Substation equipment running state monitoring control system based on data analysis
Technical Field
The invention relates to the technical field of data processing, in particular to a substation equipment running state monitoring and controlling system based on data analysis.
Background
As the voltage class in a substation increases, the impact of the operation of the power transmission and transformation equipment on the surrounding residents increases significantly. Large power transformers and substations have increasingly appeared in residential and public places, which puts certain demands on the noise of transformers and substations.
The operation failure of the power transformer is a key cause of a large-scale power failure accident of a power system. More than 30% of typical fault defects (such as over-excitation, three-phase unbalanced load, harmonic load, cooling system defect, direct current magnetic bias, loosening of windings and iron cores, loosening of accessories, overload and the like) of the transformer can be detected by using noise.
The existing fault detection method related to the noise of the transformer mainly depends on that operators judge the fault type by using human ears and experience, or the waveform and the frequency spectrum of a sound signal are used, and a certain number of sound characteristics are extracted from the waveform and the frequency spectrum to be used as a basis for distinguishing different running states. The former mainly depends on the experience of personnel, and has the problems of difficult universal application, high misjudgment probability, low identification accuracy and the like; in the latter, because of more transformer fault types, the extraction of sound characteristics which can obviously distinguish different fault types is very difficult, and the problems of insufficient characteristic information and low fault diagnosis accuracy rate are easily caused; moreover, because the occupied area of the transformer equipment area is large, a maintainer cannot accurately find a source with a fault at the first time, and the overhauling efficiency is influenced.
Disclosure of Invention
The invention aims to provide a substation equipment running state monitoring and controlling system based on data analysis, which solves the following technical problems:
how can be more quick find out the faulty equipment among the substation equipment, promote to overhaul maintenance efficiency.
The purpose of the invention can be realized by the following technical scheme:
a substation equipment running state monitoring and control system based on data analysis comprises:
the monitoring modules are distributed at the edge of the transformer equipment area;
aiming at one monitoring module, the monitoring module comprises a plurality of monitoring units, and the distances from the monitoring units to the center of the transformer equipment area are different;
each monitoring unit is internally provided with an equipment information code and is used for acquiring sound wave data associated with the equipment information code according to a preset frequency;
the acoustic signal processing unit is used for carrying out feature extraction processing on the acoustic wave data to acquire voiceprint image feature information corresponding to the acoustic wave data;
the voiceprint recognition unit is used for judging whether the voiceprint image characteristic information contains the noise data of the voltage transformation equipment;
the fault pre-estimation unit is used for acquiring corresponding first noise pre-judgment directions according to the sequence of sampling timestamps of the noise data of the transformer equipment in the same monitoring module, and then determining a first fault pre-estimation area according to first cross points of the first noise pre-judgment directions of all the monitoring modules;
the control unit is used for adjusting the power change of the target equipment according to a preset sequence; the target equipment is working equipment in the first fault prediction area;
and the verification adjusting unit is used for adjusting the coverage range of the first fault prediction area according to a preset mode when the first intersection is detected to change.
According to the technical scheme, noise collection can be simultaneously carried out in the power transformation equipment area through the plurality of monitoring modules distributed at the edge of the power transformation equipment area, the monitoring units at different positions can continuously obtain sound wave data related to equipment information codes, if the noise data of the power transformation equipment exist, due to the fact that the same noise arrives at the monitoring units at different positions in different time, the source of the noise can be judged according to the sequence of sampling time stamps corresponding to the noise data of the power transformation equipment, therefore, through the first fault estimation area determined by the intersection points of the plurality of first noise estimation directions corresponding to the plurality of monitoring modules, a worker can check the target equipment according to the range covered by the first fault estimation area, and the overhaul efficiency is greatly improved; meanwhile, the control unit can adjust the power of the related target equipment, so that the related noise is possibly changed, and the first fault estimation area can be dynamically adjusted in real time according to the first intersection point, so that the accuracy of judging the noise source is further improved.
As a further scheme of the invention: and connecting all the first cross points, taking the maximum range surrounded by the first cross points after connection as an estimation area, acquiring the gravity center position of the estimation area, and generating the first fault prediction area by taking the gravity center position as the center of a circle and a first preset distance as a radius.
Through the technical scheme, the maximum region defined by the connecting line between the first cross points does not necessarily reflect the position of the noise source really, so that a first fault estimation region with the area larger than that of the estimation region needs to be generated by taking the gravity center position of the estimation region as the center of a circle, and the accuracy of judging the noise source is ensured.
As a further scheme of the invention: the preset sequence comprises:
selecting working equipment close to the edge of the first fault pre-estimated area as target equipment, and changing the selection of the target equipment along the clockwise direction;
the preset mode comprises the following steps: and reducing the first preset distance by a preset value.
Through the technical scheme, the power of the target equipment at the edge of the first-kind area is adjusted, if the first cross point does not change within the preset time, the next target equipment can be selected clockwise until the first cross point changes, at the moment, the first preset distance is reduced by the preset value, the first fault pre-estimation area is reduced, and the overhauling efficiency is further improved.
As a further scheme of the invention: and taking all the first cross points as circle centers and taking a preset radius as a circle to obtain an area set as the first fault prediction area.
Through the technical scheme, the first fault estimation area with the irregular shape can be obtained, the target equipment can be checked by the working personnel, and the overhauling efficiency is greatly improved.
As a further scheme of the invention: the fault prediction unit is further configured to determine a second noise prediction direction according to the sound wave amplitude attenuation direction of the noise data of the voltage transformation device in the same monitoring module, and then determine a second fault prediction region according to second intersections of the second noise prediction directions of all the monitoring modules;
and determining the overlapped part of the first fault prediction area and the second fault prediction area as a high fault probability area.
Through the technical scheme, because energy attenuation exists in the noise in the transmission process, the noise amplitude attenuation change direction embodied by the monitoring units at different positions in the unified monitoring module can be used as the second noise pre-judging direction, so that the second fault estimation area can be determined through the intersection point of the second noise pre-judging direction; moreover, the second noise pre-judging direction and the first noise pre-judging direction do not necessarily point to the accurate position of the noise source really, so that the overlapping and intersecting part of the first fault pre-judging area and the second fault pre-judging area is determined to be a high fault probability area, and the accuracy of searching the noise source can be improved to a certain extent.
As a further scheme of the invention: the equipment information code comprises a serial number of the monitoring unit and a corresponding installation coordinate;
acquiring a first device information code corresponding to the earliest sampling time stamp and a second device information code corresponding to the latest sampling time stamp for all the monitoring units in one monitoring module within a preset time period;
determining the first noise pre-judging direction according to a connecting line of the installation coordinates of the first equipment information code and the second equipment information code;
acquiring a third equipment information code corresponding to the noise data of the transformer equipment with the maximum sound wave amplitude and a fourth equipment information code corresponding to the noise data of the transformer equipment with the minimum sound wave amplitude;
and determining the second noise pre-judging direction according to a connecting line of the installation coordinates of the third equipment information code and the fourth equipment information code.
Through above-mentioned technical scheme, to a monitoring module, with the line between the position of two monitoring units that obtain same transformer equipment noise data at the latest and at the latest as first noise direction of prejudging, with the line between the position of two monitoring units that the acoustic amplitude is the biggest and minimum as second noise direction of prejudging, can be fast and accurate through the information that equipment information code contains to first noise direction of prejudging and second noise direction of prejudging confirm.
As a further scheme of the invention: the device also comprises a counting module, aiming at the same monitoring module, counting the difference rate S of the first noise pre-judging direction and the second noise pre-judging direction of the monitoring unit in the monitoring module;
wherein P is the number of the monitoring modules with different first noise pre-judging directions and second noise pre-judging directions, and Q is the total number of the monitoring modules;
aiming at the monitoring unit with the first noise pre-judging direction different from the second noise pre-judging direction:
if S is larger than a preset first threshold value, adjusting the angle of the first noise pre-judging direction to the second noise pre-judging direction;
if S is larger than a preset second threshold value, adjusting the angle of the first noise pre-judging direction to the second noise pre-judging direction;
the first threshold is less than the second threshold,
。
through the technical scheme, aiming at the transformer equipment area, if the first noise pre-judging direction is different from the second noise pre-judging direction, the number of the monitoring modules is large, the number of the monitoring units in the monitoring modules is small, the accuracy is insufficient, the sound wave amplitude can be used as a main part, the timestamp wiping sequence is used as an auxiliary part, the first noise pre-judging direction is adjusted to the direction close to the second noise pre-judging direction, and the accuracy of the high fault probability area judgment is improved.
As a further scheme of the invention: if S is larger than a preset third threshold value, the first noise pre-judging direction and the second noise pre-judging direction are simultaneously mutually close to each other to adjust the angle;
wherein the second threshold is less than the third threshold,
。
by the technical scheme, the coverage area of the high failure probability area can be further corrected.
The invention has the beneficial effects that:
(1) According to the invention, noise collection can be simultaneously carried out in the power transformation equipment area through a plurality of monitoring modules distributed at the edge of the power transformation equipment area, sound wave data related to equipment information codes are continuously obtained by monitoring units at different positions, if the noise data of the power transformation equipment is judged to exist, because the same noise reaches the monitoring units at different positions for different times, the source of the noise can be judged according to the sequence of sampling timestamps corresponding to the noise data of the power transformation equipment, therefore, through a first fault estimation area determined by the intersection points of a plurality of first noise estimation directions corresponding to the plurality of monitoring modules, a worker can check target equipment according to the coverage range of the first fault estimation area, and the overhaul efficiency is greatly improved; meanwhile, the control unit can adjust the power of the related target equipment, so that the related noise is possibly changed, and the first fault estimation area can be dynamically adjusted in real time according to the first intersection point, so that the accuracy of noise source judgment is further improved;
(2) In the invention, the maximum region surrounded by the connecting lines between the first cross points does not necessarily reflect the position of the noise source really, so that a first fault estimation region with the area larger than that of the estimation region is generated by taking the gravity center position of the estimation region as the center of a circle so as to ensure the accuracy of judging the noise source;
(3) For a monitoring module, a connecting line between the positions of two monitoring units which obtain the same noise data of the transformer at the latest is used as a first noise pre-judging direction, a connecting line between the positions of two monitoring units with the largest and the smallest sound wave amplitudes is used as a second noise pre-judging direction, and the first noise pre-judging direction and the second noise pre-judging direction can be determined quickly and accurately through information contained in an equipment information code.
Drawings
The invention is further described below with reference to the accompanying drawings.
Fig. 1 is a schematic diagram of module connection of a substation equipment operation state monitoring control system according to the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Referring to fig. 1, the present invention is a substation equipment operation status monitoring and controlling system based on data analysis, including:
the monitoring modules are distributed at the edge of the power transformation equipment area;
aiming at one monitoring module, the monitoring module comprises a plurality of monitoring units, and the distance from each monitoring unit to the center of the transformer equipment area is different;
each monitoring unit is internally provided with an equipment information code and is used for acquiring sound wave data associated with the equipment information code according to a preset frequency;
the acoustic signal processing unit is used for carrying out feature extraction processing on the acoustic wave data to acquire voiceprint image feature information corresponding to the acoustic wave data;
the voiceprint recognition unit is used for judging whether the voiceprint image characteristic information contains the noise data of the voltage transformation equipment;
the fault pre-estimation unit is used for acquiring corresponding first noise pre-judgment directions according to the sequence of sampling timestamps of the noise data of the transformer equipment in the same monitoring module, and then determining a first fault pre-estimation area according to first cross points of the first noise pre-judgment directions of all the monitoring modules;
the control unit is used for adjusting the power change of the target equipment according to a preset sequence; the target equipment is working equipment in a first fault prediction area;
and the verification adjusting unit is used for adjusting the coverage range of the first fault estimation area according to a preset mode when the first intersection is detected to change.
According to the technical scheme, noise collection can be simultaneously carried out in the power transformation equipment area through the plurality of monitoring modules distributed at the edge of the power transformation equipment area, the monitoring units at different positions can continuously obtain sound wave data related to equipment information codes, if the noise data of the power transformation equipment exist, due to the fact that the same noise arrives at the monitoring units at different positions in different time, the source of the noise can be judged according to the sequence of sampling time stamps corresponding to the noise data of the power transformation equipment, therefore, through the first fault estimation area determined by the intersection points of the plurality of first noise estimation directions corresponding to the plurality of monitoring modules, a worker can check the target equipment according to the range covered by the first fault estimation area, and the overhaul efficiency is greatly improved; meanwhile, the control unit can adjust the power of the related target equipment, so that the related noise is possibly changed, and the first fault estimation area can be dynamically adjusted in real time according to the first intersection point, so that the accuracy of judging the noise source is further improved.
The acoustic signal processing unit can generate a voiceprint characteristic image according to the acoustic signal, extract voiceprint image characteristic information in the voiceprint characteristic image, compare the voiceprint image characteristic information with a preset transformer state characteristic recognition library or input the voiceprint image characteristic information into a pre-trained artificial intelligent machine learning model, namely the voiceprint recognition unit, and recognize the state of the target voltage transformation equipment related to noise. The transformer state feature recognition library comprises voiceprint image feature information and mapping relations between different states related to the transformer and noise, and the artificial intelligence machine learning model is trained in advance to establish the mapping relations between the voiceprint image feature information and the different states related to the transformer and the noise.
The generated voiceprint characteristic image is a two-dimensional color image, the horizontal axis and the vertical axis respectively represent time and frequency, and the magnitude of the signal amplitude is represented by the shade of the color image. Extracting the voiceprint image characteristic information in the voiceprint characteristic image specifically refers to extracting a noise signal frequency spectrum range in the voiceprint characteristic image and concentrated integral multiple frequency components of target frequency in the noise signal frequency spectrum range; and the step of extracting the voiceprint image feature information in the voiceprint feature image specifically means that a convolutional neural network is adopted to extract features in the voiceprint feature image to obtain the voiceprint image feature information.
In this way, the voiceprint recognition unit can give a specific device location of a noise source, but because the noise source in the substation device area is various and complex, the voiceprint recognition unit has a certain error rate, and in order to prevent a determination error, if the device location is inconsistent with the first failure prediction area, the voiceprint recognition unit may not give confidence, and if the device location is consistent with the first failure prediction area, the device in the first failure prediction area is used as a troubleshooting object.
As a further scheme of the invention: and connecting all the first cross points, taking the maximum range enclosed by the connected first cross points as an estimation area, acquiring the gravity center position of the estimation area, and generating a first fault prediction area by taking the gravity center position as the center of a circle and a first preset distance as the radius.
Through the technical scheme, the maximum region surrounded by the connecting lines between the first cross points does not necessarily reflect the position of the noise source really, so that a first fault estimation region larger than the area of the estimation region needs to be generated by taking the gravity center position of the estimation region as the center of a circle, and the accuracy of judging the noise source is ensured.
As a further scheme of the invention: the preset sequence comprises:
selecting working equipment close to the edge of the first fault prediction area as target equipment, and changing the selection of the target equipment along the clockwise direction;
the preset mode comprises the following steps: and reducing the first preset distance by a preset value.
Through the technical scheme, the power of the target equipment at the edge of the first-kind area is adjusted, if the first cross point does not change within the preset time, the next target equipment can be selected clockwise until the first cross point changes, at the moment, the first preset distance is reduced by the preset value, the first fault pre-estimation area is reduced, and the overhauling efficiency is further improved.
As a further scheme of the invention: and taking all the first cross points as the circle centers and taking a preset radius as a circle to obtain an area set as a first fault prediction area.
Through the technical scheme, the first fault estimation area with the irregular shape can be obtained, the target equipment can be checked by the working personnel, and the overhauling efficiency is greatly improved.
As a further scheme of the invention: the fault prediction unit is also used for determining a second noise prediction direction according to the sound wave amplitude attenuation direction of the noise data of the transformer equipment in the same monitoring module, and then determining a second fault prediction area according to second cross points of the second noise prediction directions of all the monitoring modules;
and determining the overlapped part of the first fault prediction area and the second fault prediction area as a high fault probability area.
Through the technical scheme, because energy attenuation exists in the noise in the transmission process, the noise amplitude attenuation change direction embodied by the monitoring units at different positions in the unified monitoring module can be used as the second noise pre-judging direction, so that the second fault estimation area can be determined through the intersection point of the second noise pre-judging direction; moreover, the second noise pre-judging direction and the first noise pre-judging direction do not necessarily point to the accurate position of the noise source really, so that the overlapping and intersecting part of the first fault pre-judging area and the second fault pre-judging area is determined to be a high fault probability area, and the accuracy of searching the noise source can be improved to a certain extent.
As a further scheme of the invention: the equipment information code comprises a serial number of the monitoring unit and a corresponding installation coordinate;
aiming at all monitoring units in one monitoring module, acquiring a first equipment information code corresponding to the earliest sampling time stamp and a second equipment information code corresponding to the latest sampling time stamp in a preset time period;
determining a first noise pre-judging direction according to a connecting line of the installation coordinates of the first equipment information code and the second equipment information code;
acquiring a third equipment information code corresponding to the transformation equipment noise data with the largest sound wave amplitude and a fourth equipment information code corresponding to the transformation equipment noise data with the smallest sound wave amplitude;
the second noise pre-judging direction can be determined according to a connecting line of the installation coordinates of the third equipment information code and the fourth equipment information code.
Through the technical scheme, aiming at one monitoring module, the connecting line between the positions of the two monitoring units which obtain the same noise data of the transformer equipment at the latest is used as the first noise pre-judging direction, the connecting line between the positions of the two monitoring units with the largest and the smallest sound wave amplitude is used as the second noise pre-judging direction, and the first noise pre-judging direction and the second noise pre-judging direction can be determined quickly and accurately through the information contained in the equipment information code.
As a further scheme of the invention: the device also comprises a statistical module, a first noise pre-judging module and a second noise pre-judging module, wherein the statistical module is used for counting the difference rate S of the first noise pre-judging direction and the second noise pre-judging direction of the monitoring units in the monitoring module aiming at the same monitoring module;
wherein, P is the number of the monitoring modules with different first noise pre-judging directions and second noise pre-judging directions, and Q is the total number of the monitoring modules;
aiming at the monitoring units with different first noise pre-judging directions and second noise pre-judging directions:
if S is larger than a preset first threshold value, adjusting the angle of the first noise pre-judging direction to the corresponding second noise pre-judging direction;
if S is larger than a preset second threshold value, adjusting the angle of the first noise pre-judging direction to the corresponding second noise pre-judging direction;
the first threshold value is less than the second threshold value,
。
through above-mentioned technical scheme, to the substation equipment region, if the first noise direction of prejudging and the different quantity of monitoring module of second noise direction of prejudging are more, explain that the monitoring unit quantity in the monitoring module is less partially, lead to the precision not enough, can the acoustic wave range be main this moment, wipe the order with the timestamp and assist, prejudge the direction with first noise and adjust to the direction that is close to second noise direction of prejudging to this promotes the regional precision of judging of high fault probability.
As a further scheme of the invention: if S is larger than a preset third threshold value, the first noise pre-judging direction and the second noise pre-judging direction are simultaneously mutually close to each other to adjust the angle;
wherein the second threshold is less than the third threshold,
。
by the technical scheme, the coverage area of the high failure probability area can be further corrected.
Although one embodiment of the present invention has been described in detail, the description is only for the purpose of illustrating the preferred embodiments of the present invention and should not be taken as limiting the scope of the invention. All equivalent changes and modifications made within the scope of the present invention shall fall within the scope of the present invention.