A traffic network resilience fuzzy control method based on adaptive experience of distribution rate

Abstract

The application discloses a traffic network resilience fuzzy control method based on adaptive experience of distribution rate, and relates to the technical field of intelligent control of traffic network; a nonlinear traffic network comprehensive dynamics model integrating multiple factors is constructed, a segmented nonlinear distribution rate is designed in combination with operation and maintenance experience, a state deviation variable is defined to convert the comprehensive dynamics model into a deviation state equation, nonlinear characteristic terms of the traffic network system are extracted for T-S fuzzy modeling, parallel distribution compensation PDC fuzzy control rules and duplex working condition closed-loop mechanisms corresponding to the T-S fuzzy model are designed, fuzzy controller gains are solved, and the performance of the traffic network system is verified, and the obtained controller gains are implanted into the traffic network control system to adjust the distribution rate and the inventory state, so that the resilience collaborative operation of the traffic network under disturbance risk is realized; the traffic network resilience fuzzy control method based on adaptive experience of distribution rate is adopted, and the universality is higher, and the adaptive capacity of the distribution rate is greatly improved.

Description

Technical Field

[0001] This invention relates to the field of intelligent control technology for transportation networks, and in particular to a resilient fuzzy control method for transportation networks based on delivery rate adaptive experience. Background Technology

[0002] Current technologies generally apply classical control theories such as optimal control, robust control, sliding mode control, and model predictive control to the coordinated regulation of transportation network operations. Early studies mostly focused on dynamic analysis and controller design based on linear transportation network models. However, considering the nonlinear characteristics of real-world transportation networks, some studies have combined single transport time delays and the bullwhip effect to design order flow regulation strategies, while others have adopted… Feedback control suppresses network oscillations. Meanwhile, the industry has begun designing delivery rate functions based on industrial operation and maintenance experience to regulate inventory fluctuations. Fuzzy modeling in Transmission System (TS) is the mainstream method for handling nonlinear systems; a few studies have introduced it into the transportation network field, but only for multi-model conversion of network structures, without combining it with actual operation and maintenance experience to optimize control logic. Traditional delivery rates are all based on safety stock as a benchmark for setting adjustment rules, passively adjusting according to the basic experience of "increasing the delivery rate when inventory is below safety stock and decreasing the delivery rate when inventory is above safety stock."

[0003] Existing transportation network dynamics models only consider single transport delays or single inventory losses, failing to simultaneously integrate three core operational factors: multiple transport delays, behavioral losses, and the propagation of intermittent interruptions. This makes them unable to accurately reflect the dynamic evolution and interruption propagation patterns of real transportation networks, resulting in poor model versatility. Traditional delivery rate adjustments use safety stock as the benchmark, neglecting to incorporate operational experience based on steady-state inventory characteristics of the transportation network, leading to weak adaptive adjustment capabilities. Furthermore, existing solutions do not integrate delivery rate operational experience into closed-loop feedback control, underutilizing its value. Conventional nonlinear control and traditional fuzzy control fail to incorporate adaptive delivery rate experience into controller design. When facing complex conditions such as sudden interruptions and large transport delays, the coordinated efficiency of the transportation network drops significantly, exhibiting weak resilience against interruptions. Existing control methods do not differentiate between normal operation and interruption / failure phases, employing a uniform control logic. When interruptions propagate through the network, problems such as drastic inventory fluctuations and system instability easily arise. Summary of the Invention

[0004] The purpose of this invention is to provide a resilient fuzzy control method for transportation networks based on adaptive delivery rate experience. This method integrates three factors: behavioral losses, multiple transport delays, and the propagation of intermittent interruptions, into a nonlinear hybrid dynamics modeling method for transportation networks. It also includes a dual-condition (normal / interruption) partitioning rule and a piecewise adaptive empirical nonlinear delivery rate design method based on inventory steady-state deviation, which differs from traditional delivery rules based on safety stock benchmarks. Furthermore, it employs a system fuzzy modeling method that embeds adaptive delivery rate operation and maintenance experience into the TS fuzzy antecedent, achieving a deep fusion of experience and fuzzy control. The invention also includes a piecewise TS fuzzy active closed-loop control architecture matching the dual-conditions of the transportation network, and a PDC parallel distributed compensation control rule design. Finally, it addresses the fuzzy controller gain solution, system stability, and other aspects based on Lyapunov-Krasovskii functionals and linear matrix inequalities. Performance evaluation method.

[0005] To achieve the above objectives, this invention provides a traffic network resilience fuzzy control method based on delivery rate adaptive experience, comprising the following steps: S1. Construct a comprehensive dynamic model of a nonlinear transportation network that integrates multiple factors; S2. Based on the model of S1 and combined with operation and maintenance experience, a segmented nonlinear delivery rate is designed. S3. Define state deviation variables and convert the comprehensive dynamic model of S1 into deviation state equations. Extract nonlinear characteristic terms of the traffic network control system to perform TS fuzzy modeling. S4, corresponding to S3, TS fuzzy model design, parallel distributed compensation PDC fuzzy control rules and dual-condition closed-loop mechanism; S5. Based on S4, solve the corresponding fuzzy controller gain under the fuzzy control rule using Lyapunov-Krasovskii and verify the performance of the traffic network system. S6. The controller gain obtained from S5 is incorporated into the traffic network control system to adjust the delivery rate and inventory status, thereby achieving resilient and coordinated operation of the traffic network under disturbance risks.

[0006] Preferably, the specific content of S1 is as follows: S11. For a transportation network containing N network nodes, combining six key factors—real-time inventory, delivery rate, market demand, behavioral losses, multiple transportation delays, and intermittent interruptions—establish dynamic equations for node inventory and evolution equations for delivery rate. The specific process of establishing the dynamic equations for node inventory is as follows: ; in, For nodes Real-time inventory, For delivery rate, As for the intensity of market demand, This is the inventory behavior loss coefficient. Assigning proportions of goods For time-varying transport delays between nodes, This is the upper bound of the delay; The specific process of the delivery rate evolution equation is as follows: ; in, For ideal time-varying delivery rate, For adaptive convergence time of delivery rate, To actively control the input amount; S12. Define disturbance state identifier quantities, divide the traffic network control system into two types of operating conditions: recovery phase and interruption fault phase, and form a unified nonlinear traffic network hybrid dynamics model.

[0007] Preferably, the specific details of the working conditions division in S12 are as follows: Through the disturbance state indicator Differentiate working conditions, During the recovery phase of the traffic network control system, full-chain cargo transportation and feedback control will be initiated. During the interruption fault phase, the node suspends cargo outbound transportation and cuts off local feedback signals.

[0008] Preferably, the specific content of S2 is as follows: S21. Based on the deviation between real-time inventory and steady-state inventory, and combined with operational experience, design a segmented nonlinear delivery rate. This replaces the traditional delivery rate based on safety stock. The specific adjustment rules for S22 and S21 are as follows: When the real-time inventory is lower than the steady-state inventory, that is... An exponential delivery rate function is used to increase the volume of goods delivered; when the real-time inventory is higher than the steady-state inventory, A fractional nonlinear function is used to gradually reduce the volume of goods delivered; among which, For inventory steady-state deviation, For steady-state inventory at nodes; S23. Design of a segmented nonlinear delivery rate based on adaptive experience of delivery rate. The specific process is as follows: ; in, , , , These are adjustable design parameters.

[0009] Preferably, the specific content of S3 is as follows: S31. Define the state deviation variable. , The comprehensive dynamic model of S1 is transformed into a deviation state equation, and the nonlinear characteristic terms of the traffic network control system are extracted. ; S32. Applying TS fuzzy rules to nonlinear terms Linear decomposition is performed, and fuzzy subsets are divided in combination with inventory deviation thresholds. Multiple fuzzy model rules are established, and the adaptive experience of delivery rate is embedded into the fuzzy antecedent. S33. Using the average defuzzification method, the global state space expression is derived to complete the fuzzy modeling of the nonlinear traffic network.

[0010] Preferably, the specific content of S4 is as follows: S41. Corresponding to the TS fuzzy model rules, design PDC fuzzy control rules of the same dimension, using inventory deviation as the judgment condition, and output active control input quantities under different operating conditions. Construct a fuzzy closed-loop control architecture; S42. Based on the two types of working conditions divided in S12, establish the closed-loop dynamic equations respectively: During normal operation , Then, the inter-node transport links and all fuzzy feedback control terms are retained, among which, It is an integer. The duration of the intermittent cycle. This represents the dividing point between the start of an interruption and the end of a recovery within each cycle; During the interruption fault phase , Then, remove cross-node transport items and retain only the local system state and disturbance items. It is an integer. The duration of the intermittent cycle. This represents the dividing point between the start of an interruption and the end of a recovery within each cycle; The closed-loop dynamic equations are as follows: ; in, For fuzzy weighting functions, For the system matrix, The fuzzy controller gain to be solved, and .

[0011] Preferably, the specific working content of S5 is as follows: Based on Lyapunov-Krasovskii functionals, Jensen inequalities, and the mutual convexity method, combined with the linear matrix inequality (LMI), the conditions for solving the controller gain are derived; the exponential stability of the traffic network control system is determined. The disturbance rejection performance is obtained by solving for the feedback gain matrix corresponding to each set of fuzzy rules. .

[0012] Preferably, the specific working content of S6 is as follows: The controller gain obtained by solving S5 Embedded in the traffic network control system, it collects real-time data on inventory, delivery rate, and disturbance status at each node, and dynamically generates active control inputs based on fuzzy rules. Adjusting delivery rates and inventory status enables resilient and coordinated operation of the transportation network under disturbance risks.

[0013] Therefore, the present invention employs a traffic network resilience fuzzy control method based on delivery rate adaptive experience, which has the following advantages compared with the prior art: 1. This application has greater versatility, integrating behavioral losses, multiple transport delays, and intermittent disturbance propagation into the dynamic model of the transportation network. This fully reflects the operational characteristics of actual transportation networks, solving the problem of poor scenario adaptability in traditional models. Furthermore, it rigorously proves the exponential stability of the system using Lyapunov-Krasovskii theory. The controller has good disturbance rejection performance and its gain can be directly solved using linear matrix inequalities, making it easy to implement in engineering. 2. The adaptive capability of the delivery rate is greatly improved. The steady-state inventory replaces the traditional safety stock as the adjustment benchmark, which expands the scope of application of operation and maintenance experience. The segmented nonlinear delivery rate can accurately match the inventory fluctuation pattern, and the inventory adjustment is smoother. The utilization rate of experience is maximized. The operation and maintenance adaptive experience of the delivery rate is embedded into the TS fuzzy closed-loop feedback control, which breaks through the limitation of the traditional solution that "experience is only used for passive adjustment and does not participate in the control design" and improves the rationality of the control logic. 3. Excellent disturbance resistance and improved collaborative efficiency: The segmented control architecture, which differentiates between normal and interrupted operating conditions, maintains system stability even under complex conditions such as interruption propagation, large transportation delays, and high inventory losses. Inventory fluctuations are significantly smaller than those of traditional methods. Fuzzy active control can quickly correct production and inventory deviations caused by disturbances, ensuring the coordinated operation of all network nodes, effectively suppressing network fluctuations, and improving overall operational efficiency.

[0014] The technical solution of the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. Attached Figure Description

[0015] Figure 1 This is a traffic network structure diagram of a traffic network resilience fuzzy control method based on delivery rate adaptive experience according to the present invention; Figure 2 This is a comparison chart of adaptive experience delivery rate and traditional delivery rate in a traffic network resilience fuzzy control method based on adaptive delivery rate of the present invention. Figure 3 This is a case study of the hydrogen energy supply chain network of the Northwest Power Grid HRP-38 power system, which is used to test a transportation network resilience fuzzy control method based on delivery rate adaptive experience according to the present invention. Figure 4 This is a regional collaborative evolution state diagram of a traffic network resilience fuzzy control method based on delivery rate adaptive experience according to the present invention. Figure 5 This is a comparative experimental result diagram of various cases of a traffic network resilience fuzzy control method based on delivery rate adaptive experience according to the present invention. Detailed Implementation

[0016] In the description of this invention, it should be noted that the terms "upper," "lower," "inner," "outer," etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings, or the orientation or positional relationship in which the product of this invention is usually placed when in use. They are only for the convenience of describing this invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limiting this invention.

[0017] Example like Figures 1-5 As shown, the present invention provides a traffic network resilience fuzzy control method based on delivery rate adaptive experience, comprising the following steps: S1. Construct a comprehensive dynamic model of a nonlinear transportation network that integrates multiple factors; The specific content of S1 is as follows: S11. For a transportation network containing N network nodes, combining six key factors—real-time inventory, delivery rate, market demand, behavioral losses, multiple transportation delays, and intermittent interruptions—establish dynamic equations for node inventory and evolution equations for delivery rate. The specific process of establishing the dynamic equations for node inventory is as follows: ; in, For nodes Real-time inventory, For delivery rate, As for the intensity of market demand, This is the inventory behavior loss coefficient. Assigning proportions of goods For time-varying transport delays between nodes, This is the upper bound of the delay; The specific process of the delivery rate evolution equation is as follows: ; in, For ideal time-varying delivery rate, For adaptive convergence time of delivery rate, To actively control the input amount; S12. Define disturbance state identifier quantities, divide the traffic network control system into two types of operating conditions: recovery phase (normal operation) and interruption fault phase, and form a unified nonlinear traffic network hybrid dynamics model.

[0018] The specific details of the working conditions in S12 are as follows: Through the disturbance state indicator Differentiate working conditions, During the recovery (normal operation) phase of the traffic network control system, full-chain freight transportation and feedback control are initiated; During the interruption fault phase, the node suspends cargo outbound transportation and cuts off local feedback signals.

[0019] S2. Based on the model of S1 and combined with operation and maintenance experience, a segmented nonlinear delivery rate is designed. The specific content of S2 is as follows: S21. Based on the deviation between real-time inventory and steady-state inventory, and combined with operational experience, design a segmented nonlinear delivery rate. This replaces the traditional delivery rate based on safety stock. The specific adjustment rules for S22 and S21 are as follows: When the real-time inventory is lower than the steady-state inventory, that is... An exponential delivery rate function is used to increase the volume of goods delivered; when the real-time inventory is higher than the steady-state inventory, A fractional nonlinear function is used to gradually reduce the volume of goods delivered; among which, For inventory steady-state deviation, For steady-state inventory at nodes; S23. Design of a segmented nonlinear delivery rate based on adaptive experience of delivery rate. The specific process is as follows: ; in, , , , These are adjustable design parameters.

[0020] S3. Define state deviation variables and convert the comprehensive dynamic model of S1 into deviation state equations. Extract nonlinear characteristic terms of the traffic network control system to perform TS fuzzy modeling. The specific details of S3 are as follows: S31. Define the state deviation variable. (Inventory Deviation) (Delivery rate deviation) The comprehensive dynamic model of S1 is transformed into a deviation state equation to eliminate the influence of steady-state quantities and extract the nonlinear characteristic terms of the traffic network control system. (Derived from adaptive delivery rate); S32. Applying TS fuzzy rules to nonlinear terms Linear decomposition is performed, and fuzzy subsets are divided in combination with inventory deviation thresholds. Multiple fuzzy model rules are established, and the adaptive experience of delivery rate is embedded into the fuzzy antecedent. S33. Using the average defuzzification method, the global state space expression is derived to complete the fuzzy modeling of the nonlinear traffic network.

[0021] S4, corresponding to S3, TS fuzzy model design, parallel distributed compensation PDC fuzzy control rules and dual-condition closed-loop mechanism; The specific content of S4 is as follows: S41. Corresponding to the TS fuzzy model rules, design PDC fuzzy control rules of the same dimension, using inventory deviation as the judgment condition, and output active control input quantities under different operating conditions. Construct a fuzzy closed-loop control architecture; S42. Based on the two types of working conditions divided in S12, establish the closed-loop dynamic equations respectively: During normal operation , Then, the inter-node transport links and all fuzzy feedback control terms are retained, among which, It is an integer. The duration of the intermittent cycle. This represents the dividing point between the start of an interruption and the end of a recovery within each cycle; During the interruption fault phase , Then, remove cross-node transport items and retain only the local system state and disturbance items. It is an integer. The duration of the intermittent cycle. This represents the dividing point between the start of an interruption and the end of a recovery within each cycle; The closed-loop dynamic equations are as follows: ; in, For fuzzy weighting functions, For the system matrix, The fuzzy controller gain to be solved, and ; S5. Based on S4, solve the corresponding fuzzy controller gain under the fuzzy control rule using Lyapunov-Krasovskii and verify the performance of the traffic network system. The specific tasks of S5 are as follows: Based on Lyapunov-Krasovskii functionals, Jensen inequalities, and the mutual convexity method, combined with the linear matrix inequality (LMI), the conditions for solving the controller gain are derived; the exponential stability of the traffic network control system is determined. The disturbance rejection performance is obtained by solving for the feedback gain matrix corresponding to each set of fuzzy rules. ; The specific process of solving the algorithm is as follows: For a given scalar (satisfy ), (satisfy And all of the above parameters satisfy ; at the same time, If a second-order matrix exists ,matrix as well as This makes the following true: ; , ; , ; And there exists a matrix with one row and two columns. The following conditions must be met: ; ; Among them, when Sometimes, And satisfy ; ; It is a fifth-order block symmetric matrix with the following elements: The specific expression is as follows: ; as well as: ; So when the disturbance At that time, the traffic network control system studied achieves internal exponential stability based on the Lyapunov-Krasovskii functional, with an exponential decay rate of: ; At the same time for a given ,in ; The system meets Performance metrics, and controller gain meets requirements. .

[0022] S6. The controller gain obtained from S5 is incorporated into the traffic network control system to adjust the delivery rate and inventory status, thereby achieving resilient and coordinated operation of the traffic network under disturbance risk. The specific tasks of S6 are as follows: The controller gain obtained by solving S5 Embedded in the traffic network control system, it collects real-time data on inventory, delivery rate, and disturbance status at each node, and dynamically generates active control inputs based on fuzzy rules. Adjusting delivery rates and inventory status enables resilient and coordinated operation of the transportation network under disturbance risks.

[0023] like Figure 2 As shown, the delivery rate experience in (a) is: ; ; (b) The improved delivery rate experience, i.e., the design of this application: ; ; ; ...; ; ; ... (a) Non-linear delivery rate: ; (b) Non-linear delivery rate: .

[0024] In the specific implementation process, the hydrogen energy supply chain network of the Northwest Power Grid HRP-38 power system is taken as the application object. This network contains 5 geographical regional nodes, and there are cross-regional hydrogen energy transportation and intermittent power outage propagation, which are typical nonlinear transportation networks.

[0025] Build a dynamic model of the hydrogen energy supply chain and divide the operating conditions: such as Figure 3 As shown, this hydrogen energy supply chain has 5 nodes (D1-D5). The model parameters are set as follows: cargo allocation ratio. Maximum shipping delay Inventory loss coefficient Interruption cycle Interruption duration Through status indicators Working conditions are divided as follows: During the normal operation phase, hydrogen energy is being transported normally throughout the region. During the disturbance fault phase (D1, D2, and D4 are affected by power outages), inter-regional transportation is suspended.

[0026] Configure adaptive non-linear delivery rate parameters: Set delivery rate parameters: =0.47、 , =0.3、 Inventory safety deviation threshold The adaptive delivery rate is generated according to the designed piecewise function.

[0027] State transition and TS fuzzy modeling: Define state deviation variables and Based on inventory deviation and threshold Four rules for defining the TS fuzzy model: Rule 1: Inventory deviation is negative and within the safety threshold; Rule 2: Inventory deviation is negative and exceeds the safety threshold; Rule 3: Inventory deviation is positive and within the safety threshold; Rule 4: Inventory deviation is positive and exceeds the safety threshold.

[0028] Design of PDC fuzzy control rules and dual-condition closed-loop system: For the four fuzzy model rules, four PDC fuzzy control rules are designed, and closed-loop equations are built for normal / interrupted operating conditions. During the interruption phase, cross-node transportation items are removed, and only local state control is retained.

[0029] Solve for the controller gain and verify the performance: Configure performance parameters: , , , , The algorithm based on the design yields four sets of fuzzy feedback gain matrices for each node. Real-time control simulation of hydrogen energy supply chain: Set time-varying demand perturbation Given the initial state, the fuzzy feedback control gain obtained by S5 is configured to each region, and the observed state fluctuations are as follows: Figure 4 As shown, the present invention can maintain real-time operational coordination among various regions, ensuring the resilience of the system.

[0030] Four sets of comparative experiments were conducted: Case 1: The delivery rate using this invention is low, and there is no active control. Case 2: Using traditional delivery methods with no proactive control; Case 3: Using traditional delivery rates and conventional controllers; Case 4: The delivery rate and fuzzy control of this invention are applied.

[0031] The system state fluctuation curves corresponding to Cases 1-4 are as follows: Figure 5 As shown in the diagram, comparing Cases 3 and 4 with Cases 1 and 2 reveals that, after applying active control, both the conventional control scheme in Case 3 and the fuzzy control scheme in Case 4 significantly outperform Cases 1 and 2 without active control in suppressing system state fluctuations, and maintain system stability, thus verifying the effectiveness and robustness of the active control strategy. Further comparison of Cases 3 and 4 shows that the fuzzy control method based on the delivery rate of this invention can further reduce system state fluctuations compared to the conventional control method using the traditional delivery rate in Case 3. Meanwhile, without applying active control, comparing Cases 1 and 2 shows that the delivery rate of this invention can still achieve fluctuation suppression. In summary, the experimental results demonstrate that the fuzzy control scheme designed in this invention possesses excellent overall control performance.

[0032] Therefore, this invention employs a traffic network resilience fuzzy control method based on adaptive delivery rate experience, integrating a nonlinear traffic network hybrid dynamics modeling method that incorporates three factors: behavioral losses, multiple transport delays, and intermittent interruption propagation; a dual-condition (normal / interruption) partitioning rule; and a piecewise adaptive empirical nonlinear delivery rate design method based on inventory steady-state deviation, which differs from traditional safety stock benchmark delivery rules. It also incorporates a system fuzzy modeling method that embeds adaptive delivery rate operation and maintenance experience into the TS fuzzy antecedent, achieving deep integration of experience and fuzzy control. Furthermore, it matches a piecewise TS fuzzy active closed-loop control architecture to the dual-condition traffic network and designs a PDC parallel distributed compensation control rule. Finally, it addresses the fuzzy controller gain solution, system stability, and... (The text abruptly ends here, so the translation stops as well.) Performance evaluation method.

[0033] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit them. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the technical solutions of the present invention, and these modifications or equivalent substitutions cannot cause the modified technical solutions to deviate from the spirit and scope of the technical solutions of the present invention.

Claims

1. A fuzzy control method for traffic network resilience based on adaptive delivery rate experience, characterized in that, Includes the following steps: S1. Construct a comprehensive dynamic model of a nonlinear transportation network that integrates multiple factors; The specific content of S1 is as follows: S11. For a transportation network containing N network nodes, combining six key factors—real-time inventory, delivery rate, market demand, behavioral losses, multiple transportation delays, and intermittent interruptions—establish dynamic equations for node inventory and evolution equations for delivery rate. The specific process of establishing the dynamic equations for node inventory is as follows: ； in, For nodes Real-time inventory, For delivery rate, As for the intensity of market demand, This represents the inventory loss coefficient. Assigning proportions of goods For time-varying transport delays between nodes, This is the upper bound of the delay; The specific process of the delivery rate evolution equation is as follows: ； in, For ideal time-varying delivery rate, For adaptive convergence time of delivery rate, To actively control the input amount; S12. Define the disturbance state identifier quantity, divide the traffic network control system into two types of operating conditions: recovery phase and interruption fault phase, and form a unified nonlinear traffic network hybrid dynamics model. The specific details of the working conditions in S12 are as follows: Through the disturbance state indicator Differentiate working conditions, During the recovery phase of the traffic network control system, full-chain cargo transportation and feedback control will be initiated. During the interruption fault phase, the node suspends cargo outbound transportation and cuts off local feedback signals; S2. Based on the model of S1 and combined with operation and maintenance experience, a segmented nonlinear delivery rate is designed. The specific content of S2 is as follows: S21. Based on the deviation between real-time inventory and steady-state inventory, and combined with operational experience, design a segmented nonlinear delivery rate. This replaces the traditional delivery rate based on safety stock. The specific adjustment rules for S22 and S21 are as follows: When the real-time inventory is lower than the steady-state inventory, that is... An exponential delivery rate function is used to increase the volume of goods delivered; when the real-time inventory is higher than the steady-state inventory, A fractional nonlinear function is used to gradually reduce the volume of goods delivered; among which, For inventory steady-state deviation, For steady-state inventory at nodes; S23. Design of a segmented nonlinear delivery rate based on adaptive experience of delivery rate. The specific process is as follows: ； in, , , , These are adjustable design parameters; S3. Define state deviation variables and convert the comprehensive dynamic model of S1 into deviation state equations. Extract nonlinear characteristic terms of the traffic network system to perform TS fuzzy modeling. S4, corresponding to S3, TS fuzzy model design, parallel distributed compensation PDC fuzzy control rules and dual-condition closed-loop mechanism; S5. Based on S4, solve the corresponding fuzzy controller gain under the fuzzy control rule using Lyapunov-Krasovskii and verify the performance of the traffic network system. S6. The controller gain obtained from S5 is incorporated into the traffic network control system to adjust the delivery rate and inventory status, thereby achieving resilient and coordinated operation of the traffic network under disturbance risks.

2. The traffic network resilience fuzzy control method based on delivery rate adaptive experience as described in claim 1, characterized in that: The specific details of S3 are as follows: S31. Define the state deviation variable. , The comprehensive dynamic model of S1 is transformed into a deviation state equation, and the nonlinear characteristic terms of the traffic network control system are extracted. ; S32. Applying TS fuzzy rules to nonlinear terms Linear decomposition is performed, and fuzzy subsets are divided in combination with inventory deviation thresholds. Multiple fuzzy model rules are established, and the adaptive experience of delivery rate is embedded into the fuzzy antecedent. S33. Using the average defuzzification method, the global state space expression is derived to complete the fuzzy modeling of the nonlinear traffic network.

3. The fuzzy control method for traffic network resilience based on adaptive delivery rate experience according to claim 2, characterized in that: The specific content of S4 is as follows: S41. Corresponding to the TS fuzzy model rules, design PDC fuzzy control rules of the same dimension, using inventory deviation as the judgment condition, and output active control input quantities under different operating conditions. Construct a fuzzy closed-loop control architecture; S42. Based on the two types of working conditions divided in S12, establish the closed-loop dynamic equations respectively: During normal operation , Then, the inter-node transport links and all fuzzy feedback control terms are retained, among which, It is an integer. The duration of the intermittent cycle. This represents the dividing point between the start of an interruption and the end of a recovery within each cycle; During the interruption fault phase , Then, remove cross-node transport items and retain only the local system state and disturbance items. It is an integer. The duration of the intermittent cycle. This represents the dividing point between the start of an interruption and the end of a recovery within each cycle; The closed-loop dynamic equations are as follows: ； in, For fuzzy weighting functions, For the system matrix, The fuzzy controller gain to be solved, and .

4. The fuzzy control method for traffic network resilience based on adaptive delivery rate experience according to claim 3, characterized in that: The specific tasks of S5 are as follows: Based on Lyapunov-Krasovskii functionals, Jensen inequalities, and the mutual convexity method, combined with the linear matrix inequality (LMI), the conditions for solving the controller gain are derived; the exponential stability of the traffic network control system is determined. The disturbance rejection performance is obtained by solving for the feedback gain matrix corresponding to each set of fuzzy rules. .

5. The fuzzy control method for traffic network resilience based on adaptive delivery rate experience according to claim 4, characterized in that: The specific tasks of S6 are as follows: The controller gain obtained by solving S5 Embedded in the traffic network control system, it collects real-time data on inventory, delivery rate, and disturbance status at each node, and dynamically generates active control inputs based on fuzzy rules. Adjusting delivery rates and inventory status enables resilient and coordinated operation of the transportation network under disturbance risks.

Images