Fuzzy path selection method for routing protocol in power internet of things

Abstract

The application discloses a fuzzy path selection method of a routing protocol in a power internet of things, and comprises the following steps: calculating the covering time between system input nodes and the system utilization, and respectively obtaining the node covering time and the system utilization language state membership function; formulating a language rule according to the step, and obtaining the relationship between the input parameter node covering time, the system utilization language state and the output fuzzy probability language state; the language rule of all input and output functions is defined as a language fuzzy limit language; calculating the system output fuzzy probability, and obtaining the fuzzy probability language state membership function; calculating the sum of the cost between each node of each route according to the fuzzy probability; comparing the route cost of multiple routes, and selecting the route with the minimum cost. Compared with the prior art, the application has the beneficial effects that the technology can improve the routing selection efficiency of the power internet of things terminal device, and reduce the service data transmission delay.

Description

Technical Field

[0001] This invention relates to the field of power Internet of Things (IoT) technology, and in particular to a fuzzy path selection method for routing protocols in the power IoT. Background Technology

[0002] With the continuous development of the national economy, the demand for electricity is constantly increasing, and the number of power grid equipment is also increasing rapidly. The traditional power industry is gradually revealing its shortcomings. The traditional power industry operation mode not only consumes a lot of manpower and material resources, but also has limitations and is not conducive to the overall coordination of the power grid.

[0003] While the Internet of Things (IoT) for the power sector holds great promise, several challenges remain. Insufficient installation of sensors and other equipment is a significant issue. Currently, power grid companies only monitor key nodes along power lines for safe operation. In some areas, terrain limitations, the large footprint of sensors, and transportation difficulties have prevented the installation of a sufficient number of measurement devices. Furthermore, the development of new sensor technologies is still necessary. Network security is another concern. With the interconnection of numerous devices and users, and the integration of various technologies, higher demands are placed on the overall cybersecurity of the power system. The power system involves a wide range of devices and users. A cyberattack or software vulnerability could lead to incalculable economic losses and impact social security and stability. Other issues include the need for continuous improvement in power grid operation scheduling, equipment control, and management systems; timely innovation in energy-saving management mechanisms and methods; and appropriate optimization of power production operation modes, resource allocation of power generation equipment, and safe operation modes of large-scale inter-provincial power grids.

[0004] The Internet of Things (IoT) can effectively improve this situation by providing services through inexpensive and ubiquitous communication technologies. Based on the internet and traditional telecommunications networks as information carriers, it enables an interconnected architecture for tens of thousands of power grid devices, allowing for unified network access management of massive amounts of power grid equipment. The IEEE 802.15.4 standard has been designed for low-power IoT networking technology. Building upon IEEE 802.15.4, the addition of the IPv6-based 6LoWPAN wireless domain area network protocol, with its widespread adoption and vast address space, has led to its application in the power IoT.

[0005] In a fast mobile network such as VANET, the commonly used routing protocol is OLSR. Typically, OLSR uses a shortest path algorithm to find a route from the source node to the routing node. However, when some nodes in the route are at the edge of their neighboring transmission range, and when a node within that range moves out of their transmission range, the route becomes disconnected, and transmission automatically drops. A reconstruction process automatically begins before the source responds, wasting time establishing routes that impact fast-moving networks or real-time services within the network. High-speed node movement leads to frequent network topology changes, short link connection times, increased network load causing node link-layer congestion, and limitations in node energy. Summary of the Invention

[0006] To address the aforementioned issues, this invention provides a fuzzy path selection method for routing protocols in the power Internet of Things.

[0007] The technical solution of this invention is: a fuzzy path selection method for routing protocols in the power Internet of Things, characterized by including the following steps:

[0008] Step 1: Calculate the system input node coverage time (S) t ), system utilization rate (R i ), and respectively obtain the coverage time (S) between nodes. t ), system utilization rate (R i Language state membership function;

[0009] Step 2: Based on the system input node coverage time (S) obtained in Step 1 t ), system utilization rate (R i) Language states are used to formulate language rules, and the inter-node coverage time (S) of input parameters is derived. t ), system utilization rate (R i The relationship between language state and output fuzzy probability language state;

[0010] The language rules for all input and output functions are defined as fuzzy language qualifiers.

[0011] Step 3: Calculate the system output fuzzy probability (P) i ), thus deriving the fuzzy probability (P) i Language state membership function;

[0012] Step 4: Based on the fuzzy probability (P) i ) Calculate the sum of costs between each node for each route, i.e., the route metric (C);

[0013] Step 5: Compare the routing costs of multiple routes and select the route with the lowest cost.

[0014] The routing metric is calculated from the sum of costs between each node of a route. It is assumed that the lower the cost of a route, the higher its quality, as expressed in formula (4).

[0015] (4);

[0016] Assuming that at least two routes have the same routing cost, this solution can be achieved by the expression in (5);

[0017] (5);

[0018] Where C is the routing metric, P i It is a fuzzy probability, where i represents the node index and P n It is a fuzzy probability link on the target node.

[0019] The coverage time between nodes is represented by equation (1);

[0020] (1);

[0021] ;

[0022] Among them, T Rmax V0 and V1 are the maximum transmission range (m), V0 and V1 are the node rate and the rate of adjacent nodes (m / s), α0 and α1 are the nodes on the X-axis and their adjacent nodes (m), β0 and β1 are the nodes on the Y-axis and their adjacent nodes (m), COG R It is the relative COG between nodes;

[0023] The coverage time between nodes is from 0 to 150 at the lower boundary, and the range of member functions is 0-1; there are four language states (L, ML, MH, and H) to represent the number of members;

[0024] System utilization is defined as the second input of the system workload, denoted by equation (2);

[0025] (2);

[0026] Among them, R i For system utilization, μ is the approximate packet processing rate, and λ is a function of the ratio of approximate packet arrival rates;

[0027] There are three state member functions: L, M, and H.

[0028] Fuzzy probabilities define four language states: L, M, H, and VH as levels of quality of inter-node connections;

[0029] The a-cut method is used to calculate the fuzzy probability density and the geometric mean is obtained by defuzzification using the geometric mean; the fuzzy probability is represented by formula (3);

[0030] (3);

[0031] Where y is the center point of each member function, and μx is the strength of the member function. The routing costs of multiple routes to the destination node are compared, and the route with the lowest cost is selected.

[0032] Compared with the prior art, the beneficial effects of the present invention are: the technology can improve the routing efficiency of power Internet of Things terminal devices and reduce the latency of business data transmission. Attached Figure Description

[0033] Figure 1 The present invention provides a flowchart of a fuzzy path selection method for routing protocols in the power Internet of Things.

[0034] Figure 2 This is a routing diagram illustrating the forwarding of RREQ and RPLY packets from node S to node D. Detailed Implementation

[0035] As shown in the figure:

[0036] The fuzzy path selection method for routing protocols in the power Internet of Things includes the following steps:

[0037] Step 1: Calculate the system input node coverage time (S) t ), system utilization rate (R i ), and respectively obtain the coverage time (S) between nodes. t ), system utilization rate (R i Language state membership function;

[0038] Step 2: Based on the system input node coverage time (S) obtained in Step 1 t ), system utilization rate (R i) Language states are used to formulate language rules, and the inter-node coverage time (S) of input parameters is derived. t ), system utilization rate (R i The relationship between language state and output fuzzy probability language state;

[0039] The language rules for all input and output functions are defined as fuzzy language qualifiers.

[0040] Step 3: Calculate the system output fuzzy probability (P) i ), thus deriving the fuzzy probability (P) i Language state membership function;

[0041] Step 4: Based on the fuzzy probability (P) i Calculate the sum of costs between each node for each route, i.e., the route metric (C);

[0042] Step 5: Compare the routing costs of multiple routes and select the route with the lowest cost.

[0043] The routing metric is calculated from the sum of costs between each node of a route. It is assumed that the lower the cost of a route, the higher its quality, as expressed in formula (4).

[0044] (4);

[0045] Assuming that at least two routes have the same routing cost, this solution can be achieved by the expression in (5);

[0046] (5);

[0047] Where C is the routing metric, P i It is a fuzzy probability, where i represents the node index and P n It is a fuzzy probability link on the target node.

[0048] The coverage time between nodes is represented by equation (1);

[0049] (1);

[0050] ;

[0051] Among them, T Rmax V0 and V1 are the maximum transmission range (m), V0 and V1 are the node rate and the rate of adjacent nodes (m / s), α0 and α1 are the nodes on the X-axis and their adjacent nodes (m), β0 and β1 are the nodes on the Y-axis and their adjacent nodes (m), COG R It is the relative COG between nodes;

[0052] The coverage time between nodes is from 0 to 150 at the lower boundary, and the range of member functions is 0-1; there are four language states (L, ML, MH, and H) to represent the number of members;

[0053] System utilization is defined as the second input of the system workload, denoted by equation (2);

[0054] (2);

[0055] Among them, R i For system utilization, μ is the approximate packet processing rate, and λ is a function of the ratio of approximate packet arrival rates;

[0056] There are three state member functions: L, M, and H.

[0057] Fuzzy probabilities define four language states: L, M, H, and VH as levels of quality of inter-node connections;

[0058] The a-cut method is used to calculate the fuzzy probability density and the geometric mean is obtained by defuzzification using the geometric mean; the fuzzy probability is represented by formula (3);

[0059] (3);

[0060] Where y is the center point of each member function, μ x It is the strength of the member function.

[0061] The process of comparing the routing costs of multiple routes to the destination node and selecting the route with the lowest cost is as follows:

[0062] 1) The RREQ packet is transmitted from the source node S to the destination node D for route discovery and route construction. The route response is unicast from node D to node S using the RPLY packet to find the forward route to the destination node D (1) S→C1→C 21 →C 31 →D,(2)S→C1→C 22 →C 32 →D and (3)S→C1→C 22 →C 23 →C 32 →D (The above uses alphanumeric symbols and arrows to represent routes, and...) Figure 2 (corresponding to the middle)

[0063] 2) Here, each link from S to D is unidirectional; in this process, each node directly receives RREQ packets through the nodes on the forward route. Since node D receives the first RREQ packet from the forward route, it starts the T4 timer to wait for REQs from other routes.

[0064] 3) When the T4 timer expires, node D receives RREQ from routes (1), (2), and (3). Then, node D starts comparing the loss of each route, assuming that route (2) has the lowest cost, and the RPLY packets sent by D are returned only by route (2).

[0065] The inter-node coverage time (S) t ), system utilization rate (R i ), fuzzy probability (P) i The routing metric (C) feature is calculated using the following formula:

[0066] The coverage time between nodes is represented by formula (1).

[0067] (1);

[0068] ;

[0069] Among them, T Rmax V is the maximum value of the transmission range (m), V and V1 are the node rate and the rate of adjacent nodes (m / s), α0 and α1 are the nodes on the X-axis and their adjacent nodes (m), β0 and β1 are the nodes on the Y-axis and their adjacent nodes (m), COG R It is the relative COG between nodes;

[0070] The inter-node coverage time ranges from 0 to 150 at the lower boundary, and the member functions range from 0 to 1. There are four language states (L, ML, MH, and H) representing the number of members, as shown in Table 1.

[0071] Linguistic <![CDATA[Lower-value,S t Range]]> Upper-value,Range L - <![CDATA[-0.067(S t - 15),[0,15]]]> ML <![CDATA[0.1(S t - 5),[5,15]]]> <![CDATA[-0.05(S t - 35),[15,35]]]> MH <![CDATA[0.1(S t - 15),[0,15]]]> <![CDATA[-0,035(S t - 55),[0,15]]]> H <![CDATA[0.009(S t - 36),[36,150]]]> <![CDATA[1,S t >150]]>

[0072] Table 1. Inter-node coverage time function

[0073] System utilization is defined as the second input to the system workload, R i R is a function of the ratio of the approximate packet processing rate to the approximate packet arrival rate. i The boundary range is 0-1. The system utilization rate is calculated and expressed by formula (2).

[0074] (2);

[0075] Among them, R i For system utilization, μ is the approximate packet processing rate, and λ is a function of the ratio of approximate packet arrival rates;

[0076] There are three state member functions: L, M, and H. The relational expressions of the state member functions are shown in Table 2.

[0077] Linguistic <![CDATA[Lower-value,S t Range]]> Upper-value,Range L - <![CDATA[-2.60(R i –0.3891),[0,0.3891]]]> M <![CDATA[19(R i –0.2976),[0,2976,0.35]]]> <![CDATA[-1.82(R i –0.9),[0,2976,0.9]]]> H <![CDATA[1.64(R i –0.4061),[0.4061,1]]]> -

[0078] Table 2 System Utilization Function

[0079] The language rules for all input and output functions are defined as linguistic fuzzy qualifiers, which have determined the relationship between inputs and outputs in Table 3.

[0080] Rule No. Inter-node Coverage System Utilization Fuzzy Probability 1 L L H 2 L M VH 3 L H VH 4 M L H 5 M M VH 6 M H VH 7 MH L M 8 MH M H 9 MH H VH 10 H L L 11 H M M 12 H H H

[0081] Table 3 Language Rules

[0082] The fuzzy probability is defined as an output function that can intuitively measure link quality based on the number of nodes between nodes. The fuzzy probability defines four states: L, M, H, and VH as the level of connection quality between nodes, as shown in Table 4.

[0083] Linguistic <![CDATA[Lower-value,S t Range]]> Upper-value,Range L - <![CDATA[-1.45(P i -0.7),[0,0.7]]]> M <![CDATA[6.68(P i -0.2),[0.2,0.35]]]> <![CDATA[-5(P i -0.55),[0.35,0.55]]]> H <![CDATA[5(P i -0.45),[0.45,0.65]]]> <![CDATA[-(P i -0.85),[0.65,0.85]]]> VH <![CDATA[5(P i -0.8),[36,150]]]> -

[0084] Table 4 Fuzzy Probability Functions

[0085] The cutting method is used to calculate the fuzzy probability degree, and the geometric mean is used to defuzzify and obtain the geometric mean. The fuzzy probability is calculated by formula (3).

[0086] (3);

[0087] Where y is the center point of each member function, μ x It is the strength of the member function.

[0088] The routing metric is calculated from the sum of costs between each node of a route. It is assumed that the lower the cost of a route, the higher its quality. The route cost is calculated and expressed in formula (4).

[0089] (4);

[0090] Assuming at least two routes have the same routing cost, then it is necessary to...

[0091] (5);

[0092] Where C is the link routing metric, P i It is a fuzzy probability, where i represents the node index and P n It is a fuzzy probabilistic link on the target node. This technology can improve the routing efficiency of power Internet of Things (IoT) terminal devices and reduce the latency of business data transmission.

[0093] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A fuzzy path selection method for routing protocols in the power Internet of Things, characterized by: Includes the following steps: Step 1: Calculate the system input node coverage time (S) t ), system utilization rate (R i ), and respectively obtain the coverage time (S) between nodes. t ), system utilization rate (R i Language state membership function; Step 2: Based on the system input node coverage time (S) obtained in Step 1 t ), system utilization rate (R i) Language states are used to formulate language rules, and the inter-node coverage time (S) of input parameters is derived. t ), system utilization rate (R i The relationship between language state and output fuzzy probability language state; The language rules for all input and output functions are defined as fuzzy language qualifiers; Step 3: Calculate the system output fuzzy probability (P) i ), thus deriving the fuzzy probability (P) i Language state membership function; Step 4: Based on the fuzzy probability (P) i Calculate the sum of costs between each node for each route, i.e., the route metric (C); Step 5: Compare the routing costs of multiple routes and select the route with the lowest cost; The coverage time between nodes is represented by equation (1); (1)； ； Among them, T Rmax V0 and V1 are the maximum transmission range (m), V0 and V1 are the node rate and the rate of adjacent nodes (m / s), α0 and α1 are the nodes on the X-axis and their adjacent nodes (m), β0 and β1 are the nodes on the Y-axis and their adjacent nodes (m), COG R It is the relative COG between nodes; The coverage time between nodes is from 0 to 150 at the lower boundary, and the range of member functions is 0-1; there are four language states (L, ML, MH, and H) to represent the number of members; System utilization is defined as the second input of the system workload, denoted by equation (2); (2)； Among them, R i For system utilization, μ is the approximate packet processing rate, and λ is a function of the ratio of approximate packet arrival rates; There are three state member functions: L, M, and H.

2. The fuzzy path selection method for routing protocols in the power Internet of Things according to claim 1, characterized in that: The routing metric is calculated from the sum of costs between each node of a route. It is assumed that the lower the cost of a route, the higher its quality, as expressed in formula (4). (4)； Assuming that at least two routes have the same routing cost, this solution can be achieved by the expression in (5); (5)； Where C is the routing metric, P i It is a fuzzy probability, where i represents the node index and P n It is a fuzzy probability link on the target node.

3. The fuzzy path selection method for routing protocols in the power Internet of Things according to claim 1, characterized in that: Fuzzy probabilities define four language states: L, M, H, and VH as levels of quality of inter-node connections; The a-cut method is used to calculate the fuzzy probability density and the geometric mean is obtained by defuzzification using the geometric mean; the fuzzy probability is represented by formula (3); (3)； Where y is the center point of each member function, μ x It is the strength of the member function.

Images