Airborne power distribution system load management method based on MILP and RHC

Abstract

The invention discloses an airborne power distribution system load management method based on MILP and RHC. The method comprises the steps that a distributed airborne power distribution system architecture and a load connection state are equivalent to an optimal control problem to be solved; constructing an electrical load power supply request matrix, and further combining an electrical load control equation to control connection or disconnection of each load; establishing a mathematical model of a load management system by adopting a mixed integer linear programming method according to the electrical load control equation and in combination with the constraint relation of the system; aiming at a mathematical model of a load management system and a corresponding constraint relationship, a rolling time domain control method is adopted, a load management problem can be converted into a problem of solving a mixed integer linear programming problem by the mathematical model of the system under a plurality of constraint relationship conditions, and finally an optimal control sequence of a bus to a load is obtained to realize automatic load management. The method has the advantages of being high in resolving speed, flexible and suitable for a complex system with multiple constraint conditions and the like.

Description

Technical Field

[0001] The present invention belongs to the field of automatic load management of aircraft power distribution systems, and in particular relates to a load management method for an airborne power distribution system based on mixed integer linear programming (MILP) and rolling horizon control (RHC). Background Art

[0002] The aircraft power distribution system is an important part of the aircraft power supply system. It integrates the transmission and distribution, control and protection functions of electric energy. It is an interactive bridge connecting the aircraft power system and the power load. The power distribution system is responsible for distributing electric energy to various electronic loads on board to ensure the normal operation of various onboard equipment. However, various power loads have certain priorities. When there are certain defects or even failures in the power supply system, the system should allow non-essential loads to be unloaded, while the necessary loads must be powered on at all times to ensure the minimum reliability requirements for the normal operation of the aircraft. Therefore, it is particularly important to design a reliable load automatic management system in the onboard power distribution system.

[0003] At present, with the development of aircraft multi-electric / all-electric technology, traditional hydraulic energy and air pressure energy are all replaced by electrical energy, and the requirements for automatic management of load priority of airborne power distribution systems are becoming higher and higher. The power load control of multi-electric aircraft is based on the flight profile and flight mission of the aircraft, and is also affected by factors such as the operating conditions of the aircraft power supply system and the components within the system. The complexity of the current flight mission stage is constantly increasing, and the mission conditions faced by aircraft loads are also becoming increasingly complex. For example, high power consumption during flight, generator and component failures, etc. may affect the safety and reliability of the system, so the requirements for load management of the power supply system are also gradually increasing.

[0004] Existing methods are mainly based on solid-state power controllers and electrical load management centers, focusing on how to establish load management control equations and request equations to achieve related load switching at different times under a specific flight profile and achieve automatic management. However, traditional research methods use electrical load management equation control and unloading management to only achieve sequential control of a single load in the system. By converting load management into a knapsack problem for solution, the number of iterations is large and the solution speed is slow. In addition, existing methods are relatively scarce in the study of automatic load management control at the system level, lacking the accumulation of mathematical models, and can no longer meet the system's requirements for rapidity and flexibility in load management. Summary of the invention

[0005] The purpose of the present invention is to provide an airborne power distribution system load management method based on MILP and RHC, which is used to dynamically control load switching according to system operating conditions and load priorities, and is adapted to the requirements of rapidity and flexibility of automatic load management.

[0006] In order to achieve the above tasks, the present invention adopts the following technical solutions:

[0007] A load management method for an airborne power distribution system based on MILP and RHC, comprising:

[0008] The distributed airborne power distribution system architecture and load connection status are equivalent to the optimal control problem to be solved;

[0009] Determine the flight mission profile based on the power distribution demand of the aircraft, divide the flight mission profile into different load management stages, obtain the power supply request matrix of the electrical load in each load management stage, and further combine the electrical load control equation to control the connection or disconnection of each load;

[0010] According to the electrical load control equation and the system constraints, the mathematical model of the load management system is established by using the mixed integer linear programming method.

[0011] Aiming at the mathematical model of the load management system and the corresponding constraint relationship, the rolling horizon control method is adopted. First, the future state is predicted, and the model is solved by constraining the difference between the predicted state and the final target to be minimized. The load management problem can be converted into a system mathematical model to solve a mixed integer linear programming problem under multiple constraint conditions, and finally the optimal control sequence of the bus for the load is obtained to realize automatic load management.

[0012] Furthermore, the distributed airborne power distribution system architecture and load connection state are equivalent to solving an optimal control problem, including:

[0013] In the distributed airborne power distribution system, the generator and auxiliary power unit distribute power through multiple buses to provide redundancy. The connection between the generator and the power load is routed through a series of electromechanical switches called contactors. The on and off of the contactors can control the connection and unloading of the load. The relationship between different components of the system is represented by a single-line diagram to construct a connection diagram of the system component relationship;

[0014] According to the system component relationship connection diagram, each generator is set to have its output voltage adjusted by the generator control unit to meet the power required by the expected load; each contactor is controlled to be opened or closed by the bus power control unit to distribute the power from the power supply to the designated bus and load as required; thus, the load management control is equivalent to the optimal control problem of the contactor, controlling the connection switching between the generator and the bus, and the connection and unloading of the load.

[0015] Furthermore, in the load management control, the equivalent processing requirements for the generator, bus and load are as follows:

[0016] The generators are not connected in parallel; according to the architecture of the distributed power distribution system, it is assumed that there is no situation where the same bus is powered by multiple AC generators at the same time;

[0017] Bus priority: Each bus has a priority list of power generators. If the first generator in the priority list is not available at the moment, the second generator should supply power to the bus, and so on.

[0018] Load priority: Set the load priority. If the power is insufficient, the load that can be removed can be removed according to the priority principle, while the load that cannot be removed must always remain powered.

[0019] Furthermore, the flight mission profile is determined based on the power distribution demand of the aircraft, the flight mission profile is divided into different load management stages, and the power supply request matrix of the electrical load in each load management stage is obtained, including:

[0020] Combined with the typical flight mission profile, the electrical load power supply request matrix R of each stage is obtained, which reflects the relationship between the electrical load management priority and the flight profile. The matrix uses rows to represent the load and columns to represent the current management stage. If R iw =1, it means that the i-th load in the w-th management priority should be powered; if R iw =0, it means that the i-th load in the w-th management priority should be disconnected.

[0021] Furthermore, the electrical load control equation is expressed as:

[0022]

[0023] Wherein: F is the status signal of the electrical load, F=0 indicates that the electrical load is faulty, and F=1 indicates that the electrical load is normal; B is the backup flag, B=1 indicates that the electrical load is a backup device and is temporarily not allowed to be powered on, and B=0 indicates that the electrical load is a main working device or a backup device that needs to be put into operation; the overline of F+B indicates a logical negation operation; P is the power supply priority of the electrical load, P=1 indicates that power is allowed to be supplied to the load, and P=0 indicates that power is not allowed to be supplied to the load; Z is the power supply request, Z=1 indicates that the power supply request is valid, and Z=0 indicates that the power supply request is invalid; S is an override signal or a power supply request signal sent from the ground; C=1 indicates that the electrical load is turned on; C=0 indicates that the system should not supply power to the electrical load.

[0024] Furthermore, the constraints of the system are specifically:

[0025] The modeling formula for busbar j is as follows:

[0026]

[0027] Where: is the total electrical load N connected to bus j at time t j Total power required; and are the power consumed by the load i that can be unloaded and the load that cannot be unloaded connected to bus j at time t; N j is the number of all electrical loads connected to bus j, n j is the number of loads that can be unloaded;

[0028] To meet the load power distribution requirements, each coefficient c corresponding to bus j and load i at time t ji (t) is a binary decision variable that optimizes the objective function to determine whether power must be provided at this moment and Determine whether the load i of bus j needs to be unloaded, c ji (t) is expressed as a mathematical constraint as follows:

[0029]

[0030] In the formula, and The index sets of the loads that cannot be unloaded and can be unloaded on bus j, N b Indicates the total number of buses.

[0031] Furthermore, the use of mixed integer linear programming method to establish a mathematical model of the load management system includes:

[0032]

[0033] Where, t * represents a discrete sampling moment, H represents the prediction time domain of discrete programming, T s is the sampling time, N b is the number of buses, N s is the number of power sources, μ is a constant weight parameter, α m (t) is a binary variable parameter indicating whether power source m is used at time t; C j (t) is the vector of binary decision variables corresponding to each load of each bus j, n j is the number of loads corresponding to bus j; where Γj(t) is a custom penalty parameter vector, and the superscript T indicates a transposition operation; Λ j (t) represents the weighted value of power allocation priority, Δ j (t) is the vector of binary decision variables corresponding to the power source connected to each bus j.

[0034] Furthermore, in the mathematical model of the system, the number of loads n corresponding to bus j is j , constant weight parameter μ, sampling time t * , discrete programming prediction time domain H, sampling time T s、Number of busbars b 、Number of power supplies s It is known that α is determined by the system architecture and time domain sampling setting parameters, where m (t), C j (t),Γj(t),Λ j (t), Δ j (t) These binary decision vectors are given an initial value according to the system state, and the optimal solution of the objective function is solved using the rolling horizon control idea, and finally the optimal control sequence c of bus j for load i is obtained. ji , realize contactor control load connection and disconnection to complete load management control.

[0035] Furthermore, the rolling horizon control first substitutes the system parameter value at the current time t into the objective function to predict the future system state, and obtains a set of control sequences c ji (t), then the sequence c ji (t) The first term is substituted into the objective function again, and the rest of the results are discarded. The first term is the first term of the sequence solution set to be sought, and then it is substituted into the next moment, and the process is iterated continuously until the end of the planning, and finally the optimal control sequence c of bus j for load i in load management control is obtained ji .

[0036] Compared with the prior art, the present invention has the following technical features:

[0037] 1. The rolling horizon control not only considers the current state of the system, but also predicts the future state. The model is iteratively solved by minimizing the difference between the predicted state and the target, and the solution speed is fast.

[0038] 2. The load management model based on mixed integer linear programming can control the automatic load according to the system operation status and priority under multiple constraints, and is suitable for complex systems with various constraints;

[0039] 3. Load management control is designed from the top-level mathematical model of the system, which overcomes the problem of lack of mathematical model accumulation in the traditional load management system that relies on solving load equations for control, and provides a valuable design method for further optimization of load management of airborne power distribution systems. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] Figure 1 It is a flowchart of the method of the present invention;

[0041] Figure 2 is a system architecture diagram of an embodiment of the present invention;

[0042] Figure 3 A schematic diagram of a flight profile of an embodiment of the present invention;

[0043] Figure 4 The basic structure of the rolling time domain control according to the embodiment of the present invention;

[0044] Figure 5 This is a diagram showing the switching result of the connection between the generator and the busbar according to an embodiment of the present invention;

[0045] Figure 6 This is a diagram showing the busbar connection load unloading result according to an embodiment of the present invention. DETAILED DESCRIPTION

[0046] See attached Figure 1 The present invention provides a load management method for an airborne power distribution system based on MILP and RHC, comprising the following steps:

[0047] Step 1: Equivalent the distributed airborne power distribution system architecture and load connection status to the optimal control problem for solution, including:

[0048] Step 101, in a distributed airborne power distribution system, the generator and the auxiliary power unit (APU) distribute power through multiple buses to provide redundancy. The connection between the generator and the power load is routed through a series of electromechanical switches (called contactors). The on and off of the contactors can control the connection and unloading of the load. The relationship between different components of the system is represented by a single-line diagram to construct a system component relationship connection diagram.

[0049] Step 102, with respect to the system component relationship connection diagram, the AC sources such as generators, busbars, contactors and power loads in the power distribution system are important components of the system. Based on the safety and redundant stability design of the power distribution system, each generator is set to have its output voltage adjusted by a generator control unit (GCU) to meet the power required by the expected load, and each contactor is controlled to be turned on or off by a busbar power control unit (BPCU), and the power from the power supply is distributed to the designated busbar and load as required to ensure that the system can still operate reliably when a fault occurs; thereby meeting the power demand and load priority constraints in the current system, and the load management control can be equivalent to the optimal control problem of the contactor, controlling the connection switching between the generator and the busbar, and the connection and unloading of the load.

[0050] In the load management control of step 102, the equivalent processing requirements for the generator, bus and load are as follows:

[0051] Requirement A1: Generators are not connected in parallel; according to the architecture of the distributed power distribution system, it is assumed that there is no situation where the same bus is powered by multiple AC generators at the same time;

[0052] Requirement A2: Bus priority; Each bus has a priority list of power generators. If the first generator in the priority list is not available at the moment, the second generator should supply power to the bus, and so on.

[0053] Requirement A3: Load priority: Set load priority. If the power supply is insufficient, unloadable loads (such as general loads such as lighting systems and auxiliary systems) are allowed to be unloaded according to the priority principle, while non-unloadable loads (such as critical loads such as flight control systems and communication systems) must always remain powered.

[0054] Step 2: The distributed airborne power distribution system must always meet the requirements of the flight mission. The flight mission profile is determined based on the power distribution requirements of the aircraft, and the flight mission profile is divided into different load management stages. The electrical load power supply request matrix in each load management stage is obtained, and the electrical load control equation is further combined to control the connection or disconnection of each load; specifically, it includes:

[0055] Step 201, the flight mission profile requirement is the basic requirement for the load management system to automatically manage the load, and is also the basic criterion for designing and optimizing the load management system.

[0056] Combined with the typical flight mission profile, the electrical load power supply request matrix R of each stage is obtained, which reflects the relationship between the electrical load management priority and the flight profile. The matrix uses rows to represent the load and columns to represent the current management stage. If R iw =1, it means that the i-th load in the w-th management priority should be powered; if R iw =0, it means that the i-th load in the w-th management priority should be disconnected.

[0057] In one embodiment of the present invention, a typical flight profile of an aircraft is mainly divided into processes such as taxiing, take-off, climbing, cruising, descending, landing and landing taxiing. Figure 3 The following is a typical flight profile diagram of an aircraft during flight. Figure 3 a to j represent different flight profile stages. According to the flight profile, the corresponding relationship between the load management stage and the electrical load power supply request of the automatic power distribution system can be obtained, as shown in Table 1:

[0058] Table 1 Power supply request matrix

[0059]

[0060] The values ​​in the matrix represent the power supply request R. R = 1 means that a signal requesting power supply is sent when entering the load management stage, otherwise no request signal is sent. The rows in the matrix represent the loads and the columns represent the current management stage. iw =1, it means that in the wth management priority, the i-th load should be powered; if R iw = 0, it means that in the wth management priority, the i-th load should be disconnected. The power request matrix is ​​used to reflect the relationship between the load priority and the flight profile.

[0061] Step 202, the electrical load control equation is the ultimate basis for the entire aircraft power distribution system to perform load on-off control. The power supply request signal effectively indicates that the load currently needs to be powered on, and the electrical load control equation is obtained based on the state of the load itself and the state of the aircraft power system. The expression is as follows:

[0062]

[0063] Wherein: F is the status signal of the electrical load, F=0 indicates that the electrical load is faulty, and F=1 indicates that the electrical load is normal; B is the backup flag, B=1 indicates that the electrical load is a backup device and is temporarily not allowed to be powered on, and B=0 indicates that the electrical load is a main working device or a backup device that needs to be put into operation; the overline of F+B indicates a logical negation operation; P is the power supply priority of the electrical load, P=1 indicates that power is allowed to be supplied to the load, and P=0 indicates that power is not allowed to be supplied to the load; Z is the power supply request, Z=1 indicates that the power supply request is valid, and Z=0 indicates that the power supply request is invalid; S is an override signal or a power supply request signal sent from the ground; C=1 indicates that the electrical load is turned on; C=0 indicates that the system should not supply power to the electrical load.

[0064] Step three, based on the electrical load control equation and combined with the system constraints, a mixed integer linear programming method is used to establish a mathematical model of the load management system.

[0065] The mathematical modeling of the system requires capturing all the demand objectives and constraints in the load management system as mathematical formula constraints, including:

[0066] Step 301: Determine constraint relationships

[0067] Both non-unloadable and unloadable loads are connected to the bus. The power required by the system at time t is modeled as the sum of the power required by all different loads connected to the specified bus. The load management system controls the on-off of general loads. The load priority in step 1 and the electrical load control equation in step 2 are combined to determine whether the load is unloaded, thereby obtaining a constraint relationship. The modeling formula for bus j is as follows:

[0068]

[0069] Where: is the total electrical load N connected to bus j at time t j Total power required; and are the power consumed by the load i that can be unloaded and the load that cannot be unloaded connected to bus j at time t; N j is the number of all electrical loads connected to bus j, n j is the number of loads that can be unloaded;

[0070] To meet the load power distribution requirements, each coefficient c corresponding to bus j and load i at time t ji (t) is a binary decision variable that optimizes the objective function to determine whether power must be provided at this moment and Determine whether load i of bus j needs to be unloaded, c ji (t) is expressed as a mathematical constraint as follows:

[0071]

[0072] In the formula, and The index sets of the loads that cannot be unloaded and can be unloaded on bus j, N b Indicates the total number of buses.

[0073] Step 302: According to the above steps, the constraint relationship of the system is obtained. The constraint is represented in the form of binary decision variables. The mathematical model of the load management system is established by mixed integer linear programming. That is, the objective function of the system is obtained according to the decision variables and the constraint relationship, and converted into an optimization problem for solving the objective function. The expression of the system objective function is as follows:

[0074]

[0075] Where, t ★ represents a discrete sampling moment, H represents the prediction time domain of discrete programming, T s is the sampling time, N b is the number of buses, N s is the number of power sources, μ is a constant weight parameter, α m (t) is a binary variable parameter indicating whether power source m is used at time t; C j (t) is the vector of binary decision variables corresponding to each load of each bus j, n j is the number of loads corresponding to bus j; where Γj(t) is a custom penalty parameter vector, and the superscript T indicates a transposition operation; Λ j (t) represents the weighted value of power allocation priority, Δ j (t) is the vector of binary decision variables corresponding to the power source connected to each bus j, and its specific form is as follows:

[0076]

[0077] In the above formula Indicates the nth j The real-time status of the contactor switch between each load and the connected busbar j; Reflects the nth j In the case of load shedding, The value of is set in order according to the load priority; Indicates the Nth s The power supply situation of bus j corresponding to each power source, Indicates the Nth s The state of a power supply being connected to bus j.

[0078] Step 4: Based on the mathematical model of the load management system and the corresponding constraints, the rolling horizon control method is used. First, the future state is predicted, and the model is solved by minimizing the difference between the predicted state and the final target. The load management problem can be converted into a mixed integer linear programming problem in step 3 by solving the system objective function under multiple constraints, and finally the optimal control sequence of the bus for the load is obtained to realize automatic load management.

[0079] Step 401: In the system objective function of step 3, the number of loads n corresponding to bus j j , constant weight parameter μ, sampling time t * , prediction time domain H of discrete programming, sampling time T s 、Number of busbars b 、Number of power supplies s It is known that α is determined by the system architecture and time domain sampling setting parameters, where m (t), C j (t),Γj(t),Λ j (t), Δ j (t) These binary decision vectors are given an initial value according to the system state, and the optimal solution of the objective function is solved using the rolling horizon control idea, and finally the optimal control sequence c of bus j for load i is obtained. ji , realize contactor control load connection and disconnection to complete load management control.

[0080] Specifically, in step 401, the rolling horizon control first substitutes the system parameter value at the current time t into the objective function to predict the future system state, and obtains a set of control sequences c ji (t), then the sequence c ji (t) The first term is substituted into the objective function again, and the rest of the results are discarded. The first term is the first term of the sequence solution set to be sought, and then it is substituted into the next time t+T s , and the process is iterated continuously until the end time of planning t+HT s Finally, the optimal control sequence c of bus j to load i in load management control is obtained ji , realizing automatic management of load.

[0081] like Figure 5As shown, the power distribution strategy is shown when only the basic load management system is active. When the load management system detects a generator power failure, it switches the connection power of bus 2 from generator 2 to the APU; Figure 6 The system demonstrates the dynamic switching results of loads at different times. Rapid load switching ensures the reliability of power demand at each stage and realizes automatic load management.

[0082] The above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. These modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application, and should all be included in the protection scope of the present application.

Claims

1. A load management method for an airborne power distribution system based on MILP and RHC, characterized in that: include: The distributed airborne power distribution system architecture and load connection status are equivalent to the optimal control problem to be solved; Determine the flight mission profile based on the power distribution demand of the aircraft, divide the flight mission profile into different load management stages, obtain the power supply request matrix of the electrical load in each load management stage, and further combine the electrical load control equation to control the connection or disconnection of each load; According to the electrical load control equation and the system constraints, the mathematical model of the load management system is established by using the mixed integer linear programming method. Aiming at the mathematical model of the load management system and the corresponding constraint relationship, the rolling horizon control method is adopted. First, the future state is predicted, and the model is solved by constraining the difference between the predicted state and the final target to be minimized. The load management problem can be converted into a system mathematical model to solve a mixed integer linear programming problem under multiple constraint conditions, and finally the optimal control sequence of the bus for the load is obtained to realize automatic load management.

2. The method for load management of an airborne power distribution system based on MILP and RHC according to claim 1, characterized in that: The method of equating the distributed airborne power distribution system architecture and load connection state to an optimal control problem for solution includes: In the distributed airborne power distribution system, the generator and auxiliary power unit distribute power through multiple buses to provide redundancy. The connection between the generator and the power load is routed through a series of electromechanical switches called contactors. The on and off of the contactors can control the connection and unloading of the load. The relationship between different components of the system is represented by a single-line diagram to construct a connection diagram of the system component relationship; According to the system component relationship connection diagram, each generator is set to have its output voltage adjusted by the generator control unit to meet the power required by the expected load; each contactor is controlled to be opened or closed by the bus power control unit to distribute the power from the power supply to the designated bus and load as required; thus, the load management control is equivalent to the optimal control problem of the contactor, controlling the connection switching between the generator and the bus, and the connection and unloading of the load.

3. The method for load management of an airborne power distribution system based on MILP and RHC according to claim 2, characterized in that: In the load management control, the equivalent processing requirements for the generator, bus and load are as follows: The generators are not connected in parallel; according to the architecture of the distributed power distribution system, it is assumed that there is no situation where the same bus is powered by multiple AC generators at the same time; Bus priority: Each bus has a priority list of power generators. If the first generator in the priority list is not available at the moment, the second generator should supply power to the bus, and so on. Load priority; Set load priority. If the power is insufficient, the load that can be removed can be removed according to the priority principle, while the load that cannot be removed must always remain powered.

4. The method for load management of an airborne power distribution system based on MILP and RHC according to claim 1, characterized in that: The flight mission profile is determined based on the power distribution demand of the aircraft, the flight mission profile is divided into different load management stages, and the power supply request matrix of the electrical load in each load management stage is obtained, including: Combined with the typical flight mission profile, the electrical load power supply request matrix R of each stage is obtained, which reflects the relationship between the electrical load management priority and the flight profile. The matrix uses rows to represent the load and columns to represent the current management stage. If R iw =1, it means that the i-th load in the w-th management priority should be powered; if R iw =0, it means that the i-th load in the w-th management priority should be disconnected.

5. The method for load management of an airborne power distribution system based on MILP and RHC according to claim 1, characterized in that: The electrical load control equation is expressed as: Wherein: F is the status signal of the electrical load, F=0 indicates that the electrical load is faulty, and F=1 indicates that the electrical load is normal; B is the backup flag, B=1 indicates that the electrical load is a backup device and is temporarily not allowed to be powered on, and B=0 indicates that the electrical load is a main working device or a backup device that needs to be put into operation; the overline of F+B indicates a logical negation operation; P is the power supply priority of the electrical load, P=1 indicates that power is allowed to be supplied to the load, and P=0 indicates that power is not allowed to be supplied to the load; Z is the power supply request, Z=1 indicates that the power supply request is valid, and Z=0 indicates that the power supply request is invalid; S is an override signal or a power supply request signal sent from the ground; C=1 indicates that the electrical load is turned on; C=0 indicates that the system should not supply power to the electrical load.

6. The method for load management of an airborne power distribution system based on MILP and RHC according to claim 1, characterized in that: The constraints of the system are specifically: The modeling formula for busbar j is as follows: Where: is the total electrical load N connected to bus j at time t j Total power required; and are the power consumed by the load i that can be unloaded and the load that cannot be unloaded connected to bus j at time t; N j is the number of all electrical loads connected to bus j, n j is the number of loads that can be unloaded; To meet the load power distribution requirements, each coefficient c corresponding to bus j and load i at time t ji (t) is a binary decision variable that optimizes the objective function to determine whether power must be provided at this moment and Determine whether load i of bus j needs to be unloaded, c ji (t) is expressed as a mathematical constraint as follows: In the formula, and The index sets of the loads that cannot be unloaded and can be unloaded on bus j, N b Indicates the total number of buses.

7. The method for load management of an airborne power distribution system based on MILP and RHC according to claim 1, characterized in that: The mathematical model of the load management system is established by using a mixed integer linear programming method, including: Where, t * represents a discrete sampling moment, H represents the prediction time domain of discrete programming, T s is the sampling time, N b is the number of buses, N s is the number of power sources, μ is a constant weight parameter, α m (t) is a binary variable parameter indicating whether power source m is used at time t; C j (t) is the vector of binary decision variables corresponding to each load of each bus j, n j is the number of loads corresponding to bus j; where Γj(t) is a custom penalty parameter vector, and the superscript T indicates a transposition operation; Λ j (t) represents the weighted value of power allocation priority, Δ j (t) is the vector of binary decision variables corresponding to the power source connected to each bus j.

8. The method for load management of an airborne power distribution system based on MILP and RHC according to claim 7, characterized in that: In the mathematical model of the system, the number of loads n corresponding to bus j is j , constant weight parameter μ, sampling time t * , prediction time domain H of discrete programming, sampling time T s 、Number of busbars b 、Number of power supplies s It is known that α is determined by the system architecture and time domain sampling setting parameters, where m (t), C j (t),Γj(t),Λ j (t), Δ j (t) These binary decision vectors are given an initial value according to the system state, and the optimal solution of the objective function is solved using the rolling horizon control idea, and finally the optimal control sequence c of bus j for load i is obtained. ji , realize contactor control load connection and disconnection to complete load management control.

9. The method for load management of an airborne power distribution system based on MILP and RHC according to claim 8, characterized in that: The rolling horizon control first substitutes the system parameter value at the current time t into the objective function to predict the future system state and obtain a set of control sequences c ji (t), then the sequence c ji (t) The first term is substituted into the objective function again, and the rest of the results are discarded. The first term is the first term of the sequence solution set to be sought, and then it is substituted into the next moment, and the process is iterated continuously until the end of the planning, and finally the optimal control sequence c of bus j for load i in load management control is obtained ji .

Images