Electric appliance type selection method and device, storage medium, terminal and program product

Abstract

The invention provides an electric appliance type selection method and device, a storage medium, a terminal and a program product. The method comprises the following steps: acquiring a user demand; the user demand comprises a demand parameter value of each type selection parameter set by a user; searching a model selection result matched with the user demand in a preset database, and / or predicting the model selection result matched with the user demand by using a pre-trained machine learning model; if a model selection result matched with the user demand is retrieved in the preset database, feeding back the model selection result to the user; if the model selection result matched with the user demand is not retrieved in the preset database or the user is not satisfied with the retrieved model selection result, adjusting the priority of each model selection parameter; and according to the adjusted priority of each type selection parameter, re-retrieving a type selection result matched with the user demand in a preset database. According to the scheme provided by the invention, model selection can be automatically carried out according to user demand data.

Description

Technical Field

[0001] The present invention relates to the field of control, and in particular to an electrical appliance selection method, device, storage medium, terminal and program product. Background Art

[0002] In the process of selecting a multi-split central air-conditioning system, it is usually necessary to consider different building types and their room information, calculate the area and load of the room, and select air-conditioning equipment according to the cooling / heating requirements of the room, such as indoor units, outdoor units, and branch pipes and other accessories, and then determine the connecting pipes (pipe diameters) between different accessories and other system auxiliary materials.

[0003] In the related technologies, the selection of multi-split air conditioners mainly relies on manual operation. This method not only requires processing a large amount of calculation data, but also has a complex selection process. It also involves knowledge and engineering applications in multiple fields. In addition, this method often does not fully consider the impact of various ambient temperatures and other working conditions on the cooling load of the model, and requires the guidance of professionals to make effective air conditioner selection. At the same time, due to the wide variety of branch pipes, connecting pipes and auxiliary materials, manual processing is also quite difficult. Summary of the invention

[0004] The main purpose of the present invention is to overcome the defects of the above-mentioned related technologies and provide a method, device, storage medium, terminal and program product for selecting an electrical appliance to solve the problem that the selection of electrical appliances in the related technologies mainly relies on manual work and requires processing of a large amount of calculation data.

[0005] On the one hand, the present invention provides a method for selecting an electrical appliance, comprising: obtaining user needs; the user needs include: demand parameter values ​​of various selection parameters set by the user; retrieving selection results matching the user needs in a preset database, and / or predicting selection results matching the user needs using a pre-trained machine learning model; if a selection result matching the user needs is retrieved from the preset database, the selection result is fed back to the user; if a selection result matching the user needs is not retrieved from the preset database, or the user is not satisfied with the retrieved selection result, the priority of each selection parameter is adjusted; and based on the adjusted priority of each selection parameter, the selection result matching the user needs is retrieved again from the preset database.

[0006] Optionally, based on the adjusted priorities of each selection parameter, the selection results that match the user needs are retrieved again in a preset database, including: normalizing the adjusted priorities of each selection parameter to obtain the weight of each selection parameter; calculating the degree of match between the actual parameter values ​​of each selection parameter of each candidate model in the database and the required parameter values ​​of each selection parameter set by the user; calculating the recommendation score of each candidate model based on the weight of each selection parameter and the degree of match between the actual parameter values ​​of each selection parameter of each candidate model and the required parameter values ​​of each selection parameter set by the user; and selecting a preset number of candidate models as the selection results based on the calculated recommendation score of each candidate model.

[0007] Optionally, the adjusted priorities of the selection parameters are normalized to obtain the weights of the selection parameters, including: the weight of a single selection parameter is equal to the ratio of the priority of the selection parameter to the sum of the priorities of all selection parameters.

[0008] Optionally, calculating the degree of match between the actual parameter value of each selection parameter of each candidate model in the database and the required parameter value of each selection parameter set by the user includes: for any selection parameter, when the selection parameter is a numerical parameter, the degree of match between the actual parameter value of the selection parameter of each candidate model and the required parameter value of the selection parameter set by the user is equal to 1 minus the ratio of the absolute value of the difference between the actual parameter value of the selection parameter and the required parameter value of the selection parameter set by the user to the required parameter value of the selection parameter set by the user; and / or, for any selection parameter, when the selection parameter is a numerical parameter, the actual parameter value of the selection parameter of each candidate model The degree of match with the required parameter value of the selection parameter set by the user is equal to the maximum value of the difference between 0 and 1 minus the absolute value of the difference between the actual parameter value of the selection parameter and the required parameter value of the selection parameter set by the user and the ratio of the required parameter value of the selection parameter set by the user to the preset tolerance of the selection parameter; and / or, for any selection parameter, when the selection parameter is a non-numeric parameter, if the actual parameter value of the selection parameter of each candidate model is consistent with the required parameter value of the selection parameter set by the user, the degree of match is 1; if the actual parameter value of the selection parameter of each candidate model is inconsistent with the required parameter value of the selection parameter set by the user, the degree of match is 0.

[0009] Optionally, the recommendation score of each candidate model is calculated according to the weight of each selection parameter and the degree of matching between the actual parameter value of each selection parameter of each candidate model and the required parameter value of each selection parameter set by the user, including: calculating the total matching degree of each candidate model with the user requirements as the recommendation score of each candidate model according to the weight of each selection parameter and the degree of matching between the actual parameter value of each selection parameter of each candidate model and the required parameter value of each selection parameter set by the user.

[0010] Optionally, it further includes: when the number of times the priority of each selection parameter is adjusted exceeds a preset number, adjusting the default weight of each selection parameter according to the priority adjusted most recently.

[0011] Optionally, adjusting the default weight of each selection parameter according to the most recently adjusted priority includes: calculating the adjusted default weight of each selection parameter by a weighted moving average algorithm according to the weight of each selection parameter after the most recently adjusted priority.

[0012] On the other hand, the present invention provides an electrical appliance selection device, comprising: an acquisition unit, used to acquire user needs; the user needs include: demand parameter values ​​of various selection parameters set by the user; a matching unit, used to retrieve selection results matching the user needs in a preset database, and / or use a pre-trained machine learning model to predict selection results matching the user needs; a feedback unit, which feeds back the selection results to the user if the matching unit retrieves selection results matching the user needs in the preset database; a first adjustment unit, used to adjust the priority of each selection parameter if the matching unit does not retrieve a selection result matching the user needs in the preset database, or the user is not satisfied with the retrieved selection result; the matching unit is also used to: re-retrieve the selection results matching the user needs in the preset database according to the priorities of each selection parameter adjusted by the adjustment unit.

[0013] Optionally, the matching unit, according to the adjusted priorities of each selection parameter, re-searches the selection results matching the user requirements in a preset database, including: normalizing the adjusted priorities of each selection parameter to obtain the weight of each selection parameter; calculating the degree of matching between the actual parameter values ​​of each selection parameter of each candidate model in the database and the required parameter values ​​of each selection parameter set by the user; calculating the recommendation score of each candidate model according to the weight of each selection parameter and the degree of matching between the actual parameter values ​​of each selection parameter of each candidate model and the required parameter values ​​of each selection parameter set by the user; and selecting a preset number of candidate models as the selection results according to the calculated recommendation scores of each candidate model.

[0014] Optionally, the matching unit normalizes the adjusted priorities of the selection parameters to obtain weights of the selection parameters, including: the weight of a single selection parameter is equal to the ratio of the priority of the selection parameter to the sum of the priorities of all selection parameters.

[0015] Optionally, the matching unit calculates the matching degree of the actual parameter value of each selection parameter of each candidate model in the database with the required parameter value of each selection parameter set by the user, including: for any selection parameter, when the selection parameter is a numerical parameter, the matching degree of the actual parameter value of the selection parameter of each candidate model and the required parameter value of the selection parameter set by the user is equal to 1 minus the ratio of the absolute value of the difference between the actual parameter value of the selection parameter and the required parameter value of the selection parameter set by the user to the required parameter value of the selection parameter set by the user; and / or, for any selection parameter, when the selection parameter is a numerical parameter, the matching degree of the actual parameter value of the selection parameter of each candidate model and the required parameter value of the selection parameter set by the user is equal to 1 minus the absolute value of the difference between the actual parameter value of the selection parameter and the required parameter value of the selection parameter set by the user The matching degree between the actual parameter value and the required parameter value of the selection parameter set by the user is equal to the maximum value of the difference between 0 and 1 minus the absolute value of the difference between the actual parameter value of the selection parameter and the required parameter value of the selection parameter set by the user and the ratio of the required parameter value of the selection parameter set by the user to the preset tolerance of the selection parameter; and / or, for any selection parameter, when the selection parameter is a non-numeric parameter, if the actual parameter value of the selection parameter of each candidate model is consistent with the required parameter value of the selection parameter set by the user, the matching degree is 1; if the actual parameter value of the selection parameter of each candidate model is inconsistent with the required parameter value of the selection parameter set by the user, the matching degree is 0.

[0016] Optionally, the matching unit calculates the recommendation score of each candidate model according to the weight of each selection parameter and the degree of matching between the actual parameter value of each selection parameter of each candidate model and the required parameter value of each selection parameter set by the user, including: calculating the total matching degree of each candidate model with the user requirements as the recommendation score of each candidate model according to the weight of each selection parameter and the degree of matching between the actual parameter value of each selection parameter of each candidate model and the required parameter value of each selection parameter set by the user.

[0017] Optionally, it further includes: a second adjustment unit, configured to adjust the default weight of each selection parameter according to the priority adjusted most recently when the first adjustment unit adjusts the priority of each selection parameter more than a preset number of times.

[0018] Optionally, the second adjustment unit adjusts the default weight of each selection parameter according to the priority of the most recent adjustment, including: calculating the adjusted default weight of each selection parameter by a weighted moving average algorithm according to the weight of each selection parameter after the priority of the most recent adjustment.

[0019] Another aspect of the present invention provides a storage medium having a computer program stored thereon, wherein the program implements the steps of any of the aforementioned methods when executed by a processor.

[0020] In another aspect, the present invention provides a terminal, comprising a processor, a memory, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of any of the aforementioned methods when executing the program.

[0021] In another aspect, the present invention provides a terminal, comprising any one of the aforementioned selection devices.

[0022] In another aspect, the present invention provides a computer program product, comprising a computer program, wherein when the computer program is executed by a processor, the steps of any of the aforementioned methods are implemented.

[0023] According to the technical solution of the present invention, the selection results are matched from the preset selection database according to the user demand data. If a matching result is found, the user evaluates and gives feedback, outputs a selection report, and updates the database with the evaluation result. If no matching result is found, the parameter weight is adjusted, and the recommended selection is re-performed according to the adjusted parameter weight. The user evaluates and scores the recommended result, and through continuous iteration, the selection result is made more accurate.

[0024] According to the technical solution of the present invention, by building an electrical appliance selection model, the difficulty of electrical appliance selection is reduced, the selection accuracy is improved, and complex calculations and data analysis can be automatically processed, which greatly reduces the time and workload of manual processing, generates selection plans in real time, improves response speed, meets rapid project requirements, reduces costs and improves user experience, and effectively solves the requirements for professional knowledge in electrical appliance selection. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] The drawings described herein are used to provide a further understanding of the present invention and constitute a part of the present invention. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention. In the drawings:

[0026] Figure 1 is a method schematic diagram of an embodiment of the method for selecting an electrical appliance provided by the present invention;

[0027] Figure 2 A schematic diagram of a specific implementation of the method of the present invention is shown, in which a step of re-searching a selection result matching the user requirement in a preset database according to the adjusted priorities of each selection parameter;

[0028] Figure 3 is a method schematic diagram of another embodiment of the method for selecting an electrical appliance provided by the present invention;

[0029] Figure 4 It is a method schematic diagram of a specific embodiment of the method for selecting an electrical appliance provided by the present invention;

[0030] Figure 5is a method schematic diagram of another specific embodiment of the method for selecting an electrical appliance provided by the present invention;

[0031] Figure 6 It is a structural block diagram of an embodiment of the device for selecting an electrical appliance provided by the present invention;

[0032] Figure 7 It is a structural block diagram of another embodiment of the device for selecting an electrical appliance provided by the present invention. DETAILED DESCRIPTION

[0033] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the specific embodiments of the present invention and the corresponding drawings. 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 making creative work are within the scope of protection of the present invention.

[0034] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, 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 that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0035] In the relevant technology, data processing is cumbersome during the air conditioning selection process, and a large amount of building information and room data, including area, orientation, etc., needs to be collected and processed. The cooling load and heating load of each room need to be calculated, which involves complex thermodynamic calculations and engineering knowledge. There are many types of multi-unit equipment, with different models and specifications. It is necessary to select suitable indoor and outdoor units according to specific needs. It is also necessary to consider the performance parameters of the equipment, such as cooling capacity, energy efficiency ratio, noise level, etc., to ensure that the selected equipment can meet the actual use needs. The selection of accessories is complicated. The selection not only includes the internal and external equipment, but also involves the selection and configuration of various accessories such as branch pipes, connecting pipes, valves, controllers, etc. The selection process involves knowledge in multiple fields such as architecture, thermodynamics, refrigeration technology, and engineering design. Professional guidance is required, and it is difficult for ordinary users to complete the selection independently.

[0036] The present invention provides a method for selecting an electrical appliance, which is particularly suitable for selecting an air conditioner, such as a multi-connected central air conditioner.

[0037] Figure 1 It is a method schematic diagram of an embodiment of the method for selecting an electrical appliance provided by the present invention.

[0038] like Figure 1 As shown, according to one embodiment of the present invention, the method for selecting an electrical appliance includes at least step S110, step S120, step S130, step S140 and step S150.

[0039] Step S110, obtaining user needs.

[0040] The user requirements may specifically include: the required parameter values ​​of each selection parameter set by the user, that is, the parameter values ​​of the selection parameters set by the user according to his or her own requirements; for example, when the electrical appliance is an air conditioner, the selection parameters may include at least one of building area, usage scenario, climate, cooling capacity, cooling capacity, energy efficiency level, noise and price.

[0041] Specifically, when a user gives a requirement, the user requirement is parsed into a query condition. The requirement parameters are parsed into a query condition. For example: building area: 400 m2, usage scenario: office building, climate: hot and humid area, cooling capacity: 250 W / m2.

[0042] Step S120, searching a preset database for a selection result that matches the user's needs, and / or using a pre-trained selection model to predict a selection result that matches the user's needs.

[0043] In a specific implementation, a selection result matching the user's needs is retrieved from a preset database. Taking air conditioner selection as an example, the structure of the database can be shown in Table 1 below.

[0044] Referring to Table 1, taking air conditioning as an example, the database list contains the unique identifier of the selection record, the selection requirement description, key parameters, recommended selection results and user feedback parameters.

[0045] Table 1

[0046]

[0047] Specifically, after the user's needs are converted into query conditions, a search is performed in the database. For example, the search is performed using the SQL query language, and the query statement is, for example:

[0048] SELECT * FROM selection_db

[0049] WHERE project_type = 'Office Building'

[0050] AND area = 400

[0051] AND equipment_type = 'General office equipment'

[0052] AND climate='hot and humid'

[0053] LIMIT 1.

[0054] In another specific embodiment, a pre-trained machine learning model is used to predict a selection result that matches the user's needs. Optionally, if a selection result that matches the user's needs is not retrieved from the preset database, a pre-trained machine learning model can be used to predict a selection result that matches the user's needs.

[0055] In a specific implementation, a machine learning model based on a KNN (K-nearest neighbor) classification algorithm is used to predict selection results that match the user's needs according to the demand parameter values ​​set by the user, and the optimal selection result is recommended through the model.

[0056] For example, if a fully matched result cannot be found in the preset database, the machine learning model can make the following recommendations:

[0057] Extract user demand features and model prediction: The model predicts the most suitable air conditioner model, such as model X of a certain brand, based on similar projects in historical data (such as projects with the same area and budget range) and user feedback. Model X is recommended to users, who can rate the recommendation results and choose to accept or adjust.

[0058] Step S130: If a selection result matching the user's requirement is retrieved from the preset database, the selection result is fed back to the user.

[0059] Optionally, if a selection result matching the user's needs is retrieved in the preset database, the selection result is fed back to the user, and the user's evaluation of the selection result is obtained and saved in the preset database.

[0060] Specifically, when the selection result is retrieved, the selection result is fed back to the user, for example, displayed, and the user is prompted to evaluate, for example, the user can score the selection result. The user evaluation can be recorded in the database and used to adjust the weight of the selection parameter. For example, the score range is set to 1-10, and a threshold is set. When the score is greater than the set threshold, it indicates that the user is satisfied with the selection result, and the selection report is output. When the score is less than the set threshold, it indicates that the user is dissatisfied with the selection result, and the subsequent step S140 is executed.

[0061] Step S140: If no selection result matching the user's requirement is retrieved from the preset database, or the user is not satisfied with the retrieved selection result, the priority of each selection parameter is adjusted.

[0062] Specifically, the priority of each selection parameter can be adjusted by the user. For example, the user is prompted to adjust the priority of the selection parameter on the selection page, and the priority of the option parameter adjusted by the user on the selection page is received to re-select. For example, the user is prompted to re-select by adjusting the parameter priority, and a slider or drop-down option is displayed for different adjustment parameters to allow the user to set the priority of different selection parameters.

[0063] For example, each selection parameter (such as cooling capacity, energy efficiency level, noise, price, etc.) is listed on the selection page, and a priority adjustment tool (such as a slider, drop-down selection or star rating option) is displayed accordingly. The priority of each parameter can be, for example, from 0 to 10, and the larger the value, the higher the user's attention to the parameter.

[0064] Step S150: Re-searching a preset database for a selection result that matches the user's requirement according to the adjusted priorities of the selection parameters.

[0065] Figure 2 The schematic diagram of a specific implementation of the method of the present invention is a step of re-searching a selection result matching the user's requirements in a preset database according to the adjusted priorities of each selection parameter. Figure 2 As shown, step S150 includes steps S151 to S154:

[0066] Step S151, normalizing the adjusted priorities of the selection parameters to obtain the weights of the selection parameters.

[0067] Specifically, if the user adjusts the priority of the selection parameter, the database will be re-matched according to the user-adjusted priority, replacing the default weight, and performing a new selection calculation. In order to ensure that the weight calculation result is not unbalanced, the set priority value needs to be normalized, that is, the priority is adjusted to a relative weight to ensure that the sum of the weights of all parameters is equal to 1.

[0068] For example, suppose the user-defined selection parameter priorities are as follows: cooling capacity: 8; energy efficiency level: 6; noise: 4; price: 9. In order to make the weight of each selection parameter between 0 and 1, and the sum of the weights of all selection parameters equal to 1, the priority of each selection parameter needs to be normalized.

[0069] In a specific implementation, the weight of a single selection parameter is equal to the ratio of the priority of the selection parameter to the sum of the priorities of all selection parameters. For example, it can be normalized and calculated according to the following formula:

[0070]

[0071] The sum of the priorities of all selection parameters is: ∑ all selection parameter priorities = 8 + 6 + 4 + 9 = 27; then, the cooling capacity weight W 制冷量 for: The weight calculation of other selection parameters is similar.

[0072] Step S152, calculating the matching degree between the actual parameter value of each selection parameter of each candidate model in the database and the required parameter value of each selection parameter set by the user;

[0073] The calculation method of the matching degree between the actual parameter value of each selection parameter of each candidate model and the required parameter value of the corresponding selection parameter set by the user can be designed according to the type of the selection parameter and the tolerance mechanism.

[0074] In a specific implementation, for any selection parameter, when the selection parameter is a numerical parameter, the matching degree between the actual parameter value of the selection parameter of each candidate model and the required parameter value of the selection parameter set by the user is equal to 1 minus the ratio of the absolute value of the difference between the actual parameter value of the selection parameter and the required parameter value of the selection parameter set by the user to the required parameter value of the selection parameter set by the user, and the calculation formula is as follows:

[0075]

[0076] The above formula is mainly applicable to the case where the selection parameters are numerical parameters. When the selection parameters are numerical parameters, the numerical parameters refer to parameters with a clear numerical range, such as area, cooling capacity, noise level, etc. For such parameters, the matching degree can be calculated based on the difference. For example, assuming that the area requirement entered by the user is 100 square meters, and the applicable area of ​​a certain air-conditioning unit is 90 square meters, then the area matching degree calculation result is:

[0077] In order to allow a certain range of tolerances, a maximum allowable deviation can be set, for example, the tolerance for area is set to ±10%. When the difference between the actual parameter value and the required parameter value is within the tolerance range, the matching degree is still close to 1, and when the difference between the actual parameter value and the required parameter value exceeds the tolerance range, the matching degree will drop rapidly. In another specific embodiment, for any selection parameter, when the selection parameter is a numerical parameter, the matching degree between the actual parameter value of the selection parameter of each candidate model and the required parameter value of the selection parameter set by the user is equal to the maximum value of 0 to 1 minus the absolute value of the difference between the actual parameter value of the selection parameter and the required parameter value of the selection parameter set by the user and the ratio of the required parameter value of the selection parameter set by the user to the preset tolerance of the selection parameter, that is,

[0078] For example, Table 1 shows the comparison of the actual matching area (actual parameter value) of the candidate model and the required area (required parameter value) set by the user when the tolerance is introduced and when the tolerance is not introduced. Assuming that the required area (i.e., the required parameter value of the area) input by the user is 100 square meters, and assuming that the room areas matched by different candidate models recorded in the database include 95, 100, 110, and 120, and the tolerance is set to 10%, the corresponding matching degrees are calculated as shown in Table 2 below:

[0079] Table 2

[0080]

[0081] That is to say, if the user inputs a required area of ​​100 square meters, the applicable area of ​​the candidate air conditioner models in the database is within the range of 90-110 square meters, and all of them can be matched. The closer the area is to 100, the higher the matching degree is. If it exceeds the 10% tolerance range, that is, less than or equal to 90 and greater than or equal to 110, the matching degree will be reduced to 0. The main purpose of introducing tolerance is to allow a certain range of errors to ensure the applicability of the recommendation results. It can also be seen from the above table that if the tolerance is not introduced, the matching degree outside the range will also have a certain score (such as 110, 120), resulting in the recommendation results being not scientific and accurate. The higher the priority of the parameter set by the user, the smaller the tolerance range. For example, the cooling capacity tolerance is ±5%, and the noise tolerance is ±10%.

[0082] In a specific embodiment, for any selection parameter, when the selection parameter is a non-numeric parameter, if the actual parameter value of the selection parameter of each candidate model is consistent with the required parameter value of the selection parameter set by the user, the matching degree is 1; if the actual parameter value of the selection parameter of each candidate model is inconsistent with the required parameter value of the selection parameter set by the user, the matching degree is 0.

[0083] Specifically, for non-numerical parameters, such as energy efficiency level, brand, and cooling mode, the matching degree is calculated based on whether the actual parameter value is completely consistent with the required parameter value. If the two parameters are consistent, the matching degree is 1; if they are inconsistent, the matching degree is 0. For example, if the user requires the energy efficiency level to be "level 1", and the energy efficiency level of an air conditioner is "level 2", the matching degree is 0.

[0084] Step S153 , calculating the recommendation score of each candidate model according to the weight of each selection parameter and the matching degree between the actual parameter value of each selection parameter of each candidate model and the required parameter value of each selection parameter set by the user.

[0085] Specifically, based on the weight of each selection parameter and the degree of matching between the actual parameter value of each selection parameter of each candidate model and the required parameter value of each selection parameter set by the user, the total matching degree between each candidate model and the user's needs is calculated as the recommended score for each candidate model to ensure that the recommended air-conditioning model better meets the user's multi-dimensional needs.

[0086] Specifically, the weighted sum of the matching degree between the actual parameter value of each selection parameter of each candidate model and the required parameter value of each selection parameter set by the user is performed according to the weight of each selection parameter, and the total matching degree between each candidate model and the user's requirement is obtained as the recommendation score of each candidate model. That is, for each candidate model, the product of the weight of each selection parameter and the matching degree between the actual parameter value of the selection parameter of the candidate model and the required parameter value of the selection parameter set by the user is summed as the recommendation score of the candidate model, which can be calculated, for example, in the following way:

[0087] Recommended score = ∑ (weight of a single selection parameter × matching degree)

[0088] The matching degree in the above formula is the matching degree between the actual parameter value of a single selection parameter of any candidate model and the required parameter value of the selection parameter set by the user, that is, the matching degree between the actual parameter value of a single selection parameter of each candidate model and the required parameter value set by the user is calculated first, and then weighted calculation is performed according to the weight of each selection parameter to obtain the total matching degree between the candidate model and the user's requirements:

[0089] If the user customizes the priorities of n selection parameters, let: Ui be the user's priority for the i-th selection parameter; Wi be the normalized weight of the i-th selection parameter; M i is the matching degree between the actual parameter value and the required parameter value of the i-th selection parameter; T is the number of selection parameters, and the calculation formula of the normalized weight Wi is:

[0090] The total matching degree calculation formula is:

[0091] Step S154, selecting a preset number of candidate models as the selection result according to the calculated recommendation score of each candidate model.

[0092] Specifically, the candidate models whose calculated recommendation scores are higher than the preset scores are taken as the selection results that match the user's needs. If there are more than two candidate models whose recommendation scores are higher than the preset scores, the candidate model with the highest recommendation score is taken as the selection result. For example, three recommended solutions are obtained, and their scores are 0.85, 0.76 and 0.91 respectively. Then, according to the order of scores, recommended solution 3 (score 0.91) is taken as the priority option. If there is no candidate model with a recommended score higher than the preset score, the selection fails, and the feedback result that no matching air conditioner model is retrieved is output, and the priority of each selection parameter is readjusted.

[0093] It should be noted that in the above step S120, the above steps S152 to S154 are also used to retrieve the selection results matching the user's needs in the preset database, that is, the actual parameter values ​​of each selection parameter of each candidate model in the database and the required parameter values ​​of each selection parameter set by the user are calculated; the recommended score of each candidate model is calculated according to the weight of each selection parameter and the actual parameter value of each selection parameter of each candidate model and the required parameter value of each selection parameter set by the user, and the preset number of candidate models are selected as the selection result according to the recommended score of each candidate model calculated. Among them, the weight of each selection parameter is the default weight. If there is no candidate model with a recommended score higher than the preset score, the selection fails, and the feedback result that no suitable air-conditioning model is matched is output, and the priority of each selection parameter is adjusted, that is, it is executed according to step S130.

[0094] Figure 3 It is a method schematic diagram of another embodiment of the method for selecting an electrical appliance provided by the present invention.

[0095] like Figure 3 As shown, according to another embodiment of the present invention, the method for selecting an electrical appliance further includes step S160.

[0096] Step S160: When the number of times the priority of each selection parameter is adjusted exceeds a preset number, the default weight of each selection parameter is adjusted according to the priority adjusted most recently.

[0097] Specifically, after the user has made multiple priority adjustments to the selection parameters, the default weight can be automatically fine-tuned based on the priority most recently adjusted by the user, providing other users with similar needs with a default weight allocation that better meets actual needs.

[0098] In a specific implementation, the default weight of each selection parameter after adjustment can be calculated using a weighted moving average algorithm based on the weight of each selection parameter after the most recent priority adjustment (i.e., the priority of the adjusted selection parameter is normalized to obtain the weight of the selection parameter).

[0099] The calculation formula of the weighted moving average algorithm is as follows:

[0100] New weights i =ɑ*recently adjusted weight i +(1-ɑ)*current default weight i

[0101] Among them, ɑ is the smoothing coefficient, ranging from 0<ɑ≤1. The larger the ɑ, the greater the impact of the recent adjustment; the recent adjustment weight i It is the weight after the user last adjusted the priority of the selection parameter, that is, the weight obtained after normalizing the priority. The current default weight i is the historical default weight, the new weight i refers to the adjusted weights.

[0102] Assume the initial weights are as follows: area: 0.4, energy efficiency: 0.3, price: 0.2, noise: 0.1;

[0103] The weights adjusted by the user most recently are as follows: area: 0.3, energy efficiency: 0.4, price: 0.2, noise: 0.1; assuming the smoothing coefficient ɑ=0.6;

[0104] New weight (area) = 0.6*0.3+0.4*0.4 = 0.34

[0105] New weight (energy efficiency) = 0.6 0.4 + 0.4 0.3 = 0.36

[0106] New weight (price) = 0.6 0.2 + 0.4 0.2 = 0.20

[0107] New weight (noise) = 0.6 0.1 + 0.4 0.1 = 0.10

[0108] That is, the adjusted weights are: area: 0.34, energy efficiency: 0.36, price: 0.20, and noise: 0.10.

[0109] In order to clearly illustrate the technical solution of the present invention, the execution process of the method for selecting an electrical appliance provided by the present invention is described below with reference to several specific embodiments.

[0110] Figure 4 FIG. 1 is a schematic diagram of a specific embodiment of the method for selecting an electrical appliance provided by the present invention. Figure 4As shown, the user demand data is obtained, and the database is searched to see whether there is a selection result matching the user demand. If there is a selection result matching the user demand, it is fed back to the user for scoring, and the database is updated according to the user's score. If the score is greater than the preset value x, the selection report is output. If the database does not retrieve a selection result matching the user demand, the parameter weight is adjusted and the match is performed again in the database.

[0111] Figure 5 FIG. 2 is a schematic diagram of another specific embodiment of the method for selecting an electrical appliance provided by the present invention. Figure 5 As shown, user demand data is obtained, and the database is searched to see whether there is a selection result matching the user demand. If there is a selection result matching the user demand, it is fed back to the user for scoring, and the database is updated according to the user's score. If the score is greater than the preset value x, a selection report is output. If no selection result matching the user demand is retrieved in the database, the selection result is predicted through the machine learning recommendation system.

[0112] The present invention also provides an electrical appliance selection device, which is particularly suitable for selecting air conditioners, such as multi-connected central air conditioners.

[0113] Figure 6 FIG. 1 is a structural block diagram of an embodiment of the device for selecting an electrical appliance provided by the present invention. Figure 6 As shown, the device 100 for selecting an electrical appliance includes: an acquisition unit 110 , a matching unit 120 , a feedback unit 130 and a first adjustment unit 140 .

[0114] The acquisition unit 110 is used to acquire user requirements.

[0115] The user requirements may specifically include: the required parameter values ​​of each selection parameter set by the user, that is, the parameter values ​​of the selection parameters set by the user according to his or her own requirements; for example, if the electrical appliance is an air conditioner, the selection parameters may include at least one of building area, usage scenario, climate, cooling capacity, cooling capacity, energy efficiency level, noise and price.

[0116] Specifically, when a user gives a requirement, the user requirement is parsed into a query condition. The requirement parameters are parsed into a query condition. For example: building area: 400 m2, usage scenario: office building, climate: hot and humid area, cooling capacity: 250 W / m2.

[0117] The matching unit 120 is used to retrieve the selection results that match the user requirements in a preset database, and / or use a pre-trained machine learning model to predict the selection results that match the user requirements.

[0118] In a specific implementation, a selection result matching the user's needs is retrieved from a preset database. Taking air conditioner selection as an example, the structure of the database can be referred to as shown in Table 1 below:

[0119] Table 1

[0120]

[0121] Referring to Table 1, taking air conditioning as an example, the database list contains the unique identifier of the selection record, the selection requirement description, key parameters, recommended selection results and user feedback parameters.

[0122] Specifically, after the user's needs are converted into query conditions, a search is performed in the database. For example, the search is performed using the SQL query language, and the query statement is, for example:

[0123] SELECT * FROM selection_db

[0124] WHERE project_type = 'Office Building'

[0125] AND area = 400

[0126] AND equipment_type = 'General office equipment'

[0127] AND climate='hot and humid'

[0128] LIMIT 1.

[0129] In another specific embodiment, a pre-trained machine learning model is used to predict a selection result that matches the user's needs. Optionally, if a selection result that matches the user's needs is not retrieved from the preset database, a pre-trained machine learning model can be used to predict a selection result that matches the user's needs.

[0130] In a specific implementation, a machine learning model based on a KNN (K-nearest neighbor) classification algorithm is used to predict selection results that match the user's needs according to the demand parameter values ​​set by the user, and the optimal selection result is recommended through the model.

[0131] For example, if a fully matched result cannot be found in the preset database, the machine learning model can make the following recommendations:

[0132] Extract user demand features and model prediction: The model predicts the most suitable air conditioner model, such as model X of a certain brand, based on similar projects in historical data (such as projects with the same area and budget range) and user feedback. Model X is recommended to users, who can rate the recommendation results and choose to accept or adjust.

[0133] Feedback unit 130, if the matching unit retrieves a selection result matching the user's needs in the preset database, the selection result is fed back to the user.

[0134] Optionally, the device 100 may further include an acquisition unit (not shown) for acquiring a user's evaluation of a selection result. If the matching unit 120 retrieves a selection result that matches the user's needs in the preset database, the feedback unit 130 feeds back the selection result to the user, acquires the user's evaluation of the selection result, and saves it in the preset database.

[0135] Specifically, when the selection result is retrieved, the selection result is fed back to the user, for example, displayed, and the user is prompted to evaluate, for example, the user can score the selection result. The user evaluation can be recorded in the database and used to adjust the weight of the selection parameter. For example, the score range is set to 1-10, and a threshold is set. When the score is greater than the set threshold, it indicates that the user is satisfied with the selection result, and a selection report is output. When the score is less than the set threshold, it indicates that the user is dissatisfied with the selection result.

[0136] The first adjustment unit 140 is configured to adjust the priority of each selection parameter if the matching unit fails to retrieve a selection result matching the user requirement in the preset database, or the user is not satisfied with the retrieved selection result.

[0137] Specifically, the priority of each selection parameter can be adjusted by the user. For example, the user is prompted to adjust the priority of the selection parameter on the selection page, and the priority of the option parameter adjusted by the user on the selection page is received to re-select. For example, the user is prompted to re-select by adjusting the parameter priority, and a slider or drop-down option is displayed for different adjustment parameters to allow the user to set the priority of different selection parameters.

[0138] For example, each selection parameter (such as cooling capacity, energy efficiency level, noise, price, etc.) is listed on the selection page, and a priority adjustment tool (such as a slider, drop-down selection or star rating option) is displayed accordingly. The priority of each parameter can be, for example, from 0 to 10, and the larger the value, the higher the user's attention to the parameter.

[0139] The matching unit 120 is further configured to: retrieve a selection result matching the user requirement from a preset database again according to the priority of each selection parameter adjusted by the adjustment unit.

[0140] In a specific implementation, the matching unit 120 re-searches the selection results matching the user requirements in the preset database according to the adjusted priorities of the selection parameters, according to the following steps S151 to S154:

[0141] Step S151, normalizing the adjusted priorities of the selection parameters to obtain the weights of the selection parameters.

[0142] Specifically, if the user adjusts the priority of the selection parameter, the database will be re-matched according to the adjusted priority. The default weight is replaced and a new selection calculation is performed. In order to ensure that the weight calculation result is not unbalanced, the set priority value needs to be normalized, that is, the priority is adjusted to a relative weight to ensure that the sum of the weights of all parameters is equal to 1.

[0143] For example, suppose the user-defined selection parameter priorities are as follows: cooling capacity: 8; energy efficiency level: 6; noise: 4; price: 9. In order to make the weight of each selection parameter between 0 and 1, and the sum of the weights of all selection parameters equal to 1, the priority of each selection parameter needs to be normalized.

[0144] In a specific implementation, the weight of a single selection parameter is equal to the ratio of the priority of the selection parameter to the sum of the priorities of all selection parameters. For example, it can be normalized and calculated according to the following formula:

[0145]

[0146] The sum of the priorities of all selection parameters is: ∑ all selection parameter priorities = 8 + 6 + 4 + 9 = 27; then, the cooling capacity weight W 制冷量 for: The weight calculation of other selection parameters is similar.

[0147] Step S152, calculating the matching degree between the actual parameter value of each selection parameter of each candidate model in the database and the required parameter value of each selection parameter set by the user;

[0148] The calculation method of the matching degree between the actual parameter value of each selection parameter of each candidate model and the required parameter value of the corresponding selection parameter set by the user can be designed according to the type of the selection parameter and the tolerance mechanism.

[0149] In a specific implementation, for any selection parameter, when the selection parameter is a numerical parameter, the matching degree between the actual parameter value of the selection parameter of each candidate model and the required parameter value of the selection parameter set by the user is equal to 1 minus the ratio of the absolute value of the difference between the actual parameter value of the selection parameter and the required parameter value of the selection parameter set by the user to the required parameter value of the selection parameter set by the user, and the calculation formula is as follows:

[0150]

[0151] The above formula is mainly applicable to the case where the selection parameters are numerical parameters. When the selection parameters are numerical parameters, the numerical parameters refer to parameters with a clear numerical range, such as area, cooling capacity, noise level, etc. For such parameters, the matching degree can be calculated based on the difference. For example, assuming that the area requirement entered by the user is 100 square meters, and the applicable area of ​​a certain air-conditioning unit is 90 square meters, then the area matching degree calculation result is:

[0152] In order to allow a certain range of tolerances, a maximum allowable deviation can be set, for example, the tolerance for area is set to ±10%. When the difference between the actual parameter value and the required parameter value is within the tolerance range, the matching degree is still close to 1, and when the difference between the actual parameter value and the required parameter value exceeds the tolerance range, the matching degree will drop rapidly. In another specific embodiment, for any selection parameter, when the selection parameter is a numerical parameter, the matching degree between the actual parameter value of the selection parameter of each candidate model and the required parameter value of the selection parameter set by the user is equal to the maximum value of 0 to 1 minus the absolute value of the difference between the actual parameter value of the selection parameter and the required parameter value of the selection parameter set by the user and the ratio of the required parameter value of the selection parameter set by the user to the preset tolerance of the selection parameter, that is,

[0153] For example, Table 1 shows the comparison of the actual matching area (actual parameter value) of the candidate model and the required area (required parameter value) set by the user when the tolerance is introduced and when the tolerance is not introduced. Assuming that the required area (i.e., the required parameter value of the area) input by the user is 100 square meters, and assuming that the room areas matched by different candidate models recorded in the database include 95, 100, 110, and 120, and the tolerance is set to 10%, the corresponding matching degrees are calculated as shown in Table 2 below:

[0154] Table 2

[0155]

[0156] That is to say, if the user inputs a required area of ​​100 square meters, the applicable area of ​​the candidate air conditioner models in the database is within the range of 90-110 square meters, and all of them can be matched. The closer the area is to 100, the higher the matching degree is. If it exceeds the 10% tolerance range, that is, less than or equal to 90 and greater than or equal to 110, the matching degree will be reduced to 0. The main purpose of introducing tolerance is to allow a certain range of errors to ensure the applicability of the recommendation results. It can also be seen from the above table that if the tolerance is not introduced, the matching degree outside the range will also have a certain score (such as 110, 120), resulting in the recommendation results being not scientific and accurate. The higher the priority of the parameter set by the user, the smaller the tolerance range. For example, the cooling capacity tolerance is ±5%, and the noise tolerance is ±10%.

[0157] In a specific embodiment, for any selection parameter, when the selection parameter is a non-numeric parameter, if the actual parameter value of the selection parameter of each candidate model is consistent with the required parameter value of the selection parameter set by the user, the matching degree is 1; if the actual parameter value of the selection parameter of each candidate model is inconsistent with the required parameter value of the selection parameter set by the user, the matching degree is 0.

[0158] Specifically, for non-numerical parameters, such as energy efficiency level, brand, and cooling mode, the matching degree is calculated based on whether the actual parameter value is completely consistent with the required parameter value. If the two parameters are consistent, the matching degree is 1; if they are inconsistent, the matching degree is 0. For example, if the user requires the energy efficiency level to be "level 1", and the energy efficiency level of an air conditioner is "level 2", the matching degree is 0.

[0159] Step S153 , calculating the recommendation score of each candidate model according to the weight of each selection parameter and the matching degree between the actual parameter value of each selection parameter of each candidate model and the required parameter value of each selection parameter set by the user.

[0160] Specifically, based on the weight of each selection parameter and the degree of matching between the actual parameter value of each selection parameter of each candidate model and the required parameter value of each selection parameter set by the user, the total matching degree between each candidate model and the user's needs is calculated as the recommended score for each candidate model to ensure that the recommended air-conditioning model better meets the user's multi-dimensional needs.

[0161] Specifically, the weighted sum of the matching degree between the actual parameter value of each selection parameter of each candidate model and the required parameter value of each selection parameter set by the user is performed according to the weight of each selection parameter, and the total matching degree between each candidate model and the user's requirement is obtained as the recommendation score of each candidate model. That is, for each candidate model, the product of the weight of each selection parameter and the matching degree between the actual parameter value of the selection parameter of the candidate model and the required parameter value of the selection parameter set by the user is summed as the recommendation score of the candidate model, which can be calculated, for example, in the following way:

[0162] Recommended score = ∑ (weight of a single selection parameter × matching degree)

[0163] The matching degree in the above formula is the matching degree between the actual parameter value of a single selection parameter of any candidate model and the required parameter value of the selection parameter set by the user, that is, the matching degree between the actual parameter value of a single selection parameter of each candidate model and the required parameter value set by the user is calculated first, and then weighted calculation is performed according to the weight of each selection parameter to obtain the total matching degree between the candidate model and the user's requirements:

[0164] If the user customizes the priorities of n selection parameters, let: Ui be the user's priority for the i-th selection parameter; Wi be the normalized weight of the i-th selection parameter; M i is the matching degree between the actual parameter value and the required parameter value of the i-th selection parameter; T is the number of selection parameters, and the calculation formula of the normalized weight Wi is:

[0165] The total matching degree calculation formula is:

[0166] Step S154, selecting a preset number of candidate models as the selection result according to the calculated recommendation score of each candidate model.

[0167] Specifically, the candidate models whose calculated recommendation scores are higher than the preset scores are taken as the selection results that match the user's needs. If there are more than two candidate models whose recommendation scores are higher than the preset scores, the candidate model with the highest recommendation score is taken as the selection result. For example, three recommended solutions are obtained, and their scores are 0.85, 0.76 and 0.91 respectively. Then, according to the order of scores, recommended solution 3 (score 0.91) is taken as the priority option. If there is no candidate model with a recommended score higher than the preset score, the selection fails, and the feedback result that no matching air conditioner model is retrieved is output, and the priority of each selection parameter is readjusted.

[0168] Figure 7 FIG. 1 is a structural block diagram of another embodiment of the device for selecting an electrical appliance provided by the present invention. Figure 7 As shown, based on the above implementation, the device 100 for selecting an electrical appliance further includes: a second adjustment unit 150 .

[0169] The second adjusting unit 150 is configured to adjust the default weight of each selection parameter according to the priority adjusted most recently when the first adjusting unit 140 adjusts the priority of each selection parameter for a number of times exceeding a preset number.

[0170] Specifically, after the user has made multiple priority adjustments to the selection parameters, the default weight can be automatically fine-tuned based on the priority most recently adjusted by the user, providing other users with similar needs with a default weight allocation that better meets actual needs.

[0171] In a specific implementation, the default weight of each selection parameter after adjustment can be calculated using a weighted moving average algorithm based on the weight of each selection parameter after the most recent priority adjustment (i.e., the priority of the adjusted selection parameter is normalized to obtain the weight of the selection parameter).

[0172] The calculation formula of the weighted moving average algorithm is as follows:

[0173] New weights i =ɑ*recently adjusted weight i +(1-ɑ)*current default weight i

[0174] Among them, ɑ is the smoothing coefficient, ranging from 0<ɑ≤1. The larger the ɑ, the greater the impact of the recent adjustment; the recent adjustment weight i It is the weight after the user last adjusted the priority of the selection parameter, that is, the weight obtained after normalizing the priority. The current default weight i is the historical default weight, the new weight i refers to the adjusted weights.

[0175] The present invention also provides a storage medium corresponding to the electrical appliance selection method, on which a computer program is stored, and when the program is executed by a processor, the steps of any of the aforementioned methods are implemented.

[0176] The present invention also provides a terminal corresponding to the electrical appliance selection method, comprising a processor, a memory, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of any of the aforementioned methods when executing the program.

[0177] The present invention also provides a terminal corresponding to the selection device of the electrical appliance, comprising any of the selection devices described above.

[0178] The present invention also provides a computer program product corresponding to the electrical appliance selection method, comprising a computer program, and when the computer program is executed by a processor, the steps of any of the aforementioned methods are implemented.

[0179] Accordingly, the solution provided by the present invention matches the selection results from the preset selection database according to the customer demand data. If a matching result is found, the user evaluates and gives feedback, outputs a selection report, and updates the database with the evaluation result. If no matching result is found, the parameter weight is adjusted, and the recommended selection is re-performed according to the adjusted parameter weight. The user evaluates and scores the recommended results, and through continuous iteration, the selection result is made more accurate.

[0180] According to the technical solution of the present invention, by building an electrical appliance selection model, the difficulty of electrical appliance selection is reduced, the selection accuracy is improved, and complex calculations and data analysis can be automatically processed, which greatly reduces the time and workload of manual processing, generates selection plans in real time, improves response speed, meets rapid project requirements, reduces costs and improves user experience. It effectively solves the requirements of air conditioner selection for professional air conditioner knowledge.

[0181] The functions described herein may be implemented in hardware, software executed by a processor, firmware, or any combination thereof. If implemented in software executed by a processor, the functions may be stored as one or more instructions or codes on a computer-readable medium or transmitted via a computer-readable medium. Other examples and implementations are within the scope and spirit of the present invention and the appended claims. For example, due to the nature of software, the functions described above may be implemented using software executed by a processor, hardware, firmware, hard wiring, or a combination of any of these. In addition, each functional unit may be integrated into a processing unit, each unit may exist physically separately, or two or more units may be integrated into one unit.

[0182] In the several embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways. Among them, the device embodiments described above are only schematic. For example, the division of the units can be a logical function division. There may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of units or modules, which can be electrical or other forms.

[0183] The units described as separate components may or may not be physically separated, and the components of the control device may or may not be physical units, that is, they may be located in one place or distributed in multiple units. Some or all of the units may be selected according to actual needs to achieve the purpose of the present embodiment.

[0184] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the relevant technology or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions to enable a computer device (which can be a personal computer, a server or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), mobile hard disk, disk or optical disk and other media that can store program codes.

[0185] The above description is only an embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent substitution, improvement, etc. made within the spirit and principle of the present invention shall be included in the scope of the claims of the present invention.

Claims

1. A method for selecting an electrical appliance, characterized in that: include: Obtain user needs; The user requirements include: the required parameter values ​​of each selection parameter set by the user; Retrieving selection results matching the user requirements in a preset database, and / or predicting selection results matching the user requirements using a pre-trained machine learning model; If a selection result matching the user's requirement is retrieved from the preset database, the selection result is fed back to the user; If no selection result matching the user's requirement is retrieved from the preset database, or the user is not satisfied with the retrieved selection result, the priority of each selection parameter is adjusted; According to the adjusted priorities of the various selection parameters, the selection results matching the user requirements are searched again in the preset database.

2. The method according to claim 1, characterized in that According to the adjusted priorities of the selection parameters, the selection results matching the user requirements are searched again in the preset database, including: Normalize the priorities of the adjusted selection parameters to obtain the weights of the selection parameters; Calculating the matching degree between the actual parameter value of each selection parameter of each candidate model in the database and the required parameter value of each selection parameter set by the user; Calculate the recommendation score of each candidate model according to the weight of each selection parameter and the matching degree between the actual parameter value of each selection parameter of each candidate model and the required parameter value of each selection parameter set by the user; According to the calculated recommendation score of each candidate model, a preset number of candidate models are selected as the selection results.

3. The method according to claim 2, characterized in that Normalize the priorities of the adjusted selection parameters to obtain the weights of the selection parameters, including: The weight of a single selection parameter is equal to the ratio of the priority of the selection parameter to the sum of the priorities of all selection parameters.

4. The method according to claim 2, characterized in that Calculating the matching degree between the actual parameter value of each selection parameter of each candidate model in the database and the required parameter value of each selection parameter set by the user, including: For any selection parameter, when the selection parameter is a numerical parameter, the matching degree between the actual parameter value of the selection parameter of each candidate model and the required parameter value of the selection parameter set by the user is equal to 1 minus the ratio of the absolute value of the difference between the actual parameter value of the selection parameter and the required parameter value of the selection parameter set by the user to the required parameter value of the selection parameter set by the user; and / or, For any selection parameter, when the selection parameter is a numerical parameter, the matching degree between the actual parameter value of the selection parameter of each candidate model and the required parameter value of the selection parameter set by the user is equal to the maximum value of the difference between 0 and 1 minus the absolute value of the difference between the actual parameter value of the selection parameter and the required parameter value of the selection parameter set by the user and the ratio of the required parameter value of the selection parameter set by the user to the preset tolerance of the selection parameter; and / or, For any selection parameter, when the selection parameter is a non-numeric parameter, if the actual parameter value of the selection parameter of each candidate model is consistent with the required parameter value of the selection parameter set by the user, the matching degree is 1; if the actual parameter value of the selection parameter of each candidate model is inconsistent with the required parameter value of the selection parameter set by the user, the matching degree is 0.

5. The method according to claim 2, characterized in that: According to the weight of each selection parameter and the matching degree between the actual parameter value of each selection parameter of each candidate model and the required parameter value of each selection parameter set by the user, the recommendation score of each candidate model is calculated, including: According to the weight of each selection parameter and the matching degree between the actual parameter value of each selection parameter of each candidate model and the required parameter value of each selection parameter set by the user, the total matching degree between each candidate model and the user requirement is calculated as the recommendation score of each candidate model.

6. The method according to any one of claims 1 to 5, characterized in that: Also includes: When the number of times the priority of each selection parameter is adjusted exceeds a preset number, the default weight of each selection parameter is adjusted according to the priority adjusted most recently.

7. The method according to claim 6, characterized in that Adjust the default weight of each selection parameter based on the priority of the most recent adjustment, including: According to the weight of each selection parameter after the most recent adjustment of the priority, the default weight of each selection parameter after adjustment is calculated by the weighted moving average algorithm.

8. An electrical appliance selection device, characterized in that: include: An acquisition unit, used to acquire user requirements; The user requirements include: the required parameter values ​​of each selection parameter set by the user; A matching unit, used to retrieve a selection result matching the user's needs from a preset database, and / or predict a selection result matching the user's needs using a pre-trained machine learning model; A feedback unit, wherein if the matching unit retrieves a selection result matching the user's needs in the preset database, the selection result is fed back to the user; A first adjustment unit, configured to adjust the priority of each selection parameter if the matching unit fails to retrieve a selection result matching the user requirement in the preset database, or the user is not satisfied with the retrieved selection result; The matching unit is further used to: according to the priorities of the selection parameters adjusted by the adjustment unit, re-search the selection results matching the user requirements in the preset database.

9. A storage medium, characterized in that: A computer program is stored thereon, and when the program is executed by a processor, the steps of any method described in claims 1-7 are implemented.

10. A terminal, characterized in that: The method comprises a processor, a memory and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of any method described in claims 1 to 7 when executing the program, or comprises the selection device described in claim 8.

11. A computer program product, characterized in that The invention comprises a computer program, which implements the steps of any one of the methods of claims 1 to 7 when being executed by a processor.

Images