Intelligent interlocking management and control system and method based on Internet of Things

Abstract

The invention discloses an intelligent interlocking management and control system and method based on the Internet of Things, and relates to the technical field of interlocking management and control. A distributed sensor network is used for collecting on-site multi-dimensional data in real time, and real-time and dynamic adjustment of traditional static interlocking logic is realized through edge processing and weight fusion of a cloud deep learning model; the problem that a traditional scheme is slow in response under environment fluctuation, equipment aging and abnormal states is effectively solved; a virtual model is constructed through a digital twinning technology, and control parameters are continuously iteratively optimized by adopting a block gradient descent algorithm, so that hidden faults can be captured in time, accurate error correction can be generated, and self-adaptive evolution of interlocking logic is realized; in addition, the system forms a closed-loop feedback network, the information island phenomenon is eliminated, and the safety, the stability and the maintenance efficiency of the whole system are remarkably improved.

Description

Technical Field

[0001] The present invention relates to the technical field of interlocking control, and in particular to an intelligent interlocking control system and method based on the Internet of Things. Background Art

[0002] In intelligent interlocking control, common self-locking, interlocking and interlocking designs are mostly based on preset electrical logic and hardware circuits, and play an important role in working conditions such as motor forward and reverse rotation, sequential start and stop, etc.

[0003] However, as the production environment and equipment operating conditions continue to change, the control logic becomes too rigid when faced with equipment aging, power supply fluctuations, or ambient temperature changes. The static design makes it difficult to respond quickly when anomalies first appear, resulting in missed early warning opportunities. Secondly, traditional solutions have limited diagnostic capabilities for subtle faults, relying solely on manual inspections or alarm devices. They are unable to capture hidden dangers hidden in tiny signal fluctuations, causing potential faults to gradually accumulate and affect overall safety. Therefore, an intelligent interlocking control system and method based on the Internet of Things is urgently needed to solve such problems. Summary of the Invention

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

[0005] The present invention provides an intelligent interlocking management and control system and method based on the Internet of Things to solve the problem that traditional static intelligent interlocking management and control solutions are difficult to respond quickly when an abnormality occurs, thus missing the opportunity for early warning.

[0006] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0007] In a first aspect, an embodiment of the present invention provides an intelligent interlocking management and control system based on the Internet of Things, which includes:

[0008] A distributed sensor network, which includes sensors arranged on interlocking nodes, including current, temperature, vibration and environmental parameter sensors, for real-time collection of field status data;

[0009] Edge data processing unit, used to pre-process and extract preliminary features of collected data, and transmit the processing results to the central data management platform;

[0010] The IoT communication module is interconnected with the edge data processing unit and the central data management platform to form a closed loop of data transmission;

[0011] A central data management platform, which includes a cloud-based data processing unit and an artificial intelligence algorithm unit. The artificial intelligence algorithm unit performs pattern recognition and predictive analysis on the interlocking control logic based on historical data and real-time feedback to form a dynamic adjustment plan;

[0012] The dynamic interlocking reconstruction unit is used to receive the adjustment plan and send the adjusted interlocking logic to each control terminal through the control bus, thus realizing the dynamic reconstruction of the traditional interlocking logic.

[0013] The control feedback unit is used to monitor the execution of the adjusted control logic and transmit feedback information to the central data management platform.

[0014] As a preferred solution of the intelligent interlocking control system based on the Internet of Things described in the present invention, the artificial intelligence algorithm unit adopts deep learning and reinforcement learning models to predict abnormal conditions, and automatically generates interlocking logic adjustment parameters based on feedback data to achieve dynamic evolution of interlocking logic.

[0015] In a second aspect, the present invention provides an intelligent interlocking control method based on the Internet of Things, comprising:

[0016] Step S1: deploy a distributed sensor network at the interlocking node to collect on-site status data in real time;

[0017] Step S2: preprocessing and feature extraction of the collected raw data, and transmitting the processed data to the central data management platform;

[0018] Step S3: On the central data management platform, historical data and real-time data are integrated, and pattern recognition and predictive analysis of interlocking logic are performed to generate dynamic adjustment plans for changes in on-site working conditions;

[0019] Step S4: transmitting the dynamic adjustment scheme to the dynamic interlocking reconfiguration unit, which updates the preset self-locking, interlocking and interlocking control logic in real time according to the scheme, and sends the updated control logic to each control terminal;

[0020] Step S5: Implement the updated control logic, monitor the execution status through the control feedback unit, and collect feedback data in real time;

[0021] Step S6: Use digital twin technology to build a twin model of the management and control system, compare it with the actual operation data in real time, and calibrate and verify the effect of the dynamic adjustment plan;

[0022] Step S7: Continuously optimize interlocking control parameters based on the feedback data and comparison results.

[0023] As a preferred solution of the intelligent interlocking control method based on the Internet of Things described in the present invention, in which: in step S3, historical data and real-time data are integrated on the central data management platform, and the interlocking logic is subjected to pattern recognition and predictive analysis using a deep learning model, and finally a dynamic adjustment plan is generated for changes in on-site working conditions.

[0024] As a preferred solution of the intelligent interlocking control method based on the Internet of Things described in the present invention, in step S3, historical data and real-time data are weightedly fused to construct an integrated feature vector D t , defined as: D t =W h ·H t +W r ·R t , where D t represents the integrated feature vector at time t, W h Represents the historical data weight matrix, H t represents the historical feature vector at time t, W r represents the real-time data weight matrix, R t represents the real-time feature vector at time t;

[0025] Through the deep learning model, the integrated feature vector is subjected to pattern recognition and predictive analysis, and the prediction interlocking logic is given. The prediction formula is: in, represents the prediction interlocking logic at time t, f(·) represents the deep learning model, and θ represents the model parameters;

[0026] The update formula for θ is:

[0027]

[0028] Among them, θ i represents the i-th model parameter, represents the initial value of the i-th model parameter, η represents the learning rate, T represents the time window length, Denotes the loss function with respect to θ i The gradient, Indicates reference interlock logic;

[0029] According to the difference between the model output and the initial interlocking logic, a dynamic adjustment plan is generated through the mapping function. The mapping formula is defined as:

[0030] Among them, Δ t represents the logical adjustment parameter at time t, g(·) represents the mapping function, To predict the interlocking logic, L 0 represents the initial interlocking logic, β represents the mapping function parameter;

[0031] The updated interlocking logic calculation formula is:

[0032] L t =L 0 +Δ t ,

[0033] Among them, Lt Represents the interlocking logic after time t is updated.

[0034] As a preferred embodiment of the intelligent interlocking control method based on the Internet of Things described in the present invention, in step S4, the step of updating the preset self-locking, interlocking and interlocking control logics in real time includes: the dynamic interlocking reconstruction unit receives the dynamic adjustment plan generated in step S3, and uses weighted update to update the preset self-locking, interlocking and interlocking control logics in real time;

[0035] The update formula of the self-locking logic is:

[0036] in, Represents the self-locking logic after time t is updated, L self,0 Represents the initial self-locking logic, α self Indicates the self-locking logic adjustment weight, Δ t represents the dynamic adjustment scheme at time t;

[0037] The interlocking logic update formula is expressed as:

[0038] in, Indicates the interlock logic after time t is updated, L mut,0 Represents the initial interlock logic, α mut Indicates interlocking logic adjustment weights;

[0039] The interlocking control logic update formula is:

[0040] in, represents the interlocking control logic after time t is updated, L int,0 Represents the initial interlocking control logic, α int Indicates the interlocking logic adjustment weight.

[0041] As a preferred solution of the intelligent interlocking control method based on the Internet of Things described in the present invention, in step S4, the step of updating the preset self-locking, interlocking and interlocking control logic in real time also includes:

[0042] The three updated control logics are combined into a comprehensive logic vector, which is expressed as:

[0043]

[0044] Among them, L t represents the comprehensive update logic vector at time t;

[0045] The updated logic is converted into instructions for each control terminal using the reconstruction mapping function. The mapping formula is: t =h(Lt ;γ), where C t represents the control instruction issued to each control terminal at time t, h(·) represents the reconstruction mapping function, and γ represents the reconstruction mapping parameter;

[0046] The update formula of γ is:

[0047] γ=γ (0) +ξ,

[0048] Among them, γ (0) represents the initial mapping parameter, and ξ represents the dynamic increment of the mapping parameter.

[0049] As a preferred solution of the intelligent interlocking control method based on the Internet of Things described in the present invention, in step S6, the step of using digital twin technology to build a twin model of the control system and performing real-time comparison with the actual operation data is as follows:

[0050] A twin model is constructed through digital twin technology to simulate the state changes of the system after receiving control instructions. The twin model state formula is defined as: T t =f s (C t ; φ), where T t represents the twin model state at time t, f s (·) represents the twin model mapping function, C t represents the control instruction, φ represents the simulation parameter of the twin model;

[0051] Compare the twin model state with the actual operating state, and construct the error vector formula as follows:

[0052] ΔE t =A t -T t , where ΔE t represents the state error at time t, A t Indicates the actual operating status at time t.

[0053] As a preferred solution of the intelligent interlocking control method based on the Internet of Things described in the present invention, in step S6, the step of using digital twin technology to construct a twin model of the control system and performing real-time comparison with actual operation data also includes:

[0054] The error is corrected using the feedback function to generate the error correction value, which is expressed as follows:

[0055] Δ′=q(ΔE t ; δ), where Δ′ represents the error correction amount, q(·) represents the feedback correction function, and δ represents the feedback correction function parameter;

[0056] Integrate the error correction amount with the original dynamic adjustment plan to obtain the updated adjustment plan, which is:

[0057] in, represents the dynamic adjustment scheme after time t is updated, Δ t Indicates the original dynamic adjustment scheme.

[0058] As a preferred solution of the intelligent interlocking control method based on the Internet of Things described in the present invention, in step S7, the step of continuously optimizing the interlocking control parameters based on the feedback data and the comparison results is as follows:

[0059] The error function is constructed using the feedback data and the twin model comparison results to calculate the deviation between the actual operating state and the simulated state. The error function is defined as:

[0060]

[0061] Among them, P t represents the interlocking control parameter vector at time t, ΔE t represents the state error at time t, |·| represents the Euclidean norm of the vector,

[0062] The block gradient descent algorithm is used to iteratively update the interlocking control parameters, and the parameter update formula is set as:

[0063]

[0064] Among them, λ represents the learning rate, Represents the error function with respect to the parameter P t gradient;

[0065] Taking the derivative of the error function we get:

[0066]

[0067] Among them, p i Represents the interlocking control parameter vector P t The i-th parameter in , m is the parameter dimension;

[0068] During the optimization process, the block update mechanism divides the interlocking control parameters into several subsets, calculates the gradient for each subset and updates it separately to improve the update efficiency and system response speed. This process adopts the block gradient descent algorithm, that is, the parameter block is updated at each moment to ensure continuous iterative optimization based on system feedback. The updated interlocking control parameters are transmitted to the digital twin model and control terminal through closed-loop feedback to achieve real-time correction and continuous optimization.

[0069] The beneficial effects of the present invention are as follows: the present invention utilizes a distributed sensor network to collect multi-dimensional data on site in real time, and through edge processing and weighted fusion with a cloud-based deep learning model, realizes real-time and dynamic adjustment of traditional static interlocking logic, effectively overcoming the problem of slow response of traditional solutions under environmental fluctuations, equipment aging and abnormal conditions; by constructing a virtual model through digital twin technology, and using a block gradient descent algorithm to continuously iterate and optimize control parameters, the solution can capture hidden faults in a timely manner and generate accurate error correction amounts, thereby realizing the adaptive evolution of interlocking logic; in addition, the system constitutes a closed-loop feedback network, eliminates the phenomenon of information islands, and significantly improves the safety, stability and maintenance efficiency of the overall system. BRIEF DESCRIPTION OF THE DRAWINGS

[0070] 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 creative work.

[0071] Figure 1 This is a schematic diagram of the framework of the intelligent interlocking control system based on the Internet of Things in Example 1.

[0072] Figure 2 This is a flow chart of the intelligent interlocking control method based on the Internet of Things in Example 1. DETAILED DESCRIPTION

[0073] 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.

[0074] 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.

[0075] 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.

[0076] Example 1, with reference to Figure 1 and Figure 2 This embodiment provides an intelligent interlocking management and control system based on the Internet of Things, including:

[0077] A distributed sensor network, which includes sensors arranged on interlocking nodes, including current, temperature, vibration and environmental parameter sensors, for real-time collection of field status data;

[0078] Edge data processing unit, used to pre-process and extract preliminary features of collected data, and transmit the processing results to the central data management platform;

[0079] The IoT communication module is interconnected with the edge data processing unit and the central data management platform to form a closed loop of data transmission;

[0080] A central data management platform, which includes a cloud-based data processing unit and an artificial intelligence algorithm unit. The artificial intelligence algorithm unit performs pattern recognition and predictive analysis on the interlocking control logic based on historical data and real-time feedback to form a dynamic adjustment plan;

[0081] The artificial intelligence algorithm unit uses deep learning and reinforcement learning models to predict abnormal conditions and automatically generates interlocking logic adjustment parameters based on feedback data to achieve dynamic evolution of interlocking logic;

[0082] The dynamic interlocking reconstruction unit is used to receive the adjustment plan and send the adjusted interlocking logic to each control terminal through the control bus, thus realizing the dynamic reconstruction of the traditional interlocking logic.

[0083] The control feedback unit is used to monitor the execution of the adjusted control logic and transmit feedback information to the central data management platform.

[0084] This embodiment also provides a control method for the above-mentioned intelligent interlocking control system based on the Internet of Things, including:

[0085] Step S1: deploy a distributed sensor network at the interlocking node to collect on-site status data in real time;

[0086] Step S2: preprocessing and feature extraction of the collected raw data, and transmitting the processed data to the central data management platform;

[0087] Step S3: On the central data management platform, historical data and real-time data are integrated, and pattern recognition and predictive analysis of interlocking logic are performed to generate dynamic adjustment plans for changes in on-site working conditions;

[0088] In step S3, historical data and real-time data are integrated on the central data management platform, and a deep learning model is used to perform pattern recognition and predictive analysis on the interlocking logic, ultimately generating a dynamic adjustment plan for changes in on-site working conditions;

[0089] In step S3, historical data and real-time data are weighted and fused to construct an integrated feature vector D t, defined as: D t =W h ·H t +W r ·R t , where D t represents the integrated feature vector at time t, W h Represents the historical data weight matrix, H t represents the historical feature vector at time t, W r represents the real-time data weight matrix, R t represents the real-time feature vector at time t;

[0090] Through the deep learning model, the integrated feature vector is subjected to pattern recognition and predictive analysis, and the prediction interlocking logic is given. The prediction formula is: in, represents the prediction interlocking logic at time t, f(·) represents the deep learning model, and θ represents the model parameters;

[0091] The update formula for θ is:

[0092]

[0093] Among them, θ i represents the i-th model parameter, represents the initial value of the i-th model parameter, η represents the learning rate, T represents the time window length, Denotes the loss function with respect to θ i The gradient, Indicates reference interlock logic;

[0094] According to the difference between the model output and the initial interlocking logic, a dynamic adjustment plan is generated through the mapping function. The mapping formula is defined as:

[0095] Among them, Δ t represents the logical adjustment parameter at time t, g(·) represents the mapping function, To predict the interlocking logic, L 0 represents the initial interlocking logic, β represents the mapping function parameter;

[0096] The updated interlocking logic calculation formula is:

[0097] L t =L 0 +Δ t ,

[0098] Among them, L t Represents the interlocking logic after time t is updated;

[0099] Specifically, weighted fusion is used to obtain an integrated feature vector, integrating historical information and real-time status, allowing the model to capture the inherent correlation of the data. A deep learning model is used to identify patterns, and a dynamic parameter update mechanism is used to optimize model performance, providing a prediction basis for the interlocking logic. The mapping function compares the prediction results with the initial logic and generates adjustment parameters for changes in on-site working conditions, making the entire control system adaptive.

[0100] Step S4: The dynamic adjustment plan is transmitted to the dynamic interlocking reconstruction unit, which updates the preset self-locking, interlocking and interlocking control logic in real time according to the plan, and sends the updated control logic to each control terminal;

[0101] In step S4, the step of updating the preset self-locking, interlocking and interlocking control logics in real time includes: the dynamic interlocking reconstruction unit receives the dynamic adjustment plan generated in step S3, and uses weighted update to update the preset self-locking, interlocking and interlocking control logics in real time;

[0102] The update formula of the self-locking logic is:

[0103] in, Represents the self-locking logic after time t is updated, L self,0 Represents the initial self-locking logic, α self Indicates the self-locking logic adjustment weight, Δ t represents the dynamic adjustment scheme at time t;

[0104] The interlocking logic update formula is expressed as:

[0105] in, Indicates the interlock logic after time t is updated, L mut,0 Represents the initial interlock logic, α mut Indicates interlocking logic adjustment weights;

[0106] The interlocking control logic update formula is:

[0107] in, represents the interlocking control logic after time t is updated, L int,0 Represents the initial interlocking control logic, α int Indicates the interlocking logic adjustment weight;

[0108] In step S4, the step of updating the preset self-locking, interlocking and interlocking control logic in real time also includes:

[0109] The three updated control logics are combined into a comprehensive logic vector, which is expressed as:

[0110]

[0111] Among them, L t represents the comprehensive update logic vector at time t;

[0112] The updated logic is converted into instructions for each control terminal using the reconstruction mapping function. The mapping formula is: t =h(L t ;γ), where C t represents the control instruction issued to each control terminal at time t, h(·) represents the reconstruction mapping function, and γ represents the reconstruction mapping parameter;

[0113] The update formula of γ is:

[0114] γ=γ (0) +ξ,

[0115] Among them, γ (0) represents the initial mapping parameter, ξ represents the dynamic increment of the mapping parameter,

[0116] Specifically, a weighted update mechanism is used here to adjust the self-locking, interlocking, and interlocking control logics in real time. The reconstructed mapping function effectively converts the updated logic into specific control instructions, ensuring that the updated instructions can be accurately delivered to each control terminal.

[0117] Step S5: Implement the updated control logic, monitor the execution status through the control feedback unit, and collect feedback data in real time;

[0118] Step S6: Use digital twin technology to build a twin model of the management and control system, compare it with the actual operation data in real time, and calibrate and verify the effect of the dynamic adjustment plan;

[0119] In step S6, the digital twin technology is used to build a twin model of the management and control system, and the steps of real-time comparison with the actual operation data are as follows:

[0120] A twin model is constructed through digital twin technology to simulate the state changes of the system after receiving control instructions. The twin model state formula is defined as: T t =f s (C t ; φ), where T t represents the twin model state at time t, f s (·) represents the twin model mapping function, C t represents the control instruction, φ represents the simulation parameter of the twin model;

[0121] Compare the twin model state with the actual operating state, and construct the error vector formula as follows:

[0122] ΔE t =At -T t , where ΔE t represents the state error at time t, A t Indicates the actual operating status at time t;

[0123] In step S6, the steps of using digital twin technology to build a twin model of the management and control system and performing real-time comparison with the actual operation data also include:

[0124] The error is corrected using the feedback function to generate the error correction value, which is expressed as follows:

[0125] Δ′=q(ΔE t ; δ), where Δ′ represents the error correction amount, Q(·) represents the feedback correction function, and δ represents the feedback correction function parameter;

[0126] Integrate the error correction amount with the original dynamic adjustment plan to obtain the updated adjustment plan, which is:

[0127] in, represents the dynamic adjustment scheme after time t is updated, Δ t Indicates the original dynamic adjustment scheme;

[0128] Specifically, digital twin technology is used here to build a virtual model of the system to achieve real-time simulation of the actual operating status. By comparing with the actual status, the deviation between the model and the field data can be discovered in a timely manner. Then, the feedback function is used to generate error corrections and correct the dynamic adjustment plan. This process not only improves the real-time responsiveness of the plan, but also provides reliable data support for system tuning. Digital twin technology builds a closed-loop feedback mechanism between actual operation and simulation, allowing the control system to continuously learn and adapt to changes in field conditions during operation, enhancing the stability and robustness of the overall system.

[0129] Step S7, continuously optimizing interlocking control parameters based on the feedback data and comparison results;

[0130] In step S7, the steps of continuously optimizing the interlocking control parameters based on the feedback data and the comparison results are as follows:

[0131] The error function is constructed using the feedback data and the twin model comparison results to calculate the deviation between the actual operating state and the simulated state. The error function is defined as:

[0132]

[0133] Among them, P t represents the interlocking control parameter vector at time t, ΔE trepresents the state error at time t, |·| represents the Euclidean norm of the vector,

[0134] The block gradient descent algorithm is used to iteratively update the interlocking control parameters, and the parameter update formula is set as:

[0135]

[0136] Among them, λ represents the learning rate, Represents the error function with respect to the parameter P t gradient;

[0137] Taking the derivative of the error function we get:

[0138]

[0139] Among them, p i Represents the interlocking control parameter vector P t The i-th parameter in , m is the parameter dimension;

[0140] During the optimization process, the block update mechanism divides the interlocking control parameters into several subsets, calculates the gradient for each subset and updates it separately to improve update efficiency and system response speed. This process uses a block gradient descent algorithm, that is, the parameter block is updated at each moment to ensure continuous iterative optimization based on system feedback. The updated interlocking control parameters are transmitted to the digital twin model and control terminal through closed-loop feedback, realizing real-time correction and continuous optimization.

[0141] Specifically, by constructing an error function, the actual operating data is quantitatively compared with the state of the digital twin model to accurately reflect the system deviation. The block gradient descent algorithm is used to realize the block iterative update of the interlocking control parameters, making the optimization process have high computational efficiency and real-time response capability.

[0142] 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. An intelligent interlocking control system based on the Internet of Things, characterized by: include, A distributed sensor network, which includes sensors arranged on interlocking nodes, including current, temperature, vibration and environmental parameter sensors, for real-time collection of field status data; Edge data processing unit, used to pre-process and extract preliminary features of collected data, and transmit the processing results to the central data management platform; The IoT communication module is interconnected with the edge data processing unit and the central data management platform to form a closed loop of data transmission; A central data management platform, which includes a cloud-based data processing unit and an artificial intelligence algorithm unit. The artificial intelligence algorithm unit performs pattern recognition and predictive analysis on the interlocking control logic based on historical data and real-time feedback to form a dynamic adjustment plan; The dynamic interlocking reconfiguration unit is used to receive the adjustment plan and send the adjusted interlocking logic to each control terminal via the control bus; The control feedback unit is used to monitor the execution of the adjusted control logic and transmit feedback information to the central data management platform.

2. The intelligent interlocking control system based on the Internet of Things according to claim 1, characterized in that: The artificial intelligence algorithm unit adopts deep learning and reinforcement learning models to predict abnormal conditions and automatically generates interlocking logic adjustment parameters based on feedback data.

3. An intelligent interlocking control method based on the Internet of Things, based on the intelligent interlocking control system based on the Internet of Things according to any one of claims 1 to 2, characterized in that: include: Step S1: deploy a distributed sensor network at the interlocking node to collect on-site status data in real time; Step S2: preprocessing and feature extraction of the collected raw data, and transmitting the processed data to the central data management platform; Step S3: On the central data management platform, historical data and real-time data are integrated, and pattern recognition and predictive analysis of interlocking logic are performed to generate dynamic adjustment plans for changes in on-site working conditions; Step S4: transmitting the dynamic adjustment scheme to the dynamic interlocking reconfiguration unit, which updates the preset self-locking, interlocking and interlocking control logic in real time according to the scheme, and sends the updated control logic to each control terminal; Step S5: Implement the updated control logic, monitor the execution status through the control feedback unit, and collect feedback data in real time; Step S6: Use digital twin technology to build a twin model of the management and control system, compare it with the actual operation data in real time, and calibrate and verify the effect of the dynamic adjustment plan; Step S7: Continuously optimize interlocking control parameters based on the feedback data and comparison results.

4. The method for intelligent interlocking control based on the Internet of Things according to claim 3, characterized in that: In step S3, historical data and real-time data are integrated on the central data management platform, and a deep learning model is used to perform pattern recognition and predictive analysis on the interlocking logic, ultimately generating a dynamic adjustment plan for changes in on-site working conditions.

5. The method for intelligent interlocking control based on the Internet of Things according to claim 4, characterized in that: In step S3, historical data and real-time data are weighted and fused to construct an integrated feature vector D t , defined as: D t =W h ·H t +W r ·R t , where D t represents the integrated feature vector at time t, W h Represents the historical data weight matrix, H t represents the historical feature vector at time t, W r represents the real-time data weight matrix, R t represents the real-time feature vector at time t; Through the deep learning model, the integrated feature vector is subjected to pattern recognition and predictive analysis, and the prediction interlocking logic is given. The prediction formula is: in, represents the prediction interlocking logic at time t, f(·) represents the deep learning model, and θ represents the model parameters; The update formula for θ is: Among them, θ i represents the i-th model parameter, represents the initial value of the i-th model parameter, η represents the learning rate, T represents the time window length, Denotes the loss function with respect to θ i The gradient, Indicates reference interlock logic; According to the difference between the model output and the initial interlocking logic, a dynamic adjustment plan is generated through the mapping function. The mapping formula is defined as: Among them, Δ t represents the logical adjustment parameter at time t, g(·) represents the mapping function, To predict the interlocking logic, L 0 represents the initial interlocking logic, β represents the mapping function parameter; The updated interlocking logic calculation formula is: L t =L 0 +D t , Among them, L t Represents the interlocking logic after time t is updated.

6. The method for intelligent interlocking control based on the Internet of Things according to claim 3, characterized in that: In step S4, the step of updating the preset self-locking, interlocking and interlocking control logics in real time includes: the dynamic interlocking reconstruction unit receives the dynamic adjustment plan generated in step S3, and uses weighted update to update the preset self-locking, interlocking and interlocking control logics in real time; The update formula of the self-locking logic is: in, Represents the self-locking logic after time t is updated, L self,0 Represents the initial self-locking logic, α self Indicates the self-locking logic adjustment weight, Δ t represents the dynamic adjustment scheme at time t; The interlocking logic update formula is expressed as: in, Indicates the interlock logic after time t is updated, L mut,0 Represents the initial interlock logic, α mut Indicates interlocking logic adjustment weights; The interlocking control logic update formula is: in, represents the interlocking control logic after time t is updated, L int,0 Represents the initial interlocking control logic, α int Indicates the interlocking logic adjustment weight.

7. The method for intelligent interlocking control based on the Internet of Things according to claim 3, characterized in that: In step S4, the step of updating the preset self-locking, interlocking and interlocking control logic in real time also includes: The three updated control logics are combined into a comprehensive logic vector, which is expressed as: Among them, L t represents the comprehensive update logic vector at time t; The updated logic is converted into instructions for each control terminal using the reconstruction mapping function. The mapping formula is: t =h(L t ;γ), where C t represents the control instruction issued to each control terminal at time t, h(·) represents the reconstruction mapping function, and γ represents the reconstruction mapping parameter; The update formula of γ is: c = c (0) +ξ, Among them, γ (0) represents the initial mapping parameter, and ξ represents the dynamic increment of the mapping parameter.

8. The method for intelligent interlocking control based on the Internet of Things according to claim 3, characterized in that: In step S6, the steps of using digital twin technology to build a twin model of the management and control system and performing real-time comparison with the actual operation data are as follows: A twin model is constructed through digital twin technology to simulate the state changes of the system after receiving control instructions. The twin model state formula is defined as: T t =f s (C t ; φ), where T t represents the twin model state at time t, f s (·) represents the twin model mapping function, C t represents the control instruction, φ represents the simulation parameter of the twin model; Compare the twin model state with the actual operating state, and construct the error vector formula as follows: ΔE t =A t -T t , where ΔE t represents the state error at time t, A t Indicates the actual operating status at time t.

9. The method for intelligent interlocking control based on the Internet of Things according to claim 8, characterized in that: In step S6, the step of using digital twin technology to build a twin model of the management and control system and performing real-time comparison with actual operation data also includes: The error is corrected using the feedback function to generate the error correction value, which is expressed as follows: Δ′=q(ΔE t ; δ), where Δ′ represents the error correction amount, q(·) represents the feedback correction function, and δ represents the feedback correction function parameter; Integrate the error correction amount with the original dynamic adjustment plan to obtain the updated adjustment plan, which is: in, represents the dynamic adjustment scheme after time t is updated, Δ t Indicates the original dynamic adjustment scheme.

10. The method for intelligent interlocking control based on the Internet of Things according to claim 9, characterized in that: In step S7, the steps of continuously optimizing the interlocking control parameters based on the feedback data and the comparison results are as follows: The error function is constructed using the feedback data and the twin model comparison results to calculate the deviation between the actual operating state and the simulated state. The error function is defined as: Among them, P t represents the interlocking control parameter vector at time t, ΔE t represents the state error at time t, |·| represents the Euclidean norm of the vector, The block gradient descent algorithm is used to iteratively update the interlocking control parameters, and the parameter update formula is set as: Among them, λ represents the learning rate, Represents the error function with respect to the parameter P t gradient; Taking the derivative of the error function we get: Among them, p i Represents the interlocking control parameter vector P t The i-th parameter in , m is the parameter dimension; During the optimization process, the block update mechanism divides the interlocking control parameters into several subsets, and calculates and updates the gradient for each subset separately. This process adopts the block gradient descent algorithm.

Images