Control method, system, program product, device and storage medium for ac-dc power supply of a substation

Abstract

The application discloses a control method of an AC / DC power supply for a transformer substation, and determines an AC / DC conversion module corresponding to an AC power supply through a long short-term neural network model, and then judges whether the corresponding AC / DC conversion module is accurate based on a deep neural network, so that the inaccuracy caused by manual selection of the AC / DC conversion module is avoided, and the stability of operation of the AC / DC power supply for the transformer substation is improved.

Description

Technical Field

[0001] This invention relates to the field of power systems, and more specifically to control methods, systems, program products, equipment, and computer-readable storage media for AC / DC power supplies used in substations. Background Technology

[0002] Substations are the most widespread and commonly installed type of power station in the current power industry. They play a crucial role in daily power operations, and their power supply systems are typically categorized into several types, including AC, DC, UPS, and communication power supplies. In general substation operation, AC power is the primary energy supply equipment, responsible for tasks such as energy storage and power operation. However, many load devices require DC power. AC / DC conversion modules are used to convert the output voltage of AC power to DC power. The suitability of the AC / DC conversion module for both AC and DC power supplies directly affects the stability of substation operation. Therefore, the selection of an AC / DC conversion module is paramount. Currently, most AC / DC conversion modules are manually selected based on experience, which is often inaccurate. An unsuitable AC / DC conversion module will directly impact the stability of the entire substation operation.

[0003] Therefore, there is an urgent need for a control method and system for AC / DC power supplies in substations to select appropriate AC / DC conversion modules in order to ensure the stable operation of substations. Summary of the Invention

[0004] According to a first aspect, one embodiment provides a control method for an AC / DC power supply for a substation, comprising:

[0005] The system acquires continuous signal data of the AC power supply in the substation; determines the corresponding AC / DC conversion module based on a long short-term neural network model (LSN), where the inputs to the LSN are the continuous signal data of the AC power supply, load type, temperature, humidity, and multiple AC / DC conversion modules, and the output is the AC / DC conversion module corresponding to the AC power supply; converts the output voltage of the AC power supply to the output voltage of the DC power supply based on the corresponding AC / DC conversion module; determines whether the output voltage of the converted DC power supply is abnormal after connecting a load based on a deep neural network model (DNN), where the inputs are the continuous output voltage, continuous output current, continuous insulation detection signal, and load type of the converted DC power supply, and the output is normal or abnormal; if the output of the deep neural network model is abnormal, the corresponding AC / DC conversion module is switched to the AC / DC conversion module with the highest similarity among multiple AC / DC conversion modules, and the output voltage of the AC power supply is converted based on the AC / DC conversion module with the highest similarity.

[0006] In one embodiment, if the output of the deep neural network model is normal, the staff is notified to indicate that the current working status is normal.

[0007] In one embodiment, the process of determining the AC / DC conversion module with the highest similarity includes: converting the instruction manual text data of the various AC / DC conversion modules into multiple SimHash values; calculating multiple similarities between the SimHash value of the instruction manual text data of the corresponding AC / DC conversion module and the SimHash values ​​of the instruction manual text data of the various AC / DC conversion modules based on Hamming distance; selecting the AC / DC conversion module with the highest similarity ranking, and using the AC / DC conversion module with the highest ranking as the AC / DC conversion module with the highest similarity.

[0008] In one embodiment, the AC / DC conversion module with the highest similarity is calculated based on the cosine angle algorithm.

[0009] According to a second aspect, one embodiment provides a control system for AC / DC power supplies used in a substation, characterized in that it includes: an acquisition module for acquiring continuous signal data of the AC power supply in the substation; a determination module for determining the AC / DC conversion module corresponding to the AC power supply based on a long short-term neural network model, wherein the input of the long short-term neural network model is the continuous signal data of the AC power supply, load type, temperature, humidity, and multiple AC / DC conversion modules, and the output of the long short-term neural network model is the AC / DC conversion module corresponding to the AC power supply among the multiple AC / DC conversion modules; and a conversion module for converting the output voltage of the AC power supply into the output voltage of the DC power supply based on the corresponding AC / DC conversion module. The output voltage judgment module is used to determine whether the output voltage of the converted DC power supply is abnormal after the load is connected, based on a deep neural network model. The input of the deep neural network model is the continuous output voltage, continuous output current, continuous insulation detection signal, and load type of the converted DC power supply. The output of the deep neural network model is normal or abnormal. The judgment module is also used to switch the corresponding AC / DC conversion module to the AC / DC conversion module with the highest similarity among a variety of AC / DC conversion modules if the output of the deep neural network model is abnormal, and to convert the output voltage of the AC power supply based on the AC / DC conversion module with the highest similarity.

[0010] In one embodiment, the judgment module is further configured to notify staff that the current working status is normal if the output of the deep neural network model is normal.

[0011] In one embodiment, the process of determining the AC / DC conversion module with the highest similarity includes: converting the instruction manual text data of the various AC / DC conversion modules into multiple SimHash values; calculating multiple similarities between the SimHash value of the instruction manual text data of the corresponding AC / DC conversion module and the SimHash values ​​of the instruction manual text data of the various AC / DC conversion modules based on Hamming distance; selecting the AC / DC conversion module with the highest similarity ranking, and using the AC / DC conversion module with the highest ranking as the AC / DC conversion module with the highest similarity.

[0012] According to a third aspect, one embodiment provides a computer program product, including a computer program, characterized in that, when the computer program is executed by a processor, it implements the steps of the control method for AC / DC power supply for substations as described in any of the first aspects above.

[0013] According to a fourth aspect, one embodiment provides an electronic device comprising: a memory; a processor; and a computer program; wherein the computer program is stored in the memory and configured to be executed by the processor to implement the method as described in the first aspect.

[0014] According to a fifth aspect, one embodiment provides a computer-readable storage medium storing a program that can be executed by a processor to implement the method as described in any one of the first aspects above.

[0015] The control method, system, program product, equipment, and computer-readable storage medium for AC / DC power supply in substations according to the above embodiments determine the AC / DC conversion module corresponding to the AC power supply through a long short-term neural network model, and then judge whether the corresponding AC / DC conversion module is accurate based on a deep neural network. This avoids the inaccuracies caused by manually selecting the AC / DC conversion module and improves the stability of AC / DC power supply operation in substations. Attached Figure Description

[0016] Figure 1 A flowchart illustrating a control method for AC / DC power supplies used in substations, provided by an embodiment of the present invention;

[0017] Figure 2 This is a schematic diagram of continuous signal data of AC power supply in an embodiment of the present invention;

[0018] Figure 3 A schematic diagram of a control system for AC / DC power supply in a substation, provided as an embodiment of the present invention;

[0019] Figure 4 This is a schematic diagram of an electronic device provided in an embodiment of the present invention. Detailed Implementation

[0020] The present invention will now be described in further detail with reference to specific embodiments and accompanying drawings. Similar elements in different embodiments are referred to by associated similar element reference numerals. In the following embodiments, many details are described to facilitate a better understanding of the invention. However, those skilled in the art will readily recognize that some features may be omitted in different situations, or may be replaced by other elements, materials, or methods. In some cases, certain operations related to the present invention are not shown or described in the specification. This is to avoid obscuring the core parts of the invention with excessive description. For those skilled in the art, detailed description of these related operations is not necessary; they can fully understand the related operations based on the description in the specification and general technical knowledge in the art.

[0021] Furthermore, the features, operations, or characteristics described in the specification can be combined in any suitable manner to form various embodiments. At the same time, the order of the steps or actions in the method description can be changed or adjusted in a manner obvious to those skilled in the art. Therefore, the various orders in the specification and drawings are only for the clear description of a particular embodiment and do not imply a necessary order, unless otherwise stated that a particular order must be followed.

[0022] The serial numbers assigned to components in this document, such as "first" and "second," are used only to distinguish the described objects and have no sequential or technical meaning. The terms "connection" and "linkage" used in this invention, unless otherwise specified, include both direct and indirect connections (linkages).

[0023] In this embodiment of the invention, the following are provided: Figure 1 The control method for AC / DC power supply in a substation, as shown, includes steps S1 to S5:

[0024] Step S1: Obtain continuous signal data of AC power supply in the substation.

[0025] Continuous signal data from an AC power source can be a sinusoidal AC voltage signal. For example, a graph showing the variation of a sinusoidal AC voltage over time... Figure 2 As shown, Figure 2 This is a schematic diagram of continuous signal data of an AC power supply in an embodiment of the present invention. The continuous signal data of the AC power supply includes the instantaneous value, maximum value, effective value, angular frequency, and initial phase of the AC power supply voltage at a certain moment.

[0026] Step S2: Determine the AC / DC conversion module corresponding to the AC power supply based on the long short-term neural network model. The input of the long short-term neural network model is the continuous signal data of the AC power supply, load type, temperature, humidity, and various AC / DC conversion modules. The output of the long short-term neural network model is the AC / DC conversion module corresponding to the AC power supply among various AC / DC conversion modules.

[0027] Long Short-Term Neural Network (LSTM) models include Long Short-Term Memory (LSTM) networks, which are a type of Recurrent Neural Network (RNN).

[0028] Long Short-Term Neural Network (LSTN) models can process sequence data of arbitrary length, capture sequence information, and output results based on the correlation between preceding and following data within the sequence. By processing continuous signal data of AC power supplies at consecutive time points using LSTN models, it is possible to output features that consider the correlation between working data at each time point, making the output features more accurate and comprehensive.

[0029] Since temperature and humidity can affect the working efficiency of the load and cause overvoltage and overtemperature protection issues during substation load operation, temperature and humidity are also used as inputs to the long short-term neural network model, thereby making the output AC / DC conversion module corresponding to the AC power supply more accurate and comprehensive.

[0030] In some embodiments, the AC / DC conversion module can use full-wave rectification or half-wave rectification to convert the output voltage of the AC power supply to the output voltage of the DC power supply. Full-wave rectification uses a diode bridge circuit structure to convert the negative voltage component of the input voltage into a positive voltage and then rectifys it into a DC voltage (pulse voltage). Half-wave rectification uses a diode to eliminate the negative voltage component of the input voltage and then rectifys it into a DC voltage (pulse voltage).

[0031] A Long Short-Term Neural Network (LSTN) model can be trained using training samples. The input to the training samples includes continuous signal data of the AC power supply, load type, temperature, humidity, and various AC / DC conversion modules. The output of the training samples is the AC / DC conversion module corresponding to the AC power supply among the various AC / DC conversion modules. In some embodiments, the LSN model can be trained using gradient descent to obtain the trained LSN model. Specifically, based on the training samples, a loss function for the LSN model is constructed. The parameters of the LSN model are adjusted using the loss function until the loss function value converges or falls below a preset threshold, at which point training is complete. The loss function can include, but is not limited to, logarithmic (log) loss function, squared loss function, exponential loss function, Hinge loss function, and absolute value loss function.

[0032] In some embodiments, the input of various AC / DC conversion modules can be the instruction manual text data of various AC / DC conversion modules.

[0033] After training, the continuous signal data of the AC power supply, load type, temperature, humidity, and various AC / DC conversion modules are input into the trained long short-term neural network model, and the output is the AC / DC conversion module corresponding to the AC power supply among various AC / DC conversion modules.

[0034] Step S3: Based on the corresponding AC / DC conversion module, the output voltage of the AC power supply is converted into the output voltage of the DC power supply.

[0035] The output voltage of an AC power supply is the effective value of the AC output voltage. An AC / DC conversion module can convert the effective value of the AC power supply output voltage into the DC power supply output voltage and connect it to the load to provide power.

[0036] Step S4: Based on a deep neural network model, determine whether the output voltage of the converted DC power supply is abnormal after the load is connected. The input of the deep neural network model is the continuous output voltage, continuous output current, continuous insulation detection signal, and load type of the converted DC power supply. The output of the deep neural network model is normal or abnormal.

[0037] The continuous insulation detection signal is obtained by the insulation monitoring module. The insulation monitoring module can issue a ground fault alarm signal when the insulation resistance to ground of the positive and negative poles of the DC system reaches the alarm setting value, making the power system operation safer. The insulation detection signal can be the voltage to ground of the two DC bus sections monitored by the insulation monitoring module. Alternatively, the insulation detection signal can be the insulation resistance to ground of the two DC bus sections monitored by the insulation monitoring module.

[0038] Load types include resistors, motors, capacitors, coils, etc. Load types can also include the values ​​of load resistance and load power.

[0039] In some embodiments, a trained deep neural network model can be used to determine whether the output voltage of the converted DC power supply is abnormal after a load is connected. The deep neural network model includes a deep neural network. The deep neural network model may include multiple processing layers, each consisting of multiple neurons, and each neuron performs matrix transformations on the data. The parameters used in the matrix can be obtained through training. The deep neural network model can also be any existing neural network model capable of processing multiple features, such as RNN, CNN, DNN, etc. The deep neural network model can also be a custom model based on requirements. The inputs to the deep neural network model are the continuous output voltage, continuous output current, continuous insulation detection signal, and load type of the converted DC power supply; the output is normal or abnormal. The trained deep neural network model is obtained through training multiple sets of training samples.

[0040] A deep neural network model can be trained using training samples. The inputs to the training samples are the continuous output voltage, continuous output current, continuous insulation detection signal, and load type of the converted DC power supply. The output of the training samples is either normal or abnormal. In some embodiments, the deep neural network model can be trained using gradient descent. Specifically, based on the training samples, a loss function for the deep neural network model is constructed. The parameters of the deep neural network model are adjusted using this loss function until the loss function value converges or falls below a preset threshold, at which point training is complete. The loss function can include, but is not limited to, logarithmic (log) loss function, squared loss function, exponential loss function, Hinge loss function, and absolute value loss function.

[0041] In some embodiments, if the output of the deep neural network model is normal, the staff is notified that the current working status is normal. In some embodiments, if the output of the deep neural network model is abnormal, the process proceeds to step S5.

[0042] Step S5: If the output of the deep neural network model is abnormal, the corresponding AC / DC conversion module is switched to the AC / DC conversion module with the highest similarity among various AC / DC conversion modules, and the output voltage of the AC power supply is converted based on the AC / DC conversion module with the highest similarity.

[0043] Because the AC / DC conversion module corresponding to the AC power supply determined by the Long Short-Term Neural Network (LSTN) model takes into account various factors such as continuous signal data of the AC power supply, load type, temperature, and humidity, the determined AC / DC conversion module is more realistic. Therefore, the AC / DC conversion module with the highest similarity to the AC power supply is more suitable for application in this substation. Furthermore, since the efficiency of substation fault handling must be high after a fault occurs, otherwise it will affect the storage and use of electrical energy, if the DC power supply output voltage is abnormal after the AC / DC conversion module is connected to the load, the AC / DC conversion module with the highest similarity to the AC power supply should be selected for replacement first. This saves time on manual selection and improves fault handling efficiency.

[0044] In some embodiments, the AC / DC conversion module with the highest similarity to the corresponding AC / DC conversion module among a variety of AC / DC conversion modules can be determined in various ways. For example, methods for calculating similarity include Jaccard similarity, Levenshtein's edit distance, and MinHash. The AC / DC conversion module with the highest similarity can be determined by calculating the similarity between the instruction manual text data of each AC / DC conversion module and the instruction manual text data of the corresponding AC / DC conversion module.

[0045] In some embodiments, the SimHash value of the instruction manual text data of the corresponding AC / DC conversion module can be calculated based on Hamming distance, and multiple similarities can be obtained between the SimHash values ​​of the instruction manual text data of the various AC / DC conversion modules; the AC / DC conversion module with the highest similarity ranking can be selected, and the AC / DC conversion module with the highest ranking can be used as the AC / DC conversion module with the highest similarity.

[0046] In some embodiments, the AC / DC conversion module with the highest similarity can also be calculated based on the cosine angle algorithm.

[0047] Based on the same inventive concept, embodiments of the present invention provide a control system for AC / DC power supplies used in substations, such as... Figure 3 As shown, it includes:

[0048] Acquisition module 31 is used to acquire continuous signal data of AC power supply in substation; determination module 32 is used to determine the AC / DC conversion module corresponding to the AC power supply based on a long short-term neural network model, wherein the input of the long short-term neural network model is the continuous signal data of the AC power supply, load type, temperature, humidity, and multiple AC / DC conversion modules, and the output of the long short-term neural network model is the AC / DC conversion module corresponding to the AC power supply among multiple AC / DC conversion modules; conversion module 33 is used to convert the output voltage of AC power supply to DC power supply output voltage based on the corresponding AC / DC conversion module; judgment module 34 is used to determine the AC / DC conversion module based on deep neural network model. The network model determines whether the output voltage of the converted DC power supply is abnormal after a load is connected. The inputs of the deep neural network model are the continuous output voltage, continuous output current, continuous insulation detection signal, and load type of the converted DC power supply. The output of the deep neural network model is normal or abnormal. The judgment module is also used to switch the corresponding AC / DC conversion module to the AC / DC conversion module with the highest similarity among a variety of AC / DC conversion modules if the output of the deep neural network model is abnormal, and to convert the output voltage of the AC power supply based on the AC / DC conversion module with the highest similarity.

[0049] In one embodiment, the judgment module is further configured to notify staff that the current working status is normal if the output of the deep neural network model is normal.

[0050] In one embodiment, the process of determining the AC / DC conversion module with the highest similarity includes: converting the instruction manual text data of the various AC / DC conversion modules into multiple SimHash values; calculating multiple similarities between the SimHash value of the instruction manual text data of the corresponding AC / DC conversion module and the SimHash values ​​of the instruction manual text data of the various AC / DC conversion modules based on Hamming distance; selecting the AC / DC conversion module with the highest similarity ranking, and using the AC / DC conversion module with the highest ranking as the AC / DC conversion module with the highest similarity.

[0051] Based on the same inventive concept, embodiments of the present invention provide an electronic device, such as... Figure 4 As shown, it includes:

[0052] Processor 41; memory 42 for storing executable program instructions in processor 41; wherein processor 41 is configured to execute a method for implementing a control method for an AC / DC power supply for a substation as described above, the method comprising:

[0053] The system acquires continuous signal data of the AC power supply in the substation; determines the corresponding AC / DC conversion module based on a long short-term neural network model (LSN), where the inputs to the LSN are the continuous signal data of the AC power supply, load type, temperature, humidity, and multiple AC / DC conversion modules, and the output is the AC / DC conversion module corresponding to the AC power supply; converts the output voltage of the AC power supply to the output voltage of the DC power supply based on the corresponding AC / DC conversion module; determines whether the output voltage of the converted DC power supply is abnormal after connecting a load based on a deep neural network model (DNN), where the inputs are the continuous output voltage, continuous output current, continuous insulation detection signal, and load type of the converted DC power supply, and the output is normal or abnormal; if the output of the deep neural network model is abnormal, the corresponding AC / DC conversion module is switched to the AC / DC conversion module with the highest similarity among multiple AC / DC conversion modules, and the output voltage of the AC power supply is converted based on the AC / DC conversion module with the highest similarity.

[0054] Based on the same inventive concept, this embodiment provides a non-transitory computer-readable storage medium. When the instructions in the storage medium are executed by the processor 41 of an electronic device, the electronic device is able to execute the control method for AC / DC power supply in a substation as described above. The method includes: acquiring continuous signal data of the AC power supply in the substation; determining the AC / DC conversion module corresponding to the AC power supply based on a long short-term neural network model, wherein the input of the long short-term neural network model is the continuous signal data of the AC power supply, load type, temperature, humidity, and multiple AC / DC conversion modules, and the output of the long short-term neural network model is the AC / DC conversion module corresponding to the AC power supply among the multiple AC / DC conversion modules; and determining the AC / DC conversion module corresponding to the AC power supply based on the corresponding AC / DC conversion module. The DC conversion module converts the output voltage of the AC power supply to the output voltage of the DC power supply. Based on a deep neural network model, it determines whether the output voltage of the converted DC power supply is abnormal after a load is connected. The inputs to the deep neural network model are the continuous output voltage, continuous output current, continuous insulation detection signal, and load type of the converted DC power supply. The output of the deep neural network model is either normal or abnormal. If the output of the deep neural network model is abnormal, the corresponding AC / DC conversion module is switched to the AC / DC conversion module with the highest similarity among various AC / DC conversion modules, and the output voltage of the AC power supply is converted based on the AC / DC conversion module with the highest similarity.

[0055] The basic concepts have been described above. Obviously, for those skilled in the art, the detailed disclosure above is merely illustrative and does not constitute a limitation of this specification. Although not explicitly stated herein, those skilled in the art may make various modifications, improvements, and corrections to this specification. Such modifications, improvements, and corrections are suggested in this specification and therefore remain within the spirit and scope of the exemplary embodiments described herein.

[0056] Furthermore, this specification uses specific terms to describe embodiments thereof. For example, "an embodiment," "one embodiment," and / or "some embodiments" refer to a particular feature, structure, or characteristic associated with at least one embodiment of this specification. Therefore, it should be emphasized and noted that references to "an embodiment," "one embodiment," or "an alternative embodiment" in different locations throughout this specification do not necessarily refer to the same embodiment. Moreover, certain features, structures, or characteristics in one or more embodiments of this specification can be appropriately combined.

[0057] Furthermore, unless expressly stated in the claims, the order of processing elements and sequences, the use of numbers and letters, or other names described in this specification are not intended to limit the order of the processes and methods described herein. Although various examples have been discussed in the foregoing disclosure of some embodiments of the invention that are currently considered useful, it should be understood that such details are for illustrative purposes only, and the appended claims are not limited to the disclosed embodiments; rather, the claims are intended to cover all modifications and equivalent combinations that conform to the spirit and scope of the embodiments described herein. For example, while the system components described above can be implemented using hardware devices, they can also be implemented solely using software solutions, such as installing the described system on existing servers or mobile devices.

[0058] Similarly, it should be noted that, in order to simplify the description disclosed herein and thus aid in the understanding of one or more embodiments of the invention, the foregoing description of embodiments in this specification may sometimes combine multiple features into a single embodiment, drawing, or description thereof. However, this method of disclosure does not imply that the subject matter of this specification requires more features than those mentioned in the claims. In fact, the embodiments contain fewer features than all the features of a single embodiment disclosed above.

[0059] Finally, it should be understood that the embodiments described in this specification are merely illustrative of the principles of the embodiments described herein. Other variations may also fall within the scope of this specification. Therefore, alternative configurations of the embodiments described herein are intended to be illustrative rather than limiting, and should be considered consistent with the teachings of this specification. Accordingly, the embodiments described herein are not limited to those explicitly introduced and described herein.

Claims

1. A control method of an AC / DC power supply for a substation, characterized by, include: Acquire continuous signal data of AC power supply in substation; The AC / DC conversion module corresponding to the AC power supply is determined based on a long short-term neural network model. The input of the long short-term neural network model is the continuous signal data of the AC power supply, load type, temperature, humidity, and multiple AC / DC conversion modules. The output of the long short-term neural network model is the AC / DC conversion module corresponding to the AC power supply among the multiple AC / DC conversion modules. Based on the corresponding AC / DC conversion module, the output voltage of the AC power supply is converted into the output voltage of the DC power supply; The deep neural network model is used to determine whether the output voltage of the converted DC power supply is abnormal after the load is connected. The input of the deep neural network model is the continuous output voltage, continuous output current, continuous insulation detection signal, and load type of the converted DC power supply. The output of the deep neural network model is normal or abnormal. If the output of the deep neural network model is abnormal, the corresponding AC / DC conversion module is switched to the AC / DC conversion module with the highest similarity among multiple AC / DC conversion modules. The output voltage of the AC power supply is then converted based on the AC / DC conversion module with the highest similarity. The process of determining the AC / DC conversion module with the highest similarity includes: The instruction manual text data of the various AC / DC conversion modules are converted into multiple SimHash values; The SimHash value of the instruction manual text data of the corresponding AC / DC conversion module and the SimHash value of the instruction manual text data of the various AC / DC conversion modules are calculated based on Hamming distance. Select the AC / DC conversion module with the highest similarity ranking, and use the AC / DC conversion module with the highest ranking as the AC / DC conversion module with the highest similarity.

2. The control method of an AC / DC power supply for a substation according to Claim 1, characterized by Also includes: If the output of the deep neural network model is normal, the staff will be notified that the current working status is normal.

3. The control method of an AC / DC power supply for a substation according to Claim 1, characterized by, The AC / DC conversion module with the highest similarity was calculated based on the cosine angle algorithm.

4. A control system for an AC / DC power supply for a substation, characterized by include: The acquisition module is used to acquire continuous signal data of the AC power supply in the substation. The determination module is used to determine the AC / DC conversion module corresponding to the AC power supply based on a long short-term neural network model. The input of the long short-term neural network model is the continuous signal data of the AC power supply, load type, temperature, humidity, and multiple AC / DC conversion modules. The output of the long short-term neural network model is the AC / DC conversion module corresponding to the AC power supply among the multiple AC / DC conversion modules. A conversion module is used to convert the output voltage of an AC power supply to the output voltage of a DC power supply based on the corresponding AC / DC conversion module. A judgment module is used to determine whether the output voltage of the converted DC power supply is abnormal after a load is connected, based on a deep neural network model. The inputs of the deep neural network model are the continuous output voltage, continuous output current, continuous insulation detection signal, and load type of the converted DC power supply. The output of the deep neural network model is normal or abnormal. The judgment module is also used to, if the output of the deep neural network model is abnormal, switch the corresponding AC / DC conversion module to the AC / DC conversion module with the highest similarity among multiple AC / DC conversion modules, and convert the output voltage of the AC power supply based on the AC / DC conversion module with the highest similarity. The process of determining the AC / DC conversion module with the highest similarity includes: The instruction manual text data of the various AC / DC conversion modules are converted into multiple SimHash values; The SimHash value of the instruction manual text data of the corresponding AC / DC conversion module and the SimHash value of the instruction manual text data of the various AC / DC conversion modules are calculated based on Hamming distance. Select the AC / DC conversion module with the highest similarity ranking, and use the AC / DC conversion module with the highest ranking as the AC / DC conversion module with the highest similarity.

5. The control system for AC / DC power supply in a substation as described in claim 4, characterized in that, The judgment module is also used to notify staff that the current working status is normal if the output of the deep neural network model is normal.

6. A computer program product comprising a computer program, characterized in that, When the computer program is executed by the processor, it implements the steps of the control method for AC / DC power supply in substations as described in any one of claims 1 to 3.

7. An electronic device, comprising: include: Memory; processor; And a computer program; wherein the computer program is stored in the memory and configured to be executed by the processor to implement the steps of the control method for AC / DC power supply for substations as described in any one of claims 1 to 3.

8. A computer-readable storage medium having stored thereon a computer program, characterized in that, When the program is executed by the processor, it implements the steps corresponding to the control method for AC / DC power supply in substations as described in any one of claims 1 to 3.

Images