Power distribution room multi-device integrated control system and method based on industrial Internet of Things architecture

Abstract

The invention relates to the technical field of power distribution room equipment integrated control, and discloses a power distribution room multi-equipment integrated control system and method based on an industrial Internet of Things architecture, and the method comprises the steps: collecting and normalizing the operation data of power distribution equipment, dividing the operation data into slow change operation data and quick change operation data, using an improved LSTM network and a sliding difference mode combined with an attention mechanism to extract time sequence features of the two; and sequentially receiving the equipment characteristic information of the power distribution equipment by adopting an improved weighted polling mode, and performing mapping processing on the equipment characteristic information based on a fuzzy logic rule to generate a control parameter. According to the invention, an edge acquisition module, a protocol conversion module and an edge gateway are deployed to construct an industrial Internet of Things architecture, classified acquisition of fast-varying and slow-varying operation data is realized, time sequence features are extracted in combination with improved LSTM and in combination with a sliding difference mode of an attention mechanism, control parameters are mapped through a fuzzy logic rule, and the operation of the industrial Internet of Things is realized. And unified scheduling and intelligent control of multiple devices are realized.

Description

Technical Field

[0001] The present invention relates to an Internet of Things (IoT) control system for industrial big data, and in particular to the technical field of integrated control of equipment in power distribution rooms. Specifically, it relates to an integrated control system and method for multiple devices in power distribution rooms based on an industrial IoT architecture. Background Art

[0002] With the continuous advancement of energy digitalization and intelligence, distribution rooms, as key nodes in power systems, are increasingly equipped with a diverse range of equipment, including smart meters, protection and measurement and control devices, circuit breakers, power quality analyzers, fault recorders, and environmental monitoring devices. The growing demand for coordinated operation of multiple devices is placing higher demands on monitoring, protection, control, and information exchange within distribution rooms. To improve operational efficiency, ensure power quality, and ensure operational safety, there is an urgent need to build an efficient and intelligent multi-device integrated control system.

[0003] The development of the Industrial Internet of Things (IIoT) provides crucial support for the intelligentization of power distribution rooms. Devices such as edge acquisition modules, protocol conversion modules, and edge gateways enable real-time collection, protocol parsing, and standardized processing of operating data from various types of power distribution equipment, providing a data foundation for subsequent intelligent control. However, efficiently integrating operating data at varying sampling granularities (milliseconds and minutes) to establish unified data processing and control logic remains a major technical challenge.

[0004] Some research has attempted to integrate and control large-scale equipment. For example, CN111651332B proposes a method for integrated control of large-scale heterogeneous equipment based on message-based middleware. This method abstracts a unified device access model and utilizes a publish / subscribe mechanism to implement read and write control of device attributes. This method is suitable for general industrial control systems and offers advantages in improving device integration flexibility and control system scalability. However, limitations remain: It fails to differentiate between fast- and slow-changing operating data in power distribution equipment, which can easily lead to loss of critical data or response delays; and the control strategy lacks timing sensitivity, making it impossible to dynamically adjust access frequency and control methods based on the operating characteristics of different devices.

[0005] To address this problem, the present invention proposes a multi-device integrated control system and method for a power distribution room based on the industrial Internet of Things architecture, constructs a unified industrial Internet of Things architecture, opens up the data interfaces between multi-source heterogeneous devices, forms a structured access system, and realizes the integrated access and interconnection of multiple devices, as well as the adjustment of control parameters. Summary of the Invention

[0006] In view of this, the present invention provides a multi-device integrated control system and method for a power distribution room based on an industrial Internet of Things architecture. Due to the wide variety of equipment in the power distribution room and the different communication protocols, the information collection standards are not unified, data docking is difficult, and efficient integrated control and unified management are difficult to achieve. Edge collection modules, protocol conversion modules, and edge gateways are deployed in the power distribution room to build a unified industrial Internet of Things architecture, open up the data interfaces between multi-source heterogeneous devices, form a structured access system, and achieve integrated access and interconnection of multiple devices. The operating data of the power distribution equipment includes two types of features, slow-changing and fast-changing. Traditional collection methods often mix them, resulting in the obscuration of important state features or analysis failure. This application introduces an improved LSTM network and a sliding difference method combined with an attention mechanism to extract refined time series features from the slow-changing and fast-changing operating data, respectively, significantly improving the accuracy and semantic expression ability of device feature modeling, thereby enhancing the foundation of state recognition and dynamic control. By improving the weighted polling mechanism and fuzzy logic rule mapping method, the feature information of each device is intelligently integrated and processed, and control parameters with strong adaptability and real-time response are dynamically generated to achieve coordinated control and efficient scheduling of multiple devices in the power distribution room.

[0007] To achieve the above objectives, the present invention provides a method for integrated control of multiple devices in a power distribution room based on an industrial Internet of Things architecture, comprising the following steps:

[0008] S1: Deploy edge data acquisition modules, protocol conversion modules, and edge gateways in the power distribution room. Power distribution equipment is connected to the multi-device integrated control system in the power distribution room through the edge data acquisition modules.

[0009] The edge acquisition module, protocol conversion module, edge gateway, and connected power distribution equipment are used as the industrial Internet of Things in the power distribution room scenario;

[0010] S2: Use the edge acquisition module to collect the operating data of the power distribution equipment, and use the protocol conversion module to regularize the collected operating data, dividing the regularized operating data into slow-changing operating data and fast-changing operating data;

[0011] S3: Using an improved LSTM network and a sliding difference method combined with an attention mechanism, the time series features of the slow-changing operation data and the fast-changing operation data are extracted respectively as the device feature information of the power distribution equipment associated with the slow-changing operation data and the fast-changing operation data;

[0012] S4: Adopting an improved weighted polling method to sequentially receive the device characteristic information of the power distribution equipment, mapping the device characteristic information based on fuzzy logic rules, generating control parameters of the power distribution equipment, and performing integrated control on multiple power distribution equipment in the power distribution room.

[0013] As a further improvement method of the present invention:

[0014] Optionally, the edge acquisition module is installed inside the power distribution equipment, the protocol conversion module and the edge gateway are integrated into a control cabinet, and the control cabinet is deployed in the power distribution room, including:

[0015] The power distribution equipment in the power distribution room is connected to the power distribution room multi-device integrated control system through the edge acquisition module for registration. After successful registration, the multi-device integrated control system of the power distribution room sends the unique device ID to the control cabinet;

[0016] After the edge gateway in the control cabinet receives the unique ID of the device, it calls the device identification service to automatically match the device model, operating data sampling frequency, proprietary communication protocol and data sampling template of the edge gateway receiving device, sends the proprietary communication protocol to the protocol conversion module, and sends the operating data sampling frequency and data sampling template to the edge acquisition module.

[0017] Optionally, the edge collection module is used to collect operating data of the power distribution equipment, including:

[0018] The edge acquisition module collects the operating data of the power distribution equipment according to the operating data sampling frequency, and preprocesses the operating data, where the preprocessing includes data caching, filtering and denoising, and sends the preprocessed operating data to the edge gateway. The edge gateway transmits the preprocessed operating data to the protocol conversion module, where the operating data includes the unique device ID of the power distribution equipment;

[0019] The protocol conversion module identifies the distribution equipment based on its unique ID, obtains its proprietary communication protocol and data frame structure, and extracts key fields from the pre-processed operating data, including register numbers, values, and status bits. It uses a standard sequence structure mapping unit to regularize the key fields into a unified data format as regularized operating data, and uses a data packet encapsulation unit to encapsulate the regularized operating data into data packets, which are then transmitted to the multi-device integrated control system in the distribution room via the edge gateway.

[0020] The multi-device integrated control system in the power distribution room parses the data packets to obtain regularized operating data.

[0021] Optionally, the regularized operating data is divided into slow-changing operating data and fast-changing operating data according to the operating data sampling frequency, wherein the operating data sampling frequency range of the slow-changing operating data is less than 2 Hz, and the operating data sampling frequency range of the fast-changing operating data is not less than 2 Hz, and the data lengths of the slow-changing operating data and the fast-changing operating data are consistent.

[0022] Optionally, an improved LSTM network and a sliding difference method combined with an attention mechanism are used to extract the time series features of the slowly changing running data and the rapidly changing running data, respectively, including:

[0023] Obtain slowly varying running data x and rapidly varying running data y, and use an improved LSTM network to extract the time series features of the slowly varying running data x. The improved LSTM network uses a multi-gated residual connection. The time series feature extraction process includes:

[0024] Use the LSTM network to calculate the current LSTM hidden state at any time step n in the slowly changing running data x Current input data x n , the LSTM hidden state of the previous time step And nonlinear mapping combination representation Where n∈[1,N], N represents the length of the slowly varying running data, x n Indicates the nth data value in the slowly changing running data x, By performing nonlinear fusion on the current LSTM hidden state and input data, the joint representation of the two is extracted to enhance the recognition of time series features. Reflects the changing trend of the slowly varying running data x at time step t;

[0025] Use multi-gating mechanism to calculate the current LSTM hidden state Current input data x n , the LSTM hidden state of the previous time step And nonlinear mapping combination representation The gate weight of the current LSTM hidden state Current input data x n , the LSTM hidden state of the previous time step And nonlinear mapping combination representation Perform weighted fusion to obtain the current LSTM hidden state The multi-gated residual connection state of the current LSTM hidden state To update:

[0026]

[0027] in, In turn, they represent the current LSTM hidden state Current input data x n , the LSTM hidden state of the previous time step And nonlinear mapping combination representation The gating weight of

[0028] Represents the current LSTM hidden state The multi-gated residual connection state of the current LSTM hidden state Updated to the multi-gated residual connection state, Represents the current LSTM hidden state The update result of

[0029] The updated current LSTM hidden state of each time step in the slowly changing running data x is extracted to form the time series features of the slowly changing running data x.

[0030] Optionally, extracting the time series features of the fast-changing running data y using the sliding difference method combined with the attention mechanism includes:

[0031] Perform sliding window difference on the fast-changing running data y to obtain the differential sequence of the fast-changing running data y, where the differential sequence is expressed as D(y)=(D k+1 (y),D k+2 (y),...,D i (y),...,D N (y)), where i∈[k+1,N], k represents the length of the sliding window, D i (y) = y i -y i-k , where y i Represents the i-th data value in the fast-changing running data y, y i-k Represents the ikth data value in the fast-changing running data y;

[0032] D k+1 (y),D k+2 (y),...,D i (y),...,D N (y) represents the difference value in the difference sequence D(y);

[0033] The nonlinear transformation method is used to perform feature mapping on the difference values ​​in the difference sequence to obtain the nonlinear transformation features of the difference values;

[0034] Calculate the time attention of the nonlinear transformation features, weight the nonlinear transformation features of the differential value according to the time attention, and obtain the time series features of the fast-changing running data y:

[0035]

[0036] Among them, D y Represents the time series characteristics of fast-changing running data y, Indicates the difference value D i (y) Temporal attention, T represents transposition, v represents the learnable mapping coefficient vector, which is used to adjust the length of the temporal feature, W att represents the learnable attention weight, b att represents the learnable attention bias, d i((y) represents the difference value D i (y) is the nonlinear transformation characteristic.

[0037] Optionally, the device characteristic information of the power distribution equipment is sequentially received using an improved weighted polling method, including:

[0038] Extract the normalized operating data of the power distribution equipment and calculate the operating risk level of the normalized operating data. The higher the operating risk level, the more abnormal the operating status of the power distribution equipment. The calculation method of the operating risk level is:

[0039]

[0040] Among them, rate represents the operational risk level of the regularized operational data, s n represents the nth data value in the regularized running data, n∈[1,N], μ represents the mean of the regularized running data, σ represents the standard deviation of the regularized running data, s j represents the jth data value in the regularized running data, s j-1 Represents the j-1th data value in the regularized running data;

[0041] Combined with the degree of operational risk, calculate the number of times the device characteristic information of the distribution equipment is received within a scheduling cycle:

[0042]

[0043] Among them, count min Indicates the preset minimum number of times, count max Indicates the preset maximum number of times, set count min is 1, count max is 5, and Sum represents the sum of the operating risk levels of all distribution equipment.

[0044] Optionally, mapping processing is performed on the device characteristic information based on fuzzy logic rules to generate control parameters of the power distribution equipment, including:

[0045] Extracting the mean and standard deviation of the device characteristic information, inputting the mean and standard deviation into the triangular membership function and the Gaussian membership function respectively, using the triangular membership function to output the membership of the device characteristic information in low, medium, and high operating states, extracting the operating state with the highest membership, where the motion state includes a high operating state, a medium operating state, and a low operating state, and using the Gaussian membership function to output the fluctuation membership value of the device characteristic information, where a higher fluctuation membership value indicates a smaller operating fluctuation of the device characteristic information, using a clustering algorithm to cluster the fluctuation membership values ​​of the received device characteristic information, and dividing the device characteristic information received in real time into the closest clusters according to the fluctuation membership values, where the clusters include high fluctuation, medium fluctuation, and low fluctuation;

[0046] The corresponding control parameters of the extracted operating state with the highest membership and the cluster to which it belongs are extracted according to fuzzy logic rules and used as the control parameters of the distribution equipment to control and adjust the operating state of the distribution equipment. The fuzzy logic rules are the correspondence rules between different types of operating states, clusters and control parameters.

[0047] In order to solve the above problems, the present invention provides a multi-device integrated control system for a power distribution room based on the industrial Internet of Things architecture. The multi-device integrated control system for a power distribution room based on the industrial Internet of Things architecture includes an auxiliary module, a time sequence feature extraction module, and a control module:

[0048] The auxiliary module includes an acquisition module, a protocol conversion module and an edge gateway. The power distribution equipment is connected to the multi-device integrated control system of the power distribution room through the edge acquisition module. The edge acquisition module is used to collect the operating data of the power distribution equipment, and the collected operating data is regularized through the protocol conversion module. The regularized operating data is uploaded to the multi-device integrated control system of the power distribution room through the edge gateway;

[0049] The time series feature extraction module is used to use an improved LSTM network and a sliding difference method combined with an attention mechanism to respectively extract the time series features of the slow-changing operation data and the fast-changing operation data as the device feature information of the associated power distribution equipment;

[0050] The control module is used to sequentially receive device characteristic information of the power distribution equipment using an improved weighted polling method, perform mapping processing on the device characteristic information based on fuzzy logic rules, generate control parameters of the power distribution equipment, and perform integrated control of multiple power distribution equipment in the power distribution room;

[0051] This is to realize the above-mentioned multi-device integrated control method for a distribution room based on the industrial Internet of Things architecture.

[0052] In order to solve the above problem, the present invention provides an electronic device, comprising:

[0053] a memory storing at least one instruction;

[0054] Communication interfaces to enable electronic equipment to communicate; and

[0055] The processor executes the instructions stored in the memory to implement the above-mentioned multi-device integrated control method for the power distribution room based on the industrial Internet of Things architecture.

[0056] In order to solve the above problems, the present invention also provides a computer-readable storage medium, which stores at least one instruction, and the at least one instruction is executed by a processor in an electronic device to implement the above-mentioned multi-device integrated control method for a distribution room based on the industrial Internet of Things architecture.

[0057] Compared with the existing technology, the present invention proposes a multi-device integrated control system and method for a power distribution room based on the industrial Internet of Things architecture. This technology has the following beneficial effects:

[0058] First, in the process of extracting time series features of slowly changing operating data, since traditional gated residual connections usually only use a single gated weight to balance the current LSTM hidden state and the current input data, it is unable to flexibly capture the simultaneous existence of multiple feature components (such as trends, fluctuations, and noise) in complex time series data. When it is necessary to simultaneously utilize hidden states, inputs, historical states, or compound transformation information, a single gated weight is difficult to ensure the full utilization of various types of information, resulting in important information being ignored or weakened. Multi-gated weights support the parallel fusion of multiple information streams, and can flexibly capture multi-dimensional features such as trends, short-term fluctuations, and historical dependencies, enriching the feature expression space. The information stream includes the current LSTM hidden state, the current input data, the LSTM hidden state of the previous time step, and the nonlinear mapping combination representation. The multi-gated weights are adaptively generated according to the context, and can dynamically adjust the fusion ratio for different time steps and input data, thereby improving adaptability and generalization capabilities. At different operating moments, the characteristics of the distribution equipment The importance of features may change, and multiple gating weights adapt to such changes. For example, during transient faults, the current input data becomes more critical. During stable operation, the LSTM hidden state of the previous time step and the current LSTM hidden state, which represent the changing trend, become more critical. Under complex coupled conditions, the nonlinear mapping combination representation can capture more detailed features. At the same time, the gating mechanism can reduce the weight of noise, abnormalities, or irrelevant information flows, reducing their interference on the model output and improving robustness. For example, when an occasional sensor failure causes abnormal input signals or sudden interference, the gating weights automatically reduce dependence on the abnormal input to prevent control strategy misjudgments. The gating weights can also flexibly adjust the influence ratio of the current LSTM hidden state and the LSTM hidden state of the previous time step, strengthening the balance between long-term dependence and short-term dynamics. For example, in a distribution room, the status of distribution equipment is often affected by several past moments. Strengthening historical dependence through gating helps accurately capture slow-changing features such as equipment aging and load gradients.

[0059] At the same time, in the process of extracting time series features from fast-changing operating data, the sliding window difference operation effectively amplifies the local change information in the fast-changing operating data, especially the instantaneous signal jumps caused by sudden anomalies, enhances the sensitivity to abnormal patterns, and avoids the masking of anomaly detection by the stable components of fast-changing operating data. At the same time, feature encoding through nonlinear mapping improves the expressive power of differential values ​​in the differential sequence, can capture richer time series nonlinear features, and helps to reveal complex dynamic changes. More importantly, the temporal attention mechanism focuses on key abnormal moments in the time series by adaptively allocating weights, thus enhancing the ability to identify sudden events and avoiding the dilution of important information caused by averaging processing.

[0060] Furthermore, the fixed-dimensional aggregate features output by this method are concise and expressive, facilitating subsequent control strategy formulation or anomaly identification. Overall, this approach combines local change capture with global temporal attention, significantly improving the efficiency of detecting and responding to millisecond-level sudden anomalies in distribution rooms. BRIEF DESCRIPTION OF THE DRAWINGS

[0061] Figure 1 A flowchart of a method for integrated control of multiple devices in a power distribution room based on an industrial Internet of Things architecture is provided in accordance with an embodiment of the present invention.

[0062] Figure 2 An industrial Internet of Things architecture diagram provided by one embodiment of the present invention.

[0063] The purpose, features and advantages of the present invention will be further described with reference to the accompanying drawings and in conjunction with the embodiments. DETAILED DESCRIPTION

[0064] It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0065] The embodiment of the present application provides a multi-device integrated control system and method for a power distribution room based on an industrial Internet of Things architecture. The execution subject of the multi-device integrated control method for a power distribution room based on an industrial Internet of Things architecture includes but is not limited to at least one of the electronic devices such as a server and a terminal that can be configured to execute the method provided by the embodiment of the present application. In other words, the multi-device integrated control method for a power distribution room based on an industrial Internet of Things architecture can be executed by software or hardware installed on a terminal device or a server device, and the software can be a blockchain platform. The server includes but is not limited to: a single server, a server cluster, a cloud server or a cloud server cluster, etc.

[0066] Reference Figure 1 as well as Figure 2 , embodiment 1 of the present invention is:

[0067] A method for integrated control of multiple devices in a power distribution room based on an industrial Internet of Things architecture comprises the following steps:

[0068] S1: Deploy edge acquisition modules, protocol conversion modules, and edge gateways in the power distribution room. The power distribution equipment is connected to the multi-device integrated control system in the power distribution room through the edge acquisition modules.

[0069] The edge acquisition module is installed inside the power distribution equipment, the protocol conversion module and the edge gateway are integrated into the control cabinet, and the control cabinet is deployed in the power distribution room, including:

[0070] The edge acquisition module includes a multi-channel signal acquisition interface, a multi-source sensor, a communication interface, and a sampling buffer. The multi-channel signal acquisition interface is used to connect to the multi-source sensor and collect the operating data of the power distribution equipment. The communication interface supports multiple communication protocols and is used to send the registration information and operating data of the power distribution equipment to the multi-device integrated control system of the power distribution room and the edge gateway in the control cabinet respectively. The sampling buffer is used to cache, filter, and denoise the collected operating data.

[0071] The protocol conversion module includes a multi-protocol parsing engine, a standard data structure mapping unit, and a data packet encapsulation unit. The multi-protocol parsing engine is used to parse the proprietary communication protocols and data frame structures of various types of power distribution equipment. The standard sequence structure mapping unit is used to regularize the operating data of the power distribution equipment into the same data format. The data packet encapsulation unit is used to encapsulate the regularized operating data into data packets and transmit them to the multi-device integrated control system of the power distribution room through the edge gateway.

[0072] The edge gateway includes a cache unit and a communication module, wherein the cache unit is used to compress and cache the data packets encapsulated by the protocol conversion module, and the communication module is used to receive the operating data of the power distribution equipment and transmit the data packets encapsulated by the protocol conversion module to the multi-device integrated control system of the power distribution room;

[0073] The power distribution equipment in the power distribution room is connected to the power distribution room multi-device integrated control system through the edge acquisition module. When the power distribution equipment is connected, it automatically sends registration information to the power distribution room multi-device integrated control system. After successful registration, the power distribution room multi-device integrated control system sends the device's unique ID to the control cabinet; the registration information includes MAC address, device type, manufacturer and model;

[0074] After the edge gateway in the control cabinet receives the unique ID of the device, it calls the device identification service to automatically match the device model, operating data sampling frequency, proprietary communication protocol and data sampling template of the edge gateway receiving device, sends the proprietary communication protocol to the protocol conversion module, and sends the operating data sampling frequency and data sampling template to the edge acquisition module.

[0075] like Figure 2 FIG. 1 is an industrial Internet of Things architecture diagram provided by an embodiment of the present invention.

[0076] In an embodiment of the present application, the power distribution equipment includes a smart meter, a protection and measurement control device (such as a microcomputer protection unit), a fault recorder, a circuit breaker, an environmental monitoring device, a power quality analyzer, a monitoring terminal, and a communication device (such as an edge gateway). The operating data of the smart meter includes voltage, current, active / reactive power, frequency, and harmonics. The operating data of the protection and measurement control device is a trip signal. The operating data of the fault recorder is voltage fluctuation data. The operating data of the circuit breaker is the action state sequence of the circuit breaker. The operating data of the environmental monitoring device is the temperature and humidity data and smoke concentration data of the distribution room. The operating data of the power quality analyzer is voltage deviation data and three-phase imbalance data. The monitoring terminal is the opening status data of the access control. The operating data of the communication device is link quality data.

[0077] It should be noted that the operating data are all in the form of data series.

[0078] S2: Use the edge acquisition module to collect the operating data of the power distribution equipment, and regularize the collected operating data through the protocol conversion module, and divide the regularized operating data into slow-changing operating data and fast-changing operating data.

[0079] Use the edge acquisition module to collect operating data of power distribution equipment, including:

[0080] The edge acquisition module collects the operating data of the power distribution equipment according to the operating data sampling frequency, and preprocesses the operating data, where the preprocessing includes data caching, filtering and denoising, and sends the preprocessed operating data to the edge gateway. The edge gateway transmits the preprocessed operating data to the protocol conversion module, where the operating data includes the unique device ID of the power distribution equipment;

[0081] The protocol conversion module identifies the distribution equipment based on its unique ID, obtains its proprietary communication protocol and data frame structure, and extracts key fields from the pre-processed operating data, including register numbers, values, and status bits. It uses a standard sequence structure mapping unit to regularize the key fields into a unified data format as regularized operating data, and uses a data packet encapsulation unit to encapsulate the regularized operating data into data packets, which are then transmitted to the multi-device integrated control system in the distribution room via the edge gateway.

[0082] In this embodiment, the proprietary communication protocols of the power distribution equipment include Modbus, IEC104, IEC61850, DNP3, PROFIBUS, etc. For example, register number 0x0010 represents the active power register, the register value 315.2 indicates that the current power is 315.2 kW, and the status bit 0x04 indicates that the power distribution equipment is in an abnormal state;

[0083] The multi-device integrated control system in the power distribution room parses the data packets to obtain regularized operating data.

[0084] It should be noted that the data length of the regularized operating data is consistent, so the higher the sampling frequency of the operating data, the shorter the collection time of the operating data.

[0085] The regularized operating data is divided into slow-changing operating data and fast-changing operating data according to the operating data sampling frequency, wherein the operating data sampling frequency range of the slow-changing operating data is less than 2 Hz, and the operating data sampling frequency range of the fast-changing operating data is not less than 2 Hz, and the data lengths of the slow-changing operating data and the fast-changing operating data are consistent.

[0086] In this embodiment, the fast-changing operating data includes regularized operating data of the smart meter, the protection and measurement control device (such as a microcomputer protection unit), the fault recorder, and the circuit breaker;

[0087] Slow-changing operation data includes regularized operation data of environmental monitoring equipment, power quality analyzers, monitoring terminals and communication equipment.

[0088] S3: Use the improved LSTM network and the sliding difference method combined with the attention mechanism to extract the time series features of the slow-changing operation data and the fast-changing operation data respectively, as the equipment feature information of the distribution equipment associated with the slow-changing operation data and the fast-changing operation data.

[0089] The improved LSTM network and the sliding difference method combined with the attention mechanism are used to extract the time series features of slow-changing running data and fast-changing running data respectively, including:

[0090] Obtain slowly varying running data x and rapidly varying running data y, and use an improved LSTM network to extract the time series features of the slowly varying running data x. The improved LSTM network uses a multi-gated residual connection. The time series feature extraction process includes:

[0091] Use the LSTM network to calculate the current LSTM hidden state at any time step n in the slowly changing running data x Current input data x n , the LSTM hidden state of the previous time step And nonlinear mapping combination representation Where n∈[1,N], N represents the length of the slowly varying running data, x n Indicates the nth data value in the slowly changing running data x, By performing nonlinear fusion on the current LSTM hidden state and input data, the joint representation of the two is extracted to enhance the recognition of time series features. Reflects the changing trend of the slowly varying running data x at time step t;

[0092] Specifically, Where δ((·) represents the activation function (the activation function is a nonlinear function, for example, the ReLU function can be used), W1, W2 represent the learnable weight matrix parameters, b represents the learnable bias parameter, and fre x Indicates the running data sampling frequency of the slowly changing running data x;

[0093] It should be noted that the current LSTM hidden state It is the deep expression of the slowly changing operation data x at time step t, reflecting the load trend and operation law of the distribution equipment. The LSTM hidden state of the previous time step Used to hide the current LSTM state Perform time series smoothing and enhance time dependence, Used to improve the recognition ability of complex patterns;

[0094] Use multi-gating mechanism to calculate the current LSTM hidden state Current input data x n , the LSTM hidden state of the previous time step And nonlinear mapping combination representation The gate weight of the current LSTM hidden state Current input data x n , the LSTM hidden state of the previous time step And nonlinear mapping combination representation Perform weighted fusion to obtain the current LSTM hidden state The multi-gated residual connection state of the current LSTM hidden state To update:

[0095]

[0096] in, In turn, they represent the current LSTM hidden state Current input data x n , the LSTM hidden state of the previous time step And nonlinear mapping combination representation The gating weight of

[0097] Represents the current LSTM hidden state The multi-gated residual connection state of the current LSTM hidden state Updated to the multi-gated residual connection state, Represents the current LSTM hidden state The update result of

[0098] In this embodiment, exp(·) represents an exponential function with a natural constant as the base, W ((1) ,W ((2) ,W ((3) ,W ((4) Represents the current LSTM hidden state, the current input data, the LSTM hidden state of the previous time step, and the learnable gated scoring matrix represented by the combination of nonlinear mappings respectively;

[0099] Extract the updated current LSTM hidden state of each time step in the slowly varying running data x to form the time series features of the slowly varying running data x. Specifically, the time series features of the slowly varying running data x are:

[0100]

[0101] Among them, h(x) represents the time series characteristics of the slowly changing running data x, They represent the current LSTM hidden states after the update of the slowly changing running data x at the 1st to Nth time steps respectively.

[0102] The sliding difference method combined with the attention mechanism is used to extract the time series features of the fast-changing running data y, including:

[0103] Perform sliding window difference on the fast-changing running data y to obtain the differential sequence of the fast-changing running data y, where the differential sequence is expressed as D(y)=(D k+1 (y),D k+2 (y),...,D i (y),...,D N (y)), where i∈[k+1,N], k represents the length of the sliding window, D i (y) = y i -y i-k , where y i Represents the i-th data value in the fast-changing running data y, y i-k Represents the ikth data value in the fast-changing running data y;

[0104] D k+1 (y),D k+2 (y),...,D i (y),...,D N (y) represents the difference value in the difference sequence D(y);

[0105] The nonlinear transformation method is used to perform feature mapping on the difference values ​​in the difference sequence to obtain the nonlinear transformation features of the difference values;

[0106] Calculate the time attention of the nonlinear transformation features, weight the nonlinear transformation features of the differential value according to the time attention, and obtain the time series features of the fast-changing running data y:

[0107]

[0108] Among them, D y Represents the time series characteristics of fast-changing running data y, Indicates the difference value D i (y) Temporal attention, T represents transpose, v represents the learnable mapping coefficient vector, W att represents the learnable attention weight, b att represents the learnable attention bias, d i ((y) represents the difference value D i (y) is a nonlinear transformation feature; specifically, the ReLU function is used to extract the nonlinear transformation features of the differential value. It should be noted that by restricting the rows and columns of the attention weights, the time series features of the slowly varying running data and the rapidly varying running data are made into feature vectors of the same length.

[0109] S4: Adopting an improved weighted polling method to sequentially receive the device characteristic information of the power distribution equipment, mapping the device characteristic information based on fuzzy logic rules, generating control parameters of the power distribution equipment, and performing integrated control on multiple power distribution equipment in the power distribution room.

[0110] The improved weighted polling method is used to sequentially receive the device characteristic information of the power distribution equipment, including:

[0111] Extract the normalized operating data of the power distribution equipment and calculate the operating risk level of the normalized operating data. The higher the operating risk level, the more abnormal the operating status of the power distribution equipment. The calculation method of the operating risk level is:

[0112]

[0113] Among them, rate represents the operational risk level of the regularized operational data, s n represents the nth data value in the regularized running data, n∈[1,N], μ represents the mean of the regularized running data, σ represents the standard deviation of the regularized running data, s j represents the jth data value in the regularized running data, s j-1 Represents the j-1th data value in the regularized running data;

[0114] It should be noted that It is used to measure the standardized cumulative deviation of the regularized operating data. The larger the standardized cumulative deviation is, the more the data value exceeds the normal range. Used to reflect the instantaneous fluctuation of regularized operating data;

[0115] Combined with the degree of operational risk, calculate the number of times the device characteristic information of the distribution equipment is received within a scheduling cycle:

[0116]

[0117] Among them, count min Indicates the preset minimum number of times, count max Indicates the preset maximum number of times, set count min is 1, count max is 5, where Sum represents the sum of the operating risk levels of all power distribution equipment. Weighted polling is performed on the equipment feature information based on the operating risk level to improve the real-time performance and sensitivity of the control, and power distribution equipment with potential risks is prioritized. The scheduling cycle is 20 minutes.

[0118] The device characteristic information is mapped based on fuzzy logic rules to generate the control parameters of the power distribution equipment, including:

[0119] Extracting the mean and standard deviation of the device characteristic information, inputting the mean and standard deviation into the triangular membership function and the Gaussian membership function respectively, using the triangular membership function to output the membership of the device characteristic information in low, medium, and high operating states, extracting the operating state with the highest membership, where the motion state includes a high operating state, a medium operating state, and a low operating state, and using the Gaussian membership function to output the fluctuation membership value of the device characteristic information, where a higher fluctuation membership value indicates a smaller operating fluctuation of the device characteristic information, using a clustering algorithm to cluster the fluctuation membership values ​​of the received device characteristic information, and dividing the device characteristic information received in real time into the closest clusters according to the fluctuation membership values, where the clusters include high fluctuation, medium fluctuation, and low fluctuation;

[0120] The corresponding control parameters of the extracted operating state with the highest membership and the cluster to which it belongs are extracted according to fuzzy logic rules and used as the control parameters of the power distribution equipment to control and adjust the operating state of the power distribution equipment.

[0121] In this embodiment, the mean, serving as a central tendency indicator representing the operating status of distribution equipment, is used through a triangular membership function to accurately classify different operating levels, such as high, medium, and low operating states, thereby finely grading equipment load or operating intensity. The standard deviation reflects the fluctuation amplitude and stability of the data, and the Gaussian membership function is used to smoothly model the degree of fluctuation, capturing abnormal fluctuations or unstable states in distribution equipment operation. This combination enables the multi-device integrated control system in the distribution room to not only perceive the load status of distribution equipment in real time, but also dynamically monitor fluctuation risks, enhancing comprehensive control over equipment operating status.

[0122] The loads of the power distribution equipment in the high operating state, the medium operating state and the low operating state decrease in sequence;

[0123] It should be noted that, based on the generated control parameters and the optimal control parameters adjusted manually, a mean square error loss function is constructed to train and optimize the learnable parameters.

[0124] Based on these two types of membership results, the constructed fuzzy logic rules enable flexible and practical control decisions. By incorporating conditions such as "If the mean is high and the fluctuation is large, reduce the load" or "If the mean is medium and the fluctuation is low, maintain the current operation," the fuzzy logic rules can automatically adjust the control strategy for different operating states, improving the accuracy and robustness of the control parameter response. This method effectively avoids the rigidity of traditional hard threshold judgments, adapts to the continuous changes in the operating state of distribution equipment, enhances the adaptability and fault tolerance of the multi-device integrated control system in the distribution room to complex operating environments, and significantly improves the intelligence level and safe and stable operation capabilities of the multi-device integrated control system in the distribution room.

[0125] Specifically, examples of the fuzzy logic rules are as follows:

[0126]

[0127] Example 2:

[0128] A multi-device integrated control system for a power distribution room based on an industrial Internet of Things architecture, to implement a multi-device integrated control method for a power distribution room based on an industrial Internet of Things architecture as described in Example 1, includes an auxiliary module, a timing feature extraction module, and a control module:

[0129] The auxiliary module includes an acquisition module, a protocol conversion module and an edge gateway. The power distribution equipment is connected to the multi-device integrated control system of the power distribution room through the edge acquisition module. The edge acquisition module is used to collect the operating data of the power distribution equipment, and the collected operating data is regularized through the protocol conversion module. The regularized operating data is uploaded to the multi-device integrated control system of the power distribution room through the edge gateway;

[0130] The time series feature extraction module is used to use an improved LSTM network and a sliding difference method combined with an attention mechanism to respectively extract the time series features of the slow-changing operation data and the fast-changing operation data as the device feature information of the associated power distribution equipment;

[0131] The control module is used to sequentially receive device characteristic information of the power distribution equipment using an improved weighted polling method, perform mapping processing on the device characteristic information based on fuzzy logic rules, generate control parameters of the power distribution equipment, and perform integrated control of multiple power distribution equipment in the power distribution room.

[0132] It should be understood that the embodiment is for illustration only and the scope of the patent application is not limited to this structure.

[0133] It should be noted that the serial numbers of the above-mentioned embodiments of the present invention are for descriptive purposes only and do not represent the advantages or disadvantages of the embodiments. In addition, the terms "including", "comprising" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, device, article or method comprising a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, device, article or method. In the absence of further restrictions, an element defined by the sentence "including a ..." does not exclude the presence of other identical elements in the process, device, article or method comprising the element.

[0134] Through the description of the above embodiments, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus the necessary general hardware platform, and of course can also be implemented by hardware, but in many cases the former is a better embodiment. Based on this understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) as described above, and includes a number of instructions for enabling a terminal device (which can be a mobile phone, computer, server, or network device, etc.) to execute the methods described in each embodiment of the present invention.

[0135] The above are only preferred embodiments of the present invention and are not intended to limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made using the contents of the present invention description and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. A method for integrated control of multiple devices in a power distribution room based on an industrial Internet of Things architecture, characterized in that: The method comprises: S1: Deploy edge data acquisition modules, protocol conversion modules, and edge gateways in the power distribution room. Power distribution equipment is connected to the multi-device integrated control system in the power distribution room through the edge data acquisition modules. S2: Use the edge acquisition module to collect the operating data of the power distribution equipment, and use the protocol conversion module to regularize the collected operating data, dividing the regularized operating data into slow-changing operating data and fast-changing operating data; S3: Using an improved LSTM network and a sliding difference method combined with an attention mechanism, the time series features of the slow-changing operation data and the fast-changing operation data are extracted respectively as the device feature information of the power distribution equipment associated with the slow-changing operation data and the fast-changing operation data; S4: Adopting an improved weighted polling method to sequentially receive the device characteristic information of the power distribution equipment, mapping the device characteristic information based on fuzzy logic rules, generating control parameters of the power distribution equipment, and performing integrated control on multiple power distribution equipment in the power distribution room.

2. The method for integrated control of multiple devices in a power distribution room based on an industrial Internet of Things architecture according to claim 1, characterized in that: The edge acquisition module is installed inside the power distribution equipment, the protocol conversion module and the edge gateway are integrated into the control cabinet, and the control cabinet is deployed in the power distribution room, including: The power distribution equipment in the power distribution room is connected to the power distribution room multi-device integrated control system through the edge acquisition module for registration. After successful registration, the multi-device integrated control system of the power distribution room sends the unique device ID to the control cabinet; After the edge gateway in the control cabinet receives the unique ID of the device, it calls the device identification service to automatically match the device model, operating data sampling frequency, proprietary communication protocol and data sampling template of the edge gateway receiving device, sends the proprietary communication protocol to the protocol conversion module, and sends the operating data sampling frequency and data sampling template to the edge acquisition module.

3. The method for integrated control of multiple devices in a power distribution room based on an industrial Internet of Things architecture according to claim 2, characterized in that: Use the edge acquisition module to collect operating data of power distribution equipment, including: The edge acquisition module collects the operating data of the power distribution equipment according to the operating data sampling frequency, and preprocesses the operating data, where the preprocessing includes data caching, filtering and denoising, and sends the preprocessed operating data to the edge gateway. The edge gateway transmits the preprocessed operating data to the protocol conversion module, where the operating data includes the unique device ID of the power distribution equipment; The protocol conversion module identifies the distribution equipment based on its unique ID, obtains its proprietary communication protocol and data frame structure, and extracts key fields from the pre-processed operating data, including register numbers, values, and status bits. It uses a standard sequence structure mapping unit to regularize the key fields into a unified data format as regularized operating data, and uses a data packet encapsulation unit to encapsulate the regularized operating data into data packets, which are then transmitted to the multi-device integrated control system in the distribution room via the edge gateway. The multi-device integrated control system in the power distribution room parses the data packets to obtain regularized operating data.

4. The method for integrated control of multiple devices in a power distribution room based on an industrial Internet of Things architecture according to claim 3, characterized in that: The regularized operating data is divided into slow-changing operating data and fast-changing operating data according to the operating data sampling frequency, wherein the operating data sampling frequency range of the slow-changing operating data is less than 2 Hz, and the operating data sampling frequency range of the fast-changing operating data is not less than 2 Hz, and the data lengths of the slow-changing operating data and the fast-changing operating data are consistent.

5. The method for integrated control of multiple devices in a power distribution room based on an industrial Internet of Things architecture according to claim 4, characterized in that: The improved LSTM network and the sliding difference method combined with the attention mechanism are used to extract the time series features of slow-changing running data and fast-changing running data respectively, including: Obtain slowly varying running data x and rapidly varying running data y, and use an improved LSTM network to extract the time series features of the slowly varying running data x. The improved LSTM network uses a multi-gated residual connection. The time series feature extraction process includes: Use the LSTM network to calculate the current LSTM hidden state at any time step n in the slowly changing running data x Current input data x n , the LSTM hidden state of the previous time step And nonlinear mapping combination representation Where n∈[1,N], N represents the length of the slowly varying running data, x n Indicates the nth data value in the slowly changing running data x, By performing nonlinear fusion on the current LSTM hidden state and input data, the joint representation of the two is extracted to enhance the recognition of time series features. Reflects the changing trend of the slowly varying running data x at time step t; Use multi-gating mechanism to calculate the current LSTM hidden state Current input data x n , the LSTM hidden state of the previous time step And nonlinear mapping combination representation The gate weight of the current LSTM hidden state Current input data x n , the LSTM hidden state of the previous time step And nonlinear mapping combination representation Perform weighted fusion to obtain the current LSTM hidden state The multi-gated residual connection state of the current LSTM hidden state To update: in, In turn, they represent the current LSTM hidden state Current input data x n , the LSTM hidden state of the previous time step And nonlinear mapping combination representation The gating weight of Represents the current LSTM hidden state The multi-gated residual connection state of Updated to the multi-gated residual connection state, Represents the current LSTM hidden state The update result of The updated current LSTM hidden state of each time step in the slowly changing running data x is extracted to form the time series features of the slowly changing running data x.

6. The method for integrated control of multiple devices in a power distribution room based on an industrial Internet of Things architecture according to claim 5, characterized in that: The sliding difference method combined with the attention mechanism is used to extract the time series features of the fast-changing running data y, including: Perform sliding window difference on the fast-changing running data y to obtain the differential sequence of the fast-changing running data y, where the differential sequence is expressed as D(y)=(D k+1 (y),D k+2 (y),...,D i (y),...,D N (y)), where i∈[k+1,N], k represents the length of the sliding window, D i (y) = y i -y i-k , where y i Represents the i-th data value in the fast-changing running data y, y i-k Represents the ikth data value in the fast-changing running data y; D k+1 (y),D k+2 (y),...,D i (y),...,D N (y) represents the difference value in the difference sequence D(y); The nonlinear transformation method is used to perform feature mapping on the difference values ​​in the difference sequence to obtain the nonlinear transformation features of the difference values; Calculate the time attention of the nonlinear transformation features, weight the nonlinear transformation features of the differential value according to the time attention, and obtain the time series features of the fast-changing running data y: Among them, D y Represents the time series characteristics of fast-changing running data y, Indicates the difference value D i (y) Temporal attention, T represents transposition, v represents the learnable mapping coefficient vector, which is used to adjust the length of the temporal feature, W att represents the learnable attention weight, b att represents the learnable attention bias, d i ((y) represents the difference value D i (y) is the nonlinear transformation characteristic.

7. The method for integrated control of multiple devices in a power distribution room based on an industrial Internet of Things architecture according to claim 1, characterized in that: The improved weighted polling method is used to sequentially receive the device characteristic information of the power distribution equipment, including: Extract the normalized operating data of the power distribution equipment and calculate the operating risk level of the normalized operating data. The higher the operating risk level, the more abnormal the operating status of the power distribution equipment. The calculation method of the operating risk level is: Among them, rate represents the operational risk level of the regularized operational data, s n represents the nth data value in the regularized running data, n∈[1,N], μ represents the mean of the regularized running data, σ represents the standard deviation of the regularized running data, s j represents the jth data value in the regularized running data, s j-1 Represents the j-1th data value in the regularized running data; Combined with the degree of operational risk, calculate the number of times the device characteristic information of the distribution equipment is received within a scheduling cycle: Among them, count min Indicates the preset minimum number of times, count max Indicates the preset maximum number of times, set count min is 1, count max is 5, and Sum represents the sum of the operating risk levels of all distribution equipment.

8. The method for integrated control of multiple devices in a power distribution room based on an industrial Internet of Things architecture according to claim 7, characterized in that: The device characteristic information is mapped based on fuzzy logic rules to generate the control parameters of the power distribution equipment, including: Extracting the mean and standard deviation of the device characteristic information, inputting the mean and standard deviation into the triangular membership function and the Gaussian membership function respectively, using the triangular membership function to output the membership of the device characteristic information in low, medium, and high operating states, extracting the operating state with the highest membership, where the motion state includes a high operating state, a medium operating state, and a low operating state, and using the Gaussian membership function to output the fluctuation membership value of the device characteristic information, where a higher fluctuation membership value indicates a smaller operating fluctuation of the device characteristic information, using a clustering algorithm to cluster the fluctuation membership values ​​of the received device characteristic information, and dividing the device characteristic information received in real time into the closest clusters according to the fluctuation membership values, where the clusters include high fluctuation, medium fluctuation, and low fluctuation; The corresponding control parameters of the extracted operating state with the highest membership and the cluster to which it belongs are extracted according to fuzzy logic rules and used as the control parameters of the distribution equipment to control and adjust the operating state of the distribution equipment. The fuzzy logic rules are the correspondence rules between different types of operating states, clusters and control parameters.

9. A multi-device integrated control system for a power distribution room based on the industrial Internet of Things architecture, characterized in that: The multi-device integrated control system for power distribution rooms based on the industrial Internet of Things architecture includes an auxiliary module, a time sequence feature extraction module, and a control module: The auxiliary module includes an acquisition module, a protocol conversion module and an edge gateway. The power distribution equipment is connected to the multi-device integrated control system of the power distribution room through the edge acquisition module. The edge acquisition module is used to collect the operating data of the power distribution equipment, and the collected operating data is regularized through the protocol conversion module. The regularized operating data is uploaded to the multi-device integrated control system of the power distribution room through the edge gateway; The time series feature extraction module is used to use an improved LSTM network and a sliding difference method combined with an attention mechanism to respectively extract the time series features of the slow-changing operation data and the fast-changing operation data as the device feature information of the associated power distribution equipment; The control module is used to sequentially receive device characteristic information of the power distribution equipment using an improved weighted polling method, perform mapping processing on the device characteristic information based on fuzzy logic rules, generate control parameters of the power distribution equipment, and perform integrated control of multiple power distribution equipment in the power distribution room; To realize the multi-device integrated control method of a power distribution room based on the industrial Internet of Things architecture as described in any one of claims 1-8.

Images