，a<t>＝Γo <t>·tanh(c<t>) σ denotes sigmoid function;
s4-5: and inputting the collected insulation resistance value of the aluminum electrolytic cell into the trained model to obtain the predicted value of the insulation resistance.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110460764.9A CN113049985A (en) | 2021-04-27 | 2021-04-27 | Aluminum cell series earth leakage fault positioning and early warning detection system and method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110460764.9A CN113049985A (en) | 2021-04-27 | 2021-04-27 | Aluminum cell series earth leakage fault positioning and early warning detection system and method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113049985A true CN113049985A (en) | 2021-06-29 |
Family
ID=76520539
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110460764.9A Pending CN113049985A (en) | 2021-04-27 | 2021-04-27 | Aluminum cell series earth leakage fault positioning and early warning detection system and method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113049985A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114217134A (en) * | 2021-11-01 | 2022-03-22 | 通富微电子股份有限公司 | Ground resistance monitor and ground resistance monitoring system |
| CN116008755A (en) * | 2023-02-07 | 2023-04-25 | 深圳新立图智能科技有限公司 | Method and system for detecting multipoint grounding of electrolytic tank based on voltage response of injection signal |
| CN116106697A (en) * | 2022-12-26 | 2023-05-12 | 深圳新立图智能科技有限公司 | Electrolytic tank earth insulation fault detection method and system based on voltage phase |
| WO2024066004A1 (en) * | 2022-09-26 | 2024-04-04 | 内蒙古霍煤鸿骏铝电有限责任公司 | Detection method for insulation-to-ground fault of aluminum electrolytic cell |
-
2021
- 2021-04-27 CN CN202110460764.9A patent/CN113049985A/en active Pending
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114217134A (en) * | 2021-11-01 | 2022-03-22 | 通富微电子股份有限公司 | Ground resistance monitor and ground resistance monitoring system |
| CN114217134B (en) * | 2021-11-01 | 2023-11-17 | 通富微电子股份有限公司 | Ground resistance monitor and ground resistance monitoring system |
| WO2024066004A1 (en) * | 2022-09-26 | 2024-04-04 | 内蒙古霍煤鸿骏铝电有限责任公司 | Detection method for insulation-to-ground fault of aluminum electrolytic cell |
| CN116106697A (en) * | 2022-12-26 | 2023-05-12 | 深圳新立图智能科技有限公司 | Electrolytic tank earth insulation fault detection method and system based on voltage phase |
| CN116106697B (en) * | 2022-12-26 | 2024-06-11 | 深圳新立图智能科技有限公司 | Electrolytic tank earth insulation fault detection method and system based on voltage phase |
| CN116008755A (en) * | 2023-02-07 | 2023-04-25 | 深圳新立图智能科技有限公司 | Method and system for detecting multipoint grounding of electrolytic tank based on voltage response of injection signal |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113049985A (en) | Aluminum cell series earth leakage fault positioning and early warning detection system and method | |
| US11674994B2 (en) | Distribution fault location using graph neural network with both node and link attributes | |
| CN101493491B (en) | Substation ground network defect synthesis diagnosis method and diagnosis system thereof | |
| US10955456B2 (en) | Method and apparatus for automatic localization of a fault | |
| JP3100757B2 (en) | Monitoring and diagnostic equipment | |
| CN106017551A (en) | Intelligent transmission line integrated monitoring analysis and early warning method | |
| CN106503884A (en) | A kind of method that health state evaluation is carried out to switch cubicle | |
| JP2021148778A (en) | Failure position identification method based on transient state for ungrounded power distribution system | |
| CN102879735A (en) | Online electric abrasion detecting method and system of vacuum circuit breaker | |
| CN107561407B (en) | Distributed transformer substation grounding grid detection and evaluation method | |
| CN104218570A (en) | Method and system for online evaluating overall measuring errors of electric energy measuring device | |
| CN107843816A (en) | A kind of transformer insulated defect state appraisal procedure for considering load factor and influenceing | |
| CN107808044A (en) | A kind of transformer insulating paper deterioration appraisal procedure for considering running temperature and influenceing | |
| CN215116704U (en) | Aluminum cell series earth leakage fault positioning detection and early warning device | |
| CN105467971A (en) | Electric power secondary equipment monitoring system and method | |
| CN107632521A (en) | A kind of potentiostat control strategy based on decision tree and neutral net | |
| CN116626543B (en) | Heating cable fault detection method and system based on skin effect | |
| CN106093516A (en) | Replay method based on the transmission line malfunction current traveling wave waveform that multiband decomposes | |
| CN108932531A (en) | Equipment state on-line checking algorithm based on model feature value | |
| CN109406923A (en) | A kind of power supply system of train earth leakage failure prediction method and device | |
| CN205861834U (en) | Failure tester | |
| CN115222160A (en) | Rail transit traction load prediction method based on measured big data | |
| CN114997375A (en) | Insulator surface metal pollution model building system in air type switch cabinet | |
| CN112464848A (en) | Information flow abnormal data monitoring method and device based on density space clustering | |
| Zhang et al. | Electric vehicle charging fault diagnosis based on bp neural network |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |