Method and system for monitoring service life of filter screen of air purification device

Abstract

The invention relates to the technical field of air purification facilities, in particular to a filter screen service life monitoring method and system for an air purification device, and the method comprises the steps: firstly obtaining the real-time operation parameters of a target air purification device, then building an initial vector set and building a fitness function, and then optimizing the initial vector set through an optimization algorithm, and finally, monitoring the service life of the target filter screen according to the optimal filter screen structure vector. Compared with the prior art, the method has the advantages that the hierarchical structure of the filter screen is described through the filter screen structure vector, and the accuracy between the filter screen structure represented by the filter screen structure vector and the real filter screen structure is represented by the fitness, namely the difference between the theoretical operation parameter represented by the filter screen structure vector and the real operation parameter; on the basis, backstepping of the filter screen structure is achieved through an optimization algorithm, and the problem that in the prior art, a filter screen service life monitoring program cannot be compatible with various filter screens of different structures is solved.

Description

Technical Field

[0001] The present invention relates to the technical field of air purification facilities, and in particular to a filter life monitoring method and system for an air purification device. Background Art

[0002] The filter is the core component of the air purification device, responsible for capturing and removing pollutants in the air, such as dust, pollen, bacteria, viruses and other harmful particles. Over time, the pollutants accumulated on the filter will gradually increase, resulting in a decrease in filtration efficiency, and may even become a breeding ground for bacteria and mold. Therefore, it is very important to monitor the service life of the filter.

[0003] Currently, air purification devices generally come with a filter life monitoring program, but most of these programs are designed based on the filter structure of a specific brand and model. This means that when users choose to replace non-original third-party filters in order to save costs or for other reasons, they may encounter compatibility issues. Since the structure and materials of third-party filters may differ from those of original filters, these differences may cause changes in the filter's air permeability, filtration efficiency, and ability to adsorb pollutants. Therefore, the original life monitoring method based on a fixed structure design may not accurately reflect the actual usage of the new filter, making the monitoring results inaccurate and unable to provide users with reliable replacement prompts.

[0004] Therefore, people need a filter life monitoring method that is compatible with a variety of filter structures with different life spans. Summary of the invention

[0005] Therefore, the present invention provides a filter life monitoring method and system for an air purification device, so as to solve the problem that the filter life monitoring program in the prior art is not compatible with filters of various structures.

[0006] The present invention provides a method for monitoring the life of a filter screen of an air purification device, comprising:

[0007] Acquire real-time operating parameters of the target air purification device, wherein the real-time operating parameters include a front-to-rear resistance difference and a front-to-rear filtration efficiency of the target air purification device;

[0008] Establishing an initial vector set, wherein the initial vector set includes a plurality of filter structure vectors, and the filter structure vectors are used to describe the hierarchical structure of the filter;

[0009] According to the filter structure vector, a fitness function is established, and the fitness function is used to calculate the corresponding theoretical operating parameters based on a filter structure vector, and calculate the difference between the theoretical operating parameters and the real-time operating parameters as the fitness of the filter structure vector, wherein the theoretical operating parameters include the front and rear resistance difference and the front and rear filtration efficiency;

[0010] Based on the initial vector set and the fitness function, the initial vector set is optimized by an optimization algorithm to obtain the optimal filter structure vector;

[0011] According to the optimal filter structure vector, the life of the target filter is monitored.

[0012] In a preferred embodiment: each element in the filter structure vector corresponds to a filter layer in the filter, and each element value represents the thickness of the corresponding filter layer.

[0013] In a preferred embodiment: based on the initial vector set and the fitness function, the initial vector set is optimized by an optimization algorithm to obtain the optimal filter structure vector, including:

[0014] A solution space is established based on the value range of the filter structure vector, and each filter structure vector in the initial filter set is used as a particle in the solution space;

[0015] Calculate the fitness of each particle based on the fitness function, and obtain the global optimal particle and the individual optimal particle;

[0016] Update the position of each particle according to the global optimal particle and the individual optimal particle;

[0017] According to the particles with updated positions, a new vector set is obtained;

[0018] Among them, the vector set corresponding to the particles finally obtained after multiple position updates is the optimal vector set, and one vector in the optimal vector set is the optimal filter structure vector.

[0019] In a preferred embodiment: each element in the filter structure vector corresponds to a unit thickness in the filter, and each element value represents the type of filter layer in the corresponding unit thickness.

[0020] In a preferred embodiment: based on the initial vector set and the fitness function, the initial vector set is optimized by an optimization algorithm to obtain the optimal filter structure vector, including:

[0021] Treat each filter structure vector in the initial filter set as an individual;

[0022] Calculate the fitness of each individual based on the fitness function, and select the parent individual according to the fitness;

[0023] Perform crossover mutation on the parent individuals to obtain the offspring individuals;

[0024] According to the parent individuals and the offspring individuals, a new vector set is obtained;

[0025] Among them, the vector set finally obtained after multiple update iterations is the optimal vector set, and one vector in the optimal vector set is the optimal filter structure vector.

[0026] In a preferred embodiment, crossover mutation is performed on the parent individuals to obtain the offspring individuals, comprising:

[0027] Obtain two first target individuals, where the first target individuals are two parent individuals to be crossed;

[0028] Perform XOR on the two first target individuals, and count the proportion of the same elements represented by the XOR result;

[0029] If the proportion of identical elements exceeds a preset threshold, the two first target individuals are crossed to obtain two offspring individuals.

[0030] In a preferred embodiment, crossover mutation is performed on the parent individuals to obtain the offspring individuals, comprising:

[0031] Obtain a second target individual, where the second target individual is the parent individual to be mutated;

[0032] Setting a mutation window of a preset length in the second target individual, and counting the types of element values ​​in the mutation window;

[0033] If there are only two types of element values ​​in the mutation window, all element values ​​in the mutation window will be mutated to one of them to obtain a descendant individual.

[0034] In a preferred embodiment: establishing an initial vector set includes:

[0035] Obtaining the historical optimal filter structure vector and the suboptimal filter structure vector;

[0036] Based on the historical optimal filter structure vector, the operating state of the target air purification device is predicted to obtain predicted operating parameters, wherein the predicted operating parameters include the front-rear resistance difference and the front-rear filtration efficiency;

[0037] According to the difference between the predicted operating parameters and the real-time operating parameters, multiple vectors are derived based on the historical optimal filter structure vector and the suboptimal filter structure vector, and an initial vector set is established, wherein the quantity ratio of the vector derived from the historical optimal filter structure vector and the multiple vectors derived from the suboptimal filter structure vector is negatively correlated with the difference between the predicted operating parameters and the real-time operating parameters.

[0038] In a preferred embodiment: according to the difference between the predicted operating parameters and the real-time operating parameters, multiple vectors are derived based on the historical optimal filter structure vector and the suboptimal filter structure vector, and an initial vector set is established, including:

[0039] According to the difference between the predicted operating parameters and the real-time operating parameters, the first neighborhood of the historical optimal filter structure vector and the second neighborhood of the suboptimal filter structure vector are obtained, wherein the range ratio of the first neighborhood and the second neighborhood is negatively correlated with the difference between the predicted operating parameters and the real-time operating parameters;

[0040] Selecting multiple vectors in a first neighborhood of a historical optimal filter structure vector to obtain a first vector;

[0041] Selecting multiple vectors in a second neighborhood of the historical suboptimal filter structure vector to obtain a second vector;

[0042] Randomly generate multiple filter structure vectors to obtain a third vector;

[0043] The first vector, the second vector and the third vector are aggregated to obtain an initial vector set.

[0044] The present invention also provides a filter life monitoring system for an air purification device, comprising:

[0045] A data monitoring module, used to obtain real-time operating parameters of the target air purification device, wherein the real-time operating parameters include the front-to-back resistance difference and the front-to-back filtration efficiency of the target air purification device;

[0046] A vector construction module is used to establish an initial vector set, wherein the initial vector set includes a plurality of filter structure vectors, and the filter structure vectors are used to describe the hierarchical structure of the filter;

[0047] A strategy building module is used to establish a fitness function according to the filter structure vector. The fitness function is used to calculate the corresponding theoretical operating parameters based on a filter structure vector, and calculate the difference between the theoretical operating parameters and the real-time operating parameters as the fitness of the filter structure vector, wherein the theoretical operating parameters include the front and rear resistance difference and the front and rear filtration efficiency;

[0048] The structural analysis module is used to optimize the initial vector set through an optimization algorithm based on the initial vector set and the fitness function to obtain the optimal filter structure vector;

[0049] The life monitoring module is used to monitor the life of the target filter according to the optimal filter structure vector.

[0050] The beneficial effects of adopting the above embodiment are:

[0051] The present invention provides a filter life monitoring method and system for an air purification device, which first obtains the real-time operating parameters of the target air purification device, then establishes an initial vector set, and establishes a fitness function according to the filter structure vector, then optimizes the initial vector set based on the initial vector set and the fitness function through an optimization algorithm to obtain the optimal filter structure vector, and finally monitors the life of the target filter according to the optimal filter structure vector. Compared with the prior art, the present invention describes the hierarchical structure of the filter through the filter structure vector, and uses the fitness, that is, the difference between the theoretical operating parameters and the real-time operating parameters represented by the filter structure vector, to characterize the accuracy between the filter structure represented by the filter structure vector and the actual filter structure. On this basis, the optimal filter structure vector is found through the optimization algorithm to achieve the reverse deduction of the filter structure, so as to predict the filter life according to the inferred filter structure, so that the target air purification device can ignore the specifications and models of the filter when detecting the filter life and still maintain accuracy, solving the problem that the filter life monitoring program in the prior art is not compatible with filters of various structures. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] Figure 1 A method flow chart of an embodiment of a method for monitoring the life of a filter screen of an air purification device provided by the present invention;

[0053] Figure 2 for Figure 1 A specific step diagram of an embodiment of step S104;

[0054] Figure 3 for Figure 1 A specific step diagram of an embodiment of step S102;

[0055] Figure 4 This is a system structure diagram of an embodiment of a filter life monitoring system for an air purification device provided by the present invention. DETAILED DESCRIPTION

[0056] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0057] Combination Figure 1 As shown, a specific embodiment of the present invention discloses a filter life monitoring method for an air purification device, comprising:

[0058] S101, obtaining real-time operating parameters of a target air purification device, wherein the real-time operating parameters include a front-to-rear resistance difference and a front-to-rear filtration efficiency of the target air purification device;

[0059] S102, establishing an initial vector set, wherein the initial vector set includes a plurality of filter structure vectors, and the filter structure vectors are used to describe the hierarchical structure of the filter;

[0060] S103, establishing a fitness function according to the filter structure vector, the fitness function is used to calculate the corresponding theoretical operating parameters based on a filter structure vector, and calculate the difference between the theoretical operating parameters and the real-time operating parameters as the fitness of the filter structure vector, wherein the theoretical operating parameters include the front and rear resistance difference and the front and rear filtration efficiency;

[0061] S104, based on the initial vector set and the fitness function, optimizing the initial vector set by an optimization algorithm to obtain an optimal filter structure vector;

[0062] S105. Monitor the life of the target filter according to the optimal filter structure vector.

[0063] The above process is mainly run when the filter has just been replaced and the filter structure is unknown. The operating parameters mainly include the front and rear resistance difference and the front and rear filtration efficiency (it can be understood that the real-time operating parameters, theoretical operating parameters and the predicted operating parameters in the following text are all operating parameters). The front and rear resistance difference is mainly used to evaluate the life of the physical filtration layer that filters dust particles in the filter, and the front and rear filtration efficiency is mainly used to evaluate the efficiency of chemical filtration in the filter.

[0064] In practice, the structure of the filter screen of an air purification device generally includes various types, such as cylindrical structure, composite structure, tower structure, etc. However, no matter what structure, the filter screen can be abstracted as a combination of multiple filter layers. The filter layer is a hierarchical structure that performs different filtering functions in the filter screen, generally including a primary filter layer, a HEPA layer, an activated carbon filter layer, etc. The "structure of the filter screen" described in this embodiment mainly refers to the arrangement and combination of different filter layers of the filter screen and the thickness of different filter layers.

[0065] It is understandable that when the filter structure is known, the operating parameters of the machine and the life of the filter can be inferred through mathematical models, but the key problem in practice is that users generally cannot know the specific structure of the filter they replaced. Therefore, the goal of the present invention is to reversely infer the structure of the filter based on the operating parameters, so as to achieve compatibility with different filter structures during life monitoring.

[0066] Compared with the prior art, the present invention describes the hierarchical structure of the filter through a filter structure vector, and uses fitness, that is, the difference between the theoretical operating parameters and the real-time operating parameters represented by the filter structure vector, to characterize the accuracy between the filter structure represented by the filter structure vector and the actual filter structure. On this basis, the optimal filter structure vector is found through an optimization algorithm to reversely infer the filter structure, thereby predicting the filter life based on the inferred filter structure, so that the target air purification device can ignore the specifications and models of the filter during filter life detection and still maintain accuracy, thereby solving the problem in the prior art that the filter life monitoring program is not compatible with filters of various different structures.

[0067] In this embodiment, the filter structure is inferred by operating parameters. It is obviously a multi-objective optimization problem, so it can be solved by an optimization algorithm. Fitness is a core concept in the optimization algorithm. It is a criterion for evaluating the quality or pros and cons of a solution. The definition of fitness may be different in different optimization problems. It is understandable that concepts such as optimization algorithms, fitness, and some vocabulary in specific optimization algorithms mentioned later, such as particles, individuals, solution spaces, etc., are all prior arts that can be understood by those skilled in the art, so this article will not explain too much.

[0068] In a preferred embodiment, each element in the filter structure vector corresponds to a filter layer in the filter, and each element value represents the thickness of the corresponding filter layer. For example, the vector (10, 20, 15) can represent that the thickness of the primary filter layer, the HEPA layer, and the activated carbon filter layer in a filter are 10 mm, 20 mm, and 15 mm, respectively.

[0069] Correspondingly, the filter structure vector of the above encoding method can be optimized by a particle swarm algorithm. In a preferred embodiment, the above step S104, based on the initial vector set and the fitness function, optimizes the initial vector set by an optimization algorithm to obtain the optimal filter structure vector, specifically includes:

[0070] A solution space is established based on the value range of the filter structure vector, and each filter structure vector in the initial filter set is used as a particle in the solution space;

[0071] Calculate the fitness of each particle based on the fitness function, and obtain the global optimal particle and the individual optimal particle;

[0072] Update the position of each particle according to the global optimal particle and the individual optimal particle;

[0073] According to the particles with updated positions, a new vector set is obtained;

[0074] Among them, the vector set corresponding to the particles finally obtained after multiple position updates is the optimal vector set, and one vector in the optimal vector set is the optimal filter structure vector.

[0075] The above process can be simply understood as adjusting the size of different element values ​​in the filter structure vector until the optimum is reached. However, it is understandable that the above process assumes that the arrangement and combination order of multiple filter layers in the filter is certain, and the only difference between different filter layers is the thickness of different filter layers. This obviously has certain limitations, and the compatible filter types are not enough. Therefore, the present invention also provides a more preferred method, which enables the present invention to be compatible with different arrangements and combinations of filter layers.

[0076] Specifically, in a preferred embodiment, each element in the filter structure vector corresponds to a unit thickness in the filter, and each element value represents the type of filter layer in the corresponding unit thickness. For example, with 2 mm as a thickness unit, the vector (1, 1, 1, 2, 2, 2, 2, 3, 3, 3, 3, 3) can represent that in a filter, the thickness of the primary filter layer, the HEPA layer and the activated carbon filter layer are 6 mm, 8 mm and 10 mm respectively. The advantage of this embodiment is that it can represent a filter with different filter layer arrangements. For example, the vector (1, 1, 1, 1, 3, 3, 3, 3, 3) can represent that in a filter, there is only an 8 mm thick primary filter layer and a 10 mm thick activated carbon filter layer, but no HEPA layer. For another example, the vector (1, 1, 1, 3, 3, 3, 4, 4, 4, 2, 2, 2) can represent that in a filter, along the flow direction of the wind, there are a 6 mm primary filter layer, an activated carbon filter layer, an electronic dust collection filter layer and a HEPA layer in sequence.

[0077] Correspondingly, the above encoding method can be optimized by genetic algorithm. Specifically, Figure 2 As shown, in a preferred embodiment, the above step S104, based on the initial vector set and the fitness function, optimizes the initial vector set by an optimization algorithm to obtain the optimal filter structure vector, specifically includes:

[0078] S201, taking each filter structure vector in the initial filter set as an individual;

[0079] S202, calculating the fitness of each individual based on the fitness function, and selecting the parent individual according to the fitness;

[0080] S203, performing crossover mutation on the parent individuals to obtain offspring individuals;

[0081] S204, obtaining a new vector set according to the parent individuals and the offspring individuals;

[0082] Among them, the vector set finally obtained after multiple update iterations is the optimal vector set, and one vector in the optimal vector set is the optimal filter structure vector.

[0083] It is understandable that there are some constraints in the above encoding method. That is, since different filter layers in the filter must be arranged continuously in practice, the same values ​​in the same filter structure vector must also be guaranteed to be adjacent. In the process of existing genetic algorithms, the crossover mutation step is prone to non-compliant individuals. For example, the vector (1, 1, 2, 1) indicates that after the 4mm primary filter layer, there is a 2mm HEPA layer, and after the HEPA is still a 2mm primary filter layer, which is obviously unreasonable. Therefore, in order to improve the optimization efficiency of the genetic algorithm, the present invention further improves the crossover mutation method in the existing genetic algorithm.

[0084] Specifically, in a preferred embodiment, the above step S203, performing crossover mutation on the parent individuals to obtain the offspring individuals, specifically includes:

[0085] Obtain two first target individuals, where the first target individuals are two parent individuals to be crossed;

[0086] Perform XOR on the two first target individuals, and count the proportion of the same elements represented by the XOR result;

[0087] If the proportion of identical elements exceeds a preset threshold, the two first target individuals are crossed to obtain two offspring individuals.

[0088] The above process mainly targets crossover behavior, and uses XOR to determine whether two individuals are similar. The degree of similarity is expressed by the proportion of the same elements represented by the XOR result. When the two individuals are very different, the proportion of the same elements represented by the XOR result is low, and crossover is not suitable at this time.

[0089] For example, for the vectors (1, 2, 2, 2, 3, 3, 3, 3) and (1, 2, 2, 2, 2, 2, 2, 3), it can be seen that the two vectors are quite different in the second half. At this time, after the crossover, it is very likely that an irregular individual such as (1, 2, 2, 2, 3, 3, 2, 3) will appear (crossover at the seventh element position).

[0090] Furthermore, in a preferred embodiment, the above step S203, performing crossover mutation on the parent individuals to obtain the offspring individuals, further includes:

[0091] Obtain a second target individual, where the second target individual is the parent individual to be mutated;

[0092] Setting a mutation window of a preset length in the second target individual, and counting the types of element values ​​in the mutation window;

[0093] If there are only two types of element values ​​in the mutation window, all element values ​​in the mutation window will be mutated to one of them to obtain a descendant individual.

[0094] The above process mainly targets mutation behavior, and limits mutation by setting a mutation window to avoid non-compliant situations. For example, for the vector (1, 2, 2, 2, 3, 3, 3, 3), traditional mutation may produce a non-compliant vector (1, 3, 2, 2, 3, 2, 3, 3). If a mutation window of length 4 is set at the position of the 3rd, 4th, 5th, and 6th elements in the vector, then based on this embodiment, the mutated structures only have two types: (1, 2, 2, 2, 2, 2, 3, 3) or (1, 2, 3, 3, 3, 3, 3), both of which are compliant individuals.

[0095] Furthermore, it is conceivable that, in practice, in the optimal vector set obtained by one optimization result, there may be multiple filter structure vectors with the same fitness but different structures, and one optimization may not be able to achieve accurate prediction. Therefore, the present invention can perform optimization multiple times to iterate, so as to continuously correct the judgment of the filter structure as the target air purification device operates, and finally achieve accurate prediction.

[0096] The present invention also provides a preferred embodiment, which further shortens the number of optimization iterations by constructing an initial vector set in a targeted manner during each optimization. Figure 3 As shown, in a preferred embodiment, the above step S102, establishing an initial vector set, specifically includes:

[0097] S301, obtaining a historical optimal filter structure vector and a suboptimal filter structure vector;

[0098] S302, predicting the operating state of the target air purification device based on the historical optimal filter structure vector to obtain predicted operating parameters, wherein the predicted operating parameters include front-to-back resistance difference and front-to-back filtration efficiency;

[0099] S303. According to the difference between the predicted operating parameters and the real-time operating parameters, multiple vectors are derived based on the historical optimal filter structure vector and the suboptimal filter structure vector, and an initial vector set is established, wherein the ratio of the number of vectors derived based on the historical optimal filter structure vector and the number of vectors derived based on the suboptimal filter structure vector is negatively correlated with the difference between the predicted operating parameters and the real-time operating parameters.

[0100] In the above process, the historical optimal filter structure vector is the optimal filter structure vector obtained in the last optimization iteration (which can be obtained by randomly selecting from the optimal vector set), and the suboptimal filter structure vector is the other vectors in the optimal vector set obtained in the last optimization iteration except the optimal filter structure vector. In theory, each vector in the optimal vector set may represent the real filter structure, but it cannot be distinguished in one optimization process. Therefore, a vector can be randomly selected for monitoring, and the accuracy of the vector can be measured by the changes in subsequent operating parameters.

[0101] This embodiment utilizes the historical optimal filter structure vector and the suboptimal filter structure vector to derive multiple vectors as the initial vector set for the next optimization, which can achieve further optimization and speed up the process of finding the optimal solution.

[0102] It is understandable that if the previous optimal filter structure vector is selected correctly, then under the condition that the environment remains unchanged, the predicted operating parameters based on the filter structure and the real-time operating parameters of the machine during subsequent actual operation should be the same. If the predicted operating parameters and the real-time operating parameters are very different, it means that the optimal filter structure vector is selected incorrectly, and the true filter structure may be other vectors in the optimal vector set. At this time, the proportion of vectors derived from the historical optimal filter structure vector in the initial vector set can be reduced, and the proportion of vectors derived from the historical suboptimal filter structure vector can be increased, further improving the probability of the subsequent optimization algorithm finding the correct vector, and significantly improving the convergence speed of monitoring accuracy.

[0103] Specifically, in a preferred embodiment, the above step S303, according to the difference between the predicted operating parameters and the real-time operating parameters, derives multiple vectors based on the historical optimal filter structure vector and the suboptimal filter structure vector, and establishes an initial vector set, specifically including:

[0104] According to the difference between the predicted operating parameters and the real-time operating parameters, the first neighborhood of the historical optimal filter structure vector and the second neighborhood of the suboptimal filter structure vector are obtained, wherein the range ratio of the first neighborhood and the second neighborhood is negatively correlated with the difference between the predicted operating parameters and the real-time operating parameters;

[0105] Selecting multiple vectors in a first neighborhood of a historical optimal filter structure vector to obtain a first vector;

[0106] Selecting multiple vectors in a second neighborhood of the historical suboptimal filter structure vector to obtain a second vector;

[0107] Randomly generate multiple filter structure vectors to obtain a third vector;

[0108] The first vector, the second vector and the third vector are aggregated to obtain an initial vector set.

[0109] The above process derives by selecting vectors in the neighborhood and introduces a randomly generated third vector to maintain the comprehensiveness of the optimal solution search.

[0110] Combination Figure 4 As shown, the present invention also provides a filter life monitoring system for an air purification device, comprising:

[0111] The data monitoring module 410 is used to obtain the real-time operating parameters of the target air purification device, wherein the real-time operating parameters include the front-to-back resistance difference and the front-to-back filtration efficiency of the target air purification device;

[0112] A vector construction module 420 is used to establish an initial vector set, wherein the initial vector set includes a plurality of filter mesh structure vectors, and the filter mesh structure vectors are used to describe the hierarchical structure of the filter mesh;

[0113] A strategy building module 430 is used to establish a fitness function according to the filter structure vector, wherein the fitness function is used to calculate the corresponding theoretical operating parameters based on a filter structure vector, and calculate the difference between the theoretical operating parameters and the real-time operating parameters as the fitness of the filter structure vector, wherein the theoretical operating parameters include the front-rear resistance difference and the front-rear filtration efficiency;

[0114] A structure analysis module 440 is used to optimize the initial vector set through an optimization algorithm based on the initial vector set and the fitness function to obtain an optimal filter structure vector;

[0115] The life monitoring module 450 is used to monitor the life of the target filter according to the optimal filter structure vector.

[0116] It should be noted here that the corresponding system provided in the above embodiments can implement the technical solutions described in the above method embodiments. The specific implementation principles of the above modules or units can be found in the corresponding contents in the above method embodiments, which will not be repeated here.

[0117] The present invention provides a filter life monitoring method and system for an air purification device, which first obtains the real-time operating parameters of the target air purification device, then establishes an initial vector set, and establishes a fitness function according to the filter structure vector, then optimizes the initial vector set based on the initial vector set and the fitness function through an optimization algorithm to obtain the optimal filter structure vector, and finally monitors the life of the target filter according to the optimal filter structure vector. Compared with the prior art, the present invention describes the hierarchical structure of the filter through the filter structure vector, and uses the fitness, that is, the difference between the theoretical operating parameters and the real-time operating parameters represented by the filter structure vector, to characterize the accuracy between the filter structure represented by the filter structure vector and the actual filter structure. On this basis, the optimal filter structure vector is found through the optimization algorithm to achieve the reverse deduction of the filter structure, so as to predict the filter life according to the inferred filter structure, so that the target air purification device can ignore the specifications and models of the filter when detecting the filter life and still maintain accuracy, solving the problem that the filter life monitoring program in the prior art is not compatible with filters of various structures.

[0118] It should be noted that the various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the various embodiments can be referenced to each other.

[0119] The above description of the disclosed embodiments enables one skilled in the art to implement or use the present invention. Various modifications to these embodiments will be apparent to one skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but rather to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for monitoring the filter life of an air purification device, characterized in that: include: Acquire real-time operating parameters of the target air purification device, wherein the real-time operating parameters include a front-to-rear resistance difference and a front-to-rear filtration efficiency of the target air purification device; Establishing an initial vector set, wherein the initial vector set includes a plurality of filter structure vectors, and the filter structure vectors are used to describe the hierarchical structure of the filter; According to the filter structure vector, a fitness function is established, and the fitness function is used to calculate the corresponding theoretical operating parameters based on a filter structure vector, and calculate the difference between the theoretical operating parameters and the real-time operating parameters as the fitness of the filter structure vector, wherein the theoretical operating parameters include the front and rear resistance difference and the front and rear filtration efficiency; Based on the initial vector set and the fitness function, the initial vector set is optimized by an optimization algorithm to obtain the optimal filter structure vector; According to the optimal filter structure vector, the life of the target filter is monitored.

2. The filter life monitoring method of an air purification device according to claim 1, characterized in that: Each element in the filter structure vector corresponds to a filter layer in the filter, and each element value represents the thickness of the corresponding filter layer.

3. The filter life monitoring method of an air purification device according to claim 2, characterized in that: Based on the initial vector set and the fitness function, the initial vector set is optimized by the optimization algorithm to obtain the optimal filter structure vector, including: A solution space is established based on the value range of the filter structure vector, and each filter structure vector in the initial filter set is used as a particle in the solution space; Calculate the fitness of each particle based on the fitness function, and obtain the global optimal particle and the individual optimal particle; Update the position of each particle according to the global optimal particle and the individual optimal particle; According to the particles with updated positions, a new vector set is obtained; Among them, the vector set corresponding to the particles finally obtained after multiple position updates is the optimal vector set, and one vector in the optimal vector set is the optimal filter structure vector.

4. The method for monitoring the filter life of an air purification device according to claim 1, characterized in that: Each element in the filter structure vector corresponds to a unit thickness in the filter, and each element value represents the type of filter layer in the corresponding unit thickness.

5. The filter life monitoring method of an air purification device according to claim 4, characterized in that: Based on the initial vector set and the fitness function, the initial vector set is optimized by the optimization algorithm to obtain the optimal filter structure vector, including: Treat each filter structure vector in the initial filter set as an individual; Calculate the fitness of each individual based on the fitness function, and select the parent individual according to the fitness; Perform crossover mutation on the parent individuals to obtain the offspring individuals; According to the parent individuals and the offspring individuals, a new vector set is obtained; Among them, the vector set finally obtained after multiple update iterations is the optimal vector set, and one vector in the optimal vector set is the optimal filter structure vector.

6. The filter life monitoring method of an air purification device according to claim 5, characterized in that: Perform crossover mutation on the parent individuals to obtain the offspring individuals, including: Obtain two first target individuals, where the first target individuals are two parent individuals to be crossed; Perform XOR on the two first target individuals, and count the proportion of the same elements represented by the XOR result; If the proportion of identical elements exceeds a preset threshold, the two first target individuals are crossed to obtain two offspring individuals.

7. The method for monitoring filter life of an air purification device according to claim 5, characterized in that: Perform crossover mutation on the parent individuals to obtain the offspring individuals, including: Obtain a second target individual, where the second target individual is the parent individual to be mutated; Setting a mutation window of a preset length in the second target individual, and counting the types of element values ​​in the mutation window; If there are only two types of element values ​​in the mutation window, all element values ​​in the mutation window will be mutated to one of them to obtain a descendant individual.

8. The method for monitoring filter life of an air purification device according to claim 1, characterized in that: Create an initial vector set, including: Obtaining the historical optimal filter structure vector and the suboptimal filter structure vector; Based on the historical optimal filter structure vector, the operating state of the target air purification device is predicted to obtain predicted operating parameters, wherein the predicted operating parameters include the front-rear resistance difference and the front-rear filtration efficiency; According to the difference between the predicted operating parameters and the real-time operating parameters, multiple vectors are derived based on the historical optimal filter structure vector and the suboptimal filter structure vector, and an initial vector set is established, wherein the quantity ratio of the vector derived from the historical optimal filter structure vector and the multiple vectors derived from the suboptimal filter structure vector is negatively correlated with the difference between the predicted operating parameters and the real-time operating parameters.

9. The method for monitoring filter life of an air purification device according to claim 8, characterized in that: According to the difference between the predicted operating parameters and the real-time operating parameters, multiple vectors are derived based on the historical optimal filter structure vector and the suboptimal filter structure vector, and an initial vector set is established, including: According to the difference between the predicted operating parameters and the real-time operating parameters, the first neighborhood of the historical optimal filter structure vector and the second neighborhood of the suboptimal filter structure vector are obtained, wherein the range ratio of the first neighborhood and the second neighborhood is negatively correlated with the difference between the predicted operating parameters and the real-time operating parameters; Selecting multiple vectors in a first neighborhood of a historical optimal filter structure vector to obtain a first vector; Selecting multiple vectors in a second neighborhood of the historical suboptimal filter structure vector to obtain a second vector; Randomly generate multiple filter structure vectors to obtain a third vector; The first vector, the second vector and the third vector are aggregated to obtain an initial vector set.

10. A filter life monitoring system for an air purification device, characterized in that: include: A data monitoring module, used to obtain real-time operating parameters of the target air purification device, wherein the real-time operating parameters include the front-to-back resistance difference and the front-to-back filtration efficiency of the target air purification device; A vector construction module is used to establish an initial vector set, wherein the initial vector set includes a plurality of filter structure vectors, and the filter structure vectors are used to describe the hierarchical structure of the filter; A strategy building module is used to establish a fitness function according to the filter structure vector. The fitness function is used to calculate the corresponding theoretical operating parameters based on a filter structure vector, and calculate the difference between the theoretical operating parameters and the real-time operating parameters as the fitness of the filter structure vector, wherein the theoretical operating parameters include the front and rear resistance difference and the front and rear filtration efficiency; The structural analysis module is used to optimize the initial vector set through an optimization algorithm based on the initial vector set and the fitness function to obtain the optimal filter structure vector; The life monitoring module is used to monitor the life of the target filter according to the optimal filter structure vector.

Images