Power distribution network explosion-proof control method, system and equipment based on Internet of Things, and medium

Abstract

The invention discloses a power distribution network explosion-proof control method, system, equipment and medium based on the Internet of Things, and relates to the technical field of power system safety control, and the method comprises the steps: constructing a state vector of each mode, calculating a cooperative tensor between the modes, using the cooperative tensor to define a potential energy relationship, and calculating a potential energy field and a flow field. Integrating the potential energy field and the flow field into a tensor potential flow collaborative graph; and mapping a space grid of the power distribution network by using the space grid, and calculating the strength of the forced flow field by using an exponential decay weighting method to obtain an optimal control path. By constructing the tensor potential flow collaborative graph and the entropy evolution equation and combining unbalanced measurement and dynamic path optimization of multi-modal data, the limitation in abnormal state recognition and dynamic control in the prior art is solved, the safety and reliability of operation of the power distribution network are improved, and through space grid mapping and morphological operation, the reliability of abnormal state recognition and dynamic control is improved. And the geometric continuity description capability of the potential risk area is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of power system safety control, and in particular to an Internet of Things-based explosion-proof control method, system, equipment, and medium for a distribution network. Background Art

[0002] With the deep integration of smart grid and Internet of Things technologies, the intelligence level of distribution networks continues to improve. Traditional distribution systems mainly rely on static monitoring and rule-based response, and lack the ability to actively identify and dynamically interact with sudden dangerous situations. Especially when facing potential explosion risk scenarios such as combustible gas leaks, arc faults, and abnormal temperature rise, problems such as delayed response, single control path, and rough regional identification become increasingly prominent.

[0003] Existing explosion-proof control methods still have shortcomings. Traditional methods focus on identifying local parameter anomalies, ignore the co-evolution mechanism between multimodal data, and fail to build energy or information flow models, resulting in delayed response and unclear control paths. Summary of the Invention

[0004] In view of the above existing problems, the present invention is proposed.

[0005] Therefore, the present invention provides a distribution network explosion-proof control method, system, equipment and medium based on the Internet of Things, which solves the problems that traditional methods focus on identifying local parameter anomalies, ignore the collaborative evolution mechanism between multimodal data, fail to build energy or information flow models, and lead to delayed response and unclear control paths.

[0006] In order to solve the above technical problems, the present invention provides the following technical solutions: In a first aspect, the present invention provides a method for explosion-proof control of a distribution network based on the Internet of Things, which comprises the following steps: Collect multimodal data of the distribution network and the third-order time structure function of the technical mode, compose a multidimensional vector, construct an unbalanced measure tensor field, perform binary encoding on the unbalanced measure tensor field, generate binary symbols, construct the state vector of each mode, calculate the synergy tensor between the modes, use the synergy tensor to define the potential energy relationship, calculate the potential energy field and flow field, and integrate the potential energy field and flow field into a tensor potential flow synergy graph; Use the spatial grid to map the spatial grid of the distribution network, define it as an abnormal area mask, extract the potential energy field corresponding to the abnormal point in the abnormal point set, calculate the entropy pressure difference between the local discrete entropies, calculate the cooperative tensor of the abnormal point set, define the entropy evolution equation, use mapping propagation to iteratively update the entropy vector, use linear interpolation to calculate the entropy value of the entropy vector grid coordinates, use the exponential decay weighted method to calculate the forced flow field intensity, use the Dijkstra algorithm to extract the path from the forced flow field intensity, and obtain the optimal control path; Monitor the optimal control path, generate and execute control instructions, and build a visual interface to display control instructions.

[0007] As a preferred solution of the explosion-proof control method for distribution network based on the Internet of Things of the present invention, the method of collecting multimodal data of the distribution network and constructing the state vector of each mode includes: Use smart sensors to collect multimodal data from the distribution network from IoT devices and perform denoising and normalization processing; The multimodal data includes gas concentration, current, temperature rise, humidity, arc frequency and device location data; Calculate the third-order time structure function for each modal data to obtain the mutation index of each mode, organize the mutation index of each mode into a multidimensional vector, use the Euclidean distance formula to calculate the Euclidean distance between modes, construct an unbalanced measure tensor field, perform histogram statistics on the unbalanced measure tensor field, calculate the entropy density, and use the percentile method to set the binarization threshold based on the entropy density. Binarize the unbalanced measure tensor field and generate binary symbols. The symbol mutation rate of the binary symbols is calculated using a two-dimensional sparse difference operator to construct the state vector of each mode.

[0008] As a preferred solution of the explosion-proof control method for distribution network based on the Internet of Things of the present invention, the integration of potential energy field and flow field into a tensor potential flow synergy graph includes: Normalize the state vector, calculate the synergy tensor between the modes, use the synergy tensor to define the potential energy relationship, and calculate the potential energy field of each mode; Based on the sign mutation rate, the flow intensity is defined, the direction vector is calculated, and the flow field is constructed; The potential energy field and flow field are integrated into a tensor potential-flow synergy graph.

[0009] As a preferred solution of the explosion-proof control method for distribution network based on the Internet of Things of the present invention, the spatial grid of the distribution network mapped using the spatial grid is defined as an abnormal area mask, including: The recognition threshold is set using statistical analysis methods. The potential energy field of mode i is extracted from the tensor potential flow synergy graph. The potential energy fields greater than the recognition threshold are screened to generate a binary anomaly point map. The binary anomaly point map is mapped to the spatial grid of the distribution network using spatial grid mapping. A morphological dilation operation is performed on the mapped binary anomaly point map to obtain an expanded anomaly point map. A morphological erosion operation is performed on the expanded anomaly point map to generate an eroded anomaly point map, which is defined as an anomaly area mask.

[0010] As a preferred solution of the explosion-proof control method for distribution network based on the Internet of Things of the present invention, wherein: the Dijkstra algorithm is used to extract the path from the forced flow field intensity to obtain the optimal control path, including: Generate an abnormal point set based on the abnormal points in the abnormal area mask, extract the potential energy field corresponding to the abnormal points in the abnormal point set, divide the frequency distribution of the potential energy field using maximum entropy frequency partitioning, calculate the local discrete entropy of mode i, calculate the entropy pressure difference between the local discrete entropies, calculate the direction vector of the abnormal point set, and combine the entropy pressure difference and the direction vector of the abnormal point set to calculate the entropy pressure difference between the modes; Calculate the collaborative tensor of the outlier set, define the entropy evolution equation based on the collaborative tensor of the outlier set, construct the entropy vector and the predicted entropy vector, and use mapping propagation to iteratively update the entropy vector; Use the fixed threshold method to set the convergence threshold, calculate the L2 norm between the information entropy vector and the predicted information entropy vector, stop the iteration when the L2 norm is less than the convergence threshold, and obtain the final entropy vector; The final entropy vector is mapped to the spatial grid of the distribution network using spatial grid mapping to obtain the grid coordinates of the entropy vector; Linear interpolation is used to calculate the entropy value of the entropy vector grid coordinates, the finite difference method is used to calculate the entropy value gradient of the grid coordinates, the exponential decay weighted method is used to calculate the forced flow field intensity, and the Dijkstra algorithm is used to extract the path from the forced flow field intensity to obtain the optimal control path.

[0011] As a preferred solution of the explosion-proof control method for distribution network based on the Internet of Things of the present invention, wherein: the monitoring of the optimal control path, generating and executing control instructions, includes: For each edge of the optimal control path, the forced flow field intensity is accumulated to obtain the path cumulative intensity. The judgment threshold is set using the empirical rule. The path with a cumulative intensity greater than the judgment threshold is determined as the priority path. Otherwise, it is determined as a normal path and monitoring continues. The path mapping method is used to map the priority path to the corresponding control device, generate control instructions, and use the Internet of Things communication protocol to send the control instructions to the smart sensor for execution.

[0012] As a preferred solution of the explosion-proof control method for distribution network based on the Internet of Things of the present invention, the step of constructing a visual interface to display control instructions includes: Use the front-end framework React.js to build a visual interface and visualize the control instructions; Users who have passed real-name verification are allowed to access the information.

[0013] In a second aspect, the present invention provides an explosion-proof control system for a distribution network based on the Internet of Things, comprising: The collection and coordination module is used to collect multimodal data of the distribution network and the third-order time structure function of the technical mode, compose a multidimensional vector, construct an unbalanced measure tensor field, perform binary encoding on the unbalanced measure tensor field, generate binary symbols, construct the state vector of each mode, calculate the coordination tensor between the modes, use the coordination tensor to define the potential energy relationship, calculate the potential energy field and flow field, and integrate the potential energy field and flow field into a tensor potential flow coordination diagram; The mask control module is used to map the spatial grid of the distribution network using a spatial grid, define the abnormal area mask, extract the potential energy field corresponding to the abnormal point in the abnormal point set, calculate the entropy pressure difference between the local discrete entropies, calculate the cooperative tensor of the abnormal point set, define the entropy evolution equation, use mapping propagation to iteratively update the entropy vector, use linear interpolation to calculate the entropy value of the entropy vector grid coordinates, use the exponential decay weighted method to calculate the forced flow field intensity, and use the Dijkstra algorithm to extract the path from the forced flow field intensity to obtain the optimal control path; The monitoring and visualization module is used to monitor the optimal control path, generate and execute control instructions, and build a visual interface to display control instructions.

[0014] In a third aspect, the present invention provides a computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: when the computer program is executed by the processor, any step of the explosion-proof control method for distribution network based on the Internet of Things as described in the first aspect of the present invention is implemented.

[0015] In a fourth aspect, the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein: when the computer program is executed by a processor, any step of the explosion-proof control method for distribution network based on the Internet of Things as described in the first aspect of the present invention is implemented.

[0016] The beneficial effects of the present invention are as follows: the present invention solves the limitations of existing technologies in abnormal state identification and dynamic control by constructing a tensor potential flow collaborative graph and an entropy evolution equation, combining the non-equilibrium measurement and dynamic path optimization of multimodal data, and improves the safety and reliability of distribution network operation. Through spatial grid mapping and morphological operations, the ability to characterize the geometric continuity of potential risk areas is improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0018] Figure 1 This is a flow chart of the explosion-proof control method for distribution network based on the Internet of Things in Example 1.

[0019] Figure 2 Schematic diagram of the explosion-proof control system for distribution network based on the Internet of Things in Example 1. DETAILED DESCRIPTION

[0020] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0021] In the following description, many specific details are set forth to facilitate a full understanding of the present invention. However, the present invention may also be implemented in other ways different from those described herein. Those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0022] Secondly, the term "one embodiment" or "embodiment" herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in various places throughout this specification does not necessarily refer to the same embodiment, nor does it refer to a separate or selective embodiment that is mutually exclusive of other embodiments.

[0023] Example 1, with reference to Figure 1 , which is the first embodiment of the present invention, provides a distribution network explosion-proof control method based on the Internet of Things, comprising the following steps: S1. Collect multimodal data of the distribution network and the third-order time structure function of the technical mode, compose a multidimensional vector, construct an unbalanced measure tensor field, perform binary encoding on the unbalanced measure tensor field, generate binary symbols, construct the state vector of each mode, calculate the synergy tensor between the modes, use the synergy tensor to define the potential energy relationship, calculate the potential energy field and flow field, and integrate the potential energy field and flow field into a tensor potential flow synergy graph; Specifically, we collect multimodal data of the distribution network and construct the state vector of each mode, including: Use smart sensors to collect multimodal data from the distribution network from IoT devices and perform denoising and normalization processing; The intelligent sensors include gas concentration, current, temperature and humidity, arc and GPS sensors; The multimodal data includes gas concentration, current, temperature rise, humidity, arc frequency and device location data; The third-order time structure function is calculated for each modal data to obtain the mutation index of each mode. The formula is: , in For multimodal data The third-order time structure function represents the sudden change index of each mode, a scalar value, is the time interval, which is set using the first decay point of the autocorrelation function, and N is the collection window length, which is set using the energy conservation sliding window method. It is multimodal data; The mutation index of each mode is composed into a multidimensional vector, and the formula is: , in The mutation index of mode i forms a multidimensional vector, which contains the mutation index of 5 data and has a dimension of 5. 、 、 、 and They are the mutation indexes of gas concentration, current, temperature rise, humidity and arc frequency respectively; The Euclidean distance formula is used to calculate the Euclidean distance between modes and construct the unbalanced measure tensor field. The formula is: , , in For modal i, at device location The non-equilibrium state measure tensor value of is the device position of modal i, is a multidimensional vector of the mutation index of mode j, is the Euclidean distance between modes i and j, is the Gaussian kernel width, set using maximum likelihood estimation to control the neighborhood influence range, is the neighborhood set of mode i, using the k-neighbor search setting; Perform histogram statistics on the non-equilibrium measure tensor field and calculate the entropy density. The formula is: , in is the entropy density, is the probability of the ath interval of the histogram, using histogram statistics; Based on the entropy density, the percentile method is used to set the binarization threshold, and the non-equilibrium measure tensor field is binarized to generate binary symbols. The formula is: , in It is a binary symbol, representing a coding diagram, 0 represents a normal state, 1 represents an abnormal state, and a potential explosion risk. is the binarization threshold; The symbol mutation rate of the binary symbol is calculated using the two-dimensional sparse difference operator. The formula is: ,in is the symbol mutation rate, reflecting the device position The local sign change of Construct the state vector of each mode using the formula: , in is the state vector of mode i, is the non-equilibrium state measure tensor value of mode i, is the binary symbol of mode i, is the sign mutation rate of mode i.

[0024] By introducing a mutation index, a scalar metric is constructed to uniformly measure the evolution rate of each modal anomaly, making different physical quantities comparable. Compared to first- or second-order functions, the introduction of a third-order time structure function can more effectively capture high-order dynamic behavior characteristics and is more sensitive to weak nonlinear changes and mutational behavior in the system. The energy conservation sliding window method is used to set the time window, enhancing the adaptability of the time scale and accommodating feature expression at multiple time and frequency levels. The mutation indexes of each modal are combined into a multidimensional vector, achieving a multimodal embedded representation of the device state space. The non-equilibrium measure tensor field constructed based on the Gaussian kernel function exhibits continuity and smoothness. A binary symbol generation mechanism, by fusing entropy density with the percentile method, enhances the robustness of threshold setting. The binary symbol reflects the demarcation between "normal" and "abnormal" at the microscopic level, forming an early identification mechanism for potential high-risk situations. The symbol mutation rate constructed using a two-dimensional sparse difference operator lays a high-resolution mutation foundation for the subsequent calculation of flow intensity and direction fields.

[0025] Furthermore, the potential energy field and flow field are integrated into a tensor potential flow synergy diagram, including: The state vector is normalized, and the cooperation tensor between the modes is calculated to measure the state similarity and incorporate it into the electrical distance. The formula is: , in is the synergy tensor between modes i and j, and are the state vectors of mode i and mode j respectively. The normalized value of the dimension, is the mean of the Euclidean distance; The potential energy relationship is defined using the cooperative tensor, and nonlinear transformation is introduced to enhance the discrimination of outliers. The formula is: , in is the potential energy relationship between modes i and j. The larger the value, the greater the state difference. is a very small constant to prevent it from being zero; Calculate the potential energy field of each mode using the formula: , in is the potential energy field of mode i, reflecting the local state heterogeneity; Based on the symbolic mutation rate, flow intensity, fusion cooperativity and structural mutation are defined as follows: , in is the flow intensity between modes i and j, is the sign mutation rate of mode j; Calculate the direction vector and construct the flow field. The formula is: , , , in is the direction vector between modes i and j, and are the spatial coordinate vectors of mode i and mode j respectively, is the device position of modality j, is the flow field vector of mode i at time t, which contains the direction and intensity of information / energy flow, is the time decay constant, set using a fixed time scale, used to control the temporal dynamics of the flow field. is the time difference, which represents the difference between the current time and the time when the multimodal data is collected; The potential energy field and flow field are integrated into a tensor potential flow synergy graph, and the formula is: , in For modal i, at device location The tensor potential flow synergy graph of , is the flow field direction angle, calculated using atan2.

[0026] The synergy tensor reflects the synergy between different modal states, effectively integrates state similarity and electrical distance constraints, and captures system coupling. Nonlinear transformation enhances the ability of potential energy relationships to distinguish anomalies, making potential "edge states" more recognizable. The potential energy field reveals the local heterogeneity of the state, which helps to determine whether the equipment is in a stable / unstable area. The flow field constructs the energy flow path and information diffusion trend through the direction vector, reflecting the spatial propagation properties of local disturbances. The introduction of the time decay constant considers the impact of historical information on the current state and realizes dynamic modeling. The modal flow intensity combines synergy and sign mutation rate to improve the response sensitivity of the flow field to structural mutations. Each modal node is embedded with multi-source semantic features, and the flow direction and amplitude are mapped to provide high-quality prior topology for the graph neural network. Atan2 is used to calculate the angle direction to avoid angle ambiguity and achieve stable encoding of direction information.

[0027] S2. Use the spatial grid to map the spatial grid of the distribution network, define it as an abnormal area mask, extract the potential energy field corresponding to the abnormal point in the abnormal point set, calculate the entropy pressure difference between the local discrete entropies, calculate the cooperative tensor of the abnormal point set, define the entropy evolution equation, use mapping propagation to iteratively update the entropy vector, use linear interpolation to calculate the entropy value of the entropy vector grid coordinates, use the exponential decay weighted method to calculate the forced flow field intensity, use the Dijkstra algorithm to extract the path from the forced flow field intensity, and obtain the optimal control path; Specifically, the spatial grid of the distribution network is mapped using a spatial grid, which is defined as an abnormal area mask, including: The statistical analysis method is used to set the recognition threshold, extract the potential energy field of mode i from the tensor potential flow synergy graph, screen the potential energy fields greater than the recognition threshold, and generate a binary outlier point map. The formula is: , in is the binary outlier point map, is the recognition threshold; Use spatial grid mapping to map the binary anomaly point map to the spatial grid of the distribution network. Perform morphological expansion operation on the mapped binary anomaly point map to obtain the expanded anomaly point map and expand the mutation point area. The formula is: , , , in is the abnormal point map after expansion, is the binary outlier map after mapping, u and v are relative to the grid points The offset coordinates, r is the radius of the structural element, and the fixed radius setting is used for the grid resolution. is a fixed coefficient, is the grid resolution of the outlier map, is the morphological dilation operation; Perform morphological corrosion operation on the expanded abnormal point map to eliminate noise and smooth the boundaries, generate the eroded abnormal point map, and form a continuous area. The formula is: , in is the abnormal point map after corrosion, It is a morphological corrosion operation; Defined as the abnormal area mask, the formula is: , in is the abnormal area mask, 1 represents a high-risk area and 0 represents a safe area.

[0028] By introducing spatial grid mapping, a spatial resolution model of the distribution network is established, so that unstructured anomalies can be regularly encoded in the spatial domain, providing a consistent reference framework for subsequent calculations. Statistical analysis is used to determine the recognition threshold, and the tensor potential field is converted into a binary anomaly map, which effectively quantifies the anomaly distribution and improves the accuracy of mutation point identification. The morphological dilation operation introduces the structural element expansion capability based on the original anomaly points, which can capture potential sub-anomalies in the local area and enhance the spatial coherence of the anomaly area. The subsequent corrosion operation further filters out isolated noise, optimizes the anomaly boundary, and makes the edge of the mask area smoother and more continuous, which is beneficial for graphics processing and flow field modeling.

[0029] Furthermore, the Dijkstra algorithm is used to extract the path from the forced flow field intensity to obtain the optimal control path, including: Generate an outlier set based on the outlier points in the outlier area mask, extract the potential energy field corresponding to the outlier points in the outlier set, divide the frequency distribution of the potential energy field using the maximum entropy frequency partitioning, and calculate the local discrete entropy of mode i. The formula is: , , in For mode i, at time t, the frequency at which the potential energy field falls into the kth interval, k is the number of intervals in the frequency distribution, is the number of potential energy fields falling into the kth interval, is the local discrete entropy of mode i at time t; Calculate the entropy pressure difference between local discrete entropies using the formula: , in is the entropy pressure difference between modes i and j, is the local discrete entropy of mode j at time t; Calculate the direction vector of the outlier set using the following formula: , , in is the direction vector between modes i and j in the outlier set, and are the spatial coordinate vectors of mode i and mode j in the outlier set, and are the device locations of mode i and mode j in the abnormal point set respectively; Combining the entropy pressure difference and the direction vector of the abnormal point set, the entropy pressure difference between modes is calculated. The formula is: , in is the entropy pressure difference between modes i and j; Calculate the cooperative tensor of the outlier set, the formula is: , in is the collaborative tensor of mode i in the set of outliers, is the neighborhood set of mode i in the set of outliers, using the k-neighbor search setting, is the tensor outer product; Based on the cooperative tensor of the outlier set, the entropy evolution equation is defined as follows: , , in is the entropy value of mode i at the next time step, is the time step, is the propagation coefficient, set using grid search, is the divergence of the cooperative tensor propagation term, calculated using the finite difference method, is the kth component of the device position in the abnormal point set, is the minimum constant of the entropy evolution equation, which is prevented from being zero; Construct the entropy vector and the predicted entropy vector, and use mapping propagation to iteratively update the entropy vector. The formula is: , , in is the entropy vector, M is the number of multimodal states, is the predicted entropy vector, Q is the transposition operation; Use the fixed threshold method to set the convergence threshold, calculate the L2 norm between the information entropy vector and the predicted information entropy vector, stop the iteration when the L2 norm is less than the convergence threshold, and obtain the final entropy vector; The final entropy vector is mapped to the spatial grid of the distribution network using spatial grid mapping to obtain the grid coordinates of the entropy vector; The entropy value of the entropy vector grid coordinate is calculated using linear interpolation, the formula is: , ,in is the entropy value of the entropy vector grid coordinate at time t, is the interpolation weight, is the grid coordinate of the entropy vector, is the grid resolution of the entropy vector, Modal , the final entropy vector at time t; The finite difference method is used to calculate the entropy gradient of the grid coordinates. The formula is: , in is the entropy gradient at time t, and The entropy values are and Directional partial derivatives; The exponential decay weighting method is used to calculate the forced flow field intensity, and the formula is: , in is the forced flow field intensity between modes i and j, is the Euclidean distance between modes i and j in the outlier set, is the decay constant, set using the average spacing method; The Dijkstra algorithm is used to extract the path from the forced flow field intensity to obtain the optimal control path. The formula is: , in is the optimal control path, is a path set, which represents all possible propagation paths starting from the mutation point in the distribution network topology.

[0030] The concept of local discrete entropy is introduced, and based on the frequency statistics of potential energy field, the complexity of mutation information in different modes is characterized, so that the model can be sensitive to the "unbalanced information distribution" of the system. The use of maximum entropy frequency partitioning improves the robustness of entropy estimation and avoids estimation bias caused by the sparse number of abnormal points. The construction of entropy pressure difference integrates the information complexity gradient between modes and provides a new mechanism for measuring the "driving force" of mutation. The entropy evolution equation models the information diffusion process through collaborative tensor modeling, characterizes the spatiotemporal propagation behavior of entropy values, and simulates the evolution mechanism of local anomalies in the distribution network to global transmission. The finite difference method is used to calculate the divergence of tensor terms, so that the model has good numerical stability and interpretability. The collaborative tensor captures the coupling synergy between modes and is a tensor expression of the high-order interaction relationship of multimodal mutation information, breaking through the traditional model based on A low-dimensional statistical method for correlation coefficients and an exponential decay weighted method are used to introduce a spatial distance modulation mechanism, so that the influence between long-distance modes decays naturally according to physical laws, thereby improving the physical rationality of the model. The forced flow field intensity is constructed as a flow-based expression of the potential-entropy mixed field, providing an energy density measurement for path extraction, showing a guiding effect similar to that of the potential flow pressure field. The predicted entropy vector and L2 norm control strategy are introduced to construct a convergence detection mechanism to ensure the controllable and stable algorithm. Linear interpolation is used to accurately restore the entropy value to the grid coordinates, and an entropy-space mapping system is constructed to support the flow field gradient calculation. The gradient field is combined with the forced flow intensity as the path graph weight, so that the path extraction takes into account both structural directionality and state strength. Through the Dijkstra algorithm, the control path from the mutation source to the optimal terminal is extracted to achieve the "shortest cost path" characterization of anomaly propagation.

[0031] S3. Monitor the optimal control path, generate and execute control instructions, and build a visual interface to display the control instructions; Specifically, the optimal control path is monitored, control instructions are generated and executed, including: For each edge of the optimal control path, the forced flow field intensity is accumulated to obtain the path cumulative intensity, which is: , in is the cumulative strength of the path; Use empirical rules to set the judgment threshold. If the cumulative strength of the path is greater than the judgment threshold, it will be determined as the priority path. Otherwise, it will be determined as a normal path and continue to be monitored. Use the path mapping method to map the priority path to the corresponding control device, such as circuit breaker, relay, gas release valve, generate control instructions, and use the Internet of Things communication protocol to send the control instructions to the smart sensor for execution.

[0032] Using forced flow field strength as a state measurement indicator with clear physical meaning can reflect the carrying strength of the path in the propagation of abnormal information in real time. The calculation method of the path cumulative strength follows the edge weight accumulation logic on the graph structure, which constitutes a quantitative assessment of the overall risk load capacity of the path. Through the path mapping method, the abstract path information is mapped to the actual physical equipment (circuit breakers, relays, etc.), and a full-chain closed loop from state identification to physical control is constructed. The control instructions are transmitted using the Internet of Things communication protocol, which enhances the modularity of the system and the interoperability between heterogeneous devices.

[0033] Furthermore, a visual interface is constructed to display control instructions, including: Use the front-end framework React.js to build a visual interface and visualize the control instructions; Users who have passed real-name verification are allowed to access the information.

[0034] Visual display improves system interactivity and explainability, helping operators quickly understand system status and master control logic. The React.js front-end framework is highly responsive and capable of component reuse, making it suitable for building monitoring interfaces with high real-time requirements. The introduction of a real-name verification mechanism ensures the security of system information and access compliance, preventing unauthorized access and misoperation. Example

[0035] Reference Figure 2 The second embodiment of the present invention is an explosion-proof control system for a distribution network based on the Internet of Things, comprising: The collection and coordination module is used to collect multimodal data of the distribution network and the third-order time structure function of the technical mode, compose a multidimensional vector, construct an unbalanced measure tensor field, perform binary encoding on the unbalanced measure tensor field, generate binary symbols, construct the state vector of each mode, calculate the coordination tensor between the modes, use the coordination tensor to define the potential energy relationship, calculate the potential energy field and flow field, and integrate the potential energy field and flow field into a tensor potential flow coordination diagram; The mask control module is used to map the spatial grid of the distribution network using a spatial grid, define the abnormal area mask, extract the potential energy field corresponding to the abnormal point in the abnormal point set, calculate the entropy pressure difference between the local discrete entropies, calculate the cooperative tensor of the abnormal point set, define the entropy evolution equation, use mapping propagation to iteratively update the entropy vector, use linear interpolation to calculate the entropy value of the entropy vector grid coordinates, use the exponential decay weighted method to calculate the forced flow field intensity, and use the Dijkstra algorithm to extract the path from the forced flow field intensity to obtain the optimal control path; The monitoring and visualization module is used to monitor the optimal control path, generate and execute control instructions, and build a visual interface to display control instructions.

[0036] This embodiment also provides a computer device, which is applicable to the distribution network explosion-proof control method based on the Internet of Things, including: a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions to implement the distribution network explosion-proof control method based on the Internet of Things proposed in the above embodiment.

[0037] The computer device may be a terminal, comprising a processor, memory, a communication interface, a display, and an input device connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores an operating system and computer programs. The internal memory provides an environment for the operating system and computer programs stored in the non-volatile storage media. The communication interface of the computer device is used to communicate with external terminals via wired or wireless communication. Wireless communication may be achieved via Wi-Fi, a carrier network, NFC (near-field communication), or other technologies. The display of the computer device may be a liquid crystal display or an electronic ink display. The input device may be a touchscreen overlay on the display, buttons, a trackball, or a touchpad on the computer device housing, or an external keyboard, touchpad, or mouse.

[0038] This embodiment also provides a storage medium having a computer program stored thereon, which, when executed by a processor, implements the explosion-proof control method for a distribution network based on the Internet of Things proposed in the above embodiment; the storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic disk or optical disk.

[0039] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A method for explosion-proof control of a distribution network based on the Internet of Things, characterized by: The steps include: Collect multimodal data of the distribution network and the third-order time structure function of the technical mode, compose a multidimensional vector, construct an unbalanced measure tensor field, perform binary encoding on the unbalanced measure tensor field, generate binary symbols, construct the state vector of each mode, calculate the synergy tensor between the modes, use the synergy tensor to define the potential energy relationship, calculate the potential energy field and flow field, and integrate the potential energy field and flow field into a tensor potential flow synergy graph; Use the spatial grid to map the spatial grid of the distribution network, define it as an abnormal area mask, extract the potential energy field corresponding to the abnormal point in the abnormal point set, calculate the entropy pressure difference between the local discrete entropies, calculate the cooperative tensor of the abnormal point set, define the entropy evolution equation, use mapping propagation to iteratively update the entropy vector, use linear interpolation to calculate the entropy value of the entropy vector grid coordinates, use the exponential decay weighted method to calculate the forced flow field intensity, use the Dijkstra algorithm to extract the path from the forced flow field intensity, and obtain the optimal control path; Monitor the optimal control path, generate and execute control instructions, and build a visual interface to display control instructions.

2. The explosion-proof control method for distribution network based on the Internet of Things according to claim 1, characterized in that: The collecting of multimodal data of the distribution network and constructing a state vector of each mode include: Use smart sensors to collect multimodal data from the distribution network from IoT devices and perform denoising and normalization processing; The multimodal data includes gas concentration, current, temperature rise, humidity, arc frequency and device location data; Calculate the third-order time structure function for each modal data to obtain the mutation index of each mode, organize the mutation index of each mode into a multidimensional vector, use the Euclidean distance formula to calculate the Euclidean distance between modes, construct an unbalanced measure tensor field, perform histogram statistics on the unbalanced measure tensor field, calculate the entropy density, and use the percentile method to set the binarization threshold based on the entropy density. Binarize the unbalanced measure tensor field and generate binary symbols. The symbol mutation rate of the binary symbols is calculated using a two-dimensional sparse difference operator to construct the state vector of each mode.

3. The explosion-proof control method for distribution network based on the Internet of Things according to claim 2, characterized in that: The integration of the potential energy field and the flow field into a tensor potential flow synergy graph includes: Normalize the state vector, calculate the synergy tensor between the modes, use the synergy tensor to define the potential energy relationship, and calculate the potential energy field of each mode; Based on the sign mutation rate, the flow intensity is defined, the direction vector is calculated, and the flow field is constructed; The potential energy field and flow field are integrated into a tensor potential-flow synergy graph.

4. The explosion-proof control method for distribution network based on the Internet of Things according to claim 3, characterized in that: The spatial grid of the power distribution network is mapped using the spatial grid, which is defined as an abnormal area mask, including: The recognition threshold is set using statistical analysis methods. The potential energy field of mode i is extracted from the tensor potential flow synergy graph. The potential energy fields greater than the recognition threshold are screened to generate a binary anomaly point map. The binary anomaly point map is mapped to the spatial grid of the distribution network using spatial grid mapping. A morphological dilation operation is performed on the mapped binary anomaly point map to obtain an expanded anomaly point map. A morphological erosion operation is performed on the expanded anomaly point map to generate an eroded anomaly point map, which is defined as an anomaly area mask.

5. The explosion-proof control method for distribution network based on the Internet of Things according to claim 4, characterized in that: The method of using the Dijkstra algorithm to extract a path from the forced flow field intensity to obtain the optimal control path includes: Generate an abnormal point set based on the abnormal points in the abnormal area mask, extract the potential energy field corresponding to the abnormal points in the abnormal point set, divide the frequency distribution of the potential energy field using maximum entropy frequency partitioning, calculate the local discrete entropy of mode i, calculate the entropy pressure difference between the local discrete entropies, calculate the direction vector of the abnormal point set, and combine the entropy pressure difference and the direction vector of the abnormal point set to calculate the entropy pressure difference between the modes; Calculate the collaborative tensor of the outlier set, define the entropy evolution equation based on the collaborative tensor of the outlier set, construct the entropy vector and the predicted entropy vector, and use mapping propagation to iteratively update the entropy vector; Use the fixed threshold method to set the convergence threshold, calculate the L2 norm between the information entropy vector and the predicted information entropy vector, stop the iteration when the L2 norm is less than the convergence threshold, and obtain the final entropy vector; The final entropy vector is mapped to the spatial grid of the distribution network using spatial grid mapping to obtain the grid coordinates of the entropy vector; Linear interpolation is used to calculate the entropy value of the entropy vector grid coordinates, the finite difference method is used to calculate the entropy value gradient of the grid coordinates, the exponential decay weighted method is used to calculate the forced flow field intensity, and the Dijkstra algorithm is used to extract the path from the forced flow field intensity to obtain the optimal control path.

6. The explosion-proof control method for distribution network based on the Internet of Things according to claim 5, characterized in that: The monitoring of the optimal control path, generating and executing control instructions includes: For each edge of the optimal control path, the forced flow field intensity is accumulated to obtain the path cumulative intensity. The judgment threshold is set using the empirical rule. The path with a cumulative intensity greater than the judgment threshold is determined as the priority path. Otherwise, it is determined as a normal path and monitoring continues. The path mapping method is used to map the priority path to the corresponding control device, generate control instructions, and use the Internet of Things communication protocol to send the control instructions to the smart sensor for execution.

7. The explosion-proof control method for distribution network based on the Internet of Things according to claim 6, characterized in that: The construction of a visual interface to display control instructions includes: Use the front-end framework React.js to build a visual interface and visualize the control instructions; Users who have passed real-name verification are allowed to access the information.

8. An explosion-proof control system for a distribution network based on the Internet of Things, used to implement the method according to any one of claims 1 to 7, characterized in that: include: The collection and coordination module is used to collect multimodal data of the distribution network and the third-order time structure function of the technical mode, compose a multidimensional vector, construct an unbalanced measure tensor field, perform binary encoding on the unbalanced measure tensor field, generate binary symbols, construct the state vector of each mode, calculate the coordination tensor between the modes, use the coordination tensor to define the potential energy relationship, calculate the potential energy field and flow field, and integrate the potential energy field and flow field into a tensor potential flow coordination diagram; The mask control module is used to map the spatial grid of the distribution network using a spatial grid, define the abnormal area mask, extract the potential energy field corresponding to the abnormal point in the abnormal point set, calculate the entropy pressure difference between the local discrete entropies, calculate the cooperative tensor of the abnormal point set, define the entropy evolution equation, use mapping propagation to iteratively update the entropy vector, use linear interpolation to calculate the entropy value of the entropy vector grid coordinates, use the exponential decay weighted method to calculate the forced flow field intensity, and use the Dijkstra algorithm to extract the path from the forced flow field intensity to obtain the optimal control path; The monitoring and visualization module is used to monitor the optimal control path, generate and execute control instructions, and build a visual interface to display control instructions.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the explosion-proof control method for distribution network based on the Internet of Things according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the explosion-proof control method for a distribution network based on the Internet of Things according to any one of claims 1 to 7 are implemented.

Images