Dispatching command system and method with intelligent anti-misoperation function

Abstract

The embodiment of the invention discloses a dispatching command system and method with an intelligent anti-misoperation function, and relates to the technical field of intelligent power grid dispatching. The method comprises the following steps: acquiring current state information of a target power grid and a related power grid dispatching instruction; comparing the power grid dispatching instruction with a target reference instruction sequence to obtain suspected abnormal instruction information; comparing the current state information with reference state information matched with the target reference instruction sequence to obtain suspected abnormal state information; inputting the suspected abnormal instruction information and the suspected abnormal state information into a feature extraction unit of a first anti-misoperation check model for processing to obtain a first abnormal feature and a second abnormal feature; inputting the first abnormal feature and the second abnormal feature into a causal logic analysis unit of the model for processing to obtain a matching relationship between the first abnormal feature and the second abnormal feature; and obtaining a first anti-misoperation verification result based on the matching relationship, and carrying out power grid dispatching command according to the first anti-misoperation verification result.

Description

Technical Field

[0001] The present application relates to the field of smart grid dispatching technology, and more specifically, to a dispatching command system and method with intelligent error prevention functions. Background Art

[0002] Grid dispatching is a key link to ensure the safe, stable and economical operation of the power system. However, during the grid dispatching process, the mis-issuance of operation instructions may cause equipment damage, grid failure or even large-scale power outages.

[0003] Traditional dispatching and commanding methods often rely on manual operation and judgment, which is not only inefficient, but also easily affected by human factors and information asymmetry (such as incorrectly sending dispatching instructions due to failure to understand node status in a timely manner), thereby increasing the risk of misoperation.

[0004] In recent years, with the continuous development of smart grid technology, higher requirements have been put forward for the intelligence and automation level of power grid dispatching and command systems. Therefore, it is necessary to study a dispatching and command system and method with intelligent error prevention function to improve the accuracy and safety of power grid dispatching. Summary of the invention

[0005] To solve the above problems, one aspect of the embodiments of this specification provides a dispatching command method with an intelligent error prevention function, the method comprising: Acquire current state information of a target power grid and a power grid dispatch instruction related to the target power grid, wherein the current state information is used to at least reflect current load data, equipment operation status, and topological connection relationship of multiple nodes included in the target power grid; Comparing the power grid dispatching instruction with a target reference instruction sequence to obtain suspected abnormal instruction information in the power grid dispatching instruction; Compare the current state information with at least one reference state information matched with the target reference instruction sequence to obtain suspected abnormal state information in the current state information; Inputting the suspected abnormal instruction information and the suspected abnormal state information into a feature extraction unit of a pre-trained first anti-mischeck model for processing, so as to extract a first abnormal feature corresponding to the suspected abnormal instruction information and a second abnormal feature corresponding to the suspected abnormal state information; Inputting the first abnormal feature and the second abnormal feature into the causal logic analysis unit of the first anti-error verification model for processing to obtain a matching relationship between the first abnormal feature and the second abnormal feature; A first anti-error verification result is obtained based on the matching relationship, and power grid dispatching and command are performed according to the first anti-error verification result; wherein the first anti-error verification result is used to reflect whether the power grid dispatching instruction contains an erroneous operation instruction, and the erroneous operation type corresponding to the power grid dispatching instruction.

[0006] In some embodiments, before acquiring the current state information of the target power grid and the power grid dispatching instruction related to the target power grid, the method further includes: A second error prevention check is performed on the operation instructions input by the dispatching operator for the target power grid according to the repeated voice data, and the operation instructions that pass the second error prevention check are used as power grid dispatching instructions related to the target power grid.

[0007] In some embodiments, the second error prevention check on the operation instructions input by the dispatch operator for the target power grid according to the repeated voice data includes: Obtain a first semantic vector according to the repeated voice data, and obtain a second semantic vector according to the operation instruction input by the dispatching operator for the target power grid; The first semantic vector and the second semantic vector are input into a pre-trained second anti-error verification model for processing to obtain a second anti-error verification result, which is used to reflect whether the operation instructions input by the dispatching operator for the target power grid are consistent with the operations corresponding to the repeated voice data.

[0008] In some embodiments, the inputting the first semantic vector and the second semantic vector into a pre-trained second anti-error verification model for processing to obtain a second anti-error verification result includes: Mapping the first semantic vector to a target space to obtain a first mapping result; Mapping the second semantic vector to the target space to obtain a second mapping result; The distance between the first mapping result and the second mapping result in the target space is calculated, and the second anti-error verification result is obtained based on the distance.

[0009] In some embodiments, the comparing the power grid dispatching instruction with the target reference instruction sequence to obtain the suspected abnormal instruction information in the power grid dispatching instruction includes: Calculating the similarity between the power grid dispatching instruction and a plurality of preset reference instruction sequences, and taking a preset reference instruction sequence having the highest similarity to the power grid dispatching instruction as a target reference instruction sequence; According to the difference between the power grid dispatching instruction and the target reference instruction sequence, the suspected abnormal instruction information in the power grid dispatching instruction is obtained, wherein the suspected abnormal instruction information includes the suspected abnormal instruction content and the corresponding node information.

[0010] In some embodiments, the comparing the current state information with at least one reference state information matching the target reference instruction sequence to obtain suspected abnormal state information in the current state information includes: Acquire at least one reference state information matching the target reference instruction sequence; Determine the difference between the current state information and each of the reference state information, and fuse the differences between the current state information and all of the reference state information to obtain suspected abnormal state information in the current state information, wherein the suspected abnormal state information includes the suspected abnormal state type and its corresponding node information.

[0011] In some embodiments, the feature extraction unit includes an instruction feature extraction submodule and a state feature extraction submodule, and the inputting of the suspected abnormal instruction information and the suspected abnormal state information into the feature extraction unit of the pre-trained first anti-mischeck model for processing to extract the first abnormal feature corresponding to the suspected abnormal instruction information and the second abnormal feature corresponding to the suspected abnormal state information includes: Inputting the suspected abnormal instruction information into the instruction feature extraction submodule for processing to obtain the first abnormal feature; The suspected abnormal state information is input into the state feature extraction submodule for processing to obtain the second abnormal feature, and the state feature extraction submodule and the instruction feature extraction submodule are obtained through collaborative training.

[0012] In some embodiments, the causal logic analysis unit includes a machine learning model, which is used to learn the potential causal logical relationship between the first abnormal feature and the second abnormal feature, and perform matching analysis on the first abnormal feature and the second abnormal feature based on the learned causal logical relationship knowledge, so as to obtain the matching relationship between the first abnormal feature and the second abnormal feature; wherein the machine learning model includes any one of a neural network model, a decision tree model, a support vector machine model or a naive Bayes model.

[0013] In some embodiments, the feature extraction unit includes a coding network, and the coding network is used to encode the suspected abnormal instruction information and the suspected abnormal state information to obtain the first abnormal feature and the second abnormal feature; The causal logic analysis unit includes a causal reasoning module, which is used to parse the causal chain between the first abnormal feature and the second abnormal feature based on a pre-built power grid dispatch causal knowledge graph to obtain a matching relationship between the first abnormal feature and the second abnormal feature.

[0014] Another aspect of the embodiments of this specification also provides a dispatching and commanding system with an intelligent error prevention function, the system comprising: An acquisition module, used to acquire current state information of a target power grid and power grid dispatching instructions related to the target power grid, wherein the current state information is used to at least reflect current load data, equipment operation status, and topological connection relationship of multiple nodes included in the target power grid; An abnormal instruction determination module is used to compare the power grid dispatching instruction with a target reference instruction sequence to obtain suspected abnormal instruction information in the power grid dispatching instruction; an abnormal state determination module, configured to compare the current state information with at least one reference state information matched with the target reference instruction sequence, and obtain suspected abnormal state information in the current state information; a feature extraction module, configured to input the suspected abnormal instruction information and the suspected abnormal state information into a feature extraction unit of a pre-trained first anti-mischeck model for processing, so as to extract a first abnormal feature corresponding to the suspected abnormal instruction information and a second abnormal feature corresponding to the suspected abnormal state information; a feature matching relationship determination module, configured to input the first abnormal feature and the second abnormal feature into a causal logic analysis unit of the first anti-error verification model for processing, and obtain a matching relationship between the first abnormal feature and the second abnormal feature; The first anti-error verification result determination module is used to obtain the first anti-error verification result based on the matching relationship, and perform power grid dispatching and command according to the first anti-error verification result; wherein the first anti-error verification result is used to reflect whether the power grid dispatching instruction contains an erroneous operation instruction, and the erroneous operation type corresponding to the power grid dispatching instruction.

[0015] The dispatching and command system and method with intelligent error prevention function provided in the embodiments of this specification may bring about beneficial effects including at least: by processing the current status information of the target power grid and the power grid dispatching instructions related to the target power grid, suspected abnormal instruction information and suspected abnormal status information are obtained, and then the suspected abnormal instruction information and the suspected abnormal status information are processed by the first error prevention verification model to analyze the matching relationship between the two. It is possible to analyze whether the power grid dispatching instructions for the target power grid during the power grid dispatching process are applicable to its current state, and identify possible misoperation situations, so as to facilitate timely taking corresponding measures to correct or adjust them, thereby greatly improving the safety and reliability of power grid dispatching and reducing power grid accidents caused by misoperation or abnormal power grid status.

[0016] Additional features will be described in part in the following description. For those skilled in the art, it will become apparent by reviewing the following and accompanying drawings, or it may be understood by the production or operation of the examples. The features of this specification may be implemented and obtained by practicing or using various aspects of the methods, tools, and combinations described in the following detailed examples. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] This specification will be further described in the form of exemplary embodiments, which will be described in detail by the accompanying drawings. These embodiments are not restrictive, and in these embodiments, the same number represents the same structure, wherein: Figure 1 is an exemplary flow chart of a dispatching and commanding method with an intelligent error prevention function according to some embodiments of this specification; Figure 2 is an exemplary structural diagram of a second anti-mischeck model according to some embodiments of this specification; Figure 3 is an exemplary structural diagram of a first anti-mischeck model according to some embodiments of this specification; Figure 4 It is an exemplary module diagram of a dispatching and commanding system with intelligent error prevention function according to some embodiments of this specification. DETAILED DESCRIPTION

[0018] In order to more clearly illustrate the technical solutions of the embodiments of this specification, the following is a brief introduction to the drawings required for the description of the embodiments. Obviously, the drawings described below are only some examples or embodiments of this specification. For ordinary technicians in this field, this specification can also be applied to other similar scenarios based on these drawings without creative work. Unless it is obvious from the language environment or otherwise explained, the same reference numerals in the figures represent the same structure or operation.

[0019] It should be understood that the "system", "device", "unit" and / or "module" used in this specification is a method for distinguishing different components, elements, parts, parts or assemblies at different levels. However, if other words can achieve the same purpose, the words can be replaced by other expressions.

[0020] As shown in this specification and claims, unless the context clearly indicates an exception, the words "a", "an", "an" and / or "the" do not refer to the singular and may also include the plural. Generally speaking, the terms "comprise" and "include" only indicate the inclusion of the steps and elements that have been clearly identified, and these steps and elements do not constitute an exclusive list. The method or device may also include other steps or elements.

[0021] Flowcharts are used in this specification to illustrate the operations performed by the system according to the embodiments of this specification. It should be understood that the preceding or following operations are not necessarily performed precisely in order. Instead, the steps may be processed in reverse order or simultaneously. At the same time, other operations may also be added to these processes, or one or more operations may be removed from these processes.

[0022] The dispatching and commanding system and method with intelligent error prevention function provided by the embodiments of this specification are described in detail below with reference to the accompanying drawings.

[0023] Figure 1 is an exemplary flow chart of a dispatching command method with intelligent error prevention function according to some embodiments of this specification. In some embodiments, the method can be executed by processing logic, which can include hardware (e.g., circuits, dedicated logic, programmable logic, microcode, etc.), software (instructions running on a processing device to perform hardware simulation), etc. or any combination thereof. In some embodiments, Figure 1 One or more operations in the flowchart of the dispatching and commanding method with intelligent error prevention function shown can be implemented by a processing device and / or a terminal device. For example, the dispatching and commanding method with intelligent error prevention function can be stored in a storage device in the form of a computer program and / or instruction, and called and / or executed by a processing device and / or a terminal device.

[0024] Reference Figure 1 The dispatching command method with intelligent error prevention function provided in the embodiment of the present application may include the following steps S110-S160: Step S110, obtaining current state information of the target power grid and power grid dispatching instructions related to the target power grid. In some embodiments, step S110 may be executed by the acquisition module 210 mentioned below.

[0025] In an embodiment of the present application, the target power grid may refer to any power grid system that requires dispatching and command. The current status information may at least be used to reflect the current load data, equipment operating status, and topological connection relationship of multiple nodes contained in the target power grid. Exemplarily, in some embodiments, the current status information may include but is not limited to real-time data such as voltage, current, power, load, fault conditions, etc. of the target power grid and the topological connection relationship between each node device in the target power grid. Among them, the load data can be obtained in real time through the SCADA (Supervisory Control And Data Acquisition) system, and the equipment operating status can be obtained through the device's own sensor or monitoring system.

[0026] The topological connection relationship can be understood as the physical connection mode and electrical connection path between each node device in the target power grid, which can describe the structural layout of the target power grid, as well as the flow path of electrical quantities such as current and voltage in the power grid. Through this topological connection relationship, the overall architecture of the power grid and the relationship between each part can be clearly grasped. In some embodiments, the topological connection relationship can be pre-configured and dynamically adjusted according to the actual working status of each node.

[0027] In the embodiment of the present application, the grid dispatching instruction related to the target grid refers to the control command issued by the dispatcher or the automation system for adjusting the operation state of the grid, such as switch operation, power adjustment, load transfer, etc. These instructions are intended to ensure the safe, stable and efficient operation of the grid. By accurately executing these instructions, it is possible to achieve precise control of the grid state, effectively respond to various emergencies and load changes, and thus improve the overall operation efficiency and reliability of the grid.

[0028] At present, the dispatch order execution process is usually as follows: first, the centralized control operator completes the repetition after receiving the order in the dispatch order system, and then transfers the order to the on-site duty personnel by phone. After performing the operation on site, the centralized control operator replies to the dispatcher by phone. Finally, the centralized control operator replies to the dispatcher through the dispatch order system.

[0029] In some embodiments of the present application, in order to achieve automatic consistency verification of the repeated voice and the dispatching operation instructions, a second anti-error verification can be performed on the operation instructions input by the dispatching operator for the target power grid based on the repeated voice data, and the operation instructions that pass the second anti-error verification can be used as the power grid dispatching instructions related to the above-mentioned target power grid.

[0030] In some embodiments of the present application, a first semantic vector can be obtained based on the repetitive voice data, and a second semantic vector can be obtained based on the operation instruction input by the dispatching operator for the target power grid; then the first semantic vector and the second semantic vector are input into a pre-trained second anti-error verification model for processing to obtain a second anti-error verification result. In an embodiment of the present application, the second anti-error verification result can be used to reflect whether the operation instruction input by the dispatching operator for the target power grid is consistent with the operation corresponding to the repetitive voice data.

[0031] Specifically, in an embodiment of the present application, the repetition voice data can be directly converted into a high-dimensional vector through models such as wav2vec and HuBERT to obtain the first semantic vector. In some embodiments, the repetition voice data can also be first converted into text data (for example, using speech recognition technology) and then the text data is converted into a text vector (for example, using models such as Word2Vec and BERT) to obtain the first semantic vector corresponding to the repetition voice data.

[0032] For the operation instructions input by the dispatching operator for the target power grid (the operation instructions are given in the form of text codes), in an embodiment of the present application, they can be parsed and vectorized through models such as Code2Vec or seq2vec to obtain the second semantic vector.

[0033] It should be noted that the above-mentioned models are only exemplary. In other embodiments of the present application, any other feasible models may be used for processing to obtain the first semantic vector and the second semantic vector. More details about the process of obtaining the first semantic vector and the second semantic vector can be regarded as prior art and will not be discussed in detail in this specification.

[0034] Figure 2 is an exemplary structural diagram of the second anti-mischeck model according to some embodiments of this specification. Figure 2 In some embodiments of the present specification, the second anti-error verification model may include a vector mapping unit and a distance calculation unit. The vector mapping unit may be used to map the first semantic vector to a target space to obtain a first mapping result, and to map the second semantic vector to the target space to obtain a second mapping result. The distance calculation unit may be used to calculate the distance between the first mapping result and the second mapping result in the target space, and obtain the second anti-error verification result based on the distance.

[0035] Specifically, in the embodiment of the present application, the above-mentioned target space can be understood as a high-dimensional vector space, each dimension of which can represent a specific semantic feature. In the embodiment of the present application, by mapping the first semantic vector and the second semantic vector to the target space, the first mapping result and the second mapping result are obtained, so that the two semantic vectors that were originally difficult to directly compare in the original space can be compared in the target space by calculating the distance. It can be understood that the smaller the distance between the first mapping result and the second mapping result in the target space, the more similar the semantics of the first semantic vector and the second semantic vector, that is, the higher the consistency between the operation instructions input by the dispatching operator for the target power grid and the operation corresponding to the recitation voice data; conversely, the larger the distance, the greater the semantic difference between the two semantic vectors, the lower the consistency between the operation instructions input by the dispatching operator for the target power grid and the operation corresponding to the recitation voice data, and the greater the possibility of misoperation.

[0036] Specifically, in an embodiment of the present application, the vector mapping unit may be a deep learning model (such as a neural network model), which can map input semantic vectors of different modalities to the target space after training. In some embodiments, a sample pair can be formed by combining a first sample semantic vector obtained based on the processing of repeated voice data and a second sample semantic vector obtained based on the operation instructions input by the dispatching operator for the target power grid (the operations corresponding to the first sample semantic vector and the second sample semantic vector in the sample pair may be consistent or inconsistent), and then a large number of similar sample pairs (including positive samples and negative samples) are used to train the deep learning model configured by the vector mapping unit so that it learns how to map semantic vectors of different modalities to the same target space and maintain the consistency of the semantics before and after the mapping. More details about the model training can be regarded as prior art and will not be repeated in this specification.

[0037] In an embodiment of the present application, the distance calculation unit may include a distance calculation module, which may use Euclidean distance, cosine similarity or other appropriate distance measurement methods to calculate the distance between the first mapping result and the second mapping result in the high-dimensional vector space (i.e., the target space). In an embodiment of the present application, a second anti-error verification result may be obtained based on the result of the distance calculation, and the second anti-error verification result may be used to reflect whether the operation instruction input by the dispatching operator for the target power grid is consistent with the operation corresponding to the repetition voice data. Specifically, if the distance calculated by the distance calculation unit is less than a preset threshold, it can be considered that the operation instruction input by the dispatching operator for the target power grid is highly consistent with the operation corresponding to the repetition voice data, and the risk of misoperation is low; conversely, if the distance is greater than or equal to the preset threshold, it may be indicated that there is a potential risk of misoperation, and further verification or confirmation is required.

[0038] It should be pointed out that in the embodiment of the present application, by performing semantic comparison processing on the operation instructions input by the dispatching operator for the target power grid and the repeated voice data through the above-mentioned second anti-error verification model, the consistency of the operation instructions input by the dispatching operator for the target power grid and the operations corresponding to the repeated voice data can be verified, thereby effectively improving the accuracy and safety of the dispatching operation, reducing the risk of human error operation, and at the same time improving the automation level and intelligence of the entire power grid dispatching and command system.

[0039] Continue to refer to Figure 1 After the above step S110, the method further includes: Step S120, comparing the power grid dispatching instruction with the target reference instruction sequence to obtain suspected abnormal instruction information in the power grid dispatching instruction. In some embodiments, step S120 can be performed by the abnormal instruction determination module 220 mentioned below.

[0040] In the embodiment of the present application, the suspected abnormal instruction information in the power grid dispatching instruction can be obtained by comparing the above-mentioned power grid dispatching instruction with a preset reference instruction sequence. Among them, the reference instruction sequence can be understood as a data sequence corresponding to the common dispatching instructions in the historical statistical data. In the embodiment of the present application, by comparing the above-mentioned power grid dispatching instruction with the reference instruction sequence, the deviation (i.e., the difference) compared with the conventional dispatching instruction can be analyzed, thereby identifying the suspected abnormal instruction information in the power grid dispatching instruction.

[0041] Specifically, in some embodiments of the present application, the similarity between the power grid dispatching instruction and multiple preset reference instruction sequences can be calculated (for example, using cosine similarity, Euclidean distance and other algorithms for calculation), and then a preset reference instruction sequence with the highest similarity to the power grid dispatching instruction is used as the target reference instruction sequence; further, the suspected abnormal instruction information in the power grid dispatching instruction can be obtained based on the difference between the power grid dispatching instruction and the target reference instruction sequence. It should be noted that in the embodiments of the present application, the suspected abnormal instruction information includes the suspected abnormal instruction content and its corresponding node information (that is, the control node information corresponding to the abnormal instruction content). In some embodiments of the present application, the suspected abnormal instruction information can be reflected in the form of marking the suspected abnormal instruction in the power grid dispatching instruction.

[0042] Step S130, compare the current state information with at least one reference state information matched by the target reference instruction sequence to obtain suspected abnormal state information in the current state information. In some embodiments, step S130 may be performed by the abnormal state determination module 230 mentioned below.

[0043] In the embodiment of the present application, a mapping relationship between several common dispatching instructions and one or more corresponding power grid states can be created through historical statistical data, thereby obtaining a mapping table. Through the mapping table, the rationality of each dispatching instruction under the corresponding power grid state can be reflected.

[0044] In an embodiment of the present application, at least one reference state information matching the target reference instruction sequence can be determined based on the mapping table. Further, the current state information of the target power grid can be compared with at least one reference state information matching the target reference instruction sequence, thereby obtaining suspected abnormal state information in the current state information.

[0045] It can be understood that in an embodiment of the present application, by comparing the current state information of the target power grid with at least one reference state information matched by the target reference instruction sequence, the deviation (i.e., difference) from one or more conventional power grid states matched with the aforementioned target reference instruction sequence can be analyzed, thereby identifying suspected state instruction information in the current state information. In an embodiment of the present application, the suspected abnormal state information may include suspected abnormal state types (e.g., excessive load, abnormal equipment state, etc.) and their corresponding node information. In some embodiments of the present application, when there are multiple reference state information matched by the target reference instruction sequence, the suspected state instruction information can be obtained by fusing the difference between the current state information and the multiple reference state information.

[0046] It should be noted that, in the embodiment of the present application, when both the above-mentioned suspected abnormal instruction information and the suspected abnormal state information reflect that there is no abnormality (that is, the above-mentioned power grid dispatching instruction is completely consistent with the target reference instruction sequence, and the above-mentioned current state information is completely consistent with one of the one or more reference state information matching the target reference instruction sequence), it can be regarded as that there is no misoperation. When one (one and only one) of the above-mentioned suspected abnormal instruction information and the suspected abnormal state information reflects the existence of an abnormality, it indicates that there may be a misoperation. When both the above-mentioned suspected abnormal instruction information and the suspected abnormal state information reflect the existence of an abnormality, it is necessary to further analyze whether the abnormality of the power grid dispatching instruction is caused by the abnormality of the current power grid state (that is, analyze whether there is a causal logical relationship between the two). The analysis process is detailed in steps S140-S160.

[0047] Step S140, input the suspected abnormal instruction information and the suspected abnormal state information into the feature extraction unit of the pre-trained first anti-mischeck model for processing, so as to extract the first abnormal feature corresponding to the suspected abnormal instruction information and the second abnormal feature corresponding to the suspected abnormal state information. In some embodiments, step S140 can be performed by the feature extraction module 240 mentioned later.

[0048] Reference Figure 3 In an embodiment of the present application, the suspected abnormal instruction information and the suspected abnormal state information can be input as input data into a feature extraction unit of a pre-trained first anti-mischeck model for processing to extract a first abnormal feature corresponding to the suspected abnormal instruction information and a second abnormal feature corresponding to the suspected abnormal state information.

[0049] For details, please refer to Figure 3 The feature extraction unit may include an instruction feature extraction submodule and a state feature extraction submodule. In the embodiment of the present application, the suspected abnormal instruction information may be input into the instruction feature extraction submodule for processing to obtain the first abnormal feature; the suspected abnormal state information may be input into the state feature extraction submodule for processing to obtain the second abnormal feature.

[0050] It should be noted that in an embodiment of the present application, the instruction feature extraction submodule and the state feature extraction submodule can be two independent deep learning networks (such as convolutional neural networks), which are used to extract the features of suspected abnormal instruction information and suspected abnormal state information, respectively. Among them, the instruction feature extraction submodule can perform deep learning and feature extraction on the characteristics of the instruction information, such as instruction type, control object (such as node device), operation parameters, etc., so as to obtain the first abnormal feature. The state feature extraction submodule can perform deep learning and feature extraction on the characteristics of the state information, such as equipment status, load parameters, environmental parameters, topological connection relationship between nodes, etc., so as to obtain the second abnormal feature.

[0051] In some embodiments of the present application, the state feature extraction submodule and the instruction feature extraction submodule can be obtained through collaborative training. The specific training process can refer to step S150.

[0052] Step S150, input the first abnormal feature and the second abnormal feature into the causal logic analysis unit of the first anti-error verification model for processing, and obtain a matching relationship between the first abnormal feature and the second abnormal feature. In some embodiments, step S150 can be performed by the feature matching relationship determination module 250 mentioned later.

[0053] Continue to refer to Figure 3 After obtaining the first abnormal feature and the second abnormal feature, the first abnormal feature and the second abnormal feature can be input into the causal logic analysis unit of the first anti-mischeck model for processing to obtain a matching relationship between the first abnormal feature and the second abnormal feature.

[0054] Specifically, in an embodiment of the present application, the causal logic analysis unit may include a machine learning model, which can be used to learn the potential causal logical relationship between the first abnormal feature and the second abnormal feature, and perform matching analysis on the first abnormal feature and the second abnormal feature based on the learned causal logical relationship knowledge, so as to obtain the matching relationship between the first abnormal feature and the second abnormal feature.

[0055] Exemplarily, in some embodiments, the machine learning model may include any one of a neural network model, a decision tree model, a support vector machine model, or a naive Bayes model, and these machine learning models may automatically learn and extract the potential causal logical relationship therein based on the first abnormal feature and the second abnormal feature processed by the feature extraction unit, and then match and analyze the two abnormal features to determine whether there is a causal relationship. If there is no causal relationship, it means that the power grid dispatch instruction may not match the current state of the target, and there may be an erroneous operation, which requires further verification.

[0056] The following is a brief description of the training process of the first anti-error verification model in the embodiment of the present application: In the embodiment of the present application, a training set can be constructed by a large number of sample scheduling instructions and sample power grid states. Specifically, the sample suspected abnormal instruction information corresponding to each sample scheduling instruction can be determined (the determination process can refer to the above suspected abnormal instruction information), and the sample suspected abnormal state information corresponding to each sample power grid state can be determined (the determination process can refer to the above suspected abnormal state information), and then the sample suspected abnormal instruction information and the sample suspected abnormal state information are used as a set of training data.

[0057] Furthermore, in the embodiment of the present application, for each set of training data, a corresponding training label can be determined according to the matching relationship between the corresponding sample scheduling instruction and the sample power grid state. Specifically, when the sample scheduling instruction and the sample power grid state match correctly (that is, the sample scheduling instruction is applicable to the sample power grid state), the training label of the set of training data is determined as "matched"; on the contrary, if there is a mismatch between the sample scheduling instruction and the sample power grid state, the training label of the set of training data is determined as "mismatched", indicating that the sample scheduling instruction is not applicable to the sample power grid state, and there may be a misoperation at this time.

[0058] Through the above method, a large number of training samples and their corresponding training labels can be obtained. Furthermore, each group of training samples can be input into the feature extraction unit of the first anti-error verification model to obtain the first abnormal feature and the second abnormal feature, and then the first abnormal feature and the second abnormal feature can be input into the causal logic analysis unit of the first anti-error verification model to obtain a predicted matching relationship. In an embodiment of the present application, the model parameters corresponding to the instruction feature extraction submodule and the state feature extraction submodule contained in the feature extraction unit and the parameters of the machine learning model contained in the causal logic analysis unit can be adjusted according to the difference between the predicted matching relationship and the above training label, until the difference between the predicted matching relationship and the above training label reaches the preset requirement, the training process of the first anti-error verification model can be completed, and the above-mentioned trained first anti-error verification model can be obtained.

[0059] After completing the above training, the above suspected abnormal instruction information and suspected abnormal state information can be processed by the trained first anti-error verification model to analyze the matching relationship between the two. In an embodiment of the present application, the matching relationship can reflect whether there is a causal logical relationship between the above suspected abnormal instruction information and the suspected abnormal state information. If there is a causal logical relationship, it means that the power grid dispatching instruction is applicable to the current state of the target power grid. If there is no causal logical relationship, it means that the power grid dispatching instruction is not applicable to the current state of the target power grid, which means that the power grid dispatching instruction may be misoperated.

[0060] In some embodiments of the present application, the above-mentioned training labels may also include corresponding misoperation instructions and misoperation types, and the output of the causal logic analysis unit may also include corresponding predicted misoperation instructions and predicted misoperation types. In other words, in some embodiments of the present application, by adopting different training methods, the causal logic analysis unit can not only analyze whether there is a causal logic relationship between the suspected abnormal instruction information and the suspected abnormal state information, but also further predict the possible misoperation instructions and their types. As an example only, in some embodiments of the present application, the matching relationship obtained by the causal logic analysis unit can be expressed as: "R1: 0 / 1; R2: NULL / O1 / O2...; R3: NULL / L1 / L2 / L3..." Among them, "R1:0 / 1" indicates that the matching relationship between the first abnormal feature and the second abnormal feature is mismatch (0) or match (1); "R2:NULL / K1 / K2..." indicates that the predicted possible erroneous operation instruction is NULL (when the matching relationship between the first abnormal feature and the second abnormal feature is match (1), this item output is null value) / first instruction K1 / second instruction K2...; "R3:NULL / L1 / L2 / L3..." indicates that the predicted possible erroneous operation type is NULL (when the matching relationship between the first abnormal feature and the second abnormal feature is match (1), this item output is null value) / first type L1 / second type L2 / third type L3...

[0061] As an alternative implementation, in some embodiments of the present application, the feature extraction unit may also include a coding network, which may be used to encode the suspected abnormal instruction information and the suspected abnormal state information to obtain the first abnormal feature and the second abnormal feature. The causal logic analysis unit may also include a causal reasoning module, which may be used to parse the causal chain between the first abnormal feature and the second abnormal feature based on a pre-built power grid dispatch causal knowledge graph to obtain a matching relationship between the first abnormal feature and the second abnormal feature.

[0062] Specifically, the above-mentioned power grid dispatching causal knowledge graph can be understood as a graph that describes the causal relationship between various instruction information and status information in the power grid dispatching process. It can be constructed based on a large amount of historical data (such as historical dispatching instructions and historical power grid status) and expert knowledge, and can accurately reflect the internal connection between various instruction information and status information in the power grid dispatching process. In an embodiment of the present application, by using the power grid dispatching causal knowledge graph, the causal chain between the first abnormal feature and the second abnormal feature can be efficiently and accurately parsed, thereby deriving a matching relationship between them. The matching relationship can reflect whether there is a causal logical relationship between the above-mentioned suspected abnormal instruction information and the above-mentioned suspected abnormal status information. If not, it indicates that there may be a misoperation situation, and the above-mentioned power grid dispatching instruction needs to be further verified.

[0063] Specifically, in some embodiments of the present application, a feature association matrix can be constructed based on the first abnormal feature and the second abnormal feature obtained by the above-mentioned encoding network processing, and then the feature association matrix can be parsed based on the pre-constructed power grid dispatch causal knowledge graph to obtain the matching relationship between the first abnormal feature and the second abnormal feature. For example, in some embodiments, the elements in the feature association matrix can be mapped to the nodes in the power grid dispatch causal knowledge graph according to the node and edge information in the power grid dispatch causal knowledge graph, and then the shortest path between the first abnormal feature and the second abnormal feature can be found by traversing the edges in the power grid dispatch causal knowledge graph, and the matching relationship between the first abnormal feature and the second abnormal feature can be determined based on the shortest path. Among them, the shortest path represents the causal chain between the first abnormal feature and the second abnormal feature in the power grid dispatch process.

[0064] It can be understood that in practical applications, by determining the matching relationship between the above-mentioned first abnormal feature and the second abnormal feature, it is possible to analyze whether the grid dispatching instructions for the target grid during the grid dispatching process are applicable to its current state, identify possible misoperations, and facilitate timely adoption of corresponding measures to correct or adjust them, thereby greatly improving the safety and reliability of grid dispatching and reducing grid accidents caused by misoperations or abnormal conditions.

[0065] Step S160, obtaining a first error prevention check result based on the matching relationship, and performing power grid dispatching command according to the first error prevention check result. In some embodiments, step S160 may be performed by the first error prevention check result determination module 260 mentioned below.

[0066] In an embodiment of the present application, after obtaining the matching relationship between the first abnormal feature and the second abnormal feature through the above steps, a first error prevention check result can be obtained based on the matching relationship. In some embodiments of the present application, the first error prevention check result is used to reflect whether the power grid dispatching instruction contains an erroneous operation instruction, and the type of erroneous operation corresponding to the power grid dispatching instruction.

[0067] As an example only, in some embodiments of the present application, when the matching relationship between the first abnormal feature and the second abnormal feature obtained through the above process is "1; NULL; NULL", the first anti-error check result can be obtained that there is no erroneous operation in the power grid dispatching instruction; when the matching relationship between the first abnormal feature and the second abnormal feature obtained through the above process is "0; K2; L3", the first anti-error check result can be obtained that there may be an erroneous operation in the power grid dispatching instruction, the possible erroneous operation instruction is K2, and the possible erroneous operation type is L3.

[0068] It should be noted that the above matching relationship and the first anti-error check result are represented by way of example only. In some other embodiments, any other reasonable and feasible method may be used for representation, which is not specifically limited in this specification.

[0069] In an embodiment of the present application, after obtaining the first anti-error verification result in the above manner, it is possible to determine whether the power grid dispatching instruction matches the current state of the target power grid, and perform power grid dispatching command based on this. It should be pointed out that in this manner, not only can misoperation caused by human factors be avoided, but also misoperation caused by information asymmetry (for example, misissuing dispatching instructions due to failure to timely understand the node status) can be avoided, thereby greatly improving the accuracy, safety and intelligence of power grid dispatching, reducing the risk of misoperation, and improving the overall efficiency of power grid operation.

[0070] Figure 4 is a schematic diagram of a module of a dispatching and commanding system with intelligent error prevention function according to some embodiments of this specification. Figure 4 The dispatching and command system 200 shown with intelligent error prevention function can be implemented in the form of software and / or hardware. For example, it can be configured in the form of software and / or hardware to a processing device and / or a terminal device to process the current status information of the target power grid and the power grid dispatching instructions related to the target power grid, thereby performing error prevention verification on the power grid dispatching instructions.

[0071] Reference Figure 4 In some embodiments, the dispatching and command system 200 with intelligent error prevention function may include an acquisition module 210, an abnormal instruction determination module 220, an abnormal state determination module 230, a feature extraction module 240, a feature matching relationship determination module 250 and a first error prevention verification result determination module 260.

[0072] The acquisition module 210 may be used to acquire current status information of a target power grid and power grid dispatching instructions related to the target power grid.

[0073] The abnormal instruction determination module 220 can be used to compare the power grid dispatching instruction with the target reference instruction sequence to obtain suspected abnormal instruction information in the power grid dispatching instruction.

[0074] The abnormal state determination module 230 may be configured to compare the current state information with at least one reference state information that matches the target reference instruction sequence, and obtain suspected abnormal state information in the current state information.

[0075] The feature extraction module 240 can be used to input the suspected abnormal instruction information and the suspected abnormal state information into the feature extraction unit of the pre-trained first anti-mischeck model for processing to extract the first abnormal feature corresponding to the suspected abnormal instruction information and the second abnormal feature corresponding to the suspected abnormal state information.

[0076] The feature matching relationship determination module 250 can be used to input the first abnormal feature and the second abnormal feature into the causal logic analysis unit of the first anti-error verification model for processing to obtain the matching relationship between the first abnormal feature and the second abnormal feature.

[0077] The first error prevention verification result determination module 260 can be used to obtain a first error prevention verification result based on the matching relationship, and perform power grid dispatching and command according to the first error prevention verification result.

[0078] Continue to refer to Figure 4In some embodiments of the present application, the dispatching and command system 200 with intelligent error prevention function may also include a second error prevention verification result determination module 270. The second error prevention verification result determination module 270 can be used to perform a second error prevention verification on the operation instructions input by the dispatching operator for the target power grid according to the repeated voice data, and use the operation instructions that pass the second error prevention verification as the power grid dispatching instructions related to the target power grid.

[0079] For more details about the above modules, please refer to other places in this manual (for example Figure 1~Figure 3 part and its related description) will not be repeated here.

[0080] It should be understood that Figure 4 The dispatching and commanding system 200 with intelligent error prevention function and its modules shown can be implemented in various ways. For example, in some embodiments, the system and its modules can be implemented by hardware, software, or a combination of software and hardware. Among them, the hardware part can be implemented by using dedicated logic; the software part can be stored in a memory and executed by an appropriate instruction execution system, such as a microprocessor or a dedicated design hardware. Those skilled in the art can understand that the above-mentioned system and method can be implemented using computer executable instructions and / or included in a processor control code, for example, such as a carrier medium such as a disk, CD or DVD-ROM, a programmable memory such as a read-only memory (firmware), or a data carrier such as an optical or electronic signal carrier. Such code is provided on the carrier medium such as a disk, CD or DVD-ROM, a programmable memory such as a read-only memory (firmware), or a data carrier such as an optical or electronic signal carrier. The system and its modules of this specification can not only be implemented by hardware circuits such as ultra-large-scale integrated circuits or gate arrays, semiconductors such as logic chips, transistors, or programmable hardware devices such as field programmable gate arrays, programmable logic devices, etc., but can also be implemented by software executed by various types of processors, and can also be implemented by a combination of the above-mentioned hardware circuits and software (for example, firmware).

[0081] It should be noted that the above description of the dispatching and commanding system 200 with intelligent error prevention function is provided for illustrative purposes only and is not intended to limit the scope of this specification. It is understandable that for those skilled in the art, according to the description of this specification, without departing from this principle, the modules can be arbitrarily combined, or a subsystem can be formed to connect with other modules. For example, Figure 4 The acquisition module 210, abnormal instruction determination module 220, abnormal state determination module 230, feature extraction module 240, feature matching relationship determination module 250, first error-proofing result determination module 260, and second error-proofing result determination module 270 described in the embodiment can be different modules in one system, or one module can realize the functions of two or more modules mentioned above. Such variations are all within the protection scope of this specification.

[0082] In summary, the beneficial effects that may be brought about by the embodiments of this specification include but are not limited to: (1) In the dispatching command system and method with intelligent error prevention function provided in some embodiments of this specification, by processing the current state information of the target power grid and the power grid dispatching instructions related to the target power grid, suspected abnormal instruction information and suspected abnormal state information are obtained, and then the suspected abnormal instruction information and the suspected abnormal state information are processed by the first error prevention verification model to analyze the matching relationship between the two. It is possible to analyze whether the power grid dispatching instructions for the target power grid in the power grid dispatching process are suitable for its current state, and identify possible misoperation situations, so as to facilitate timely taking corresponding measures to correct or adjust them, thereby greatly improving the safety and reliability of power grid dispatching and reducing power grid accidents caused by misoperation or abnormal power grid status; (2) In the dispatching command system and method with intelligent error prevention function provided in some embodiments of this specification, suspected abnormal instruction information and suspected abnormal state information are obtained by processing the current state information of the target power grid and the power grid dispatching instructions related to the target power grid, and then the suspected abnormal instruction information and the suspected abnormal state information are processed by the first error prevention verification model to analyze the matching relationship between the two. It is possible to analyze whether the power grid dispatching instructions for the target power grid in the power grid dispatching process are suitable for its current state, and identify possible misoperation situations, so as to facilitate timely taking corresponding measures to correct or adjust them, thereby greatly improving the safety and reliability of power grid dispatching and reducing power grid accidents caused by misoperation or abnormal power grid status. In the dispatching and commanding system and method, the causal logic analysis unit in the first anti-error verification model is used to analyze the causal logic relationship between the suspected abnormal instruction information and the suspected abnormal state information, so as to more accurately determine whether the power grid dispatching instruction matches the current state of the target power grid, thereby effectively avoiding the risk of misoperation caused by instruction errors or abnormal power grid state; (3) In the dispatching and commanding system and method with intelligent anti-error function provided in some embodiments of the present specification, the second anti-error verification model is used to perform semantic comparison processing on the operation instructions input by the dispatching operator for the target power grid and the repeated voice data, so as to verify the consistency of the operation instructions input by the dispatching operator for the target power grid and the operation corresponding to the repeated voice data, which can further improve the accuracy and safety of the dispatching operation, reduce the risk of human misoperation, and at the same time improve the automation level and intelligence level of the entire power grid dispatching and commanding system.

[0083] It should be noted that different embodiments may produce different beneficial effects. In different embodiments, the beneficial effects that may be produced may be any one or a combination of the above, or may be any other beneficial effects that may be obtained.

[0084] The basic concepts have been described above. Obviously, for those skilled in the art, the above detailed disclosure is only for example and does not constitute a limitation of this specification. Although not explicitly stated here, 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, so such modifications, improvements and corrections still belong to the spirit and scope of the exemplary embodiments of this specification.

[0085] At the same time, this specification uses specific words to describe the embodiments of this specification. For example, "one embodiment", "an embodiment", and / or "some embodiments" refer to a certain feature, structure or characteristic related to at least one embodiment of this specification. Therefore, it should be emphasized and noted that "one embodiment" or "an embodiment" or "an alternative embodiment" mentioned twice or more in different positions in this specification does not necessarily refer to the same embodiment. In addition, certain features, structures or characteristics in one or more embodiments of this specification can be appropriately combined.

[0086] In addition, it will be understood by those skilled in the art that various aspects of this specification may be illustrated and described by a number of patentable categories or situations, including any new and useful process, machine, product or combination of substances, or any new and useful improvements thereto. Accordingly, various aspects of this specification may be performed entirely by hardware, entirely by software (including firmware, resident software, microcode, etc.), or by a combination of hardware and software. The above hardware or software may be referred to as "data blocks", "modules", "engines", "units", "components" or "systems". In addition, various aspects of this specification may be represented as a computer product located in one or more computer-readable media, which includes computer-readable program code.

[0087] A computer storage medium may include a propagated data signal containing computer program code, for example, in baseband or as part of a carrier wave. The propagated signal may be in a variety of forms, including electromagnetic, optical, etc., or a suitable combination. A computer storage medium may be any computer-readable medium other than a computer-readable storage medium, which can be connected to an instruction execution system, device or apparatus to communicate, propagate or transmit the program for use. The program code on the computer storage medium may be transmitted via any suitable medium, including radio, cable, fiber optic cable, RF, or similar media, or any combination of the above media.

[0088] The computer program codes required for the operation of each part of this specification can be written in any one or more programming languages, including object-oriented programming languages ​​such as Java, Scala, Smalltalk, Eiffel, JADE, Emerald, C++, C#, VB.NET, Python, etc., conventional procedural programming languages ​​such as C language, Visual Basic, Fortran2003, Perl, COBOL2002, PHP, ABAP, dynamic programming languages ​​such as Python, Ruby and Groovy, or other programming languages, etc. The program code can be run entirely on the user's computer, or run on the user's computer as an independent software package, or run partly on the user's computer and partly on a remote computer, or run entirely on a remote computer or processing device. In the latter case, the remote computer can be connected to the user's computer through any network form, such as a local area network (LAN) or a wide area network (WAN), or connected to an external computer (e.g., via the Internet), or in a cloud computing environment, or used as a service such as software as a service (SaaS).

[0089] In addition, unless explicitly stated in the claims, the order of the processing elements and sequences described in this specification, the use of alphanumeric characters, or the use of other names are not intended to limit the order of the processes and methods of this specification. Although the above disclosure discusses some invention embodiments that are currently considered useful through various examples, it should be understood that such details are only for illustrative purposes, and the attached claims are not limited to the disclosed embodiments. On the contrary, the claims are intended to cover all modifications and equivalent combinations that are consistent with the essence and scope of the embodiments of this specification. For example, although the system components described above can be implemented by hardware devices, they can also be implemented only by software solutions, such as installing the described system on an existing processing device or mobile device.

[0090] Similarly, it should be noted that in order to simplify the description disclosed in this specification and thus help understand one or more embodiments of the invention, in the above description of the embodiments of this specification, multiple features are sometimes combined into one embodiment, figure or description thereof. However, this disclosure method does not mean that the features required by the subject matter of this specification are more than the features mentioned in the claims. In fact, the features of the embodiments are less than all the features of the single embodiment disclosed above.

[0091] In some embodiments, numbers describing the number of components and attributes are used. It should be understood that such numbers used in the description of the embodiments are modified by the modifiers "about", "approximately" or "substantially" in some examples. Unless otherwise specified, "about", "approximately" or "substantially" indicate that the numbers are allowed to vary by ±20%. Accordingly, in some embodiments, the numerical parameters used in the specification and claims are approximate values, which may change according to the required features of individual embodiments. In some embodiments, the numerical parameters should take into account the specified significant digits and adopt the general method of retaining the digits. Although the numerical domains and parameters used to confirm the breadth of the range in some embodiments of this specification are approximate values, in specific embodiments, the setting of such numerical values ​​is as accurate as possible within the feasible range.

[0092] Finally, it should be understood that the embodiments described in this specification are only used to illustrate the principles of the embodiments of this specification. Other variations may also fall within the scope of this specification. Therefore, as an example and not a limitation, alternative configurations of the embodiments of this specification may be considered consistent with the teachings of this specification. Accordingly, the embodiments of this specification are not limited to the embodiments explicitly introduced and described in this specification.

Claims

1. A dispatching and commanding method with intelligent error prevention function, characterized in that: include: Acquire current state information of a target power grid and a power grid dispatch instruction related to the target power grid, wherein the current state information is used to at least reflect current load data, equipment operation status, and topological connection relationship of multiple nodes included in the target power grid; Comparing the power grid dispatching instruction with a target reference instruction sequence to obtain suspected abnormal instruction information in the power grid dispatching instruction; Compare the current state information with at least one reference state information matched with the target reference instruction sequence to obtain suspected abnormal state information in the current state information; Inputting the suspected abnormal instruction information and the suspected abnormal state information into a feature extraction unit of a pre-trained first anti-mischeck model for processing, so as to extract a first abnormal feature corresponding to the suspected abnormal instruction information and a second abnormal feature corresponding to the suspected abnormal state information; Inputting the first abnormal feature and the second abnormal feature into the causal logic analysis unit of the first anti-error verification model for processing to obtain a matching relationship between the first abnormal feature and the second abnormal feature; A first anti-error verification result is obtained based on the matching relationship, and power grid dispatching and command are performed according to the first anti-error verification result; wherein the first anti-error verification result is used to reflect whether the power grid dispatching instruction contains an erroneous operation instruction, and the erroneous operation type corresponding to the power grid dispatching instruction.

2. The method according to claim 1, characterized in that Before acquiring the current state information of the target power grid and the power grid dispatching instruction related to the target power grid, the method further includes: A second error prevention check is performed on the operation instructions input by the dispatching operator for the target power grid according to the repeated voice data, and the operation instructions that pass the second error prevention check are used as power grid dispatching instructions related to the target power grid.

3. The method according to claim 2, characterized in that The second error prevention check on the operation instruction input by the dispatching operator for the target power grid according to the repeated voice data includes: Obtain a first semantic vector according to the repeated voice data, and obtain a second semantic vector according to the operation instruction input by the dispatching operator for the target power grid; The first semantic vector and the second semantic vector are input into a pre-trained second anti-error verification model for processing to obtain a second anti-error verification result, which is used to reflect whether the operation instructions input by the dispatching operator for the target power grid are consistent with the operations corresponding to the repeated voice data.

4. The method according to claim 3, characterized in that The step of inputting the first semantic vector and the second semantic vector into a pre-trained second anti-error verification model for processing to obtain a second anti-error verification result includes: Mapping the first semantic vector to a target space to obtain a first mapping result; Mapping the second semantic vector to the target space to obtain a second mapping result; The distance between the first mapping result and the second mapping result in the target space is calculated, and the second anti-error verification result is obtained based on the distance.

5. The method according to claim 1, characterized in that The comparing the power grid dispatching instruction with the target reference instruction sequence to obtain suspected abnormal instruction information in the power grid dispatching instruction includes: Calculating the similarity between the power grid dispatching instruction and a plurality of preset reference instruction sequences, and taking a preset reference instruction sequence having the highest similarity to the power grid dispatching instruction as a target reference instruction sequence; According to the difference between the power grid dispatching instruction and the target reference instruction sequence, the suspected abnormal instruction information in the power grid dispatching instruction is obtained, wherein the suspected abnormal instruction information includes the suspected abnormal instruction content and the corresponding node information.

6. The method according to claim 5, characterized in that The comparing the current state information with at least one reference state information matched with the target reference instruction sequence to obtain suspected abnormal state information in the current state information includes: Acquire at least one reference state information matching the target reference instruction sequence; Determine the difference between the current state information and each of the reference state information, and fuse the differences between the current state information and all of the reference state information to obtain suspected abnormal state information in the current state information, wherein the suspected abnormal state information includes the suspected abnormal state type and its corresponding node information.

7. The method according to claim 6, characterized in that The feature extraction unit includes an instruction feature extraction submodule and a state feature extraction submodule, and the feature extraction unit of the first pre-trained anti-mischeck model inputs the suspected abnormal instruction information and the suspected abnormal state information for processing to extract a first abnormal feature corresponding to the suspected abnormal instruction information and a second abnormal feature corresponding to the suspected abnormal state information, including: Inputting the suspected abnormal instruction information into the instruction feature extraction submodule for processing to obtain the first abnormal feature; The suspected abnormal state information is input into the state feature extraction submodule for processing to obtain the second abnormal feature, and the state feature extraction submodule and the instruction feature extraction submodule are obtained through collaborative training.

8. The method according to claim 7, characterized in that The causal logic analysis unit includes a machine learning model, which is used to learn the potential causal logic relationship between the first abnormal feature and the second abnormal feature, and perform matching analysis on the first abnormal feature and the second abnormal feature based on the learned causal logic relationship knowledge, so as to obtain the matching relationship between the first abnormal feature and the second abnormal feature; wherein the machine learning model includes any one of a neural network model, a decision tree model, a support vector machine model or a naive Bayes model.

9. The method according to claim 6, characterized in that The feature extraction unit includes a coding network, and the coding network is used to encode the suspected abnormal instruction information and the suspected abnormal state information to obtain the first abnormal feature and the second abnormal feature; The causal logic analysis unit includes a causal reasoning module, which is used to parse the causal chain between the first abnormal feature and the second abnormal feature based on a pre-built power grid dispatch causal knowledge graph to obtain a matching relationship between the first abnormal feature and the second abnormal feature.

10. A dispatching and commanding system with intelligent error prevention function, characterized in that: include: An acquisition module, used to acquire current state information of a target power grid and power grid dispatching instructions related to the target power grid, wherein the current state information is used to at least reflect current load data, equipment operation status, and topological connection relationship of multiple nodes included in the target power grid; An abnormal instruction determination module is used to compare the power grid dispatching instruction with a target reference instruction sequence to obtain suspected abnormal instruction information in the power grid dispatching instruction; an abnormal state determination module, configured to compare the current state information with at least one reference state information matched with the target reference instruction sequence, and obtain suspected abnormal state information in the current state information; a feature extraction module, configured to input the suspected abnormal instruction information and the suspected abnormal state information into a feature extraction unit of a pre-trained first anti-mischeck model for processing, so as to extract a first abnormal feature corresponding to the suspected abnormal instruction information and a second abnormal feature corresponding to the suspected abnormal state information; a feature matching relationship determination module, configured to input the first abnormal feature and the second abnormal feature into a causal logic analysis unit of the first anti-error verification model for processing, and obtain a matching relationship between the first abnormal feature and the second abnormal feature; The first anti-error verification result determination module is used to obtain the first anti-error verification result based on the matching relationship, and perform power grid dispatching and command according to the first anti-error verification result; wherein the first anti-error verification result is used to reflect whether the power grid dispatching instruction contains an erroneous operation instruction, and the erroneous operation type corresponding to the power grid dispatching instruction.

Images