A multi-controller cyclic redundancy switching method based on dynamic priority adjustment and state mapping

Abstract

The present invention discloses a multi-controller cyclic redundancy switching method based on dynamic priority adjustment and state mapping to solve the problem that existing redundant control systems cannot well meet the needs of industrial production. The method includes: a first step of judging whether the controller is in a normal or faulty state; a second step of dynamically adjusting the priority after the controller fails; and a third step of realizing dynamic redundant switching of the faulty controller based on the controller priority. The proposed multi-controller cyclic redundancy switching method based on dynamic priority adjustment and state mapping can well meet the needs of industrial production, realize dynamic priority adjustment and redundant switching under complex faults of multiple controllers, effectively solve the problem of maintaining high priority after an actuator fails, and ensure that the controller can be re-connected to the system after the fault is repaired, thereby improving the continuity and stability of production and operation.

Description

Technical Field

[0001] The present invention relates to the field of automation control technology, and in particular to a multi-controller cyclic redundancy switching method based on dynamic priority adjustment and state mapping. Background Art

[0002] With the continuous advancement of industrial automation and the intensification of competition in the manufacturing industry, the cost of downtime caused by equipment failure is immeasurable. Consequently, the reliability of production equipment is becoming increasingly important, and control system reliability is becoming increasingly crucial. For applications requiring high reliability and safety, especially for critical equipment that plays a vital role in production, redundant design is often necessary. Analyzing key aspects of the system, proposing solutions to improve control system reliability, and implementing redundancy in key areas can significantly enhance overall system reliability and stability.

[0003] The controller is a key component in industrial automation systems, responsible for controlling and monitoring the operation of various machines and equipment. However, during system operation, the controller will inevitably experience problems such as aging and failure. If there is only one controller in the system, when the controller fails, the entire production line will be forced to stop, thus affecting productivity and efficiency. In order to ensure the continuous and stable operation of the system and reduce downtime and failure losses, it is usually necessary to design a redundant controller. Currently, redundant control of controllers mostly adopts a one-master-one-backup or one-master-multiple-backup method, connecting two or more controllers in parallel to achieve fault switching, and automatically switching to the backup controller after the main controller fails to ensure the continuous and stable operation of the control system. For example, the controller redundancy and switching method provided by patent CN103455005A adopts a one-master-one-backup redundant control system. This method uses an arbitration mechanism to downgrade the main controller to a backup controller after a failure, and at the same time upgrade the backup controller to the main control. This method can only achieve redundant switching of two controllers, one master and one backup. Patent CN112445127A provides a redundant control method for active and standby controllers. This method suffers from a cumbersome switching process and slow dynamic response, hindering production operations. Furthermore, a controller failure can easily lead to a dual-active state, causing system failure. Patent CN107733684A discloses a redundant switching method for multiple controllers with shared storage space. A slave controller monitors all master controllers. This method selects a master controller from a multi-controller computational redundancy cluster and sets a fixed controller priority. If the failed controller is the master, the highest-priority controller among the remaining controllers automatically switches to the master. If the failed controller is a slave, the other controllers identify the failed controller as faulty. This method only addresses redundant switching after a single controller failure and cannot address switching under combined failures of multiple controllers. Furthermore, the slave controller's priority is a fixed, manually set value that cannot be updated or dynamically adjusted after an actuator failure, resulting in a paradoxical situation where the actuator maintains a high priority despite failure. Furthermore, this method is designed for sequential controller redundant switching and does not address controllers that are subsequently repaired. Summary of the Invention

[0004] The purpose of the present invention is to overcome the shortcomings of the existing technology and provide a multi-controller cyclic redundancy switching method based on dynamic priority adjustment and state mapping. The controller priority is automatically and dynamically adjusted according to the access order, fault status and fault recovery status of the controller, and the ability of the controller to re-access the system after the fault is repaired is taken into account to ensure long-term reliable operation of the system. The process of the present invention is as follows Figure 1 As shown, the real-time acquisition controller collects three input signals α with a time interval of Δt i and the feedback signal γ i ,like Figure 3Perform XOR operation on the collected signals, as shown in Figure 4 As shown, the system then performs a two-out-of-three logic judgment on the results of the XOR operation to determine the state of each controller, as shown in Figure 5 As shown. Based on the controller status, the relationship between the number of normal controllers and the number of faulty controllers is compared. When the number of normal controllers is less than the number of actuators, the system cannot replace the faulty controller through redundancy adjustment, and the remote monitoring system alarms; when the number of normal controllers is greater than or equal to the number of actuators, the system replaces the faulty controller through redundancy adjustment. The specific steps for replacing a faulty controller are as follows: First, update the controller priority based on the controller status, as shown in the following example: Figure 6 As shown; secondly, according to the updated controller status and priority, the system sends a switching signal to the dynamic switch to respond quickly. By switching the dynamic switch, the normal hot standby controller replaces the faulty controller.

[0005] 1. When the controller fails, perform the following operations:

[0006] The first step is to determine whether the controller is in normal or faulty state:

[0007] (1) n controllers, m actuators, where n ≥ m, that is, there are nm redundant controllers. The physical serial number matrix of the controller is where a i represents the i-th controller, i = 1, 2, ..., n, the physical serial number of the controller is determined by its installation position; the controller state matrix is where b i Indicates that the physical serial number is a i The state value of the controller, b i =1 indicates controller a i Normal, status value b i =0 means controller a i Fault. Perform real-time status judgment on each controller and update the status matrix B in real time based on the judgment result. The update steps are as follows:

[0008] 1) Controller a i The input signal is α i , the output signal is β i , controller a i The input signal α i The feedback signal generated by the feedback circuit is γ i , synchronized acquisition controller a by remote monitoring system i Input and feedback signals;

[0009] II) Considering the delay in signal transmission, a two-out-of-three logic judgment method is used to judge the controller state. The sampling interval is set to Δt, and the input signal and feedback signal are collected three times continuously to form a logic matrix. in The superscript j=1, 2, 3 indicates the order in which the signals are collected three times. Indicates that the collected input signal and feedback signal are high level, Indicates that the collected input signal and feedback signal are low level;

[0010] III) For the matrix C i The first and second columns of the same OR operation, that is in ☉ represents the XOR operation, judgment matrix Find the judgment matrix D based on the calculation results i The sum of all elements X≥2, then the controller a i Normal, let the state value b i =1; X<2, then the controller a is determined i Fault, let the state value b i =0;

[0011] (2) Repeat the above steps I, II, and III to judge the state of each controller, obtain the state values ​​of n controllers, and update the state matrix B.

[0012] The second step is to dynamically adjust the priority after the controller fails:

[0013] (1) Controller priority matrix e i Represents controller a i Priority, e i ∈{1, 2, 3, ..., n}, ∈ is the inclusion relation, that is, the set {1, 2, 3, ..., n} contains e i , e i =1 indicates controller a i The highest priority is e i =n represents controller a i The priority is the lowest. The higher the priority of the controller, the higher the priority is. u, v∈{1,2,3,…,n}, u≠v, The symbols Indicates any choice, When the controller is first installed, its initial priority is determined by the controller physical serial number mapping, that is, the controller installed for the first time The corresponding priority is Controller Priority Adjustment Matrix

[0014] (2) Find the sum Y of the elements in the first column of matrix F, that is, the number of normal controllers If Y≥m, the normal hot standby controller will replace all the faulty working controllers through the switching of the dynamic switch; if Y<m, the dynamic switch cannot be switched to replace all the faulty working controllers, and the remote monitoring system will alarm.

[0015] (3) If Y ≥ m, the priority of each controller is dynamically adjusted according to the following steps:

[0016] 1) Controller a i The status value b i =1, the priority of all controllers remains unchanged;

[0017] II) Controller a i The status value b i =0, the priority is higher than e i The controller priority remains unchanged and is lower than e i The controller priority of each controller is increased by one level, and controller a i The priority of is reduced to n;

[0018] (4) According to the physical sequence number from a1→a n Poll the status value b of each controller in turn i , repeat steps I and II to complete the dynamic adjustment of the priorities of n controllers and update the matrix F.

[0019] The third step is to implement dynamic redundant switching of failed controllers based on controller priority:

[0020] (1) Physical serial number matrix of actuator where z l represents the lth actuator, l = 1, 2, ..., m; define the connection matrix between the controller and the actuator H = n controllers h il =1 represents controller a i With actuator z l Connected, h il =0 represents controller a i Not with actuator z il are connected.

[0021] (2) If Y ≥ m, the normal hot standby controller replaces all the faulty working controllers through the switching of dynamic switches. The steps are as follows:

[0022] I) Multiply each column of the state matrix B with the connection matrix H in turn to obtain the connection matrix H0 after the state is updated, and calculate the sum of all elements of the H0 matrix The connection matrix H0 and the priority matrix E form an augmented matrix Retrieve the elements with values ​​of 1 from the 1st column to the mth column in the augmented matrix, and delete the rows and columns corresponding to element 1 to obtain the reduced matrix The priority size corresponding to each row in the filter selects the top mk rows of priority to form a local connection matrix

[0023] II) The matrix The main diagonal elements are assigned a value of 1 and the other elements remain unchanged, and the updated local connection matrix is ​​obtained. The controller in the row corresponding to element 1 is connected to the actuator in the column through the switching of the dynamic switch;

[0024] III) The matrix The elements in replace the corresponding elements in the H matrix in turn according to the position relationship of the corresponding rows and columns, and update the matrix H;

[0025] (3) After the faulty controller is repaired or replaced, the controller status is judged again by the access system, and the status matrix B is updated. The controller priority is adjusted again according to the updated status matrix, and the priority matrix E is updated.

[0026] (4) Define the full information matrix G = (H|BE). The full information matrix G includes the connection status of each controller and actuator, the status of each controller, and the priority level. The newly connected controller performs real-time status judgment like other controllers. If the controller fails again, the second and third steps are performed again. This cycle is repeated to ensure the normal operation of the control system. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] Figure 1 A flowchart of a multi-controller redundant switching method based on dynamic priority adjustment and state mapping proposed by the present invention;

[0028] Figure 2 The input signal α of the method of the present invention is i The controller generates a feedback signal γ i and the output signal β i Schematic diagram of;

[0029] Figure 3 The remote monitoring system collects input signal α three times at intervals of Δt i and the feedback signal γi Schematic diagram;

[0030] Figure 4 The input signal α collected by the remote monitoring system at the same time i and the feedback signal γ i Perform XNOR operation truth table;

[0031] Figure 5 The basis table for the remote monitoring system to perform three-out-of-two logic judgment on the normal or fault status of the controller;

[0032] Figure 6 A schematic diagram of the dynamic adjustment of the priority of the remote monitoring system after a controller failure;

[0033] Figure 7 This is a schematic diagram of the switching process of the dynamic switch based on the controller status and priority of Example 1;

[0034] Figure 8 This is a schematic diagram of the switching process of the dynamic switch based on the controller status and priority in Example 2; DETAILED DESCRIPTION

[0035] In order to better present the purpose, technical solutions and advantages of the present invention in actual production, it is necessary to further explain the present invention more clearly in combination with production examples and related drawings. Obviously, the specific implementation examples described in the text are only part of the specific implementation examples of the present invention, rather than all the implementation examples. Based on the content of the present invention, all other results obtained by technical personnel in related fields through interpretation of the present invention without making creative work are within the scope of protection of the present invention.

[0036] The specific technical solutions provided by the implementation examples of this application are described in detail below with reference to the accompanying drawings.

[0037] Example 1:

[0038] A control system with five controllers a1, a2, a3, a4, and a5 and three actuators z1, z2, and z3, where a1, a2, and a3 are working controllers and a4 and a5 are standby controllers. When two working controllers a1 and a3 fail, the operating steps of the present invention are as follows: Step 1: Determine whether the controller is in a normal or faulty state:

[0039] (1) Controller physical serial number matrix Controller State Matrix Perform real-time status judgment on each controller and update the state matrix B in real time based on the judgment results. The update steps are as follows:

[0040] I) The input signal α1 and output signal β1 of the controller a1, and the feedback signal γ1 generated by the feedback circuit of the input signal α1 of the controller a1, and the input and feedback signals of the controller a1 are synchronously collected by the remote monitoring system;

[0041] II) Considering the delay in signal transmission, a two-out-of-three logic judgment method is used to judge the controller state. The sampling interval is set to Δt, and the input signal and feedback signal are collected three times continuously to form a logic matrix.

[0042] III) Perform XOR operation on the first and second columns of matrix C1, that is, Judgment Matrix According to the calculation results, calculate the sum of all elements of the judgment matrix D1 If X=1≤m=2, it is determined that the controller a1 is faulty and the state value b1=0;

[0043] IV) the controller a2 has an input signal α2 and an output signal β2. The input signal α2 of the controller a2 is fed back through the feedback circuit to generate a feedback signal γ2. The input and feedback signals of the controller a2 are synchronously collected by the remote monitoring system.

[0044] V) Considering that there will be delays in the signal transmission process, a two-out-of-three logic judgment method with three independent measurements at equal intervals is used to judge the controller state. Set the collection time interval to Δt, and continuously collect three input signals and feedback signals to form a logic matrix

[0045] VI) Perform XOR operation on the first column and the second column of matrix C2, that is, Judgment Matrix According to the calculation results, find the sum of all elements of the judgment matrix D2 If X=3≥m=2, the controller a2 is judged to be normal and the state value b2=1;

[0046] VII) Controller a3 has an input signal α3 and an output signal β3. The input signal α3 of controller a3 is fed back through the feedback circuit to generate a feedback signal γ3. The input and feedback signals of controller a3 are synchronously collected by a remote monitoring system.

[0047] VIII) Considering the delay in signal transmission, a two-out-of-three logic judgment method is used to judge the controller state. The sampling interval is set to Δt, and the input signal and feedback signal are collected three times continuously to form a logic matrix.

[0048] IX) Perform an XOR operation on the first column and the second column of the matrix C3, that is, Judgment Matrix According to the calculation results, find the sum of all elements of the judgment matrix D3 If X=1≤m=2, it is determined that the controller a3 is faulty and the state value b3=0;

[0049] X) The input signal of controller a4 is α4, and the output signal is β4. The feedback signal generated by the feedback circuit of the input signal α4 of controller a4 is γ4. The input and feedback signals of controller a4 are synchronously collected by the remote monitoring system;

[0050] XI) Considering the delay in signal transmission, a two-out-of-three logic judgment method is used to judge the controller state. The sampling interval is set to Δt, and the input signal and feedback signal are collected three times continuously to form a logic matrix.

[0051] XII) Perform an XOR operation on the first and second columns of the matrix C4, that is, Judgment Matrix According to the calculation results, calculate the sum of all elements of the judgment matrix D4 If X=3≥m=2, the controller a4 is judged to be normal and the state value b4=1;

[0052] XIII) controller a5 receives an input signal α5 and outputs a signal β5. The input signal α5 of controller a5 is fed back through a feedback circuit to generate a feedback signal γ5. The input and feedback signals of controller a5 are synchronously collected by a remote monitoring system.

[0053] XIV) Considering the delay in signal transmission, a two-out-of-three logic judgment method is used to judge the controller state. The sampling interval is set to Δt, and the input signal and feedback signal are collected three times continuously to form a logic matrix.

[0054] XV) Perform an XOR operation on the first and second columns of the matrix C5, that is, Judgment Matrix According to the calculation results, calculate the sum of all elements of the judgment matrix D5 If X=3≥m=2, the controller a5 is judged to be normal and the state value b5=1;

[0055] (2) Perform state judgment on each controller, obtain the state values ​​of the five controllers, and update the state matrix

[0056] The second step is to dynamically adjust the priority after the controller fails:

[0057] (1) Priority matrix for initial controller installation Establish the controller priority adjustment matrix based on the state judgment in the first step

[0058] (2) Find the sum Y of the elements in the first column of matrix F, that is, the number of normal controllers If Y=3≥m=2, the normal hot standby controller will replace all the failed working controllers through the switching of dynamic switches;

[0059] (3) If Y = 3 ≥ m = 2, the priority of each controller is dynamically adjusted according to the following steps:

[0060] I) Retrieve the status value b1 of controller a1, b1=0, the current priority of controller a1 is 1, the priorities of controllers a2, a3, a4, and a5 with lower priorities than 1 are increased by one level, and the priority of controller a1 is reduced to 5. The priority update matrix

[0061] II) Retrieve the state value b2 of controller a2, b2=1, the priorities of all controllers remain unchanged, and the priority update matrix

[0062] III) Retrieve the status value b3 of controller a3, b3 = 0. The current priority of controller a3 is 2. The priority of controller a2, which has a higher priority than 2, remains unchanged. The priorities of controllers a1, a4, and a5, which have a lower priority than 2, are increased by one level. The priority of controller a3 is reduced to 5. The priority update matrix

[0063] IV) Retrieve the state value b4 of controller a4. If b4=1, the priority of all controllers remains unchanged and the priority matrix is ​​updated.

[0064] V) Retrieve the state value b5 of controller a5. If b5=1, the priority of all controllers remains unchanged and the priority update matrix

[0065]

[0066] (4) Complete the dynamic adjustment of the priorities of the five controllers and obtain the updated matrix The third step is to implement dynamic redundant switching of failed controllers based on controller priority:

[0067] (1) Physical serial number matrix of actuator The connection matrix between controller and actuator H = 5 controllers

[0068] (2) From the second step, we know that Y = 3 ≥ m = 2. Then, by switching the dynamic switch, the normal hot standby controller replaces all the faulty working controllers. The steps are as follows:

[0069] I) State Matrix Multiply each column of the connection matrix H in turn to obtain the connection matrix after the state is updated Find the sum of all elements of the H0 matrix The connection matrix H0 and the priority matrix E form an augmented matrix In the augmented matrix Retrieve the elements with values ​​of 1 in columns 1 to 3, and delete the rows and columns corresponding to element 1 to obtain the reduced matrix according to The priority size corresponding to each row of the matrix selects the top mk=3-1=2 rows of priority and obtains the local connection matrix

[0070] II) The matrix The main diagonal elements are assigned a value of 1 and the other elements remain unchanged, and the updated local connection matrix is ​​obtained. The controllers a4 and a5 in the row corresponding to element 1 and the actuators z1 and z3 in the column corresponding to element 1 complete the connection between controller a4 and actuator z1 and between controller a5 and actuator z3 through the switching of dynamic switches, as shown in the following example: Figure 7 As shown;

[0071] III) The matrix The elements in replace the corresponding elements in the H matrix according to the position relationship of the corresponding rows and columns, and the updated matrix H = 5 controllers are obtained.

[0072] (3) After the faulty controller is repaired or replaced, the controller is connected to the system to determine its status again and update the status matrix. Adjust the controller priority again according to the updated state matrix and update the priority matrix

[0073] (4) Full information matrix G = (H|BE) = 5 controllers The full information matrix G contains the connection status of each controller and actuator, as well as the status and priority of each controller. A newly connected controller performs real-time status checks just like the other controllers. If a controller fails again, the second and third steps are repeated, and this cycle repeats to ensure the normal operation of the control system.

[0074] Example 2:

[0075] A control system with five controllers a1, a2, a3, a4, and a5 and three actuators z1, z2, and z3, where a1, a2, and a3 are active controllers and a4 and a5 are standby controllers. When a working controller a2 and a standby controller a5 fail, the following steps are used based on the present invention:

[0076] The first step is to determine whether the controller is in normal or faulty state:

[0077] (1) Controller physical serial number matrix Controller State Matrix Perform real-time status judgment on each controller and update the state matrix B in real time based on the judgment results. The update steps are as follows:

[0078] I) The controller a1 has an input signal α1 and an output signal β1. The input signal α1 of the controller a1 is fed through a feedback circuit to generate a feedback signal γ1. The input and feedback signals of the controller a1 are synchronously collected by a remote monitoring system.

[0079] II) Considering the delay in signal transmission, a two-out-of-three logic judgment method is used to judge the controller state. The sampling interval is set to Δt, and the input signal and feedback signal are collected three times continuously to form a logic matrix.

[0080] III) Perform XOR operation on the first and second columns of matrix C1, that is, Judgment Matrix According to the calculation results, calculate the sum of all elements of the judgment matrix D1 If X=2≥m=2, the controller a1 is judged to be normal and the state value b1=1;

[0081] IV) the controller a2 has an input signal α2 and an output signal β2. The input signal α2 of the controller a2 is fed back through the feedback circuit to generate a feedback signal γ2. The input and feedback signals of the controller a2 are synchronously collected by the remote monitoring system.

[0082] V) Considering that there will be a delay in the transmission of the feedback signal γ2, a two-out-of-three logic judgment method with three independent measurements at equal intervals is used to judge the controller state. Set the collection time interval to Δt, and continuously collect three input signals and feedback signals to form a logic matrix

[0083] VI) Perform XOR operation on the first column and the second column of matrix C2, that is, Judgment Matrix According to the calculation results, find the sum of all elements of the judgment matrix D2 If X=1≤m=2, it is determined that the controller a2 is faulty and the state value b2=0;

[0084] VII) Controller a3 has an input signal α3 and an output signal β3. The input signal α3 of controller a3 is fed back through the feedback circuit to generate a feedback signal γ3. The input and feedback signals of controller a3 are synchronously collected by a remote monitoring system.

[0085] VIII) Considering the delay in signal transmission, a two-out-of-three logic judgment method is used to judge the controller state. The sampling interval is set to Δt, and the input signal and feedback signal are collected three times continuously to form a logic matrix.

[0086] IX) Perform an XOR operation on the first column and the second column of the matrix C3, that is, Judgment Matrix According to the calculation results, find the sum of all elements of the judgment matrix D3 If X=3≥m=2, the controller a3 is judged to be normal and the state value b3=1;

[0087] X) The controller a4 has an input signal α4 and an output signal β4. The input signal α4 of the controller a4 is fed back through the feedback circuit to generate a feedback signal γ4. The input and feedback signals of the controller a4 are synchronously collected by the remote monitoring system.

[0088] XI) Considering that there will be a delay in the transmission of the feedback signal γ4, a two-out-of-three logic judgment method with three independent measurements at equal intervals is used to judge the controller state. Set the sampling time interval to Δt, and continuously collect three input signals and feedback signals to form a logic matrix

[0089] XII) Perform an XOR operation on the first and second columns of the matrix C4, that is, Judgment Matrix According to the calculation results, calculate the sum of all elements of the judgment matrix D4 If X=3≥m=2, the controller a4 is judged to be normal and the state value b4=1;

[0090] XIII) controller a5 receives an input signal α5 and an output signal β5. The input signal α5 of controller a5 is fed back through a feedback circuit to generate a feedback signal γ5. The input, output, and feedback signals of controller a5 are synchronously collected by a remote monitoring system.

[0091] XIV) Considering that there will be a delay in the transmission of the feedback signal γ5, a two-out-of-three logic judgment method with three independent measurements at equal intervals is used to judge the controller state. Set the collection time interval to Δt, and continuously collect three input signals and feedback signals to form a logic matrix

[0092] XV) Perform an XOR operation on the first and second columns of the matrix C5, that is, Judgment Matrix According to the calculation results, calculate the sum of all elements of the judgment matrix D5 If X=0≤m=2, it is determined that the controller a5 is faulty and the state value b5=0;

[0093] (2) Perform state judgment on each controller, obtain the state values ​​of the five controllers, and update the state matrix

[0094] The second step is to dynamically adjust the priority after the controller fails:

[0095] (1) Priority matrix for initial controller installation Establish the controller priority adjustment matrix based on the state judgment in the first step

[0096] (2) Find the sum Y of the elements in the first column of matrix F, that is, the number of normal controllers If Y=3≥m=2, the normal hot standby controller replaces all failed working controllers through the switching of dynamic switches.

[0097] (3) If Y = 3 ≥ m = 2, the priority of each controller is dynamically adjusted according to the following steps:

[0098] I) Retrieve the state value b1 of controller a1, b1=1, the priority of all controllers remains unchanged, and the priority update matrix

[0099]

[0100] II) Retrieve the status value b2 of controller a2, b2 = 0, the current priority of controller a2 is 2, the priority of controller a1 with a higher priority than 2 remains unchanged, the priority of controllers a2, a4, and a5 with a lower priority than 2 is increased by one level, and the priority of controller a2 is reduced to 5. The priority update matrix

[0101] III) Retrieve the state value b3 of controller a3, b3=1, the priorities of all controllers remain unchanged, and the priority update matrix

[0102] IV) Retrieve the state value b4 of controller a4, b4=1, the priorities of all controllers remain unchanged, and the priority update matrix

[0103] V) Retrieve the status value b5 of controller a5, b5 = 0. The current priority of controller a5 is 4. The priorities of controllers a1, a3, and a4 with higher priorities than 4 remain unchanged. The priority of controller a2 with lower priority than 4 is increased by one level. The priority of controller a5 is reduced to 5. The priority update matrix

[0104] (4) Complete the dynamic adjustment of the priorities of the five controllers and obtain the updated matrix The third step is to implement dynamic redundant switching of failed controllers based on controller priority:

[0105] (1) Physical serial number matrix of actuator Connection matrix H between controllers and actuators, H = 5 controllers

[0106] (2) From the second step, we know that Y = 3 ≥ m = 2. Then, by switching the dynamic switch, the normal hot standby controller replaces all the faulty working controllers. The steps are as follows:

[0107] I) State Matrix Multiply each column of the connection matrix H in turn to obtain the connection matrix after the state is updated Find the sum of all elements of the H0 matrix The connection matrix H0 and the priority matrix E form an augmented matrix In the augmented matrix Retrieve the elements with values ​​of 1 in columns 1 to 3, and delete the rows and columns corresponding to element 1 to obtain the reduced matrix according to The priority size corresponding to each row of the matrix selects the top mk=3-2=1 rows of priority and obtains the local connection matrix

[0108] II) The matrix The main diagonal elements are assigned a value of 1 and the other elements remain unchanged, and the updated local connection matrix is ​​obtained. The controller a4 in the row corresponding to element 1 and the actuator z2 in the column corresponding to element 1 are connected by switching the dynamic switch, as shown in the following example: Figure 8 As shown;

[0109] III) The matrix The elements in replace the corresponding elements in the H matrix according to the position relationship of the corresponding rows and columns, and the updated matrix H = 5 controllers are obtained.

[0110] (3) After the faulty controller is repaired or replaced, the controller is connected to the system to determine its status again and update the status matrix. Adjust the controller priority again according to the updated state matrix and update the priority matrix

[0111] (4) Full information matrix G = (HB E) = 5 controllers The full information matrix G contains the connection status of each controller and actuator, as well as the status and priority of each controller. A newly connected controller performs real-time status checks just like the other controllers. If a controller fails again, the second and third steps are repeated, and this cycle repeats to ensure the normal operation of the control system.

Claims

1. A multi-controller cyclic redundancy switching method based on dynamic priority adjustment and state mapping, characterized by: The first step is to determine whether the controller is in normal or faulty state: (1) n controllers, m actuators, where n ≥ m, that is, there are nm redundant controllers; the physical serial number matrix of the controllers where a i Represents the i-th controller, i = 1, 2, ..., n, the physical serial number of the controller is determined by its installation location; the controller state matrix where b i Indicates that the physical serial number is a i The state value of the controller, b i =1 indicates controller a i Normal, status value b i =0 means controller a i Fault; perform real-time status judgment on each controller and update the status matrix B in real time based on the judgment result. The update steps are as follows: 1) Controller a i The input signal is α i , the output signal is β i , controller a i The input signal α i The feedback circuit generates a feedback signal γ i , synchronized acquisition controller a by remote monitoring system i Input and feedback signals; II) Considering the delay in signal transmission, a two-out-of-three logic judgment method with three independent measurements at equal intervals is used to judge the controller state; the acquisition time interval is set to Δt, and the input signal and feedback signal are collected three times continuously to form a logic matrix in The superscript j=1, 2, 3 indicates the order in which the signals are collected three times. Indicates that the collected input signal and feedback signal are high level, Indicates that the collected input signal and feedback signal are low level; III) For the matrix C i The first and second columns of the same OR operation, that is in ☉ represents the XOR operation, judgment matrix Find the judgment matrix D based on the calculation results i The sum of all elements X≥2, then the controller a i Normal, let the state value b i =1; X<2, then the controller a is determined i Fault, let the state value b i =0; (2) Repeat steps I, II, and III to determine the state of each controller, obtain the state values ​​of n controllers, and update the state matrix B; The second step is to dynamically adjust the priority after the controller fails: (1) Controller priority matrix e i Represents controller a i Priority, e i ∈{1, 2, 3, ..., n}, ∈ is the inclusion relation, that is, the set {1, 2, 3, ..., n} contains e i , e i =1 indicates controller a i The highest priority is e i =n represents controller a i The priority is the lowest. The higher the priority of the controller, the higher the priority is. The symbols Indicates any choice, Indicates existence; when the controller is first installed, its initial priority is determined by the controller physical serial number mapping, that is, the controller installed for the first time The corresponding priority is Controller Priority Adjustment Matrix (2) Find the sum Y of the elements in the first column of matrix F, that is, the number of normal controllers If Y≥m, the normal hot standby controller will replace all the faulty working controllers through the switching of the dynamic switch; if Y<m, the dynamic switch cannot be switched to replace all the faulty working controllers, and the remote monitoring system will alarm. (3) If Y ≥ m, the priority of each controller is dynamically adjusted according to the following steps: 1) Controller a i The status value b i =1, the priority of all controllers remains unchanged; II) Controller a i The status value b i =0, the priority is higher than e i The controller priority remains unchanged and is lower than e i The controller priority of each controller is increased by one level, and controller a i The priority of is reduced to n; (4) According to the physical sequence number from a1→a n Poll the status value b of each controller in turn i , repeat steps I and II to complete the dynamic adjustment of the priorities of n controllers and update the matrix F; The third step is to implement dynamic redundant switching of failed controllers based on controller priority: (1) Physical serial number matrix of actuator where z l represents the lth actuator, l = 1, 2, ..., m, the connection matrix between the controller and the actuator H = n controllers h il =1 indicates controller a i With actuator z l Connected, h il =0 means controller a i Not with actuator z il connected; (2) If Y ≥ m, the normal hot standby controller replaces all the faulty working controllers through the switching of dynamic switches. The steps are as follows: I) Multiply each column of the state matrix B with the connection matrix H in turn to obtain the connection matrix H0 after the state is updated, and calculate the sum of all elements of the H0 matrix The connection matrix H0 and the priority matrix E form an augmented matrix Retrieve the elements with values ​​of 1 from the 1st column to the mth column in the augmented matrix, and delete the rows and columns corresponding to element 1 to obtain the reduced matrix According to the matrix The priority size corresponding to each row in the filter selects the top mk rows of priority to form a local connection matrix II) The matrix The main diagonal elements are assigned a value of 1 and the other elements remain unchanged, and the updated local connection matrix is ​​obtained. The controller in the row corresponding to element 1 is connected to the actuator in the column through the switching of the dynamic switch; III) The matrix The elements in replace the corresponding elements in the matrix H in turn according to the position relationship of the corresponding rows and columns, and update the matrix H; (3) After the faulty controller is repaired or replaced, the controller status is determined again by the access system, and the status matrix B is updated. The controller priority is adjusted again according to the updated status matrix, and the priority matrix E is updated. (4) Full information matrix G = (H|BE). The full information matrix G includes the connection status of each controller and actuator, the status and priority of each controller. The newly connected controller performs real-time status judgment like other controllers. When the controller fails again, the second and third steps are performed again. This cycle is repeated to ensure the normal operation of the control system.

Images