Electric power emergency material optimal configuration method based on quantum genetic algorithm

Abstract

The invention discloses an electric power emergency material optimal configuration method based on a quantum genetic algorithm, and belongs to the technical field of electric power system resource emergency response, and the method comprises the steps: carrying out the statistics of the number of disaster points and material points of an electric power system, and obtaining the positions of all disaster points and all material points; external factors of current disaster-affected points and material points are obtained, and the number of various emergency materials of the disaster-affected points is increased in a graded mode; constructing a power system emergency material allocation model based on external factors; and solving the emergency material allocation model by using a quantum genetic algorithm, updating the population through a quantum revolving door, and iterating an optimal emergency material allocation scheme. According to the method, the emergency material allocation model is constructed, the quantum genetic algorithm is used for solving, and a feasible solution thought and scheme are provided for coping with a rapid emergency plan of a power system.

Description

Technical Field

[0001] The present invention relates to the technical field of power system resource emergency response, and in particular to a method for optimizing configuration of power emergency materials based on quantum genetic algorithm. Background Art

[0002] In order to timely grasp the operating mechanism of sudden disasters and deploy emergency response scheduling plans. Existing traditional emergency management methods mostly use case-based reasoning (CBR) methods, naive Bayesian classification algorithms, Markov decision-making, etc. These technologies often have high requirements on the quality of training data, lack fast convergence and optimal solution efficiency in high-dimensional data computing space, and have certain defects in dealing with power grid system resource emergency response plans and decisions. Traditional emergency plan generation algorithms have been unable to meet the increasingly complex digital environment requirements of power grid systems. Summary of the invention

[0003] The purpose of the present invention is to provide a method for optimizing the allocation of power emergency materials based on a quantum genetic algorithm. First, the external emergency environment factors are extracted by physical observation means, an emergency material allocation model is constructed based on external factors, and a quantum genetic algorithm is used to solve it to obtain the optimal emergency material allocation result.

[0004] To achieve the above object, the present invention provides a method for optimizing the configuration of power emergency materials based on a quantum genetic algorithm, the steps comprising:

[0005] S1. Count the number of disaster-affected points and material points in the power system, and obtain the location of each disaster-affected point and each material point;

[0006] S2. Obtain the external factors of each disaster-stricken point and each material point, and increase the quantity of various emergency materials in the disaster-stricken point by level;

[0007] S3. Construct an emergency material deployment model for the power system based on external factors;

[0008] S4. Use quantum genetic algorithm to solve the emergency material allocation model, update the population through quantum revolving door, and iterate the optimal emergency material allocation plan.

[0009] Preferably, obtaining the external factors of each current disaster-stricken point and each material point in step S2 includes: observing and evaluating the current weather level through physical means, obtaining the quantity of various emergency materials required for each disaster-stricken point according to rough set theory and expert advice, observing the road congestion and length from each material point to the disaster-stricken point, and observing whether any special emergency situation occurs at each disaster-stricken point, wherein the weather includes wind and rain.

[0010] Preferably, in step S3, constructing a power system emergency material allocation model based on external factors includes:

[0011] The first goal is to minimize the route for transporting emergency supplies and maintain the continuity of transportation. The formula is:

[0012] minF1=∑ i∈BS ∑ j∈SS ∑ k∈TS LG ij *PA ijk ;

[0013] Where F1 represents the first objective function, PA ijk represents the quantity of k types of materials that material point i plans to deliver to disaster-stricken point j, i represents the material point, BS represents the material point set, k represents the material type, T represents the material type set, j represents the disaster-stricken point, and SS represents the disaster-stricken point set;

[0014] Determine the second goal, considering that the longer the time, the greater the harm, and set the penalty function to suppress this situation. The formula is:

[0015] minF2=∑ j∈SS ∑ k∈TS (ND jk -∑ i∈BS PA ijk )*X jk ;

[0016] Where F2 represents the second objective function, ND jk X represents the number of emergency supplies of type k required at disaster site j for a weather with wind level and rain level. jk It means that if disaster site j lacks material type k, it is recorded as 1, otherwise it is 0;

[0017] Determine the third goal, taking into account the road length and vehicle speed, and penalize vehicles that exceed the road threshold. The formula is:

[0018] f 3＝ ∑ i∈BS ∑ j∈SS ∑ k∈TS ;

[0019] (LG ij -maxlength)*Y ij *carnum i ;

[0020] carnum i =∑ j∈SS [(∑ k∈TS PA ijk *X jk ) / ND jk ];

[0021] Where F3 represents the third objective function, LG ij represents the distance from material point i to disaster point j, maxlength represents the road threshold, Y ij If the distance from material point i to disaster point j exceeds the distance threshold, it is 1; otherwise, it is 0. jk If the disaster site j lacks material type k, it is recorded as 1, otherwise it is 0. i It indicates the number of vehicles to be used in the plan to transport each emergency material point to the disaster point;

[0022] The overall goal of the emergency material allocation model is determined according to the first objective function, the second objective function and the third objective function, and its formula is:

[0023] minF=F1+F2*fb+F3*fc;

[0024] carnum <SC i i∈BS;

[0025] ∑ j∈SS PA ijk <SC ik i∈BS,k∈TS;

[0026] Where fb and fc represent hyperparameters, respectively, and represent the influence of F2 excitation and F3 excitation SC. ik represents the storage capacity of emergency supplies of type k stored at supply point i, SC i Indicates the number of standard transport vehicles in resource point i.

[0027] Preferably, in step S4, a quantum genetic algorithm is used to solve the emergency material allocation model, and the population is updated through a quantum revolving door to iterate the optimal emergency material allocation plan, including:

[0028] S41. Input preconditions, including constraints, incentive functions, disaster category levels, power material distribution, environmental parameters, and communication levels;

[0029] S42, initializing the population: using quantum bits to encode and generate quantum chromosomes;

[0030] S43, define the number of iterations, and add 1 each time the number of iterations;

[0031] S44, converting the quantum code into a classical state, and then converting it into an actual emergency material allocation plan, scoring it according to the emergency material allocation model, and calculating the fitness value;

[0032] S45, based on the comparison between the fitness value and the global optimal fitness value, it is more like a quantum revolving door angle θ, and θ is used to perform a revolving door operation on the quantum chromosome to obtain a new population;

[0033] S46: Whether the number of iterations meets the end condition, if so, continue the iteration and enter step S43; if not, return to step S44 and perform a new iteration.

[0034] Preferably, the quantum rotating door strategy in step S45 is:

[0035] △θ=θmin+f'*(θmax-θmin);

[0036] f' = (fx - fbest) / fx;

[0037] θ'=|△θ*f'|;

[0038] In the formula, θmin and θmax represent pre-given parameters, representing the upper and lower limits of the rotation angle θ, fx and fbest represent the fitness value of the current quantum chromosome measured after observation and the global optimal fitness value, respectively, θ' represents the required quantum rotation angle, and f' represents the fitness impact value.

[0039] Therefore, the present invention adopts the above-mentioned method for optimizing the configuration of power emergency materials based on quantum genetic algorithm, which has the following beneficial effects:

[0040] (1) Collect emergency situation characteristics at the physical level, build an emergency material allocation model based on external factors, and use the excellent global search ability of quantum genetic algorithm in the search for the optimal process to solve the problem and obtain the optimal emergency material allocation result;

[0041] (2) Using quantum genetic algorithm to solve the problem avoids the situation where the quality of training data is too high and improves efficiency;

[0042] (3) Adapt to the problems of suddenness, complexity and uncertainty that occur in current power system accidents.

[0043] The technical solution of the present invention is further described in detail below through the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] Figure 1 is a flow chart of a method according to an embodiment of the present invention;

[0045] Figure 2 Schematic diagram of quantum bit calculation in an embodiment of the present invention. DETAILED DESCRIPTION

[0046] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, rather than all of the embodiments. The components of the embodiments of the present invention generally described and shown in the drawings here can be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present invention provided in the drawings is not intended to limit the scope of the claimed invention, but merely represents selected embodiments of the present invention. Based on the embodiments in the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0047] Example

[0048] Reference Figure 1 The present invention provides a method for optimizing the configuration of power emergency materials based on a quantum genetic algorithm, the steps comprising:

[0049] S1. Count the number of disaster-affected points and material points in the power system, obtain the location of each disaster-affected point and each material point, and process the obtained data.

[0050] S2. Obtain the external factors of each disaster-stricken point and each material point, and increase the quantity of various emergency materials in the disaster-stricken points in different levels.

[0051] Obtaining external factors includes: observing and evaluating the weather at the current disaster-stricken points and various material points through physical means, such as the levels of wind and rain. Each type is defined as three levels. According to rough set theory and expert advice, the quantity of various emergency materials required for each disaster-stricken point is obtained, the road congestion and length from each material point to the disaster-stricken point are observed, and whether special emergency situations occur at each disaster-stricken point.

[0052] S3. Construct an emergency material allocation model for the power system based on external factors.

[0053] In order to facilitate the operation of the model, it is assumed that the vehicles are the same, and the running speed and carrying capacity are the same. The construction of the power system emergency material allocation model specifically includes:

[0054] The first goal is to minimize the route for transporting emergency supplies and maintain the continuity of transportation. The formula is:

[0055] minF1=∑ i∈BS ∑ j∈SS ∑ k∈TS LG ij *PA ijk :

[0056] Where F1 represents the first objective function, PA ijkIt represents the quantity of k types of materials that material point i plans to deliver to disaster-stricken point j, i represents the material point, BS represents the set of material points, k represents the type of material, T represents the set of material types, j represents the disaster-stricken point, and SS represents the set of disaster-stricken points.

[0057] Determine the second goal, considering that the longer the time, the greater the harm, and set the penalty function to suppress this situation. The formula is:

[0058] minF2=∑ j∈SS ∑ k∈TS (ND jk -∑ i∈BS PA ijk )*X jk ;

[0059] Where F2 represents the second objective function, ND jk X represents the number of emergency supplies of type k required at disaster site j for a weather with wind level and rain level. jk It means that if disaster site j lacks material type k, it is recorded as 1, otherwise it is 0;

[0060] Determine the third goal, taking into account the road length and vehicle speed, and penalize vehicles that exceed the road threshold. The formula is:

[0061] f 3＝ ∑ i∈BS ∑ j∈SS ∑ k∈TS ;

[0062] (LG ij -maxlength)*Y ij *carnum i ;

[0063] carnum i =∑ j∈SS [(∑ k∈TS PA ijk *X jk ) / ND jk ];

[0064] Where F3 represents the third objective function, LG ij represents the distance from material point i to disaster point j, maxlength represents the road threshold, Y ij If the distance from material point i to disaster point j exceeds the distance threshold, it is 1; otherwise, it is 0. jk If the disaster site j lacks material type k, it is recorded as 1, otherwise it is 0. i It indicates the number of vehicles to be used in the plan to transport each emergency material point to the disaster point;

[0065] The overall goal of the emergency material allocation model is determined according to the first objective function, the second objective function and the third objective function, and its formula is:

[0066] minF=F1+F2*fb+F3*fc;

[0067] carnum <SC i i∈BS;

[0068] ∑ j∈SS PA ijk <SC ik i∈BS,k∈TS;

[0069] Where fb and fc represent hyperparameters, respectively, and represent the influence of F2 excitation and F3 excitation SC. ik represents the storage capacity of emergency supplies of type k stored at supply point i, SC i represents the number of standard transport vehicles in the material point i. F represents the overall goal. The overall goal is to minimize the transportation distance of emergency materials while ensuring sufficient materials and saving time.

[0070] S4. Use quantum genetic algorithm to solve the emergency material allocation model, update the population through quantum revolving door, and iterate the optimal emergency material allocation plan, including:

[0071] S41. Input preconditions, including constraints, incentive functions, disaster category levels, power material distribution, environmental parameters, and communication levels.

[0072] S42. Initialize the population: use quantum bits to encode and generate quantum chromosomes.

[0073] like Figure 2 The figure shows a schematic diagram of the quantum bit computing method. The state of the quantum bit can be represented as a vector on the surface of the Bloch sphere. Any point on the surface of the sphere corresponds to a unique state of the quantum bit. The north pole of the sphere represents the |0> state, the south pole represents the |1> state, and the equator represents the superposition of the |0> and |1> states. In quantum theory, a quantum bit is represented by the superposition of the |0> and |1> states, and its formula is expressed as:

[0074] |ψ>=α|0>+β|1>;

[0075] |α| 2 +|β| 2 =1;

[0076] Among them, |α| 2 and |β| 2Represents the probability of selecting the states |0> and |1> when observing the quantum bit, and α and β satisfy the normalization condition.

[0077] The quantum encoding method is as follows:

[0078]

[0079] In the formula, q represents a quantum chromosome with n quantum bits, αk 2 and βk 2 It represents the probability that the kth quantum bit in the quantum chromosome q is in the transition state of '0' and '1'.

[0080] S43. Define the number of iterations, which is initially 1 and increases by 1 each time.

[0081] S44, converting the quantum code into a classical state, and then converting it into an actual emergency material allocation plan, scoring it according to the emergency material allocation model, and calculating the fitness value;

[0082] S45. According to the comparison between the fitness value and the global optimal fitness value, it is more like a quantum revolving door angle θ, and θ is used to perform a revolving door operation on the quantum chromosome to obtain a new population. The quantum revolving door strategy is:

[0083] △θ=θmin+f'*(θmax-θmin);

[0084] f' = (fx - fbest) / fx;

[0085] θ'=|△θ*f'|;

[0086] In the formula, θmin and θmax represent pre-given parameters, representing the upper and lower limits of the rotation angle θ, fx and fbest represent the fitness value of the current quantum chromosome measured after observation and the global optimal fitness value, respectively, θ' represents the required quantum rotation angle, and f' represents the fitness impact value.

[0087] S46: Whether the number of iterations meets the end condition, if so, continue the iteration and enter step S43; if not, return to step S44 and perform a new iteration.

[0088] In order to verify the effectiveness of the method of the present invention, a simulation experiment is set up as shown below.

[0089] The experiment uses simulation data, setting 6 material points, 6 disaster-affected points, and three types of materials (people, resources, and materials), that is, 36 roads, each of which is shown in Table 1. The storage capacity of the three types of materials and vehicles stored in the six material points is shown in Table 2. The weather is set to wind and rain (each divided into three levels 1, 2, and 3, the higher the level, the greater the harm). According to the wind and rain levels, the quantity of materials required by the six disaster-affected points under nine weather conditions is given according to the experts' suggestions, as shown in Table 3.

[0090] Table 1. Road distances

[0091]

[0092] Table 2. Material point inventory

[0093]

[0094]

[0095] Table 3 Demand of disaster-affected areas

[0096]

[0097] The particle swarm optimization algorithm (Pso) and genetic algorithm (GA) are compared with the quantum genetic algorithm in this application. 100 groups of tests are performed on the simulation data and the algorithm, with 5 rounds in each group. The comparison results are shown in Table 4.

[0098] Table 4 Comparison results

[0099] algorithm TSN AV_L MA_L MI_L HC QC LE IG Ga 20 4.75243 7.28571 2.2381 56.6928 94.7965 80 81 Pso 30 5.01362 6.99076 2.02752 64.3282 94.7778 100 91 QGA 8 4.65165 7.14231 2.24567 45.8333 92.7125 65 100

[0100] The concepts of parameter evaluation indicators are as follows:

[0101] TSN: Total shortage of emergency supplies;

[0102] AV_L: average road length for transporting emergency supplies;

[0103] MA_L: the longest road length for transporting emergency supplies;

[0104] MI_L: shortest road length for transporting emergency supplies;

[0105] HC: Fifty percent of the material arrival rate in the current round;

[0106] QC: 75% material arrival rate in the current round;

[0107] LE: The final epoch number to reach the optimal value of the activation function;

[0108] IG: unusual number of groups.

[0109] From the results, the quantum genetic algorithm converges faster than the general genetic algorithm; and it chooses a relatively shorter path, which can improve the timeliness of the emergency plan in the face of emergency situations; at the same time, the general number of groups shows that the quantum genetic algorithm is more stable. Through simulation experiments, it is effectively demonstrated that the quantum genetic algorithm has certain advantages and practicality over the ordinary genetic algorithm and particle swarm algorithm.

[0110] Therefore, the present invention adopts the above-mentioned method for optimizing the configuration of power emergency materials based on quantum genetic algorithm. By constructing an emergency material allocation model and using quantum genetic algorithm to solve it, it provides a feasible solution and plan for responding to the rapid emergency plan of the power system.

[0111] Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that they can still modify or replace the technical solution of the present invention with equivalents, and these modifications or equivalent replacements cannot cause the modified technical solution to deviate from the spirit and scope of the technical solution of the present invention.

Claims

1. A method for optimizing the configuration of power emergency materials based on quantum genetic algorithm, characterized in that the steps include: S1. Count the number of disaster-affected points and material points in the power system, and obtain the location of each disaster-affected point and each material point; S2. Obtain the external factors of each disaster-stricken point and each material point, and increase the quantity of various emergency materials in the disaster-stricken point by level; S3. Construct an emergency material deployment model for the power system based on external factors; S4. Use quantum genetic algorithm to solve the emergency material allocation model, update the population through quantum revolving door, and iterate the optimal emergency material allocation plan.

2. According to claim 1, a method for optimizing the configuration of power emergency materials based on quantum genetic algorithm is characterized in that: The external factors of each current disaster-stricken point and each material point are obtained in step S2, including: observing and evaluating the current weather level through physical means, obtaining the quantity of various emergency materials required for each disaster-stricken point according to rough set theory and expert advice, observing the road blockage and length from each material point to the disaster-stricken point, and observing whether any special emergency situation occurs at each disaster-stricken point, wherein the weather includes wind and rain.

3. According to claim 2, a method for optimizing the configuration of power emergency materials based on quantum genetic algorithm is characterized in that: In step S3, the power system emergency material allocation model is constructed based on external factors, including: The first goal is to minimize the route for transporting emergency supplies and maintain the continuity of transportation. The formula is: minF1=∑ i∈BS ∑ j∈SS ∑ k∈TS LG ij *PA ijk ; Where F1 represents the first objective function, PA ijk represents the quantity of k types of materials that material point i plans to deliver to disaster-stricken point j, i represents the material point, BS represents the material point set, k represents the material type, T represents the material type set, j represents the disaster-stricken point, and SS represents the disaster-stricken point set; Determine the second goal, considering that the longer the time, the greater the harm, and set the penalty function to suppress this situation. The formula is: minF2=∑ j∈SS ∑ k∈TS (ND jk -∑ i∈BS PA ijk )*X jk ; Where F2 represents the second objective function, ND jk X represents the number of emergency supplies of type k required at disaster site j for a weather with wind level and rain level. jk It means that if disaster site j lacks material type k, it is recorded as 1, otherwise it is 0; Determine the third goal, taking into account the road length and vehicle speed, and penalize vehicles that exceed the road threshold. The formula is: minF 3＝ ∑ i∈BS ∑ j∈SS ∑ k∈TS ; (LG ij -maxlength)*Y ij *carnum i ; carnum i =∑ j∈SS [(∑ k∈TS PA ijk *X jk ) / ND jk ]; Where F3 represents the third objective function, LG ij represents the distance from material point i to disaster point j, maxlength represents the road threshold, Y ij If the distance from material point i to disaster point j exceeds the distance threshold, it is 1; otherwise, it is 0. jk If the disaster site j lacks material type k, it is recorded as 1, otherwise it is 0. i It indicates the number of vehicles to be used in the plan to transport each emergency material point to the disaster point; The overall goal of the emergency material allocation model is determined according to the first objective function, the second objective function and the third objective function, and its formula is: minF=F1+F2*fb+F3*fc; fleshy <SC i i∈BS; ∑ j∈SS PA ijk <SC ik i∈BS,k∈TS; Where fb and fc represent hyperparameters, respectively, and represent the influence of F2 excitation and F3 excitation SC. ik represents the storage capacity of emergency supplies of type k stored at supply point i, SC i Indicates the number of standard transport vehicles in resource point i.

4. The method for optimizing the configuration of power emergency materials based on quantum genetic algorithm according to claim 3 is characterized in that: In step S4, a quantum genetic algorithm is used to solve the emergency material allocation model, and the population is updated through a quantum revolving door to iterate the optimal emergency material allocation plan, including: S41. Input preconditions, including constraints, incentive functions, disaster category levels, power material distribution, environmental parameters, and communication levels; S42, initializing the population: using quantum bits to encode and generate quantum chromosomes; S43, define the number of iterations, and add 1 each time the number of iterations; S44, converting the quantum code into a classical state, and then converting it into an actual emergency material allocation plan, scoring it according to the emergency material allocation model, and calculating the fitness value; S45, based on the comparison between the fitness value and the global optimal fitness value, it is more like a quantum revolving door angle θ, and θ is used to perform a revolving door operation on the quantum chromosome to obtain a new population; S46: Whether the number of iterations meets the end condition, if so, continue the iteration and enter step S43; if not, return to step S44 and perform a new iteration.

5. The method for optimizing the configuration of power emergency materials based on quantum genetic algorithm according to claim 4 is characterized in that: The quantum revolving door strategy in step S45 is: △θ=θmin+f'*(θmax-θmin); f' = (fx - fbest) / fx; θ'=|△θ*f'|; In the formula, θmin and θmax represent pre-given parameters, representing the upper and lower limits of the rotation angle θ, fx and fbest represent the fitness value of the current quantum chromosome measured after observation and the global optimal fitness value, respectively, θ' represents the required quantum rotation angle, and f' represents the fitness impact value.

Images