Freight rate optimization method based on route optimization

Abstract

The invention discloses a freight rate optimization method based on route optimization, and belongs to the technical field of route optimization, and the method comprises the steps: constructing an electricity-coal logistics network, and carrying out the minimum solving of the transportation cost through employing an improved particle swarm algorithm based on the electricity-coal logistics network, thereby rapidly determining an optimal path between a transportation starting node and a transportation ending node, and improving the transportation efficiency. By optimizing the electricity and coal transportation route, the transportation cost is reduced, the transportation efficiency is improved, so that the transportation price is optimized, and the method has a wide application prospect and is of great significance in promoting the development of the electricity and coal logistics industry.

Description

Technical Field

[0001] The present invention belongs to the technical field of route optimization, and in particular relates to a freight rate optimization method based on route optimization. Background Art

[0002] With the rapid development of my country's economy, the demand for electricity continues to grow. As the main fuel for thermal power generation, the purchase volume of thermal coal has a significant impact on the production cost and economic benefits of power companies. Thermal coal refers to coal used for thermal power generation. It is the main fuel source for thermal power plants, and the transportation problem of thermal coal is becoming increasingly prominent. At present, there are the following problems in the transportation of thermal coal: unreasonable transportation routes lead to high transportation costs; low transportation efficiency affects the supply of thermal coal; the formulation of freight rates lacks scientific basis, which is not conducive to the healthy development of the thermal coal logistics market. In view of the above problems, it is necessary to study a thermal coal freight rate optimization method based on route optimization to reduce transportation costs, improve transportation efficiency, and achieve reasonable freight rates. Summary of the invention

[0003] The present invention provides a freight rate optimization method based on route optimization, which is used to solve the technical problem of high transportation cost caused by unreasonable coal transportation routes in the prior art.

[0004] A freight rate optimization method based on route optimization, comprising:

[0005] Obtaining logistics route data involved in thermal coal transportation, and constructing a logistics network including multiple logistics nodes based on the logistics route data;

[0006] According to the logistics network, determine the transportation mode between any two nodes, and determine the unit weight transportation price corresponding to each transportation mode, and obtain the unit weight transportation price between any two nodes with transportation routes;

[0007] Obtaining the logistics data to be planned, and encoding it in a binary encoding manner based on the logistics network to generate a plurality of coded particles corresponding to the logistics data to be planned;

[0008] Based on the multiple coded particles, an improved particle swarm algorithm is used to iterate the coded particles, with the goal of minimizing the freight rate, to obtain the optimal coded particles that meet the preset conditions;

[0009] The optimal coded particle is decoded, and a freight rate optimization result is determined according to the unit weight transportation price between any two nodes with transportation routes.

[0010] Furthermore, the logistics route data involved in the transportation of thermal coal is obtained, and a logistics network including multiple logistics nodes is constructed according to the logistics route data, including:

[0011] Obtaining logistics route data, wherein the logistics route data includes a plurality of logistics nodes and paths between the logistics nodes;

[0012] A logistics network is constructed based on multiple logistics nodes and the paths between logistics nodes. If there is a path between any two logistics nodes, direct access is allowed; if there is no path between any two logistics nodes, direct access is not allowed.

[0013] Further, according to the logistics network, the transportation mode between any two nodes is determined according to the logistics network, and the unit weight transportation price corresponding to each transportation mode is determined, and the unit weight transportation price between any two nodes with transportation routes is obtained, including:

[0014] Based on the logistics network, determine the mode of transportation and the length of the path between any two logistics nodes with a connecting path; the mode of transportation includes road transportation, railway transportation and / or air transportation;

[0015] According to the transportation mode between any two logistics nodes, determine the unit cost required to transport unit weight of electric coal per unit distance under this transportation mode;

[0016] According to the path length and unit cost, the unit weight transportation price between any two logistics nodes with a connecting path is determined, and the unit weight transportation price between any two nodes with a transportation route is obtained; wherein each transportation mode has a corresponding unit weight transportation price.

[0017] Furthermore, the logistics data to be planned is obtained, and based on the logistics network, it is encoded in a binary encoding manner to generate a plurality of coded particles corresponding to the logistics data to be planned, including:

[0018] Acquire the logistics data to be planned; wherein the logistics data to be planned includes a transportation start node and a transportation end node;

[0019] Based on the logistics network, the states corresponding to the transport start node and the transport end node are set to 1, the states of other logistics nodes are randomly initialized to 1 or 0, and the states corresponding to all nodes in all logistics networks are combined into a string to obtain a binary number;

[0020] The binary number is converted into a decimal number to obtain a coded particle corresponding to the logistics data to be planned, and a plurality of coded particles are repeatedly obtained.

[0021] Furthermore, based on the plurality of coded particles, an improved particle swarm algorithm is used to iterate the coded particles, with the goal of minimizing the freight rate, to obtain the optimal coded particles that meet the preset conditions, including:

[0022] Based on the freight rate, obtain the objective function value corresponding to each coded particle;

[0023] According to the objective function values ​​corresponding to all coded particles, all coded particles are arranged in the order of the objective function from small to large, and then divided into the first particle group, the second particle group and the third particle group according to the ratio of 2:8:2;

[0024] For the first coded particle in the first particle group, another random first coded particle is used to perform adaptive collaborative search, and the states corresponding to the transport start node and the transport end node are set to 1 to obtain the updated first coded particle;

[0025] For the second coded particles in the second particle group, a double-guiding method is used to quickly search for the second coded particles, and the states corresponding to the transport start node and the transport end node are set to 1 to obtain the updated second coded particles;

[0026] For the third coded particle in the third particle group, a probability decision method is used to make the third coded particle follow a first coded particle, and the states corresponding to the transport start node and the transport end node are set to 1, so as to obtain the updated third coded particle;

[0027] Determine whether the preset condition is met, if so, obtain the optimal coded particle according to the updated first coded particle, the updated second coded particle and the updated third coded particle, otherwise return the step of the objective function value corresponding to each coded particle.

[0028] Furthermore, based on the freight rate, the objective function value corresponding to each coded particle is obtained, including:

[0029] For any coded particle, round the coded particle to an integer and convert it into a binary number to obtain the logistics node to be passed corresponding to the coded particle;

[0030] Remove the isolated nodes in the logistics nodes to be passed corresponding to the coded particles to obtain the target logistics nodes; wherein the isolated nodes are used to represent nodes that have only a connection path with one other logistics node or have no connection path with other logistics nodes;

[0031] Based on the target logistics node, all paths between the transport start node and the transport end node are traversed, and the path with the smallest sum of unit weight transport prices is determined as the target path;

[0032] The sum of the unit weight transportation prices on the target path is used as the objective function value corresponding to the encoded particle; wherein the smaller the objective function value corresponding to the encoded particle, the better.

[0033] Furthermore, before updating the first coded particle, the method further includes: determining the first adaptive control factor, the second adaptive control factor and the adaptive inertia weight corresponding to each coded particle as follows:

[0034]

[0035] in, represents the first adaptive control factor corresponding to the i-th coded particle in the t-th iteration, c1 max represents the maximum value corresponding to the first adaptive control factor, c1 min represents the minimum value corresponding to the first adaptive control factor, T represents the preset maximum number of iterations, represents the normalization factor corresponding to the i-th encoded particle in the t-th iteration, Indicates the number of encoded particles corresponding to the i-th particle in the t-th iteration. Second adaptive control factor, c2 max represents the maximum value corresponding to the second adaptive control factor, c2 min represents the minimum value corresponding to the second adaptive control factor, represents the inertia weight corresponding to the i-th encoded particle in the t-th iteration, w max Indicates the maximum inertia weight, w min represents the minimum inertia weight, Indicates the quality of the position corresponding to the i-th coded particle in the t-th iteration process, Indicates the minimum position advantage, Indicates the maximum position advantage, represents the i-th encoded particle in the t-th iteration The corresponding fitness is obtained by taking the inverse of the objective function value. represents the optimal coding particle The corresponding fitness, represents the i-th coded particle The corresponding historical optimal value The corresponding fitness, i=1,2,…,I, I represents the total number of encoded particles.

[0036] Furthermore, for the first coded particle in the first particle group, another random first coded particle is used to perform adaptive collaborative search, and the states corresponding to the transport start node and the transport end node are set to 1, to obtain the updated first coded particle, including:

[0037] For any first coded particle, a different other first coded particle is randomly matched to the first coded particle to obtain a cooperative particle corresponding to each first coded particle;

[0038] Based on the cooperative particle corresponding to the first coded particle, the adaptive cooperative search is performed as follows:

[0039]

[0040] in, represents the mth first coded particle in the tth iteration, m = 1, 2, ..., M, M represents the total number of first coded particles, Represents the first coded particle The corresponding collaborative search value, represents the search speed corresponding to the mth first coded particle in the t+1th iteration, Represents the first coded particle The corresponding inertia weight, represents the search speed corresponding to the mth first coded particle in the tth iteration, Represents the first coded particle The corresponding second adaptive control factor, Represents the first coded particle The corresponding collaborative particles;

[0041] Collaborative search value The states corresponding to the transport start node and the transport end node are set to 1, and the first coded particle after update is obtained.

[0042] Furthermore, for the second coded particles in the second particle group, a double-guiding method is used to quickly search for the second coded particles, and the states corresponding to the transport start node and the transport end node are set to 1 to obtain the updated second coded particles, including:

[0043] For any second coded particle, the double-guide method is used to quickly search for the second coded particle:

[0044]

[0045]

[0046] in, represents the nth second coded particle in the tth iteration, n=1,2,…,N, N represents the total number of second coded particles, Represents the second coded particle The corresponding quick search value, represents the search speed corresponding to the nth second-coded particle in the t+1th iteration, represents the search speed corresponding to the nth second-coded particle in the tth iteration, Represents the second coded particle The corresponding inertia weight, Represents the second coded particle The corresponding first adaptive control factor, r1 represents a random number between (0,1), Represents the second coded particle The corresponding historical optimal value is Represents the second coded particle The corresponding second adaptive control factor, r2 represents a random number between (0,1), represents the best coding particle among all coding particles;

[0047] Will quickly search for values The states corresponding to the transport start node and the transport end node are set to 1, and the updated second coded particle is obtained.

[0048] Furthermore, for the third coded particle in the third particle group, a probability decision method is adopted to make the third coded particle follow a first coded particle, and the states corresponding to the transport start node and the transport end node are set to 1, so as to obtain the updated third coded particle, including:

[0049] For the first coded particles in the first particle group, determine the fitness corresponding to each first coded particle, and obtain the sum of the fitness corresponding to all the first coded particles;

[0050] The fitness corresponding to the first coded particle is divided by the total fitness to obtain the probability of the first coded particle being followed;

[0051] For any third coded particle, based on the probability of being followed, a roulette method is used to determine the following coded particle corresponding to the third coded particle;

[0052] Based on the following coded particle corresponding to the third coded particle, the guided search for the third coded particle is:

[0053]

[0054] in, represents the kth third-coded particle in the tth iteration, k = 1, 2, ..., K, K represents the total number of third-coded particles, Represents the third coded particle The corresponding guided search value, represents the search speed corresponding to the kth third-coded particle in the t+1th iteration, represents the search speed corresponding to the kth third-coded particle in the tth iteration, Represents the third coded particle The corresponding inertia weight, Represents the third coded particle The corresponding second adaptive control factor, Indicates following the coded particle;

[0055] Will guide the search value The states corresponding to the transport start node and the transport end node are set to 1, and the updated third coded particle is obtained.

[0056] The present invention provides a freight rate optimization method based on route optimization. By constructing a power coal logistics network, an improved particle swarm algorithm is used to minimize the transportation cost based on the power coal logistics network, so that the optimal path between the transportation start node and the transportation end node can be quickly determined. By optimizing the power coal transportation route, the transportation cost is reduced and the transportation efficiency is improved, thereby realizing the optimization of the freight rate. The method has broad application prospects and is of great significance for promoting the development of the power coal logistics industry. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the invention and, together with the description, serve to explain the principles of the invention.

[0058] Figure 1 A flowchart of a freight rate optimization method based on route optimization provided in an embodiment of the present invention.

[0059] Figure 2 A flow chart of obtaining unit weight transportation price provided by an embodiment of the present invention.

[0060] Figure 3 A flowchart of obtaining optimal coded particles that meet preset conditions is provided in an embodiment of the present invention.

[0061] The above drawings have shown clear embodiments of the present invention, which will be described in more detail below. These drawings and text descriptions are not intended to limit the scope of the present invention in any way, but to illustrate the concept of the present invention for those skilled in the art by referring to specific embodiments. DETAILED DESCRIPTION

[0062] Exemplary embodiments will be described in detail herein, examples of which are shown in the accompanying drawings. When the following description refers to the drawings, the same numbers in different drawings represent the same or similar elements unless otherwise indicated. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present invention. Instead, they are merely examples of devices and methods consistent with some aspects of the present invention as detailed in the appended claims.

[0063] The embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0064] like Figure 1As shown, an embodiment of the present invention provides a method for optimizing freight rates based on route optimization, comprising:

[0065] S101, obtaining logistics route data involved in thermal coal transportation, and constructing a logistics network including multiple logistics nodes according to the logistics route data;

[0066] For example, it is possible to collect which logistics nodes can realize the transportation of thermal coal, and determine the logistics network based on the connection relationship between these logistics nodes. At the same time, in order to facilitate the implementation of subsequent steps, the transportation mode between logistics nodes can also be collected, and the transportation mode can be one or more of road transportation, railway transportation and air transportation.

[0067] S102, determining the transportation mode between any two nodes according to the logistics network, and determining the unit weight transportation price corresponding to each transportation mode, and obtaining the unit weight transportation price between any two nodes with transportation routes;

[0068] After constructing the logistics network, the mode of transportation can be determined from the logistics network. The transportation price of each mode of transportation is relatively fixed, and the unit weight transportation price can be determined based on this, which facilitates subsequent path optimization.

[0069] S103, obtaining the logistics data to be planned, and encoding it in a binary encoding manner based on the logistics network to generate a plurality of coded particles corresponding to the logistics data to be planned;

[0070] The logistics data to be planned include the transportation start node and the transportation end node. These two logistics nodes must be passed, while other logistics nodes may be passed or not passed, so path planning is required. The embodiment of the present invention first uses binary coding to encode, and each element position corresponds to a logistics node. The element is 1 or 0. When the element is 1, it means that the transportation node needs to be passed, and when the element is 0, it means that the node does not need to be passed. However, binary coding is difficult to achieve continuous optimization, so the binary code can be converted into a decimal number, so as to obtain a numerical value that can be solved continuously and obtain a coded particle.

[0071] S104, based on the multiple coded particles, adopting an improved particle swarm algorithm to iterate the coded particles, with the goal of minimizing the freight rate, to obtain the optimal coded particles that meet the preset conditions;

[0072] After the path optimization is converted into coded particles that can be solved continuously, an intelligent optimization algorithm can be used to solve the coded particles, such as using a particle swarm algorithm to iterate the coded particles. However, the traditional particle swarm algorithm has the problem of being easily trapped in local optimality. Therefore, an embodiment of the present invention provides an improved particle swarm algorithm to achieve optimal path planning.

[0073] S105: Decode the optimal coded particle, and determine a freight rate optimization result according to the unit weight transportation price between any two nodes with transportation routes.

[0074] The optimal coded particles are decoded to obtain binary codes. According to the correspondence between the binary codes and the logistics nodes, the target logistics nodes can be obtained. Then, useless nodes in the target logistics nodes are eliminated to obtain the final target logistics nodes. These target logistics nodes represent the nodes that need to be passed through to minimize the freight rate. Useless nodes are nodes that have no connection paths with other target logistics nodes or nodes that have only one path.

[0075] Traverse the final target logistics nodes, determine the transportation path with the lowest freight rate, and obtain the freight rate optimization result. Then, the final freight rate can be determined by the unit weight transportation price between any two nodes with transportation routes and the weight of the electric coal to be transported.

[0076] The present invention provides a freight rate optimization method based on route optimization. By constructing a power coal logistics network, an improved particle swarm algorithm is used to minimize the transportation cost based on the power coal logistics network, so that the optimal path between the transportation start node and the transportation end node can be quickly determined. By optimizing the power coal transportation route, the transportation cost is reduced and the transportation efficiency is improved, thereby realizing the optimization of the freight rate. The method has broad application prospects and is of great significance for promoting the development of the power coal logistics industry.

[0077] In an embodiment of the present invention, the logistics route data involved in the transportation of thermal coal is obtained, and a logistics network including a plurality of logistics nodes is constructed according to the logistics route data, including:

[0078] Obtaining logistics route data, wherein the logistics route data includes a plurality of logistics nodes and paths between the logistics nodes;

[0079] A logistics network is constructed based on multiple logistics nodes and the paths between logistics nodes. If there is a path between any two logistics nodes, direct access is allowed; if there is no path between any two logistics nodes, direct access is not allowed.

[0080] like Figure 2 As shown, according to the logistics network, the transportation mode between any two nodes is determined according to the logistics network, and the unit weight transportation price corresponding to each transportation mode is determined, and the unit weight transportation price between any two nodes with transportation routes is obtained, including:

[0081] S201. Based on the logistics network, determine the mode of transportation and the length of the path between any two logistics nodes with a connecting path; the mode of transportation includes road transportation, railway transportation and / or air transportation;

[0082] This data can be entered by staff or collected by third-party maps.

[0083] S202, according to the transportation mode between any two logistics nodes, determining the unit cost required to transport unit weight of electric coal per unit distance under the transportation mode;

[0084] The unit time may be the average time required to transport a unit distance (such as one kilometer) in historical transportation data, and the unit cost may be the average cost required to transport a unit distance in historical transportation data.

[0085] S203. Determine the unit weight transportation price between any two logistics nodes with a connecting path based on the path length and the unit cost, and obtain the unit weight transportation price between any two nodes with a transportation route; wherein each mode of transportation has a corresponding unit weight transportation price.

[0086] After multiplying the path length and the unit cost, we can get the unit cost required to transport a unit weight of thermal coal from one node to another, that is, the unit weight transportation price, which can be used as a basis to determine the freight rate required to transport the thermal coal between different logistics nodes.

[0087] In an embodiment of the present invention, the logistics data to be planned is obtained, and based on the logistics network, it is encoded in a binary encoding manner to generate a plurality of coded particles corresponding to the logistics data to be planned, including:

[0088] Acquire the logistics data to be planned; wherein the logistics data to be planned includes a transportation start node and a transportation end node;

[0089] Based on the logistics network, the states corresponding to the transport start node and the transport end node are set to 1, the states of other logistics nodes are randomly initialized to 1 or 0, and the states corresponding to all nodes in all logistics networks are combined into a string to obtain a binary number;

[0090] The binary number is converted into a decimal number to obtain a coded particle corresponding to the logistics data to be planned, and a plurality of coded particles are repeatedly obtained.

[0091] like Figure 3 As shown, based on the multiple coded particles, an improved particle swarm algorithm is used to iterate the coded particles, with the goal of minimizing the freight rate, to obtain the optimal coded particles that meet the preset conditions, including:

[0092] S301, based on the freight rate, obtaining the objective function value corresponding to each coded particle;

[0093] S302, according to the objective function values ​​corresponding to all the encoded particles, after arranging all the encoded particles in the order of the objective function from small to large, they are divided into a first particle group, a second particle group and a third particle group according to a 2:8:2 conversion;

[0094] S303, for the first coded particle in the first particle group, use another random first coded particle to perform adaptive collaborative search, and set the states corresponding to the transport start node and the transport end node to 1, to obtain the updated first coded particle;

[0095] S304: for the second coded particles in the second particle group, a double-guiding method is used to quickly search for the second coded particles, and the states corresponding to the transport start node and the transport end node are set to 1, so as to obtain the updated second coded particles;

[0096] S305, for the third coded particle in the third particle group, a probability decision method is used to make the third coded particle follow a first coded particle, and the states corresponding to the transport start node and the transport end node are set to 1, so as to obtain the updated third coded particle;

[0097] S306, determining whether a preset condition is met, if so, obtaining the optimal coding particle according to the updated first coding particle, the updated second coding particle and the updated third coding particle, otherwise returning the step of the objective function value corresponding to each coding particle.

[0098] The preset condition may be: the current number of iterations reaches the maximum number of iterations.

[0099] In the embodiment of the present invention, based on the freight rate, the objective function value corresponding to each encoded particle is obtained, including:

[0100] For any coded particle, round the coded particle to an integer and convert it into a binary number to obtain the logistics node to be passed corresponding to the coded particle;

[0101] Remove the isolated nodes in the logistics nodes to be passed corresponding to the coded particles to obtain the target logistics nodes; wherein the isolated nodes are used to represent nodes that have only a connection path with one other logistics node or have no connection path with other logistics nodes;

[0102] Based on the target logistics node, all paths between the transport start node and the transport end node are traversed, and the path with the smallest sum of unit weight transport prices is determined as the target path; wherein, during the traversal process, the transport mode with the smallest unit weight transport price should be selected;

[0103] The unit weight transportation price refers to the price required to transport a unit weight of thermal coal from one node to another.

[0104] The sum of the unit weight transportation prices on the target path is used as the objective function value corresponding to the encoded particle; wherein the smaller the objective function value corresponding to the encoded particle, the better.

[0105] In the embodiment of the present invention, before updating the first coded particle, the method further includes: determining the first adaptive control factor, the second adaptive control factor and the adaptive inertia weight corresponding to each coded particle as follows:

[0106]

[0107]

[0108] in, represents the first adaptive control factor corresponding to the i-th coded particle in the t-th iteration, c1 max represents the maximum value corresponding to the first adaptive control factor, c1 min represents the minimum value corresponding to the first adaptive control factor, T represents the preset maximum number of iterations, represents the normalization factor corresponding to the i-th encoded particle in the t-th iteration, Indicates the number of encoded particles corresponding to the i-th particle in the t-th iteration. Second adaptive control factor, c2 max represents the maximum value corresponding to the second adaptive control factor, c2 min represents the minimum value corresponding to the second adaptive control factor, represents the inertia weight corresponding to the i-th encoded particle in the t-th iteration, w max Indicates the maximum inertia weight, w min represents the minimum inertia weight, Indicates the quality of the position corresponding to the i-th coded particle in the t-th iteration process, Indicates the minimum position advantage, Indicates the maximum position advantage, represents the i-th encoded particle in the t-th iteration The corresponding fitness is obtained by taking the inverse of the objective function value. represents the optimal coding particle The corresponding fitness, represents the i-th coded particle The corresponding historical optimal value The corresponding fitness, i=1,2,…,I, I represents the total number of encoded particles.

[0109] It is worth noting that, in order to ensure the distribution of the particle group, I can be an integer of 10, so as to facilitate the division of the coded particles into 2:8:2.

[0110] The first adaptive control factor decreases with the increase of iteration number, which meets the requirement of the algorithm to attach importance to individual search in the early stage; the second adaptive control factor increases with the increase of iteration number, which meets the requirement of the algorithm to attach more importance to collective learning experience in the later stage. The inertia weight decreases linearly with the increase of iteration number, gradually increasing the search accuracy and improving the convergence of the algorithm.

[0111] In the embodiment of the present invention, for the first coded particle in the first particle group, another random first coded particle is used to perform adaptive collaborative search, and the states corresponding to the transport start node and the transport end node are set to 1 to obtain the updated first coded particle, including:

[0112] For any first coded particle, a different other first coded particle is randomly matched to the first coded particle to obtain a cooperative particle corresponding to each first coded particle;

[0113] Based on the cooperative particle corresponding to the first coded particle, the adaptive cooperative search is performed as follows:

[0114]

[0115] in, represents the mth first coded particle in the tth iteration, m = 1, 2, ..., M, M represents the total number of first coded particles, Represents the first coded particle The corresponding collaborative search value, represents the search speed corresponding to the mth first coded particle in the t+1th iteration, Represents the first coded particle The corresponding inertia weight, represents the search speed corresponding to the mth first coded particle in the tth iteration, Represents the first coded particle The corresponding second adaptive control factor, Represents the first coded particle The corresponding collaborative particles;

[0116] Collaborative search value The states corresponding to the transport start node and the transport end node are set to 1, and the first coded particle after update is obtained.

[0117] The first coded particle search method provided by the embodiment of the present invention enables the coded particle clock with a better position to be developed in a known optimal area, which is beneficial to improving the iterative efficiency of the algorithm.

[0118] In the embodiment of the present invention, for the second coded particle in the second particle group, a double-guiding method is used to quickly search the second coded particle, and the states corresponding to the transport start node and the transport end node are set to 1 to obtain the updated second coded particle, including:

[0119] For any second coded particle, the double-guide method is used to quickly search for the second coded particle:

[0120]

[0121] in, represents the nth second coded particle in the tth iteration, n=1,2,…,N, N represents the total number of second coded particles, Represents the second coded particle The corresponding quick search value, represents the search speed corresponding to the nth second-coded particle in the t+1th iteration, represents the search speed corresponding to the nth second-coded particle in the tth iteration, Represents the second coded particle The corresponding inertia weight, Represents the second coded particle The corresponding first adaptive control factor, r1 represents a random number between (0,1), Represents the second coded particle The corresponding historical optimal value is Represents the second coded particle The corresponding second adaptive control factor, r2 represents a random number between (0,1), represents the best coding particle among all coding particles;

[0122] Will quickly search for values The states corresponding to the transport start node and the transport end node are set to 1, and the updated second coded particle is obtained.

[0123] The dual-guidance method provided by the embodiment of the present invention can quickly search for unknown areas between the normal area and the optimal area, thereby providing the possibility of searching for the optimal position. At the same time, with the help of adaptive inertia weight and control coefficient, different search strategies can be effectively assigned to each particle, and the early stage tends to global search, and the later stage tends to fine search, thereby improving the efficiency of the algorithm.

[0124] In the embodiment of the present invention, for the third coded particle in the third particle group, a probability decision method is adopted to make the third coded particle follow a first coded particle, and the states corresponding to the transport start node and the transport end node are set to 1, so as to obtain the updated third coded particle, including:

[0125] For the first coded particles in the first particle group, determine the fitness corresponding to each first coded particle, and obtain the sum of the fitness corresponding to all the first coded particles;

[0126] The fitness corresponding to the first coded particle is divided by the total fitness to obtain the probability of the first coded particle being followed;

[0127] For any third coded particle, based on the probability of being followed, a roulette method is used to determine the following coded particle corresponding to the third coded particle;

[0128] Based on the following coded particle corresponding to the third coded particle, the guided search for the third coded particle is:

[0129]

[0130] in, represents the kth third-coded particle in the tth iteration, k = 1, 2, ..., K, K represents the total number of third-coded particles, Represents the third coded particle The corresponding guided search value, represents the search speed corresponding to the kth third-coded particle in the t+1th iteration, represents the search speed corresponding to the kth third-coded particle in the tth iteration, Represents the third coded particle The corresponding inertia weight, Represents the third coded particle The corresponding second adaptive control factor, Indicates following the coded particle;

[0131] Will guide the search value The states corresponding to the transport start node and the transport end node are set to 1, and the updated third coded particle is obtained.

[0132] The search method for the third coded particle provided in an embodiment of the present invention can enable the third coded particle to learn from the first coded particle, and at the same time, combined with the inertia weight, perform a global search in the early stage of the algorithm and a fine search in the later stage of the algorithm, and at the same time, enable the particle group to generate diversified offspring in different search directions, thereby improving the diversity of samples and avoiding falling into a local optimum.

[0133] Those skilled in the art will readily appreciate other embodiments of the present invention after considering the specification and practicing the invention disclosed herein. The present invention is intended to cover any variations, uses or adaptations of the present invention that follow the general principles of the present invention and include common knowledge or customary techniques in the art that are not disclosed by the present invention. It should be understood that the present invention is not limited to the precise structure described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from the scope thereof. The scope of the present invention is limited only by the appended claims.

Claims

1. A freight rate optimization method based on route optimization, characterized in that: include: Obtaining logistics route data involved in thermal coal transportation, and constructing a logistics network including multiple logistics nodes based on the logistics route data; According to the logistics network, determine the transportation mode between any two nodes, and determine the unit weight transportation price corresponding to each transportation mode, and obtain the unit weight transportation price between any two nodes with transportation routes; Obtaining the logistics data to be planned, and encoding it in a binary encoding manner based on the logistics network to generate a plurality of coded particles corresponding to the logistics data to be planned; Based on the multiple coded particles, an improved particle swarm algorithm is used to iterate the coded particles, with the goal of minimizing the freight rate, to obtain the optimal coded particles that meet the preset conditions; The optimal coded particle is decoded, and a freight rate optimization result is determined according to the unit weight transportation price between any two nodes with transportation routes.

2. The method for optimizing freight rates based on route optimization according to claim 1, characterized in that: Obtaining logistics route data involved in the transportation of thermal coal, and constructing a logistics network including multiple logistics nodes based on the logistics route data, including: Obtaining logistics route data, wherein the logistics route data includes a plurality of logistics nodes and paths between the logistics nodes; A logistics network is constructed based on multiple logistics nodes and the paths between logistics nodes. If there is a path between any two logistics nodes, direct access is allowed; if there is no path between any two logistics nodes, direct access is not allowed.

3. The method for optimizing freight rates based on route optimization according to claim 1, characterized in that: According to the logistics network, the transportation mode between any two nodes is determined according to the logistics network, and the unit weight transportation price corresponding to each transportation mode is determined, and the unit weight transportation price between any two nodes with transportation routes is obtained, including: Based on the logistics network, determine the mode of transportation and the length of the path between any two logistics nodes with a connecting path; the mode of transportation includes road transportation, railway transportation and / or air transportation; According to the transportation mode between any two logistics nodes, determine the unit cost required to transport unit weight of electric coal per unit distance under this transportation mode; According to the path length and unit cost, the unit weight transportation price between any two logistics nodes with a connecting path is determined, and the unit weight transportation price between any two nodes with a transportation route is obtained; wherein each transportation mode has a corresponding unit weight transportation price.

4. The method for optimizing freight rates based on route optimization according to claim 3, characterized in that: Obtain the logistics data to be planned, and encode it in a binary encoding manner based on the logistics network to generate a plurality of coded particles corresponding to the logistics data to be planned, including: Acquire the logistics data to be planned; wherein the logistics data to be planned includes a transportation start node and a transportation end node; Based on the logistics network, the states corresponding to the transport start node and the transport end node are set to 1, the states of other logistics nodes are randomly initialized to 1 or 0, and the states corresponding to all nodes in all logistics networks are combined into a string to obtain a binary number; The binary number is converted into a decimal number to obtain a coded particle corresponding to the logistics data to be planned, and a plurality of coded particles are repeatedly obtained.

5. The method for optimizing freight rates based on route optimization according to claim 4, characterized in that: Based on the multiple coded particles, an improved particle swarm algorithm is used to iterate the coded particles, with the goal of minimizing the freight rate, to obtain the optimal coded particles that meet the preset conditions, including: Based on the freight rate, obtain the objective function value corresponding to each coded particle; According to the objective function values ​​corresponding to all coded particles, all coded particles are arranged in the order of the objective function from small to large, and then divided into the first particle group, the second particle group and the third particle group according to the ratio of 2:8:2; For the first coded particle in the first particle group, another random first coded particle is used to perform adaptive collaborative search, and the states corresponding to the transport start node and the transport end node are set to 1 to obtain the updated first coded particle; For the second coded particles in the second particle group, a double-guiding method is used to quickly search for the second coded particles, and the states corresponding to the transport start node and the transport end node are set to 1 to obtain the updated second coded particles; For the third coded particle in the third particle group, a probability decision method is used to make the third coded particle follow a first coded particle, and the states corresponding to the transport start node and the transport end node are set to 1, so as to obtain the updated third coded particle; Determine whether the preset condition is met, if so, obtain the optimal coded particle according to the updated first coded particle, the updated second coded particle and the updated third coded particle, otherwise return the step of the objective function value corresponding to each coded particle.

6. The method for optimizing freight rates based on route optimization according to claim 5, characterized in that: Based on the freight rate, the objective function value corresponding to each coded particle is obtained, including: For any coded particle, round the coded particle to an integer and convert it into a binary number to obtain the logistics node to be passed corresponding to the coded particle; Remove the isolated nodes in the logistics nodes to be passed corresponding to the coded particles to obtain the target logistics nodes; wherein the isolated nodes are used to represent nodes that have only a connection path with one other logistics node or have no connection path with other logistics nodes; Based on the target logistics node, all paths between the transport start node and the transport end node are traversed, and the path with the smallest sum of unit weight transport prices is determined as the target path; The sum of the unit weight transportation prices on the target path is used as the objective function value corresponding to the encoded particle; wherein the smaller the objective function value corresponding to the encoded particle, the better.

7. The method for optimizing freight rates based on route optimization according to claim 5, characterized in that: Before updating the first coded particle, the method further includes: determining the first adaptive control factor, the second adaptive control factor and the adaptive inertia weight corresponding to each coded particle as follows: in, represents the first adaptive control factor corresponding to the i-th coded particle in the t-th iteration, c1 max represents the maximum value corresponding to the first adaptive control factor, c1 min represents the minimum value corresponding to the first adaptive control factor, T represents the preset maximum number of iterations, represents the normalization factor corresponding to the i-th encoded particle in the t-th iteration, Indicates the number of encoded particles corresponding to the i-th particle in the t-th iteration. Second adaptive control factor, c2 max represents the maximum value corresponding to the second adaptive control factor, c2 min represents the minimum value corresponding to the second adaptive control factor, represents the inertia weight corresponding to the i-th encoded particle in the t-th iteration, w max Indicates the maximum inertia weight, w min represents the minimum inertia weight, Indicates the quality of the position corresponding to the i-th coded particle in the t-th iteration process, Indicates the minimum position advantage, Indicates the maximum position advantage, represents the i-th encoded particle in the t-th iteration The corresponding fitness is obtained by taking the inverse of the objective function value. represents the optimal coding particle The corresponding fitness, represents the i-th coded particle The corresponding historical optimal value The corresponding fitness, i=1,2,…,I, I represents the total number of encoded particles.

8. The method for optimizing freight rates based on route optimization according to claim 7, characterized in that: For the first coded particle in the first particle group, another random first coded particle is used to perform adaptive collaborative search, and the states corresponding to the transport start node and the transport end node are set to 1, to obtain the updated first coded particle, including: For any first coded particle, a different other first coded particle is randomly matched to the first coded particle to obtain a cooperative particle corresponding to each first coded particle; Based on the cooperative particle corresponding to the first coded particle, the adaptive cooperative search is performed as follows: in, represents the mth first coded particle in the tth iteration, m = 1, 2, ..., M, M represents the total number of first coded particles, Represents the first coded particle The corresponding collaborative search value, represents the search speed corresponding to the mth first coded particle in the t+1th iteration, Represents the first coded particle The corresponding inertia weight, represents the search speed corresponding to the mth first coded particle in the tth iteration, Represents the first coded particle The corresponding second adaptive control factor, Represents the first coded particle The corresponding collaborative particles; Collaborative search value The states corresponding to the transport start node and the transport end node are set to 1, and the first coded particle after update is obtained.

9. The method for optimizing freight rates based on route optimization according to claim 8, characterized in that: For the second coded particles in the second particle group, a double-guiding method is used to quickly search for the second coded particles, and the transport start node and The state corresponding to the transport termination node is set to 1, and the updated second coded particle is obtained, including: For any second coded particle, the double-guide method is used to quickly search for the second coded particle: in, represents the nth second coded particle in the tth iteration, n=1,2,…,N, N represents the total number of second coded particles, Represents the second coded particle The corresponding quick search value, represents the search speed corresponding to the nth second-coded particle in the t+1th iteration, represents the search speed corresponding to the nth second-coded particle in the tth iteration, Represents the second coded particle The corresponding inertia weight, Represents the second coded particle The corresponding first adaptive control factor, r1 represents a random number between (0,1), Represents the second coded particle The corresponding historical optimal value is Represents the second coded particle The corresponding second adaptive control factor, r2 represents a random number between (0,1), represents the best coding particle among all coding particles; Will quickly search for values The states corresponding to the transport start node and the transport end node are set to 1, and the updated second coded particle is obtained.

10. The method for optimizing freight rates based on route optimization according to claim 9, characterized in that: For the third coded particle in the third particle group, a probability decision method is adopted to make the third coded particle follow a first coded particle, and the states corresponding to the transport start node and the transport end node are set to 1, so as to obtain the updated third coded particle, including: For the first coded particles in the first particle group, determine the fitness corresponding to each first coded particle, and obtain the sum of the fitness corresponding to all the first coded particles; The fitness corresponding to the first coded particle is divided by the total fitness to obtain the probability of the first coded particle being followed; For any third coded particle, based on the probability of being followed, a roulette method is used to determine the following coded particle corresponding to the third coded particle; Based on the following coded particle corresponding to the third coded particle, the guided search for the third coded particle is: in, represents the kth third-coded particle in the tth iteration, k = 1, 2, ..., K, K represents the total number of third-coded particles, Represents the third coded particle The corresponding guided search value, represents the search speed corresponding to the kth third-coded particle in the t+1th iteration, represents the search speed corresponding to the kth third-coded particle in the tth iteration, Represents the third coded particle The corresponding inertia weight, Represents the third coded particle The corresponding second adaptive control factor, Indicates following the coded particle; Will guide the search value The states corresponding to the transport start node and the transport end node are set to 1, and the updated third coded particle is obtained.

Images