Transformer substation on-load fault detection method and device and storage medium

Abstract

The invention discloses a transformer substation on-load fault detection method and device and a storage medium, and the method comprises the steps: carrying out the on-load test of a transformer substation, obtaining the load information related to a current, loading a preset on-load fault detection model and a support vector machine, and enabling the on-load fault detection model to comprise an input layer, a hidden layer and an output layer, the method comprises the following steps: receiving load information in an input layer, carrying out linear transformation on the load information in a hidden layer to obtain a load vector, mapping the load vector into a fault type when a transformer substation is loaded in an output layer, and inputting a plurality of sample pairs into a support vector machine for decision making to obtain the fault type when the transformer substation is loaded. The sample pair includes load information and a fault category. The on-load test is performed on the transformer substation, and the fault type is accurately diagnosed in combination with the preset on-load fault test model and the support vector machine, so that the stability and the safety of the power system are ensured, and the accuracy and the efficiency of fault detection when the transformer substation is on load are improved.

Description

Technical Field

[0001] The embodiments of the present invention relate to the technical field of substation operation and maintenance, and in particular to a substation load fault detection method, device and storage medium. Background Art

[0002] As the scale of power systems continues to expand, substations, as one of the components of power systems, are usually related to the reliability and stability of power supply. Therefore, fault diagnosis and detection of substation equipment is one of the factors for the stable operation of power systems.

[0003] Traditional substation fault detection methods mainly rely on manual inspections, current and voltage monitoring, and judgments based on empirical rules. Manual inspections are easily influenced by experience, and simple parameter monitoring of minor faults is difficult. These methods have low accuracy and efficiency in detecting faults under load in substations. Summary of the invention

[0004] The present invention provides a method, a device and a storage medium for detecting faults in a substation under load, so as to improve the accuracy and efficiency of detecting faults in a substation under load.

[0005] In a first aspect, an embodiment of the present invention provides a method for detecting faults in a substation under load, comprising:

[0006] Carry out load test on the substation to obtain load information related to the current;

[0007] Loading a preset load-based fault detection model and a support vector machine; the load-based fault detection model includes an input layer, a hidden layer, and an output layer;

[0008] The load information is received in the input layer, the load information is linearly transformed in the hidden layer to obtain a load vector, and the load vector is mapped to a fault category of the substation when it is loaded in the output layer;

[0009] A plurality of sample pairs are input into the support vector machine for decision making to obtain the fault type of the substation under load; the sample pairs include the load information and the fault category.

[0010] In a second aspect, an embodiment of the present invention further provides a substation load fault detection device, comprising:

[0011] The load information acquisition module is used to perform load testing on the substation to obtain load information related to the current;

[0012] A model loading module, used to load a preset load-based fault detection model and a support vector machine; the load-based fault detection model includes an input layer, a hidden layer, and an output layer;

[0013] A model processing module, configured to receive the load information in the input layer, perform linear transformation on the load information in the hidden layer to obtain a load vector, and map the load vector to a fault category of the substation when it is loaded in the output layer;

[0014] The fault type acquisition module is used to input multiple sample pairs into the support vector machine for decision-making to obtain the fault type of the substation when it is under load; the sample pair includes the load information and the fault category.

[0015] In a third aspect, an embodiment of the present invention further provides a computer device, the computer device comprising:

[0016] one or more processors;

[0017] A storage device for storing one or more programs;

[0018] When the one or more programs are executed by the one or more processors, the one or more processors implement the substation load fault detection method provided by the first aspect of the present invention.

[0019] In a fourth aspect, an embodiment of the present invention further provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the substation load fault detection method provided in the first aspect of the present invention.

[0020] In a fifth aspect, an embodiment of the present invention further provides a computer program product, the computer program product comprising a computer program, and when the computer program is executed by a processor, the substation load fault detection method provided in the first aspect of the present invention is implemented.

[0021] In an embodiment of the present invention, a load test is performed on the substation to obtain load information related to the current, and a preset load-based fault detection model and a support vector machine are loaded. The load-based fault detection model includes an input layer, a hidden layer, and an output layer. The load information is received in the input layer, and the load information is linearly transformed in the hidden layer to obtain a load vector. The load vector is mapped to a fault category when the substation is loaded in the output layer. Multiple sample pairs are input into the support vector machine for decision-making to obtain the fault type when the substation is loaded. The sample pairs include load information and fault categories. By combining the load-based fault detection model and the support vector machine, the accuracy of substation fault diagnosis can be effectively improved. The load-based fault detection model receives current load information through the input layer, performs linear transformation in the hidden layer, extracts load feature vectors, and thus identifies abnormal points in the load information. The load-based fault detection model mainly focuses on local anomaly detection in single load information. The support vector machine uses multiple sample pairs to make decisions, and identifies global anomalies in the sequence based on the load information and fault categories of these samples. The combination of the two, using the load fault detection model for local anomaly detection and the support vector machine for global anomaly decision-making, has clear levels and complements each other, effectively improving the accuracy and efficiency of load fault detection in substations. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figure 1 A flow chart of a method for detecting faults in a substation under load provided in the first embodiment of the present invention;

[0023] Figure 2 A schematic diagram of a load measurement vector circuit for a substation provided in the first embodiment of the present invention;

[0024] Figure 3 A structural block diagram of a substation load fault detection device provided in the second embodiment of the present invention;

[0025] Figure 4 A schematic diagram of the structure of a computer device provided in Embodiment 3 of the present invention. DETAILED DESCRIPTION

[0026] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.

[0027] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present invention described herein can cover sequential implementations other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0028] Embodiment 1

[0029] See also Figure 1 , shows a flow chart of a substation load fault detection method provided by the first embodiment of the present invention, the method can be performed by a substation load fault detection device, the substation load fault detection device can be implemented in the form of hardware and / or software, and the substation load fault detection device can be configured in a computer device. Figure 1 As shown, the method includes:

[0030] Step 101: Perform a load test on the substation to obtain load information related to the current.

[0031] Substations are one of the facilities in the power system. Their main function is to convert high-voltage transmission power into low-voltage power suitable for distribution and use. Substations increase or decrease voltage through transformers, and distribute, regulate and protect power to ensure the stable operation of the power system.

[0032] Load testing refers to loading a certain load on the power equipment of the substation to simulate the working status under actual load operating conditions. It tests the performance of the power equipment of the substation under different load conditions. It can help understand the electrical characteristics of the power equipment of the substation when it is running under load and provide a basis for fault detection and diagnosis.

[0033] In this embodiment, under actual operating conditions, the working state of the substation is simulated, and load information related to the current is obtained. This load information is the basis for analyzing and detecting substation faults. By analyzing the load information, potential fault types can be identified, thereby realizing fault prediction and prevention of the substation.

[0034] Exemplarily, the load information includes the magnitude of the current, the phase angle of the current, the active power and the reactive power.

[0035] In one embodiment of the present invention, step 101 may include the following steps:

[0036] Step 1011: Perform a load test on the substation to obtain the current of each phase.

[0037] In this embodiment, the current information of each phase in the substation is obtained through actual load testing. The change of each phase current is usually more important for analyzing whether there is a fault in the substation. The current characteristics of each phase under different loads can reflect potential abnormalities or faults in the operation of the substation.

[0038] Step 1012: When a confirmation operation triggered by the current information is received, the voltage of phase A is selected from the current information as a reference voltage.

[0039] In this embodiment, the voltage of phase A is selected as the reference voltage to ensure the accuracy and consistency of phase measurement. By confirming the voltage of phase A, the substation can ensure that the phase relationship between current and voltage is established on a consistent reference point, which helps the accuracy of subsequent phase angle calculation.

[0040] Exemplarily, through the reference source acquisition circuit device, which is composed of transformation, filtering, AD acquisition and pulse forming circuit and pulse acquisition, the reference source acquisition circuit device collects the circuit current, the current phase angle, the active power and the reactive power, and then confirms it as the reference source by the operation button. After the reference source is confirmed, the central processing unit (CPU) will control the continuous output of the synchronization signal synchronized with the reference source to control the output inverter power supply to synchronize with it. After the load information is collected, the collection input channel is switched to the signal channel to be collected through CPU control to facilitate the next load information collection.

[0041] Step 1013: Set the angle at which the reference voltage leads the current as the phase angle of the current.

[0042] In this embodiment, the phase angle of the current is calculated to ensure that the phase difference between the current and the voltage is correctly measured and represented. The phase angle is one of the important parameters for analyzing substation faults. Through this calculation, the relative relationship between the current and the voltage can be determined, thereby revealing whether there is an abnormality or fault.

[0043] For example, Figure 2 The figure shows the load measurement circuit diagram of the substation. A Indicates that the voltage of phase A is used as the reference voltage, I A Indicates the current of phase A, I B Indicates the current of phase B, I Crepresents the current of phase C, P represents active power, Q represents reactive power, 120° represents the phase angle of the current of phase B with the voltage of phase A as the reference voltage, and 240° represents the phase angle of the current of phase C with the voltage of phase A as the reference voltage; with phase A as the reference voltage, the angles of each phase are expressed by the following formula:

[0044]

[0045] in, is the phase angle, is the A phase angle, Indicates the B phase angle, represents the C phase angle, 120° represents the phase angle of the B phase current, 240° represents the phase angle of the C phase current, Indicates a certain phase.

[0046] For example, when the polarity and phase sequence of the secondary current loop are connected incorrectly, the error characteristics are reflected in the current phase. According to the phase sequence and polarity characteristics, the incorrect wiring of any phase of the secondary current loop can be divided into the following five types: 1. The leading phase is connected to the current phase after the polarity is reversed; 2. The lagging phase is connected to the current phase; 3. The current phase is connected to the reverse polarity; 4. The leading phase is connected to the current phase; 5. The lagging phase is connected to the current phase after the polarity is reversed.

[0047] 1. The leading phase is connected with the current phase after the polarity is reversed. This means that in the phase sequence connection, the leading angle of a phase current is mistakenly connected in reverse, affecting the accuracy of the phase sequence.

[0048] 2. The lagging phase is connected to the real phase, which means that the lagging phase is directly connected to the real phase, which may cause phase error.

[0049] 3. Reverse polarity of this phase means that the electrical polarity of this phase is reversed, causing the current direction to be opposite to the expected direction.

[0050] 4. The leading phase is connected to the real phase, which means that the leading phase is connected to the real phase, which may lead to inconsistent phase relationship.

[0051] 5. The lagging phase is connected to the current phase after the polarity is reversed, which means that the polarity of the lagging phase is reversed and then reconnected to the current phase, which will result in the wrong current direction and phase.

[0052] Step 102: Load the preset load-test fault model and support vector machine.

[0053] In this embodiment, the preset load-based fault detection model and support vector machine (SVM) are loaded to detect the load information of the substation and classify the fault type when the substation is tested under load. The load-based fault detection model includes an input layer, a hidden layer and an output layer.

[0054] Exemplarily, the load-detection fault model may be a neural network model such as a generalized regression neural network model, and the present invention is not limited thereto.

[0055] The generalized regression neural network (GRNN) model is a neural network model based on nonlinear regression, which is mainly used for function approximation and regression problems. GRNN can effectively fit and predict high-dimensional data by using radial basis function (RBF) as the activation function. The main advantages of GRNN are simple structure, high computational efficiency, and strong generalization ability. In the fault detection problem, GRNN can perform regression analysis through the input load information and output a prediction of the fault type.

[0056] Support vector machine (SVM) is a supervised learning model based on statistical learning theory, which is widely used in classification and regression tasks. SVM classifies data points by finding an optimal hyperplane to separate different categories. For regression tasks, SVM uses support vector regression (SVR) to achieve fitting. The core advantage of SVM is that it can handle high-dimensional data.

[0057] Step 103: receiving load information in the input layer, performing linear transformation on the load information in the hidden layer to obtain a load vector, and mapping the load vector to a fault category when the substation is loaded in the output layer.

[0058] In this embodiment, the input load information is converted into the output of the fault category through the hierarchical structure of the load-testing fault model. The load information related to the load test of the substation is received at the input layer, and the load information is the characteristic data of the substation operation status; the load information is linearly transformed in the hidden layer, and a new representation (i.e., load vector) is obtained through a specific calculation method (such as radial basis function transformation), and the load vector can more effectively express the working state of the substation under load; in the output layer, based on the calculation result of the load vector, it is mapped to the fault category of the substation under load, so as to realize the fault detection and realize the automatic and accurate fault diagnosis.

[0059] In one embodiment of the present invention, step 103 may include the following steps:

[0060] Step 311: Receive load information recorded in a matrix in the input layer.

[0061] In this embodiment, the matrix record can efficiently represent and process multi-dimensional load information. The load information of the substation usually involves multiple variables, such as current magnitude, current phase angle, power, etc., and there may be complex relationships between these load information. Through the matrix structure, these multi-dimensional load information can be organized into a mathematical object to ensure the efficiency of the whole process.

[0062] The matrix is ​​represented as:

[0063]

[0064] Among them, P is the matrix of load information, P is the matrix of load information, P 11 To P RQ All represent load information, R is the dimension of load information, Q is the number of load information, P RQ It is the Rth load information in the Qth load information.

[0065] In one embodiment of the present invention, step 103 may include the following steps:

[0066] Step 321: Transpose the matrix to construct radial basis centers.

[0067] In this embodiment, the matrix transposition process is to adjust the representation of the load information to a form suitable for subsequent calculations. By transposing, the radial basis center can be effectively constructed, so that the load-detecting fault model can locally optimize the load information in the feature space.

[0068] Step 322: Query the offset of the load information.

[0069] In this embodiment, the offset query is to translate or adjust the load information so that the data can adapt to the training requirements of the load-measurement fault model during the processing. The offset usually represents the deviation of the load information in a certain dimension. By adjusting it, the deviation in the load information can be eliminated. Each load information corresponds to an offset, and the offset can be set according to the actual situation to ensure that the load-measurement fault model can learn the relationship between the load information and the fault more accurately.

[0070] Step 323: Perform linear transformation on the load information according to the matrix, radial basis center and offset to obtain a load vector.

[0071] In this embodiment, the load information is transformed into a feature space that is more suitable for the load-measurement fault model and support vector machine by combining the matrix, radial basis center and offset to perform linear transformation. This transformation maps the load information to a new space, so that the load-measurement fault model can more clearly identify the type of fault that may occur when the substation is loaded. Linear transformation can extract the key features of the data, reduce redundant information, and thus improve the efficiency and accuracy of the classifier.

[0072] Exemplarily, the load vector is expressed as:

[0073] a i =exp(-[dist(C,P i)] 2 b i );

[0074] Among them, a i represents the load vector of the i-th load information, i represents the number of load information, C represents the radial basis center, P i represents the i-th load information in the matrix, b i Indicates the offset of the i-th load information, dist(C,P i ) represents the radial basis center to P i exp represents the exponential function.

[0075] In one embodiment of the present invention, step 103 may include the following steps:

[0076] Step 331: In the output layer, query the connection weight between the hidden layer and the input layer.

[0077] In this embodiment, in the structure of the load-measurement fault model, the input layer is responsible for receiving the load information, while the hidden layer performs a certain linear transformation to convert the load information into an intermediate representation (i.e., a load vector). The task of the output layer is to determine the final fault category based on the output of the hidden layer. The output layer needs to know the connection weight between the hidden layer and the input layer. The connection weight reflects the degree of influence of the load information of the input layer on the neurons (i.e., the offset) in the hidden layer.

[0078] Step 332: In the output layer, weighted sum is performed based on the connection weight and the load vector to obtain the fault category of the substation when it is under load.

[0079] In this embodiment, in the output layer, weighted summation is performed based on the connection weights and the load vector in order to integrate the hidden layer information and finally output the fault category of the substation under load. The load vector represents the load information after the hidden layer linear transformation, and each load information corresponds to the activation value of a hidden layer neuron. Through weighted summation (i.e., weighting the load vector using the connection weights), the value of each fault category can be obtained. The result of this weighted summation, after further processing (such as mapping through an identity function), finally obtains the fault category of the substation under load, ensuring that the load-measurement fault model can output an accurate fault type based on the input load information.

[0080] For example, in the output layer, the connection weight and the load vector are substituted into the following formula for weighted summation to obtain the fault category when the substation is loaded:

[0081]

[0082] y i =purelin(ni )=n i ;

[0083] Among them, n i The network output vector representing the i-th load information, W 2,1 represents the connection weight, a i represents the load vector of the i-th load information, n represents the number of load information, y i represents the fault category of the i-th load information, and purelin represents the identity function.

[0084] Step 104: Input multiple sample pairs into a support vector machine for decision making to obtain the fault type of the substation under load.

[0085] In this embodiment, since the current of the substation is unstable when the substation is under load, resulting in inaccurate test results, multiple measurements can be performed to obtain multiple sample pairs, thereby obtaining output results that are more in line with the actual situation. In order to further improve the accuracy of fault measurement when the substation is under load, a support vector machine (SVM) is used to make decisions on multiple sample pairs to determine the fault type when the substation is under load. Each sample pair consists of load information and a corresponding fault category. The SVM can classify according to the input load information and fault category, and accurately determine the fault type when the substation is under load in actual applications, thereby improving the accuracy of fault identification.

[0086] In one embodiment of the present invention, step 104 may include the following steps:

[0087] Step 1041: Construct a decision function in the support vector machine.

[0088] In this embodiment, a support vector machine (SVM) constructs a decision function through training samples. The decision function is based on support vectors and related hyperplanes to distinguish different types of faults. By constructing the decision function, the SVM can effectively make classification decisions based on the input load information and map the load information to the corresponding fault category.

[0089] Exemplarily, the decision function is expressed as:

[0090]

[0091] K(x i ,y i ) = exp(-||x i -y i || 2 );

[0092] Among them, α i , α j Both represent Lagrange multipliers, yi ,y j All represent fault categories, C represents the penalty parameter, δ ij represents the slack variable, n represents the number of load information, x i ,x j Both represent load information, st represents constraint conditions, and K represents kernel function.

[0093] Step 1042: Substitute multiple sample pairs into the decision function to make a decision, and obtain the fault type when the substation is under load.

[0094] In this embodiment, multiple sample pairs (including load information and fault categories) are substituted into the constructed decision function for prediction and classification. Through the calculation of the decision function, SVM obtains the corresponding fault type according to each load information, thereby realizing accurate judgment of the fault type when the substation is under load.

[0095] In an embodiment of the present invention, a load test is performed on the substation to obtain load information related to the current, and a preset load-based fault detection model and a support vector machine are loaded. The load-based fault detection model includes an input layer, a hidden layer, and an output layer. The load information is received in the input layer, and the load information is linearly transformed in the hidden layer to obtain a load vector. The load vector is mapped to a fault category when the substation is loaded in the output layer. Multiple sample pairs are input into the support vector machine for decision-making to obtain the fault type when the substation is loaded. The sample pairs include load information and fault categories. By combining the load-based fault detection model and the support vector machine, the accuracy of substation fault diagnosis can be effectively improved. The load-based fault detection model receives current load information through the input layer, performs linear transformation in the hidden layer, extracts load feature vectors, and thus identifies abnormal points in the load information. The load-based fault detection model mainly focuses on local anomaly detection in single load information. The support vector machine uses multiple sample pairs to make decisions, and identifies global anomalies in the sequence based on the load information and fault categories of these samples. The combination of the two, using the load fault detection model for local anomaly detection and the support vector machine for global anomaly decision-making, has clear levels and complements each other, effectively improving the accuracy and efficiency of load fault detection in substations.

[0096] Embodiment 2

[0097] Figure 3 A schematic diagram of a substation load fault detection device provided in the second embodiment of the present invention is shown in FIG. Figure 3 As shown, the device comprises:

[0098] The load information acquisition module 301 is used to perform a load test on the substation to obtain load information related to the current;

[0099] The model loading module 302 is used to load the preset load-based fault detection model and support vector machine; the load-based fault detection model includes an input layer, a hidden layer, and an output layer;

[0100] The model processing module 303 is used to receive the load information in the input layer, perform linear transformation on the load information in the hidden layer to obtain a load vector, and map the load vector to a fault category of the substation when it is loaded in the output layer;

[0101] The fault type acquisition module 304 is used to input multiple sample pairs into the support vector machine for decision making to obtain the fault type of the substation when it is under load; the sample pair includes the load information and the fault category.

[0102] In one embodiment of the present invention, the load information includes the magnitude of the current, the phase angle of the current, the active power and the reactive power; the load information acquisition module 301 includes:

[0103] Load test module, used to perform load test on the substation to obtain the current of each phase;

[0104] A reference voltage acquisition module, configured to select a voltage of phase A from the current information as a reference voltage when receiving a confirmation operation triggered by the current information;

[0105] The phase angle setting module is used to set the angle at which the reference voltage leads the current as the phase angle of the current.

[0106] In one embodiment of the present invention, the model processing module 303 includes:

[0107] A load information receiving module, used for receiving the load information recorded in a matrix in the input layer;

[0108] Wherein, the matrix is ​​expressed as:

[0109]

[0110] Wherein, P is the matrix of the load information, P 11 To P RQ represents the load information, R is the dimension of the load information, Q is the number of load information, P RQ It is the Rth load information in the Qth load information.

[0111] In one embodiment of the present invention, the model processing module 303 includes:

[0112] A matrix transposition module, used for performing a transposition process on the matrix to construct a radial basis center;

[0113] An offset query module, used to query the offset of the load information;

[0114] The load vector acquisition module is used to perform linear transformation on the load information according to the matrix, the radial basis center and the offset to obtain a load vector.

[0115] In one embodiment of the present invention, the load vector is expressed as:

[0116] a i =exp(-[dist(C,P i )] 2 b i );

[0117] Among them, a i represents the load vector of the i-th load information, i represents the number of load information, C represents the radial basis center, P i represents the load information of the ith element in the matrix, b i represents the offset of the i-th load information, dist(C,P i ) represents the radial basis center to the P i exp represents the exponential function.

[0118] In one embodiment of the present invention, the model processing module 303 includes:

[0119] A connection weight query module, used for querying the connection weight between the hidden layer and the input layer in the output layer;

[0120] The weighted summation module is used to perform weighted summation in the output layer according to the connection weight and the load vector to obtain the fault category of the substation when it is under load.

[0121] In one embodiment of the present invention, the weighted summation module includes:

[0122] The fault category calculation module is used to substitute the connection weight and the load vector into the following formula in the output layer for weighted summation to obtain the fault category of the substation when it is under load:

[0123]

[0124] y i =purelin(n i )=n i ;

[0125] Among them, n i The network output vector representing the load information of the ith 2,1represents the connection weight, a i represents the load vector of the i-th load information, n represents the number of load information, y i represents the fault category of the i-th load information, and purelin represents the identity function.

[0126] In one embodiment of the present invention, the fault type acquisition module 304 includes:

[0127] A decision function construction module, used for constructing a decision function in the support vector machine;

[0128] A sample pair decision module, used for substituting a plurality of sample pairs into the decision function to make a decision and obtain the fault type of the substation when it is under load;

[0129] Wherein, the decision function is expressed as:

[0130]

[0131] K(x i ,y i ) = exp(-||x i -y i || 2 );

[0132] Among them, α i , α j Both represent Lagrange multipliers, y i ,y j All represent the fault categories, C represents the penalty parameter, δ ij represents the slack variable, n represents the number of load information, x i ,x j Both represent the load information, st represents the constraint condition, and K represents the kernel function.

[0133] The substation load fault detection device provided in the embodiment of the present invention can execute the substation load fault detection method provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects for executing the substation load fault detection method.

[0134] Embodiment 3

[0135] See also Figure 4, showing a schematic diagram of the structure of a computer device provided by an embodiment of the present invention. The computer device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workbenches, personal digital assistants, blade servers, mainframe computers, and other suitable computers. The computer device may also represent various forms of mobile devices, such as personal digital processing, cellular phones, smart phones, wearable devices (such as helmets, glasses, watches, etc.), and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present invention described and / or required herein.

[0136] like Figure 4 As shown, the computer device 10 includes at least one processor 11, and a memory connected to the at least one processor 11, such as a read-only memory (ROM) 12, a random access memory (RAM) 13, etc., wherein the memory stores a computer program that can be executed by at least one processor, and the processor 11 can perform various appropriate actions and processes according to the computer program stored in the read-only memory (ROM) 12 or the computer program loaded from the storage unit 18 to the random access memory (RAM) 13. In the RAM 13, various programs and data required for the operation of the computer device 10 can also be stored. The processor 11, ROM 12 and RAM 13 are connected to each other through a bus 14. An input / output (I / O) interface 15 is also connected to the bus 14.

[0137] A number of components in the computer device 10 are connected to the I / O interface 15, including: an input unit 16, such as a keyboard, a mouse, etc.; an output unit 17, such as various types of displays, speakers, etc.; a storage unit 18, such as a disk, an optical disk, etc.; and a communication unit 19, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 19 allows the computer device 10 to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks.

[0138] The processor 11 may be a variety of general and / or special processing components with processing and computing capabilities. Some examples of the processor 11 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, digital signal processors (DSPs), and any appropriate processors, controllers, microcontrollers, etc. The processor 11 executes the various methods and processes described above, such as the substation load measurement fault detection method.

[0139] In some embodiments, the substation load measurement fault detection method can be implemented as a computer program, which is tangibly contained in a computer-readable storage medium, such as a storage unit 18. In some embodiments, part or all of the computer program can be loaded and / or installed on the computer device 10 via the ROM 12 and / or the communication unit 19. When the computer program is loaded into the RAM 13 and executed by the processor 11, one or more steps of the substation load measurement fault detection method described above can be performed. Alternatively, in other embodiments, the processor 11 can be configured to execute the substation load measurement fault detection method in any other appropriate manner (for example, by means of firmware).

[0140] Various implementations of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), load programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various implementations can include: being implemented in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0141] Computer programs for implementing the methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, so that when the computer program is executed by the processor, the functions / operations specified in the flow chart and / or block diagram are implemented. The computer program may be executed entirely on the machine, partially on the machine, partially on the machine and partially on a remote machine as a stand-alone software package, or entirely on a remote machine or server.

[0142] In the context of the present invention, a computer-readable storage medium may be a tangible medium that may contain or store a computer program for use by or in combination with an instruction execution system, device or equipment. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices or equipment, or any suitable combination of the foregoing. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0143] To provide interaction with a user, the systems and techniques described herein can be implemented on a computer device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and a pointing device (e.g., a mouse or trackball) through which the user can provide input to the computer device. Other types of devices can also be used to provide interaction with the user; for example, the feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including acoustic input, voice input, or tactile input).

[0144] The systems and techniques described herein may be implemented in a computing system that includes backend components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes frontend components (e.g., a user computer with a graphical user interface or a web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such backend components, middleware components, or frontend components. The components of the system may be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: a local area network (LAN), a wide area network (WAN), a blockchain network, and the Internet.

[0145] A computing system may include a client and a server. The client and the server are generally remote from each other and usually interact through a communication network. The client and server relationship is generated by computer programs running on the corresponding computers and having a client-server relationship with each other. The server may be a cloud server, also known as a cloud computing server or cloud host, which is a host product in the cloud computing service system to solve the defects of difficult management and weak business scalability in traditional physical hosts and VPS services.

[0146] Embodiment 4

[0147] An embodiment of the present invention further provides a computer program product, which includes a computer program. When the computer program is executed by a processor, the substation load fault detection method provided by any embodiment of the present invention is implemented.

[0148] In the process of implementation, the computer program product can be written in one or more programming languages ​​or a combination thereof to perform the computer program code of the present invention, including object-oriented programming languages, such as Java, Smalltalk, C++, and conventional procedural programming languages, such as "C" language or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a separate software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer can be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or can be connected to an external computer (for example, using an Internet service provider to connect through the Internet).

[0149] It should be understood that the various forms of processes shown above can be used to reorder, add or delete steps. For example, the steps described in the present invention can be executed in parallel, sequentially or in different orders, as long as the desired results of the technical solution of the present invention can be achieved, and this document does not limit this.

[0150] The above specific implementations do not constitute a limitation on the protection scope of the present invention. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions can be made according to design requirements and other factors. Any modification, equivalent substitution and improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for detecting faults in a substation under load, characterized in that: include: Carry out load test on the substation to obtain load information related to the current; Load the preset load-test fault model and support vector machine; The load-based fault detection model includes an input layer, a hidden layer and an output layer; The load information is received in the input layer, the load information is linearly transformed in the hidden layer to obtain a load vector, and the load vector is mapped to a fault category of the substation when it is loaded in the output layer; A plurality of sample pairs are input into the support vector machine for decision making to obtain the fault type of the substation under load; the sample pairs include the load information and the fault category.

2. The method according to claim 1, characterized in that The load information includes the magnitude of the current, the phase angle of the current, the active power and the reactive power; the load information related to the current obtained by performing a load test on the substation includes: Carry out load test on the substation to obtain the current of each phase; When receiving a confirmation operation triggered by the current information, filtering out the voltage of phase A from the current information as a reference voltage; The angle at which the reference voltage leads the current is set as the phase angle of the current.

3. The method according to claim 1, characterized in that: The receiving load information in the input layer comprises: receiving the load information recorded in a matrix in the input layer; Wherein, the matrix is ​​expressed as: Wherein, P is the matrix of the load information, P 11 To P RQ represents the load information, R is the dimension of the load information, Q is the number of load information, P RQ It is the Rth load information in the Qth load information.

4. The method according to claim 3, characterized in that The linear transformation of the load information in the hidden layer to obtain a load vector includes: Transposing the matrix to construct radial basis centers; Querying the offset of the load information; The load information is linearly transformed according to the matrix, the radial basis center and the offset to obtain a load vector.

5. The method according to claim 4, characterized in that The load vector is expressed as: a i =exp(-[dist(C,P i )] 2 ·b i ); Among them, a i represents the load vector of the i-th load information, i represents the number of load information, C represents the radial basis center, P i represents the load information of the ith element in the matrix, b i represents the offset of the i-th load information, dist(C,P i ) represents the radial basis center to the P i exp represents the exponential function.

6. The method according to claim 1, characterized in that Mapping the load vector to the fault category of the substation when it is loaded in the output layer includes: In the output layer, querying the connection weight between the hidden layer and the input layer; In the output layer, a weighted sum is performed according to the connection weight and the load vector to obtain the fault category of the substation when it is under load.

7. The method according to claim 6, characterized in that In the output layer, weighted summation is performed according to the connection weight and the load vector to obtain the fault category of the substation when it is loaded, including: In the output layer, the connection weight and the load vector are substituted into the following formula for weighted summation to obtain the fault category of the substation under load: y i =I chewed(n i )=n i ; Among them, n i The network output vector representing the load information of the ith 2,1 represents the connection weight, a i represents the load vector of the i-th load information, n represents the number of load information, y i represents the fault category of the i-th load information, and purelin represents the identity function.

8. The method according to any one of claims 1 to 7, characterized in that: The inputting of multiple sample pairs into the support vector machine for decision making to obtain the fault type of the substation under load includes: constructing a decision function in the support vector machine; Substituting multiple sample pairs into the decision function to make a decision, and obtaining the fault type of the substation under load; Wherein, the decision function is expressed as: K(x i ,y i )=exp(-||xi-yi|| 2 ); Among them, α i , α j Both represent Lagrange multipliers, y i ,y j All represent the fault categories, C represents the penalty parameter, δ ij represents the slack variable, n represents the number of load information, x i ,x j Both represent the load information, st represents the constraint condition, and K represents the kernel function.

9. A computer device, characterized in that: The computer device comprises: one or more processors; A storage device for storing one or more programs; When the one or more programs are executed by the one or more processors, the one or more processors implement the substation load fault detection method as described in any one of claims 1 to 8.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the substation load fault detection method as described in any one of claims 1 to 8 is implemented.

Images