A micro energy gateway deployment method, device and medium based on multi-scene matching

Abstract

The application discloses a micro-energy gateway deployment method and device based on multi-scene matching, and a medium, and belongs to the technical field of Internet of Things. The method comprises the following steps: encoding a demand deployment scene into a demand DNA sequence, performing accurate searching, if a hit is obtained, outputting a hit deployment scheme, if no hit is obtained, entering fuzzy matching, converting the demand DNA sequence into a hash signature, performing fuzzy searching, obtaining a similar DNA sequence list, calculating the Hamming distance between each similar DNA sequence and the demand DNA sequence, and determining an intermediate output best candidate scheme or generating a scheme through an optimization model according to the distribution of the minimum and second minimum Hamming distances, and simulating and verifying the intermediate output deployment scheme, and outputting the passed deployment scheme as a final scheme. The application can realize rapid response by reusing historical experience, and can flexibly adapt to variable scenes under the premise of ensuring scheme quality by combining dynamic decision and optimization generation.

Description

Technical Field

[0001] This invention belongs to the field of Internet of Things (IoT) technology, specifically relating to a micro energy gateway deployment method, device, and medium based on multi-scenario matching. Background Technology

[0002] In the actual operation of microgrids, there are problems such as dynamic fluctuations in energy supply and demand (e.g., random output of distributed photovoltaic / wind power and significant peak-valley characteristics of user-side load), complex multi-objective optimization constraints (needing to simultaneously consider deployment costs, energy load balance, and data interaction latency), and strong heterogeneity in application scenarios (large differences in energy consumption patterns between urban and rural power distribution networks and industrial / commercial / residential users). These problems make it difficult for traditional gateway deployment methods to achieve efficient coordination of the entire microgrid process (source-grid-load-storage) under the premise of controllable costs. Specifically, these problems manifest as follows:

[0003] Low deployment efficiency: Traditional methods often rely on human experience or a single mathematical optimization model. For each new scenario, calculations need to be started from scratch. This makes it impossible to effectively utilize historical success cases, resulting in a large amount of repetitive work, long decision-making cycles, and difficulty in dealing with frequently occurring standardized scenarios.

[0004] Poor scenario adaptability: Real-world deployment scenarios are complex and varied, including requirements that are exactly the same as historical cases, as well as new situations that are only partially similar. Existing technologies can either only handle exact matches or rely entirely on fuzzy search, lacking a mechanism to dynamically switch strategies based on the degree of matching. This makes it difficult to quickly obtain high-quality solutions in similar but not identical scenarios. Summary of the Invention

[0005] This invention addresses the shortcomings of existing technologies by providing a micro-energy gateway deployment method, device, and medium based on multi-scenario matching. It can achieve rapid response by reusing historical experience, and at the same time, it combines dynamic decision-making and optimization generation to flexibly adapt to changing scenarios while ensuring the quality of the solution.

[0006] This invention provides the following technical solution: Firstly, a method for deploying a micro energy gateway based on multi-scenario matching is provided, including the following steps: The deployment scenarios are encoded as demand DNA sequences, and the DNA sequences are searched in the scenario library. If a match is found, the deployment scheme corresponding to the matched scenario is output; otherwise, fuzzy matching is performed. Once fuzzy matching is initiated, the required DNA sequence is converted into a hash signature, and the hash signature is searched in the scenario library to obtain a list of similar DNA sequences that match the hash signature. Calculate the Hamming distance between each similar DNA sequence and the required DNA sequence, and select the deployment scheme intermediate output corresponding to the minimum Hamming distance based on the distribution of the minimum Hamming distance and the second smallest Hamming distance, or generate the deployment scheme intermediate output through the optimization model; The intermediate output deployment scheme is simulated and verified. If the verification is successful, the deployment scheme is output as the final scheme. If the verification fails, the deployment scheme is regenerated by optimizing the model until the verification is successful.

[0007] Optionally, encoding the demand deployment scenario into a demand DNA sequence specifically involves digitizing the terminal density, terrain complexity, cost sensitivity, business real-time performance, and business reliability of the demand deployment scenario, and then concatenating them in sequence to form a digital string, which is then used as the DNA sequence.

[0008] Optionally, the required DNA sequence can be converted into a hash signature using a locality-sensitive hash function.

[0009] Optionally, the step of calculating the Hamming distance between each similar DNA sequence and the required DNA sequence, and selecting the intermediate output of the deployment scheme corresponding to the minimum Hamming distance based on the distribution of the minimum and second-smallest Hamming distances, or generating the intermediate output of the deployment scheme through an optimization model, specifically involves: Calculate the Hamming distance between each similar DNA sequence and the required DNA sequence, and sort them in ascending order to obtain a list of candidate solutions. The DNA sequence corresponding to the smallest Hamming distance is the best candidate. Before comparing the best candidate with the list of candidate solutions If the Hamming distance difference between the candidate solutions is less than the set value and the difference is significant, then the deployment solution corresponding to the best candidate will be output as an intermediate solution; otherwise, the deployment solution will be generated by optimizing the model.

[0010] Optionally, the step of generating a deployment plan through model optimization specifically includes: Obtain scenario information for the required deployment scenarios, including terminal information, communication technology constraints, and gateway selection library; Construct an optimization model with the objective functions of minimizing deployment cost, maximum gateway load rate, and total latency, and with constraints of coverage, capacity, distance, and gateway uniqueness. Based on the scenario information of the acquired demand scenarios, a genetic algorithm combined with greedy local search is used to solve the optimization model, and the optimal deployment scheme is output as an intermediate solution.

[0011] Optionally, the objective function of the optimization model for: ; in, , and Candidate deployment schemes The corresponding deployment cost, maximum gateway load rate, and total latency; , and The weights are respectively the deployment cost, the maximum gateway load rate, and the total latency, which are obtained by extracting scenario features and fuzzy inference.

[0012] Optionally, the weights corresponding to deployment cost, maximum gateway load rate, and total latency are obtained by extracting scene features and using fuzzy inference, specifically as follows: Based on the scenario information of the deployment scenario, quantify the scenario characteristics, including terrain complexity, cost sensitivity, and business real-time requirements; Three quantified scene features are used as inputs to the fuzzy system, and membership functions and fuzzy rule bases are set accordingly. Based on membership functions and fuzzy rule bases, the input scene features are fuzzified, activated, aggregated, and defuzzified to obtain the weights corresponding to deployment cost, maximum gateway load rate, and total latency.

[0013] Optionally, it also includes building a rule knowledge base to provide initial solutions when solving the optimization model; The construction of the rule knowledge base specifically involves: analyzing the mapping relationship between the deployment scheme and its corresponding scene features for the deployment scheme that has been verified through simulation and generated based on the optimization model, generating rules, and storing them in the rule knowledge base.

[0014] In a second aspect, a computer device is provided, including a processor and a memory; wherein, when the processor executes a computer program stored in the memory, it implements the steps of the micro-energy gateway deployment method based on multi-scenario matching as described in any one of the first aspects.

[0015] Thirdly, a computer-readable storage medium is provided for storing a computer program; when the computer program is executed by a processor, it implements the steps of the micro-energy gateway deployment method based on multi-scenario matching as described in any one of the first aspects.

[0016] Compared with the prior art, the beneficial effects of the present invention are: This invention achieves a perfect balance between efficiency and accuracy through a progressive screening mechanism. First, the precise matching stage encodes the required scenario as a DNA sequence, directly searching for completely identical cases in the scenario library. Once a match is found, historically mature solutions can be instantly reused, avoiding redundant calculations and providing extremely fast response to frequently occurring standardized scenarios. If no match is found, the fuzzy matching stage begins, using LSH hash signatures to quickly filter out a small number of similar scenarios, compressing the massive candidate set to a controllable range and significantly reducing subsequent computational complexity. Next, Hamming distance is calculated, and intelligent decision-making is based on the distribution of the minimum and second-minimum distances: if a clearly superior candidate exists, historical experience is directly adopted; these solutions are usually proven in practice and highly reliable. If no significant advantage is found, an optimization model is activated to generate a customized new solution, ensuring a feasible solution for any scenario. Finally, closed-loop testing is performed on all intermediate outputs. Only solutions that pass verification are ultimately implemented; those that fail trigger the optimization model to regenerate until the requirements are met. This process fully utilizes accumulated historical knowledge, making the micro-energy gateway deployment method increasingly intelligent with use, while mathematical optimization ensures adaptability to new scenarios. Overall, the method in this application achieves synergistic optimization among rapid response, solution quality, scenario coverage, and system robustness, and is particularly suitable for fields such as micro energy gateway deployment where requirements are highly variable and reliability is extremely demanding. Attached Figure Description

[0017] Figure 1 This is a flowchart of the micro energy gateway deployment method based on multi-scenario matching of the present invention.

[0018] Figure 2 This is a process example of how the present invention outputs a deployment solution when a deployment requirement scenario occurs. Detailed Implementation

[0019] The present invention will be further described below with reference to the accompanying drawings. The following embodiments are only used to more clearly illustrate the technical solutions of the present invention and should not be used to limit the scope of protection of the present invention. It should be noted that the term "comprising" and any variations thereof in the specification, claims and the above-mentioned drawings of the present invention are intended to cover non-exclusive inclusion. For example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to these processes, methods, products or devices.

[0020] Example 1: like Figure 1 As shown, a micro energy gateway deployment method based on multi-scenario matching includes the following steps: S1: Encode the deployment scenario into a requirement DNA sequence and search for the DNA sequence in the scenario library. If a match is found, output the deployment plan corresponding to the matched scenario; otherwise, proceed to fuzzy matching. S2: When entering fuzzy matching, the required DNA sequence is converted into a hash signature, and the hash signature is searched in the scenario library to obtain a list of similar DNA sequences that match the hash signature; S3: Calculate the Hamming distance between each similar DNA sequence and the required DNA sequence, and select the deployment scheme intermediate output of the similar DNA sequence corresponding to the minimum Hamming distance based on the distribution of the minimum Hamming distance and the second smallest Hamming distance, or generate the deployment scheme intermediate output through the optimization model; S4: Simulate and verify the intermediate output deployment scheme. If the verification is successful, output the deployment scheme as the final scheme. If the verification fails, regenerate the deployment scheme through the optimized model until the verification is successful.

[0021] In this implementation, step S1, which encodes the demand deployment scenario into a demand DNA sequence, specifically involves digitizing the terminal density, terrain complexity, cost sensitivity, business real-time performance, and business reliability of the demand deployment scenario, and then concatenating them in sequence to form a digital string, which is then used as the DNA sequence.

[0022] Specifically, digitization is based on expert experience: Terminal density: Low=0, Medium=1, High=2; Terrain complexity: Simple=0, Medium=1, Complex=2; Cost sensitivity: Low=0, Medium=1, High=2; Business real-time performance: Delay tolerance=0, Medium=1, Real-time sensitivity=2; Business reliability: Optional=0, Important=1, Critical=2.

[0023] Here are specific examples of encoding deployment scenarios into requirement DNA sequences: Scenario A (Remote Rural Area): [Terminal Density = 0 (Low), Terrain Complexity = 2 (High), Cost Sensitivity = 2 (High), Business Real-Time Performance = 1 (Medium), Business Reliability = 1 (Important)], the requirement DNA sequence for Scenario A is: 02211. Scenario B (Urban Industrial Area): [Terminal Density = 2 (High), Terrain Complexity = 0 (Simple), Cost Sensitivity = 0 (Low), Business Real-Time Performance = 2 (Real-Time), Business Reliability = 2 (Critical)], the DNA sequence for Scenario B is: 20222.

[0024] In this embodiment, step S2 converts the required DNA sequence into a hash signature using a locality-sensitive hash function.

[0025] Before performing DNA sequence lookup in the scene library in step S1 and hash signature lookup in the scene library in step S2, a two-level hash index needs to be built for the scene library.

[0026] Specifically: Create a primary index (exact matching database): Key: The complete DNA sequence of the scenario (e.g., 02211). Value: The optimal configuration for this scenario. Query speed: Instantaneous.

[0027] Create a secondary index (fuzzy matching library): Locality-Sensitive Hashing (LSH) hashes similar DNA sequences (i.e., similar scenarios) into the same bucket with high probability. Specifically, the complete DNA sequence (e.g., 02211) is input into the LSH function. The function outputs a shorter hash signature. Similar DNA sequences will have common parts in their signatures. This hash signature is used as the key for a secondary index. The value is all DNA sequences with the same or similar hash signatures and their corresponding schemes.

[0028] In this embodiment, step S3 specifically includes the following sub-steps: Step S31: Calculate the Hamming distance between each similar DNA sequence and the required DNA sequence, and sort them in ascending order to obtain a list of candidate solutions. The DNA sequence corresponding to the smallest Hamming distance is the best candidate. Step S32: Compare the best candidate with the candidate solution list beforehand If the Hamming distance difference between the candidate solutions is less than the set value and the difference is significant, then the deployment solution corresponding to the best candidate will be output as an intermediate solution; otherwise, the deployment solution will be generated by optimizing the model.

[0029] After obtaining the list of candidate solutions, the Hamming distance between each similar DNA sequence and the required DNA sequence is: ,in, To minimize the Hamming distance, The maximum distance in the candidate solution list is calculated before... The distance difference between each candidate (e.g., the first 3). If the quality of the first few candidates is significantly higher than the others, a "cliff" will appear. In this case, the deployment scheme corresponding to the best candidate is selected as the intermediate output; otherwise, the deployment scheme is generated by optimizing the model.

[0030] like Figure 2 As shown, when a deployment scenario arises, the intermediate output deployment solution process is as follows: First-level matching: exact matching (instantaneous).

[0031] Generate its DNA sequence for the required deployment scenario, such as 02211.

[0032] Use this DNA sequence directly as the key to query the main index (exact match database).

[0033] Result: Match: 02211 exists in the database. The corresponding deployment plan will be returned immediately. Matching ends.

[0034] Miss: Proceed to the next matchmaking stage.

[0035] Second-level matching: Fuzzy matching (fast).

[0036] The DNA sequence (02211) of the deployment scenario is used to generate a hash signature, such as A3F9, using the Locality Sensitive Hash (LSH) function. A3F9 is then used to query the secondary index (fuzzy matching library). This will return a list of all DNA sequences with hash signatures that are the same as or similar to A3F9.

[0037] For example, the returns are: 02211 (exact match not found, indicating a possible LSH collision); 02210 (very similar, only the last digit is different: reliability requirement changes from critical to critical); 01211 (very similar, the second digit is different: terrain complexity changes from high to medium).

[0038] Third-level matching: similarity ranking (intelligent).

[0039] Calculate the Hamming distance between the DNA (02211) of the deployment scenario and all candidate DNAs (02210, 01211) returned by the second-level matching. The Hamming distance between 02211 and 02210 is 1 (only one bit different), and the Hamming distance between 02211 and 01211 is also 1 (only one bit different). Sort the candidate solutions in ascending order of Hamming distance. The smaller the distance, the more similar the scenarios.

[0040] Before calculation The distance difference between each candidate (e.g., the first 3). If the quality of the first few candidates is significantly higher than the others, a cliff will appear. In this case, the deployment scheme corresponding to the best candidate is selected as the intermediate output; otherwise, the deployment scheme is generated by optimizing the model.

[0041] Scenario A (significantly better): If Very small (e.g., =1), and gap ( If the value is large (e.g., greater than 2), it indicates the best candidate. Significantly superior to others. At this point, even... Even if the absolute value isn't perfect, it's still considered usable. The threshold (used to select the best candidate as the deployment plan or reject all candidates) is actually... (A very small buffer) Only this optimal solution is accepted. Case B (mediocre candidate quality): If the distance between the sorted candidates is... ; And the gap is very small ( This indicates that there is no particularly outstanding solution, and all candidates are not significantly different. At this point, it is considered that there is no highly available solution, and the threshold is [value missing]. (More stringent than the best option) causes all candidates to be deemed unusable, thus triggering the generation of deployment options through model optimization.

[0042] In this embodiment, the deployment scheme is generated by optimizing the model, specifically including: Step E1: Obtain scenario information for the required deployment scenario, including terminal information, communication technology constraints, and gateway selection library.

[0043] Terminal information: Total number of terminals M, geographical coordinates of each terminal Data generation rate of each terminal Service type (e.g., control commands: low latency; metering data: high reliability).

[0044] Communication technology constraints: Maximum communication radius of the selected communication method (e.g., LoRa) Link bandwidth (B) and typical transmission delay (C) .

[0045] Gateway Options Library: A collection of gateway models to choose from, along with the purchase cost of each model. Processing capacity (e.g., maximum number of connections, number of packets processed per second), power consumption) .

[0046] Step E2: Construct an optimization model with the objective functions of minimizing deployment cost, maximum gateway load rate, and total latency, and with constraints of coverage, capacity, distance, and gateway uniqueness.

[0047] Optimize the objective function of the model for: ; in, , and Candidate deployment schemes The corresponding deployment cost, maximum gateway load rate, and total latency; , and The weights are respectively the deployment cost, the maximum gateway load rate, and the total latency, which are obtained by extracting scenario features and fuzzy inference.

[0048] Candidate deployment schemes The decision variable is: number of gateways. , indicates position Should the first be deployed? Binary variables of various gateway models , indicates in the terminal Is it connected to a location? binary variables of the gateway .

[0049] ; ; ; in, Candidate deployment schemes The Middle The procurement cost of various gateway models The operating energy cost per unit of data traffic. For position Total processing capacity of the gateway and Positions Gateway processing latency and queuing latency, This refers to the transmission delay of the communication link.

[0050] Constraints of the optimization model: Coverage constraints: Coverage constraints ensure that each terminal must and can only connect to one gateway.

[0051] Capacity constraints: Capacity constraints ensure that the total amount of data processed by the gateway cannot exceed its capacity.

[0052] Distance constraints: ;in, Indicates terminal Connect to location The Euclidean distance to the gateway, As a terrain discount factor, The terrain complexity is considered; the distance constraint ensures that if the terminal connects to the gateway, the distance must be within the effective communication range.

[0053] Gateway uniqueness: The gateway uniqueness constraint ensures that at most one type of gateway can be deployed at a given location.

[0054] Step E3: Based on the acquired scenario information, a genetic algorithm combined with greedy local search is used to solve the optimization model and obtain the optimal deployment scheme as an intermediate output.

[0055] Encoding (using Genetic Algorithm (GA): A deployment scheme is encoded as a chromosome. The chromosome contains two genes: Gateway Deployment Gene: a list indicating which type of gateway is deployed in which candidate locations; Terminal Association Gene: a list indicating the gateway index associated with each terminal.

[0056] Initial population generation (injecting prior knowledge): Instead of being completely random, a heuristic method is used. For example, the K-means clustering algorithm is first used to generate N clusters based on the terminal location. Individuals in the initial population prioritize associating terminals within the same cluster with the same gateway, thus accelerating convergence.

[0057] Fitness calculation: The objective function value constructed in step E2 is used directly as the fitness value. The smaller the value, the higher the fitness.

[0058] Genetic manipulation: selection, crossover, and mutation are performed to produce a new generation of population.

[0059] Innovative Local Search: After each generation of genetic operations, a greedy local search is performed on the best individuals in the population to quickly improve the quality of the solution. For example: Gateway Merging Strategy: Attempt to merge two lightly loaded gateways and replace them with a higher-capacity gateway, evaluating whether this reduces costs without violating constraints. Terminal Reassociation Strategy: For the gateway with the highest load, attempt to switch some of its terminals to a nearby, less loaded gateway to optimize the load balancing objective.

[0060] Termination and Output: When the maximum number of iterations is reached or the quality of the solution no longer improves significantly, the algorithm terminates and outputs the optimal gateway deployment scheme and terminal-gateway relationship.

[0061] In some other embodiments, if the genetic algorithm combined with greedy local search fails to solve the problem, an emergency rule base and expert intervention can be enabled.

[0062] Emergency Rule Base: Automatically degrades using a more lenient emergency rule base, containing conservative but feasible basic configuration schemes (e.g., ensuring full coverage and basic performance regardless of the scenario). Expert Intervention Interface: Triggers alarms and submits key parameters and solution status to the expert decision-making interface. Experts can: adjust model parameters (e.g., relax certain constraints), manually specify an initial solution for the algorithm to recalculate, or directly provide a configuration scheme based on experience.

[0063] In this embodiment, step E2 involves extracting scene features and using fuzzy inference to obtain the weights corresponding to deployment cost, maximum gateway load rate, and total latency. Specifically, this involves quantifying scene features, including terrain complexity and cost sensitivity, based on the scene information of the required deployment scenario. and business real-time requirements The three quantified scene features are used as inputs to the fuzzy system, and membership functions and fuzzy rule bases are set accordingly. Based on the membership functions and fuzzy rule bases, the input scene features are fuzzified, activated, aggregated, and defuzzified to obtain the weights corresponding to deployment cost, maximum gateway load rate, and total latency.

[0064] When quantifying scenario characteristics, cost sensitivity can be categorized into low, medium, and high levels based on deployment requirements and expert scoring, corresponding to values ​​of 0.2, 0.5, and 0.8 (or continuous values ​​from 0 to 1). When quantifying real-time requirements, latency sensitivity can be defined for each service type on each terminal (e.g., control commands: 0.9, video surveillance: 0.6, metering data: 0.2). The service distribution of all terminals can be statistically analyzed using a weighted average or by taking the maximum value. When quantifying terrain complexity, the elevation standard deviation or undulation of the area where all terminals are located can be used.

[0065] Construction of the fuzzy rule base: covering all possible input combinations, with the core rules as follows: Cost weight rule: If cost sensitivity Low level, terrain complexity The level is medium and the business real-time requirements are high. If it is high, then High value; If cost sensitivity The level is high and the terrain complexity is high. The level is low and the business real-time requirement is high. If it is in the middle, then It is a low value; Other rules must ensure that the trade-off logic among multiple objectives is consistent, such as when load is prioritized. When prioritizing improvement and delay promote.

[0066] Load weight rule: If the terrain complexity High level, cost sensitive The level is medium and the business real-time requirements are high. If it is low, then High value; If the terrain complexity Low level, cost sensitive The level is high and the business real-time requirements are high. If it is in the middle, then It is a low value.

[0067] Delay weight rule: If business real-time requirements For high, cost sensitive The level is medium and the terrain complexity is low. If the level is low, then High value; If business real-time requirements For low cost sensitivity The level is high and the terrain complexity is high. If the level is medium, then It is a low value.

[0068] The quantized input feature values ​​are then subjected to fuzzy inference, and the fuzzy output of each weight is obtained through the following steps: Fuzzification: Calculate the membership degree of each input value to each fuzzy set based on the membership function.

[0069] Rule activation: For each rule, calculate the satisfaction of its antecedent (condition part) (usually take the minimum value or product), and then apply it to the consequent (conclusion part) of the rule to obtain the truncation or scaling result of the fuzzy set of each output variable.

[0070] Aggregation: The results of all rules applied to the same output variable are superimposed (the union is taken) to obtain a comprehensive fuzzy set of that output variable.

[0071] The combined fuzzy set of each output variable is transformed into precise numerical weights through defuzzification. Common methods include the centroid method (calculating the centroid of the fuzzy set) or the maximum membership method. Finally, three weights are obtained. , and Once the normalization condition is met (e.g., the sum of the three is 1), they are directly used as the coefficients of the objective function.

[0072] In some other embodiments, a rule knowledge base is also included to provide an initial solution when solving the optimization model.

[0073] The rule knowledge base is constructed by analyzing the mapping relationship between the deployment scheme and its corresponding scenario features, which is verified by simulation and generated based on the optimization model, generating rules, and storing them in the rule knowledge base.

[0074] For example, if the scenario characteristics are low terminal density, high terrain complexity, high cost sensitivity, and moderate real-time business requirements, then the optimization rule is: recommended gateway model = basic capacity type, gateway to terminal ratio of 1:30~1:50, and optimization target weight. , =0.2, The association strategy is based on proximity and manageable load.

[0075] In step S4, the intermediate output deployment scheme is rigorously simulated and tested using a digital twin or simulation simulator to evaluate its expected performance (cost, latency, load balancing, etc.) in actual deployment. Ultimately, the user obtains a verified, optimal or suboptimal scheme for the number of gateways, capacity configuration, and network topology, which can be directly used to guide hardware procurement, software configuration, and on-site deployment.

[0076] Example 2: The present invention provides a computer device, including a processor and a memory; wherein, when the processor executes a computer program stored in the memory, it implements the steps of the above-described micro-energy gateway deployment method based on multi-scenario matching.

[0077] For more detailed information on the above methods, please refer to the relevant content disclosed in the foregoing embodiments, which will not be repeated here.

[0078] Example 3: The present invention provides a computer-readable storage medium for storing a computer program; when the computer program is executed by a processor, it implements the steps of the above-described micro-energy gateway deployment method based on multi-scenario matching.

[0079] For a more detailed explanation of the above method, please refer to the relevant content disclosed in the foregoing embodiments, which will not be repeated here.

[0080] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on its differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. Regarding the devices and storage media disclosed in the embodiments, since they correspond to the methods disclosed in the embodiments, the descriptions are relatively simple; relevant details can be found in the method section. Those skilled in the art will clearly understand that the technologies in the embodiments of this invention can be implemented using software plus necessary general-purpose hardware platforms. Based on this understanding, the technical solutions in the embodiments of this invention, in essence or the parts that contribute to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or certain parts of the embodiments of this invention.

[0081] The above are merely preferred embodiments of the present invention. The scope of protection of the present invention is not limited to the above embodiments. All technical solutions falling within the scope of the present invention's concept are within the scope of protection of the present invention. It should be noted that for those skilled in the art, any improvements and modifications made without departing from the principles of the present invention should be considered within the scope of protection of the present invention.

Claims

1. A method for deploying a micro energy gateway based on multi-scenario matching, characterized in that, Includes the following steps: The deployment scenarios are encoded as demand DNA sequences, and the DNA sequences are searched in the scenario library. If a match is found, the deployment scheme corresponding to the matched scenario is output; otherwise, fuzzy matching is performed. Once fuzzy matching is initiated, the required DNA sequence is converted into a hash signature, and the hash signature is searched in the scenario library to obtain a list of similar DNA sequences that match the hash signature. Calculate the Hamming distance between each similar DNA sequence and the required DNA sequence, and select the deployment scheme intermediate output corresponding to the minimum Hamming distance based on the distribution of the minimum Hamming distance and the second smallest Hamming distance, or generate the deployment scheme intermediate output through the optimization model; The intermediate output deployment scheme is simulated and verified. If the verification is successful, the deployment scheme is output as the final scheme. If the verification fails, the deployment scheme is regenerated by optimizing the model until the verification is successful.

2. The micro energy gateway deployment method based on multi-scenario matching according to claim 1, characterized in that, The process of encoding the demand deployment scenario into a demand DNA sequence involves digitizing the terminal density, terrain complexity, cost sensitivity, business real-time performance, and business reliability of the demand deployment scenario, and then concatenating them in sequence to form a digital string, which is then used as the DNA sequence.

3. The micro-energy gateway deployment method based on multi-scenario matching according to claim 1, characterized in that, The required DNA sequence is transformed into a hash signature using a locality-sensitive hash function.

4. The micro-energy gateway deployment method based on multi-scenario matching according to claim 1, characterized in that, The process involves calculating the Hamming distance between each similar DNA sequence and the desired DNA sequence, and selecting the deployment scheme intermediate output corresponding to the minimum Hamming distance based on the distribution of the minimum and second-minimum Hamming distances, or generating the deployment scheme intermediate output through an optimization model. Specifically: Calculate the Hamming distance between each similar DNA sequence and the required DNA sequence, and sort them in ascending order to obtain a list of candidate solutions. The DNA sequence corresponding to the smallest Hamming distance is the best candidate. Before comparing the best candidate with the list of candidate solutions If the Hamming distance difference between the candidate solutions is less than the set value and the difference is significant, then the deployment solution corresponding to the best candidate will be output as an intermediate solution. Otherwise, a deployment plan is generated by optimizing the model.

5. The micro-energy gateway deployment method based on multi-scenario matching according to claim 1, characterized in that, The process of generating a deployment plan through model optimization specifically includes: Obtain scenario information for the required deployment scenarios, including terminal information, communication technology constraints, and gateway selection library; Construct an optimization model with the objective functions of minimizing deployment cost, maximum gateway load rate, and total latency, and with constraints of coverage, capacity, distance, and gateway uniqueness. Based on the scenario information of the acquired demand scenarios, a genetic algorithm combined with greedy local search is used to solve the optimization model, and the optimal deployment scheme is output as an intermediate solution.

6. The micro-energy gateway deployment method based on multi-scenario matching according to claim 5, characterized in that, The objective function of the optimization model for: ； in, , and Candidate deployment schemes The corresponding deployment cost, maximum gateway load rate, and total latency; , and The weights are respectively the deployment cost, the maximum gateway load rate, and the total latency, which are obtained by extracting scenario features and fuzzy inference.

7. The micro-energy gateway deployment method based on multi-scenario matching according to claim 6, characterized in that, The deployment cost, maximum gateway load rate, and total latency are obtained by extracting scene features and using fuzzy inference, specifically as follows: Based on the scenario information of the deployment scenario, quantify the scenario characteristics, including terrain complexity, cost sensitivity, and business real-time requirements; Three quantified scene features are used as inputs to the fuzzy system, and membership functions and fuzzy rule bases are set accordingly. Based on membership functions and fuzzy rule bases, the input scene features are fuzzified, activated, aggregated, and defuzzified to obtain the weights corresponding to deployment cost, maximum gateway load rate, and total latency.

8. The micro-energy gateway deployment method based on multi-scenario matching according to claim 1, characterized in that, It also includes building a rule knowledge base to provide initial solutions when solving optimization models; The construction of the rule knowledge base specifically involves: analyzing the mapping relationship between the deployment scheme and its corresponding scene features for the deployment scheme that has been verified through simulation and generated based on the optimization model, generating rules, and storing them in the rule knowledge base.

9. A computer device, characterized in that, It includes a processor and a memory; wherein, when the processor executes the computer program stored in the memory, it implements the steps of the micro energy gateway deployment method based on multi-scenario matching as described in any one of claims 1-8.

10. A computer-readable storage medium, characterized in that, Used to store computer programs; when the computer programs are executed by a processor, they implement the steps of the micro energy gateway deployment method based on multi-scenario matching as described in any one of claims 1-8.

Images