A target matching analysis method, system and storage medium

Abstract

The embodiments of this specification provide a target matching analysis method, system and storage medium, including: determining the degree of matching between the target order information and the multimodal data of the beverage processing equipment based on the target order information and the multimodal data; in response to the degree of matching satisfying the matching condition, determining matching fault information based on the multimodal data and the target order information; and determining fault handling measures based on the matching fault information.

Description

Technical Field

[0001] The present invention relates to the field of matching analysis, and in particular to a target matching analysis method, system and storage medium. Background Art

[0002] With the development of the beverage processing industry, there are more and more types of beverages. During the beverage production process, the beverage processing equipment may not work properly due to various factors, such as mechanical failure, improper parameter settings (such as mismatch between order requirements and parameter settings), external interference, etc. If only relying on simple manual inspection, there is a lack of accurate judgment of the working status of the equipment, which makes it difficult to detect potential problems with the equipment (such as matching failure, etc.).

[0003] Therefore, it is necessary to provide a target matching analysis method, system and storage medium that can comprehensively and accurately monitor the working status of the equipment and quickly take processing measures when there are problems with the equipment, so as to improve the quality and efficiency of beverage production and enhance the user experience. Summary of the invention

[0004] One or more embodiments of the present specification provide a target matching analysis method, comprising: determining a degree of matching between the target order information and the multimodal data of a beverage processing device based on the target order information and the multimodal data; in response to the degree of matching satisfying a matching condition, determining matching fault information based on the multimodal data and the target order information; and determining fault handling measures based on the matching fault information.

[0005] One or more embodiments of the present specification provide a target matching analysis system, the system comprising: a first determination module, configured to determine the degree of matching between the target order information and the multimodal data of the beverage processing equipment based on the target order information and the multimodal data; a second determination module, configured to determine matching fault information based on the multimodal data and the target order information in response to the matching degree satisfying a matching condition; and a third determination module, configured to determine fault handling measures based on the matching fault information.

[0006] One or more embodiments of the present specification provide a computer-readable storage medium, wherein the storage medium stores computer instructions. When a computer reads the computer instructions in the storage medium, the computer runs the target matching analysis method described in any one of the above embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0007] This specification will be further described in the form of exemplary embodiments, which will be described in detail by the accompanying drawings. These embodiments are not restrictive, and in these embodiments, the same number represents the same structure, wherein:

[0008] Figure 1 is a schematic diagram of an application scenario of a target matching analysis system according to some embodiments of this specification;

[0009] Figure 2 is an exemplary module diagram of a target matching analysis system according to some embodiments of this specification;

[0010] Figure 3 is an exemplary flow chart of a target matching analysis method according to some embodiments of this specification;

[0011] Figure 4 is an exemplary flow chart for determining matching fault information according to some embodiments of this specification;

[0012] Figure 5 It is an exemplary schematic diagram of an order prediction model according to some embodiments of the present specification. DETAILED DESCRIPTION

[0013] In order to more clearly illustrate the technical solutions of the embodiments of this specification, the following is a brief introduction to the drawings required for the description of the embodiments. Obviously, the drawings described below are only some examples or embodiments of this specification. For ordinary technicians in this field, this specification can also be applied to other similar scenarios based on these drawings without creative work. Unless it is obvious from the language environment or otherwise explained, the same reference numerals in the figures represent the same structure or operation.

[0014] It should be understood that the "system", "device", "unit" and / or "module" used herein are a method for distinguishing different components, elements, parts, portions or assemblies at different levels. However, if other words can achieve the same purpose, the words can be replaced by other expressions.

[0015] Flowcharts are used in this specification to illustrate the operations performed by the system according to the embodiments of this specification. It should be understood that the preceding or following operations are not necessarily performed precisely in order. Instead, the steps may be processed in reverse order or simultaneously. At the same time, other operations may also be added to these processes, or one or more operations may be removed from these processes.

[0016] Figure 1 Schematic diagram of application scenarios of target matching analysis systems according to some embodiments of this specification. Figure 1 As shown, the application scenario 100 of the target matching analysis system may include a beverage processing device 110 , a network 120 , a processor 130 , a storage device 140 and a user terminal 150 .

[0017] The network 120 can connect the components of the application scenario 100 of the target matching analysis system and / or connect the system with external resources. The network 120 enables communication between the components and with other components outside the system to facilitate the exchange of data and / or information. For example, the processor 130 can obtain target order information from the user terminal 150 through the network 120.

[0018] The processor 130 may process data and / or information obtained from other devices or system components. The processor may execute program instructions based on the data, information and / or processing results to perform one or more functions described in the present application. For example, the processor 130 may determine the degree of match between the target order information and the multimodal data based on the target order information and the multimodal data of the beverage processing equipment. For another example, the processor 130 may determine matching fault information based on the multimodal data and the target order information in response to the matching degree satisfying the matching condition; and determine fault handling measures based on the matching fault information.

[0019] In some embodiments, the processor 130 may include one or more sub-processing devices (e.g., a single-core processing device or a multi-core multi-core processing device). As an example only, the processor 130 may include a central processing unit (CPU), an application-specific integrated circuit (ASIC), an application-specific instruction processor (ASIP), a graphics processing unit (GPU), a physical processing unit (PPU), a digital signal processor (DSP), a microprocessor, etc., or any combination thereof.

[0020] The storage device 140 may be used to store data and / or instructions. The storage device 140 may include one or more storage components, each of which may be an independent device or a part of another device. In some embodiments, the storage device 140 may include a random access memory (RAM), a read-only memory (ROM), a mass storage device, etc., or any combination thereof.

[0021] In some embodiments, when the processor 130 obtains the target order information from the user terminal 150 , it may store the data on the storage device 140 , or obtain the data on the storage device 140 , etc.

[0022] The user terminal 150 refers to one or more terminal devices or software used by the user. In some embodiments, the user terminal 150 may be used by one or more users. In some embodiments, the user terminal 150 may include a mobile phone 150-1, a tablet 150-2, and a computer 150-3. In some embodiments, the user may be at least one of a customer placing an order, a clerk operating the beverage processing equipment 110, and the like.

[0023] It should be noted that the application scenario 100 of the target matching analysis system is provided only for the purpose of illustration and is not intended to limit the scope of this specification. For those of ordinary skill in the art, various modifications or variations can be made according to the description of this specification. For example, the application scenario 100 of the target matching analysis system can also include a sensor. However, variations and modifications will not deviate from the scope of this specification.

[0024] Figure 2 is an exemplary module diagram of a target matching analysis system according to some embodiments of this specification. Figure 2 As shown, the target matching analysis system 200 may include a first determination module 210, a second determination module 220, and a third determination module 230. In some embodiments, one or more of the first determination module 210, the second determination module 220, and the third determination module 230 may be integrated into the processor 130.

[0025] In some embodiments, the first determination module 210 may be configured to determine the degree of match between the target order information and the multimodal data based on the target order information and the multimodal data of the beverage processing equipment. In some embodiments, the first determination module 210 may be further configured to determine the raw material consistency data based on the execution information of the beverage processing equipment and the target order information; and determine the degree of match based on the raw material consistency data, the target order information, and the multimodal data.

[0026] In some embodiments, the first determination module 210 may be further configured to determine matching fault information based on the raw material consistency data, the multimodal data, and the target order information.

[0027] In some embodiments, the second determination module 220 may be configured to determine matching failure information based on the multimodal data and the target order information in response to the matching degree satisfying the matching condition.

[0028] In some embodiments, the third determination module 230 may be configured to determine the fault handling measures based on the matching fault information. In some embodiments, the third determination module 230 may be further configured to determine the current product information based on the multimodal data and the execution information of the beverage processing equipment; and determine the fault handling measures based on the current product information, the target order information and the matching fault information. In some embodiments, the fault handling measures include product handling measures, and the third determination module 230 may be further configured to determine the product handling measures based on the pending order and the current product information.

[0029] For more information on the above, see Figures 3 to 5 Related description.

[0030] It should be understood that Figure 2The system and its modules shown can be implemented in various ways. It should be noted that the above description of the target matching analysis system 200 and its modules is only for the convenience of description and cannot limit this specification to the scope of the embodiments. It is understandable that for those skilled in the art, after understanding the principle of the system, it is possible to arbitrarily combine the modules or form a subsystem to connect with other modules without deviating from this principle. In some embodiments, Figure 2 The first determination module 210, the second determination module 220, and the third determination module 230 disclosed in the specification may be different modules in a system, or one module may implement the functions of two or more modules. For example, each module may share a storage module, or each module may have its own storage module. Such variations are within the scope of protection of this specification.

[0031] Figure 3 is an exemplary flow chart of a target matching analysis method according to some embodiments of this specification. In some embodiments, process 300 may be executed by processor 130. For an explanation of processor 130, see Figure 1 Related description.

[0032] In some embodiments, Figure 3 As shown, the processor 130 can determine the degree of matching between the target order information and the beverage processing equipment based on the target order information and the multimodal data of the beverage processing equipment; in response to the degree of matching satisfying the matching condition, determine matching fault information based on the multimodal data and the target order information; and determine fault handling measures based on the matching fault information.

[0033] Step 310: Based on the target order information and the multimodal data of the beverage processing equipment, determine the matching degree between the target order information and the multimodal data. For more information about beverage processing equipment, see Figure 1 The corresponding description.

[0034] The target order information refers to order information related to the target beverage. The target order information can be used to instruct the beverage processing equipment to make the target beverage according to the customer's needs. In some embodiments, the target order information may include the order time of the target beverage and the target beverage information. Among them, the target beverage refers to the beverage being made by the beverage processing equipment. For example, the target beverage may include at least one of coffee, soy milk, milk tea, etc.

[0035] The target beverage information refers to the relevant parameters related to the target beverage. For example, the target beverage information includes the type of target beverage, sugar content, ice content, and other parameters related to the target beverage. For example, taking the target beverage coffee as an example, the beverage types may include latte, Americano, cappuccino, etc.; the sugar content may include different amounts of sugar added, such as sugar-free, less sugar, full sugar, etc.; the ice content may include different amounts of ice cubes added, such as ice-free, less ice, normal ice, etc.

[0036] In some embodiments, the target order information may be generated based on the collected customer needs. For example, the customer may input the beverage parameters of his / her preference through the user terminal 150. The processor 130 may obtain the beverage parameters and order time input by the user terminal 150 through the network 120 and determine them as the target order information.

[0037] In some embodiments, when the beverage processing equipment is used to make the target beverage, the beverage processing equipment may be a multifunctional beverage processing equipment or an intelligent self-service beverage processing equipment, etc. For example, taking a beverage processing equipment coffee machine as an example, the multifunctionality includes multiple functions such as automatic addition of coffee beans, grinding, automatic addition of water, and automatic addition of sugar.

[0038] Modal data refers to data collected when the beverage processing equipment makes a target beverage. In some embodiments, the processor 130 can collect modal data through a sensor.

[0039] In some embodiments, different modal data represent different types of sensor data acquired by the processor 130. For example, the processor 130 may collect different modal data through different types of sensors. The sensor may include at least one of a current sensor, a vibration sensor, a sound sensor, and a valve sensor.

[0040] In some embodiments, the above-mentioned sensors can be pre-installed at multiple preset positions of the beverage processing equipment. Among them, the preset positions can be pre-set by the processor and / or the technician. For example, the preset position of the current sensor includes the power input terminal of the beverage processing equipment, the electrical component circuit of the beverage processing equipment, etc.; the preset position of the valve sensor includes the valve component of the beverage processing equipment, etc.; the preset position of the vibration sensor includes the motor or pump of the beverage processing equipment, etc. The preset position of the sound sensor can include the outside of the body of the beverage processing equipment, the container wall, etc. The above-mentioned sensors can also be pre-installed at any feasible position of the beverage processing equipment.

[0041] Multimodal data refers to data collected in multiple dimensions about the beverage processing equipment when making the target beverage. Multimodal data can be used to comprehensively evaluate the operating status of the beverage processing equipment when making the target beverage. For example, multimodal data can include at least one of electrical signal data, vibration data, sound data, etc.

[0042] The electrical signal data refers to electrical information related to the beverage processing equipment. For example, the electrical signal data includes current, voltage, power, etc. In some embodiments, the electrical signal data can reflect the electrical operating conditions of the beverage processing equipment.

[0043] In some embodiments, the electrical signal data may be collected by a current sensor of a beverage processing device, wherein the voltage, power and other data in the electrical signal data may be calculated by the processor 130 by using the resistance and current of a known circuit in combination with Ohm's law.

[0044] Vibration data refers to vibration information related to the beverage processing equipment. For example, the vibration data includes vibration amplitude, frequency, and vibration period, etc. In some embodiments, the vibration data can reflect the vibration operation status of the beverage processing equipment.

[0045] In some embodiments, the processor 130 may collect vibration data via a vibration sensor.

[0046] Sound data refers to sound information related to the beverage processing equipment. For example, the sound data includes the loudness, frequency, and pitch of the sound. In some embodiments, the processor 130 can collect the sound data through a sound sensor.

[0047] In some embodiments, changes in the sound data may reflect the operating conditions of the beverage processing equipment, for example, abnormal noise may mean that mechanical parts are worn or stuck.

[0048] In some embodiments, the processor may combine different modal data and represent them in sequence or vector form to form multimodal data. For example, the multimodal data may be represented as {I, V, P, f, N...}, where I represents current, V represents voltage, P represents power, f represents vibration frequency, N represents loudness of sound, etc., or other data related to beverage processing equipment.

[0049] In some embodiments, the sensor continuously collects data during the process of the beverage processing device making the target beverage. Since the data may fluctuate during operation (for example, the current may slightly change due to the workload when the beverage processing device is running), the processor 130 can calculate the average value or mode of each modal data and use the average value or mode or modal data at multiple time points as one of the multi-modal data to reflect the state of the beverage processing device during the process of making the target beverage.

[0050] In some embodiments, the multimodal data includes actual modal data and reference modal data. The actual modal data refers to multimodal data collected by the beverage processing equipment when the beverage processing equipment actually produces the target beverage. For example, the actual modal data can be represented as {I a 、V a , P a 、f a 、N a …}. Among them, I a It is expressed as the current of beverage processing equipment during actual operation, V a , P a 、f a 、N a ,...and I a Same reason.

[0051] Reference modal data refers to preset standard multimodal data. For example, reference modal data can be represented as {I b 、V b , P b 、f b 、N b …}. Among them, I b Expressed as a preset standard current, V b , P b 、f b 、N b ,...and I b Same reason.

[0052] In some embodiments, the reference modality data and the actual modality data are in the same order, ensuring that the processor can accurately compare the reference modality data with the actual modality data.

[0053] In some embodiments, the processor may determine the reference modal data by querying a first preset comparison table based on the target order information. The first preset comparison table may include a correspondence between the target beverage information and the reference modal data. The first preset comparison table may be pre-constructed based on historical data or prior knowledge. In some embodiments, the processor 130 and / or a technician may construct the first preset comparison table based on multi-modal data when the beverage processing equipment is working normally.

[0054] The matching degree refers to the matching degree between the target order information and the multimodal data. In some embodiments, the matching degree can be represented by a numerical value within a preset range (e.g., within 0-1). The matching degree can also be represented in any feasible way.

[0055] In some embodiments, the matching degree includes an initial matching degree between the actual multimodal data and the reference multimodal data, wherein the initial matching degree refers to a matching degree initially determined by the processor 130 .

[0056] In some embodiments, the processor 130 may determine the degree of matching in a variety of ways. For example, the processor 130 may determine the reference multimodal data by querying the first preset comparison table based on the target order information, and determine the similarity between the actual multimodal data and the reference multimodal data, and determine the similarity as the initial degree of matching. The similarity may be represented by cosine similarity, vector distance, etc.

[0057] In some embodiments, the higher the similarity, the higher the initial matching degree, indicating that the actual multimodal data is more correlated with the reference multimodal data.

[0058] In some embodiments, the processor may determine the raw material consistency data based on the execution information of the beverage processing equipment and the target order information; and determine the degree of matching based on the raw material consistency data, the target order information and the multimodal data.

[0059] Raw materials refer to ingredients or seasonings used to prepare drinks. For example, raw materials may include different types of coffee beans, tea leaves, juice, sugar, milk, cream or other seasonings.

[0060] In some embodiments, the raw materials include execution raw materials and target raw materials. The execution raw materials refer to the raw materials added by the beverage processing equipment when the target beverage is actually made. For example, taking a coffee machine as an example, the execution raw materials may include the types of raw materials added when the coffee is actually made. The target raw materials refer to the standard raw materials preset for the target beverage. For example, the target raw materials may include the types of standard raw materials preset for the target beverage.

[0061] The execution information refers to information related to the execution raw materials. For example, taking a coffee machine as an example, the execution information may include one or more groups of execution raw materials, the amount of execution raw materials added, and the order of adding the execution raw materials.

[0062] In some embodiments, the processor 130 may determine the execution information based on the electrical signals in the beverage processing equipment.

[0063] For example, taking a coffee machine as an example, different coffee bean cans correspond to different types of raw materials. When the valve of a coffee bean can is opened, the valve sensor will detect the opening operation and opening duration of the valve and generate a corresponding electrical signal. The electrical signal will be transmitted to the processor 130 through the signal transmission path. The processor 130 can receive the signal and identify the corresponding execution raw material and the amount of execution raw material added. The processor can obtain the order of adding execution raw materials, etc. through different signals received multiple times in succession.

[0064] In some embodiments, the execution information may be sequence data consisting of a plurality of execution materials and the amount of the execution materials added. For example, the execution information may be represented as {x1, x2, x3…x i}. Where x represents a certain execution raw material and the amount of the execution raw material added, x i It can be a sequence element containing the i-th sequential execution material x and the added amount of the execution material, for example, x i It may include 2 sub-elements, etc. The execution raw material data sequence is the sequence in which the raw materials are added when the beverage processing device actually makes the target beverage. The sequence in which the raw materials are added may be determined by the raw materials sequentially acquired by the processor 130.

[0065] The target raw material information refers to information related to the target raw materials of the target beverage. For example, the target raw material information may include one or more groups of target raw materials, standard addition amounts of the target raw materials, and standard order of adding the target raw materials.

[0066] In some embodiments, the processor can determine the target raw material information in a variety of ways. For example, the processor can determine the target raw material information based on the target beverage information in the target order information by querying the second preset comparison table. Among them, the second preset comparison table may include the correspondence between each group of target beverage information and the target raw material and the standard addition amount of the target raw material. The second preset comparison table can be pre-constructed based on historical data or prior knowledge. For example, if the target order information is a cup of low-sugar latte, the processor can query the second preset comparison table to identify the type of coffee beans, standard addition amount and standard order corresponding to the low-sugar latte.

[0067] In some embodiments, the target raw material information may be sequence data consisting of a plurality of target raw materials and standard addition amounts of the target raw materials. For example, the target raw material information may be represented as {y1, y2, y3…y i}. Where y represents a target raw material and the standard addition amount of the target raw material, y i It can be a sequence element containing the i-th target raw material y and the standard addition amount of the target raw material, for example, y i It may include 2 sub-elements, etc. The target raw material data sequence is a preset standard sequence of adding raw materials, which may be set by default by the processor and / or pre-set by a technician according to requirements.

[0068] Raw material consistency data refers to data reflecting the consistency between the target raw material and the standard addition amount in the target raw material information and the execution raw material and the addition amount in the execution information of the beverage processing equipment. Raw material consistency data can reflect the matching degree between the execution raw material and the addition amount corresponding to the target order and the target raw material and the standard addition amount in the target raw material information.

[0069] In some embodiments, the raw material consistency data may be determined in a variety of ways. For example, the processor 130 may determine the raw material consistency data by the ratio between the total number of matching items and the total number of items to be matched.

[0070] A matching item refers to a sequence element in the execution information that is identical to the target raw material information. For example, for each item x in the sequence data i And the order corresponding to y i If the execution raw material of the i-th order is the same as the target raw material, and the amount of the execution raw material added is equal to the standard amount of the target raw material added, then the item x i The total number of matching items is the number of all items that meet the above requirements, where the execution raw materials are the same as the target raw materials and the addition amount is equal to the standard addition amount.

[0071] The item to be matched refers to each item in the target raw material information. For example, the target raw material information {y1, y2, y3…y i} in, y i It can also be expressed as the i-th order of items to be matched. The total number of items to be matched is the number of items in the target raw material information. i The total number of items.

[0072] In some embodiments, the processor 130 may obtain the matching degree in a variety of ways. For example, the processor 130 may obtain the matching degree based on the initial matching degree and the raw material consistency data. Exemplarily, the matching degree is positively correlated with the initial matching degree and the raw material consistency data.

[0073] In some embodiments, the processor 130 may perform a weighted combination of the raw material consistency data and the initial matching degree based on the initial matching degree, the raw material consistency data and a preset weight coefficient, and calculate the matching degree using the following formula (1). M = K1 × C + K2 × I (1) Wherein, M is the matching degree, C is the raw material consistency data, I is the initial matching degree, K1 and K2 are preset weight coefficients, K1 represents the influence weight of the raw material consistency data in the calculation process, and K2 represents the influence weight of the initial matching degree. The value of the preset weight coefficient can be set by the processor by default and / or pre-set by the technician according to the requirements. For example, the values ​​of K1 and K2 can be 0.7 and 0.3 respectively, or the values ​​of K1 and K2 can be 0.5.

[0074] In some embodiments of the present specification, the raw material consistency data can be used to determine whether the raw materials added by the beverage processing equipment during the actual production of the target beverage are consistent with the raw materials in the target order information, further improving the accuracy of the determined matching degree.

[0075] Step 320 , in response to the matching degree satisfying the matching condition, determining matching fault information based on the multimodal data and the target order information.

[0076] The matching condition refers to a condition that must be met to determine a matching failure. In some embodiments, the matching condition can be preset. In some embodiments, the matching condition is that the matching degree is lower than a first preset threshold. The first preset threshold is negatively correlated to the usage time of the beverage processing equipment, and the first preset threshold is not lower than a preset minimum limit.

[0077] The first preset threshold refers to the maximum preset threshold of the matching failure. In some embodiments, the first preset threshold can be set by the processor by default and / or pre-set by the technician according to the requirements. For example, the first preset threshold can be 0.8, etc. When the matching degree is less than 0.8, the processor needs to determine a matching failure.

[0078] The minimum limit refers to the lowest threshold of the first preset threshold. In some embodiments, the minimum limit can be set by default by the processor and / or pre-set by a technician according to requirements.

[0079] The usage time refers to the cumulative time of the beverage processing equipment from the date when it was first put into use to the current time. In some embodiments, the longer the usage time of the beverage processing equipment is, the smaller the first preset threshold is. Among them, the first use time and the current time of the beverage processing equipment can be obtained through the clock module. The clock module refers to a module for providing time and timing functions. In some embodiments, the clock module can be integrated in the processor 130.

[0080] In some embodiments of the present specification, after long-term use of beverage processing equipment, mechanical parts may be worn out, precision may be reduced, and other problems may occur, and thus the old equipment may not be able to maintain the same precision as the new equipment. By using different first preset thresholds (for example, the first preset threshold corresponding to the new equipment is greater than the first preset threshold corresponding to the old equipment), the performance of equipment with different usage times can be adapted, and the rationality of the determined matching conditions can be further improved. By setting the minimum preset limit, it is possible to prevent potential equipment problems from being missed due to the first preset threshold being too low, thereby ensuring that matching failures can still be correctly identified.

[0081] Matching failure information refers to relevant information about the reason why the actual multimodal data and the target order information do not match. For example, the matching failure information includes at least one of an execution order error, a machine failure, a raw material error, a working parameter setting error, and the like.

[0082] In some embodiments, in response to the matching degree satisfying the matching condition, the processor can determine the matching fault information in a variety of ways. For example, the processor 130 can determine the reference modality data based on the target order information. For more information, see Figure 3The relevant description of step 310 is shown in the figure. The processor can determine the target difference sequence based on the actual multimodal data and the reference multimodal data; the processor 130 can determine the reference matching fault information corresponding to the reference difference sequence with the highest similarity as the matching fault information by querying the third preset comparison table based on the target difference sequence and determining the similarity between the target difference sequence and the reference difference sequence in the table. Among them, the third preset comparison table may include the correspondence between each group of reference difference sequences and the reference matching fault information. The third preset comparison table can be pre-constructed based on historical data or prior knowledge. The similarity can be represented by cosine similarity, vector distance, etc.

[0083] The target difference sequence refers to the relative deviation sequence between the actual multimodal data and the reference multimodal data. The target difference sequence can be used to characterize the degree of deviation between the actual state of the beverage processing equipment and the preset standard state during operation.

[0084] In some embodiments, the processor 130 may determine a target difference sequence based on the actual multimodal data and the reference multimodal data. For example, the actual multimodal data may be represented as {I a 、V a , P a 、f a 、N a …}, the reference multimodal data can be expressed as {I b 、V b , P b 、f b 、N b …}, then the target difference sequence can be expressed as in, It is expressed as the difference between the current of the beverage processing equipment during actual operation and the preset standard current. …and Same reason.

[0085] The reference difference value sequence refers to the standard difference value sequence pre-stored in the third preset comparison table.

[0086] Reference matching fault information refers to known matching fault information corresponding to the reference difference sequence in the third preset comparison table. In some embodiments, the processor and / or technicians can obtain a large number of reference difference sequences and reference matching fault information based on historical data or prior knowledge to construct the third preset comparison table.

[0087] In some embodiments, the processor 130 may also determine the matching fault information in a variety of ways based on the raw material consistency data, the multimodal data, and the target order information. Figure 3For example, the processor 130 may determine the reference modal data based on the target order information. The processor 130 may determine the target difference sequence in the above manner based on the reference modal data and the actual multimodal data. The processor may determine the matching fault information by querying a third preset comparison table based on the raw material consistency data and the target difference sequence.

[0088] The processor 130 may add a column of raw material consistency data to the third preset comparison table, and the raw material consistency data may reflect the corresponding relationship between each group of raw material consistency data, the target difference sequence and the matching fault information (such as incorrect raw material input, etc.). The third preset comparison table may be pre-constructed based on historical data or prior knowledge.

[0089] In some embodiments of the present specification, by comprehensively analyzing multimodal data, raw material consistency data and target orders, the processor can more accurately identify and locate possible causes of failure of beverage processing equipment (for example, whether the cause of the failure is raw material error, etc.).

[0090] In some embodiments, the processor may also determine the idle time difference based on the start time and the order time; determine the expected working time based on the target order information; determine the execution time difference based on the expected working time and the actual working time; determine the matching fault information based on the idle time difference, the execution time difference, the multimodal data and the target order information. For more information, please refer to Figure 4 The processor 130 may also determine the matching fault information in any other feasible manner.

[0091] Step 330: Determine a fault handling measure based on the matching fault information.

[0092] Fault handling measures refer to handling measures taken to correct matching fault information. For example, fault handling measures may include at least one of overhauling beverage processing equipment, re-introducing raw materials, restarting beverage processing equipment, or notifying manual intervention.

[0093] In some embodiments, the processor 130 may determine the fault handling measures by a variety of methods. For example, the processor 130 may determine the fault handling measures based on a preset correspondence between matching fault information and fault handling measures. The preset correspondence refers to a correspondence between matching fault information and fault handling measures. The preset correspondence may be set by default by the processor and / or pre-built by a technician based on historical data or prior knowledge.

[0094] Exemplarily, if the matching fault information is a motor fault, the processor 130 may select a fault handling measure according to a preset corresponding relationship, such as notifying a technician to perform machine maintenance.

[0095] In some embodiments, the multimodal data further includes grinding image data. The processor 130 may also determine current product information based on the multimodal data and execution information of the beverage processing equipment, and determine fault handling measures based on the current product information, target order information and matching fault information.

[0096] The grinding image data refers to image data that can reflect the grinding result of the beverage processing equipment. In some embodiments, when the beverage processing equipment completes the grinding, the processor 130 can obtain the grinding image data by photographing the grinding result through the camera module.

[0097] The camera module refers to a camera device installed in the beverage processing equipment. For example, the camera module includes an image sensor, a lens, a processing unit, etc. The camera module can be used to capture the grinding results and transmit high-quality grinding image data to the processor 130.

[0098] In some embodiments, the camera module may be integrated in multiple locations of the beverage processing equipment, such as above the collection container of the beverage processing equipment, near the grinding disc of the beverage processing equipment, and other locations.

[0099] For example, taking a coffee machine as an example, the camera module can be integrated into a position of the coffee machine close to the grinding disk. When the coffee machine completes grinding, the camera module can capture an image of the grinding result.

[0100] The current product refers to the beverage currently being produced by the beverage processing equipment.

[0101] The current product information refers to information related to the current product. For example, the current product information may include at least one of the type, sugar content, ice content, grinding degree, etc. of the currently produced beverage.

[0102] In some embodiments, the processor 130 may determine the current product information based on the multimodal data and the execution information through a plurality of preset algorithms. The preset algorithms may be set by default by the processor and / or pre-set by a technician according to requirements. For example, the preset algorithms may include a matching algorithm and a grinding degree confirmation algorithm.

[0103] The matching algorithm refers to an algorithm used by the processor 130 to determine the type, sugar content, ice content, etc. in the current product information. For example, when the coffee machine actually makes coffee, the execution information is Arabica beans, the amount of sugar added, the amount of ice cubes, etc. After the coffee is made, the processor 130 can extract the current product information as Arabica beans, the amount of added sugar, the amount of ice cubes, etc. from the execution information through the matching algorithm.

[0104] The grinding degree confirmation algorithm refers to an algorithm used by the processor 130 to determine the grinding degree in the current product information. In some embodiments, the processor can determine the grinding degree through the following steps S1-S3 according to the grinding degree confirmation algorithm.

[0105] Step S1 : The processor 130 may determine a first grinding degree based on the grinding image data by using a preset image recognition algorithm.

[0106] The first grinding degree refers to a preliminary grinding degree obtained by a preset image recognition algorithm.

[0107] The preset image recognition algorithm may be set by default by the processor and / or pre-set by a technician according to requirements. For example, the preset image recognition algorithm may include an edge detection algorithm (EDA), an image segmentation algorithm (ISA), and the like.

[0108] Step S2, the processor 130 can determine the second grinding degree based on other multimodal data (such as current, vibration, rotation speed, etc.) and execution information other than the grinding image data through a fourth preset comparison table.

[0109] The second grinding degree refers to the grinding degree further determined by the fourth preset comparison table.

[0110] In some embodiments, the fourth preset comparison table may include other multimodal data, execution information and the correspondence between the grinding degree in addition to each set of grinding image data. The fourth preset comparison table may be pre-constructed based on historical data or prior knowledge.

[0111] In step S3, the processor 130 may obtain a matching degree based on the first grinding degree and the second grinding degree.

[0112] In some embodiments, the grinding degree is positively correlated with the first grinding degree and the second grinding degree. The processor may weight the first grinding degree and the second grinding degree according to the first grinding degree, the second grinding degree and a preset weight coefficient, and calculate the grinding degree by the following formula (2). G=t1×G1+t2×G2 (2) Wherein, G is the grinding degree, G1 is the first grinding degree, G2 is the second grinding degree, t1 and t2 are preset weight coefficients, t1 represents the influence weight of the first grinding degree in the calculation process, and t2 represents the influence weight of the second grinding degree. The value of the preset weight coefficient can be set by the processor by default and / or pre-set by the technician according to the requirements. For example, the values ​​of t1 and t2 can be 0.7 and 0.3 respectively, or the values ​​of t1 and t2 can be both 0.5.

[0113] In some embodiments, the processor 130 may determine whether to abandon the process of modifying the current product so that the modified product meets the requirements of the target order based on the raw material consistency data and the second preset threshold value; in response to the fact that the current product can be modified so that the modified product meets the requirements of the target order, the processor may determine that the fault handling measure is to modify the current product so that the modified product meets the requirements of the target order. The above operation of "abandoning the process of modifying the current product so that the modified product meets the requirements of the target order" is referred to as "temporarily abandoning the current product". The above operation of "the current product can be modified so that the modified product meets the requirements of the target order" is referred to as "not abandoning the current product".

[0114] The second preset threshold refers to a threshold used to determine whether to temporarily abandon the current product. In some embodiments, the second preset threshold can be set by default by the processor and / or pre-set by a technician according to demand. For example, the second preset threshold can be 0.5, etc.

[0115] In some embodiments, the processor or technician can determine whether to temporarily abandon the current product in a variety of ways. For example, in response to the raw material consistency data being less than the second preset threshold, the current product can be temporarily abandoned. After temporarily abandoning the current product, the processor can determine the product handling measures for the current product based on the pending orders and the current product information. For more information, see Figure 3 The following instructions.

[0116] In some embodiments, the processor may also determine whether the grinding degree meets the preset requirements based on the grinding degree in the current product information, and determine whether to temporarily abandon the current product. For example, in response to the grinding degree not meeting the preset requirements, the current product is temporarily abandoned; in response to the grinding degree meeting the preset requirements, the current product may be chosen not to be abandoned. Among them, meeting the preset requirements may be that the grinding degree in the current product information is consistent with the preset grinding degree. The preset grinding degree may be set by default by the processor or by a technician according to demand, for example, the preset grinding degree includes medium grinding degree, etc.

[0117] Taking coffee as an example, a medium grind size means that the coffee powder has a medium particle size. A medium grind size is similar to coarse sand particles. For example, a medium grind size can be 500 microns to 800 microns, etc.

[0118] In some embodiments, taking coffee as an example, after the coffee is processed and produced, since the grinding degree directly affects the final taste of the coffee, too large a grinding particle size will result in insufficient extraction of the coffee during the brewing process, resulting in a significant deviation in the taste of the finished coffee, which cannot meet the expected quality requirements. And after the coffee is processed and produced, the grinding degree of the coffee beans is no longer adjustable, so if the grinding degree in the current product information does not meet the preset requirements, it cannot be corrected through subsequent fault handling measures. By judging the grinding degree, the current product information with excessive grinding can be identified in a timely manner, and the current product can be discarded to avoid wasting resources in subsequent processes.

[0119] In some embodiments of the present specification, by comparing the raw material consistency data with the second preset threshold, the processor can more accurately determine whether the current product meets the quality standard. If the raw material consistency data is lower than the second preset threshold (such as 50%), the current product is temporarily abandoned. This ensures that the current product that does not meet the requirements will no longer continue to determine whether it can obtain the target order through correction, thereby improving the processing efficiency of the processor. It should be noted that comparing the raw material consistency data with the second preset threshold to determine whether to temporarily abandon the current product can also be an optional step, that is, regardless of the size of the raw material difference, the processor can determine whether to temporarily abandon the current product.

[0120] In some embodiments, in response to the grinding degree meeting the preset requirement, the processor may determine whether to not abandon the current product based on the current product information and the target order information.

[0121] In some embodiments, the processor can determine whether multiple groups of first preset conditions are met based on the current product information and the target order information, and determine whether the current product is not abandoned. The first preset conditions include: the beverage type of the current product information is the same as the beverage type of the target order information; the ice amount of the current product information is less than or equal to the ice amount in the target order information; the amount of sugar added in the current product information is less than or equal to the amount of sugar added in the target order information, etc. If any of the above first preset conditions is not met, the current product is temporarily abandoned.

[0122] In some embodiments, in response to temporarily abandoning the current product, the processor may determine a product handling measure for the current product based on the pending order and the current product information. For more information, see Figure 3 The following instructions.

[0123] In some embodiments, in response to temporarily abandoning the current product, the processor 130 may enable the beverage processing equipment to re-prepare the target beverage based on the target order information.

[0124] In some embodiments, in response to not abandoning the current product, the processor may determine, based on the current product information, the target order information, and the matching fault information, that the fault handling measure is to modify the current product so that the modified product meets the requirements of the target order. The modification includes adding sugar or ice to the current product.

[0125] In some embodiments, when making adjustments such as adding ice or sugar, the processor 130 can determine the standard amount of ice and sugar added for the target order based on the target order by querying a preset standard table. The processor calculates the adjustment amount required to correct the current product to the product corresponding to the target order based on the difference between the amount of ice and sugar added corresponding to the current product information and the standard amount of ice and sugar added corresponding to the target order. Among them, the preset standard table may include each amount of ice (such as less ice, normal ice, etc.) and standard amount of sugar added (such as less sugar, normal sugar, etc.) for different orders. The preset standard table can be pre-constructed based on historical data or prior knowledge. For more information on how to adjust the current product, please refer to Figure 3 The following instructions.

[0126] In some embodiments of the present specification, by combining multimodal data (such as grinding image data) with execution information, the processor can be independent of single multimodal data such as vibration data or current data, and can further comprehensively analyze potential problems of the beverage processing equipment during operation. By providing multiple fault handling options, the flexibility of the fault handling measures is enhanced, making the fault handling measures more targeted, reducing the situation of completely discarding the current product, and ensuring the quality of the beverages actually made by the beverage processing equipment.

[0127] In some embodiments, the fault handling measures include product handling measures. The processor can determine the product handling measures based on the pending orders and current product information.

[0128] Product handling measures refer to the handling measures of the current product. For example, the product handling measures include adjusting the current product information or discarding the current product.

[0129] A pending order refers to an order related to a pending drink. For example, a pending order may include the order time of the pending drink and the pending drink information. In some embodiments, the pending order may be one or more. Among them, a pending drink refers to a drink that the drink processing equipment has not yet started to make.

[0130] In some embodiments, the processor may obtain and store pending orders through the order management module.

[0131] The order management module refers to a module for managing pending orders. For example, the order management module includes a Shopify order management module (Shopify Order Management System), a SAP order management module (SAP Order Management System, SAP OMS), an Oracle order management module (Oracle Order Management System, Oracle OMS), etc. In some embodiments, the order management module can be integrated into the processor 130.

[0132] The pending beverage information refers to parameters related to the pending beverage. For example, the pending beverage information includes the type of the pending beverage, the amount of raw materials added, the amount of ice, and other parameters related to the pending beverage. In some embodiments, the pending beverage information can be generated based on the collected customer needs. For example, the customer can input the preferred beverage parameters through the user terminal 150 and store them in the order management module. The processor 130 can obtain the beverage parameters and order time in the order management module through the network 120 and determine them as the pending beverage information.

[0133] In some embodiments, the processor may determine the product processing measures in a variety of ways. For example, the processor may filter out the matching orders from the pending orders based on the pending beverage information in the pending orders, and determine the product processing measures based on the matching orders.

[0134] A matchable order refers to an order whose current product information matches an order to be processed. In some embodiments, the matchable order includes a first matching order, a second matching order, and a target matching order.

[0135] The first matching order refers to the matching order initially screened by the processor.

[0136] In some embodiments, the processor can filter out orders with the same beverage type as the current product information in the beverage information to be processed according to the first matching rule, and determine them as the first matching orders. The first matching rule refers to a rule for matching based on the same beverage type. The first matching rule can be set by default by the processor or pre-set by a technician based on prior experience.

[0137] In some embodiments, in response to the absence of a first matching order, determining the product handling action is to discard the current product.

[0138] The second matching order refers to a matching order further screened out by the processor 130 based on the first matching order.

[0139] In some embodiments, the processor 130 can screen out orders whose ice amount, sugar amount, etc. in the beverage information to be processed are not less than the corresponding ice amount, sugar amount, etc. in the current product information based on the first matching order through the second matching rule, and determine them as the second matching order. The second matching rule refers to a rule that further matches the ice amount and sugar amount, etc. in the first matching order to be not less than the current product information. The second matching rule can be set by default by the processor or pre-set by a technician based on prior experience.

[0140] In some embodiments, in response to the absence of a second matching order, the product handling measure is determined to be discarding the current product. The target matching order refers to a matching order that is finally determined.

[0141] In some embodiments, the processor 130 may obtain the order with the earliest order placement time in the second matching orders based on the second matching orders, and determine it as the target matching order.

[0142] In some embodiments, in response to the current product being modified so that the modified product meets the requirements of the target matching order, the processor may determine, based on the to-be-processed beverage information in the target matching order, that the product processing measure is to modify the current product so that the modified product meets the requirements of the target matching order. The modification includes adjusting the current product such as adding sugar or ice.

[0143] In some embodiments, when making adjustments such as adding ice or adding sugar, the processor 130 may determine the amount of adjustment required to correct the current product to the product corresponding to the target matching order based on the pending beverage information in the target matching order. For example, the processor 130 may determine the standard amount of ice and sugar added to the target matching order by querying a preset standard table based on the target matching order. The processor may calculate the amount of adjustment required to correct the current product information to the product corresponding to the target matching order based on the difference between the amount of ice and sugar added corresponding to the current product information and the standard amount of ice and sugar added corresponding to the target matching order.

[0144] For example, the amount of sugar added in the current product information is less sugar (10 grams), while the standard amount of sugar added in the target matching order is normal sugar (40 grams). The amount of sugar added in the current product information is less than the standard amount of sugar added in the target matching order. The processor can calculate that the amount of sugar adjustment for the current product to be corrected to the product corresponding to the target matching order is to increase 30 grams of sugar. The adjustment of adding ice, etc. is the same as the adjustment of adding sugar.

[0145] In some embodiments, the processor 130 may estimate future orders in a future time period; and determine product handling measures based on pending orders, future orders, and current product information.

[0146] Future orders refer to pending orders that may appear in a future time period. The length of the future time period can be preset, for example, the future time period is 0.5 or 1 hour from the current time point. In some embodiments, the number of future orders can be one or more.

[0147] In some embodiments, the processor 130 can estimate future orders in a variety of ways. For example, the processor 130 can determine a historical order period. The historical order period refers to a historical time period that is the same as the future time period. For example, the future time period is from 1:00 p.m. to 2:00 p.m., and the time between 1:00 p.m. and 2:00 p.m. every day in the past preset time is a historical order period. The length of the past preset time can be preset according to actual needs, for example, within the past month, etc.

[0148] In some embodiments, the processor 130 may use multiple historical orders in multiple historical order periods within a preset time in the past as multiple target historical orders. The processor 130 may calculate the average number of orders corresponding to the historical order period based on the number of orders in the multiple historical order periods, and use the average value as the number N of future orders. There may be multiple identical target historical orders in multiple target historical orders. The identical target historical order means that the beverage information corresponding to the target historical order is the same. The beverage information may include the type of beverage, sugar content, ice amount, etc. The processor 130 may classify the identical target historical orders in multiple target historical orders, classify the identical target historical orders into one category, and determine the number of target historical orders in each category as the number of orders for the target historical orders of that category. The processor may use the target historical orders ranked in the top N by the number of orders as future orders.

[0149] In some embodiments, the processor 130 may predict future orders through an order prediction model based on historical order information, historical external information, and future external information. For more information, see Figure 5 and related instructions.

[0150] In some embodiments, the processor 130 can filter out future orders with the same beverage type as the current product information according to the first matching rule and determine them as first matching orders. For details about the first matching rule and the first matching order, please refer to Figure 3 The description of step 330 above will not be repeated here.

[0151] In some embodiments, in response to the absence of a first matching order, determining the product handling action is to discard the current product.

[0152] In some embodiments, the processor 130 can filter out future orders whose ice volume, sugar content, etc. are not less than the current product information based on the first matching order and through the second matching rule, and determine them as second matching orders. For details about the second matching rule and the second matching order, please refer to Figure 3 The description of step 330 above will not be repeated here.

[0153] In some embodiments, in response to the absence of a second matching order, determining the product handling action is to discard the current product.

[0154] In some embodiments, in response to the existence of the second matching order, the product handling measure is determined to be to retain the current product until the predicted future order is converted into an actual pending order, and the product handling measure is determined to be to adjust the current product information. The adjustment includes adding sugar or ice to the current product. Retaining the current product includes refrigerating the current product.

[0155] In some embodiments of this specification, by considering future orders, the processor can take into account possible future order requirements when making product processing measures for current product information, thereby effectively reducing the waste of current products, optimizing resource utilization, and improving the production efficiency and order fulfillment rate of beverage processing equipment.

[0156] In some embodiments, the processor may determine a target matching order based on future orders and current product information; determine a hold time based on the target matching order; determine a future flavor sequence based on current product information and the hold time; and determine a product handling measure based on the future flavor sequence and the current flavor sequence. For more information, see Figure 5 and related instructions.

[0157] In some embodiments of this specification, by screening pending orders and adjusting the current product information, it is ensured that even if the current product cannot be used for the target order, there is still a chance to adjust it by matching the pending orders, rather than directly discarding the current product. This optimizes resource use and improves the production efficiency of beverage processing equipment.

[0158] In some embodiments of the present specification, the introduction of matching degree enables the processor to accurately determine whether the beverages actually produced by the beverage processing equipment meet the requirements of the target order information, ensuring that each target order is produced according to the customer's customized requirements, thereby improving the accuracy of order execution; by matching fault information, the processor can quickly execute corresponding processing measures and select more efficient fault handling measures, thereby ensuring the quality and efficiency of beverage production.

[0159] It should be noted that the above description of process 300 is only for illustration and description, and does not limit the scope of application of this specification. For those skilled in the art, various modifications and changes can be made to the process hand-eye calibration under the guidance of this specification. However, these modifications and changes are still within the scope of this specification.

[0160] Figure 4 FIG. 4 is an exemplary flow chart of determining matching fault information according to some embodiments of the present specification. In some embodiments, process 400 may be executed by processor 130 .

[0161] In some embodiments, the execution information may include the start working time and the actual working time of the beverage processing equipment corresponding to the target order information. The processor 130 may determine the idle time difference based on the start working time and the order time; determine the expected working time based on the target order information; determine the execution time difference based on the expected working time and the actual working time; and determine the matching fault information based on the idle time difference, the execution time difference, the multimodal data, and the target order information.

[0162] For more information about beverage processing equipment, please refer to Figure 1 For more information about target order information, matching failure information, and multimodal data, see Figure 3 and its related description.

[0163] Step 410, determining the idle time difference based on the start working time and the order placement time.

[0164] The start working time refers to the time point when the beverage processing device (such as a coffee machine, etc.) executes the target order. In some embodiments, the start working time can be obtained by the time when the processor receives the instruction signal sent by the beverage processing device (such as a coffee machine, etc.) to start processing the target order.

[0165] The actual working time refers to the length of time that the beverage processing equipment (such as a coffee machine, etc.) actually takes to complete the target order. In some embodiments, the actual working time can be obtained by detecting the time difference between the beverage processing equipment (such as a coffee machine, etc.) sending a command signal to start processing and obtaining a processing completion signal.

[0166] The order time refers to the time point when the beverage processing equipment (such as a coffee machine, etc.) receives the target order. In some embodiments, the order time can be obtained by the user terminal 150 that generates the target order.

[0167] The idle time difference refers to the length of time required for a beverage processing device (such as a coffee machine, etc.) to switch from a non-working state to a working state. In some embodiments, the idle time difference may be the length of time between a beverage processing device (such as a coffee machine, etc.) receiving a target order and executing the target order. In some embodiments, the idle time difference does not include the queuing time of the target order in the beverage processing device (such as a coffee machine, etc.).

[0168] In some embodiments, the idle time difference can be determined based on the start working time and the order placement time. For example, the idle time difference can be the absolute difference between the order placement time and the start working time. If the beverage processing equipment is processing a queued order, the idle time difference can be the absolute difference between the order placement time and the start working time minus the queue time for processing the queued order. The queued order is an order that is placed before the target order and needs to be processed by the beverage processing equipment. The number of queued orders is at least one. The processor can determine the queue time of the queued order by the above-mentioned method of determining the actual working time. Through the above-mentioned determination method, the idle time difference can not include the queue time of the target order due to the busy beverage processing equipment.

[0169] Step 420, determine the estimated working time based on the target order information.

[0170] The estimated working time refers to the estimated length of time required for beverage processing equipment (such as coffee machines, etc.) to complete a target order. In some embodiments, the estimated working time can be determined based on the target order information. For example, taking the target beverage coffee as an example, the estimated working time can be determined by querying the fifth preset comparison table based on the coffee type in the target order information. The fifth preset comparison table refers to a table that records the mapping relationship between coffee types and estimated working time. The fifth preset comparison table can be constructed based on the coffee types in the historical coffee making order information and the actual working time. For more information about target order information, please see Figure 3 and its related description.

[0171] Step 430, determining the execution time difference based on the expected working time and the actual working time.

[0172] The execution time difference refers to the deviation between the time actually taken by the beverage processing equipment (such as a coffee machine, etc.) to complete the target order and the time expected to complete the target order. In some embodiments, the execution time difference can be used to determine whether the beverage processing equipment (such as a coffee machine, etc.) has been stuck, thereby facilitating more accurate determination of the fault information of the beverage processing equipment (such as a coffee machine, etc.).

[0173] In some embodiments, the execution time difference can be determined based on the expected working time and the actual working time. For example, the execution time difference can be the time difference between the expected working time and the actual working time.

[0174] Step 440, determining matching fault information based on the idle time difference, the execution time difference, the multimodal data and the target order information.

[0175] In some embodiments, matching fault information can be determined based on idle time difference, execution time difference, multimodal data, and target order information. For example, processor 130 can add two columns of idle time difference and execution time difference on the basis of the third preset comparison table to form a mapping relationship between matching fault information and idle time difference, execution time difference, multimodal data, and target order information. Processor 130 can query the third preset comparison table to determine matching fault information based on idle time difference, execution time difference, multimodal data, and target order information. For more information about the third preset comparison table, please refer to Figure 3 and its related description.

[0176] In some embodiments, the execution information also includes the unit execution time of the execution step. The processor 130 can also determine the unit estimated time of the execution step based on the target order information; determine the time difference sequence based on the unit execution time and the unit estimated time; and determine the matching fault information based on the idle time difference, the time difference sequence, the multimodal data, and the target order information.

[0177] The execution steps refer to sequence data consisting of the steps required for the beverage processing equipment (such as a coffee machine, etc.) to complete the target order. For example, taking coffee as an example, the execution steps may include steps such as adding, grinding, and stirring raw materials. Exemplarily, the steps required to complete the target order are, in sequence, adding raw material A, grinding, adding raw material B, adding raw material C, and stirring. Steps 1 to 5 of the execution steps may be expressed as (addition of raw material A, grinding, addition of raw material B, addition of raw material C, and stirring). In some embodiments, the execution steps may be steps that the user has set in advance for the beverage processing equipment (such as a coffee machine, etc.) under each order information based on operating experience.

[0178] The estimated unit duration refers to the sequence data of the estimated length of time required to complete the execution of the execution step. The estimated unit durations corresponding to different execution steps can be the same or different. For example, the estimated unit duration (2, 180, 4, 5, 120) can mean that the estimated execution time of the beverage processing equipment (such as a coffee machine, etc.) to complete the above steps 1 to 5 is 2 seconds, 180 seconds, 4 seconds, 5 seconds, and 120 seconds.

[0179] In some embodiments, the estimated duration of the unit can be determined based on the target order information. For example, the estimated duration of the unit can be determined based on the target order information by querying the sixth preset comparison table. The sixth preset comparison table refers to a table that records the mapping relationship between the target order information and the estimated duration of the unit. The sixth preset comparison table can be obtained through a third-party information source and / or data source. For example, the processor 130 can construct the sixth preset comparison table by obtaining the unit execution time of each execution step under each order information set by the user in advance according to the operating experience for the beverage processing equipment (such as a coffee machine, etc.).

[0180] Unit execution time refers to the sequence data of the actual length of time taken by the beverage processing equipment (such as a coffee machine, etc.) to actually complete the execution steps. For example, the unit estimated time (2, 180, 4, 5, 120) may mean that the actual time taken by the beverage processing equipment (such as a coffee machine, etc.) to complete steps 1 to 5 is 2 seconds, 180 seconds, 4 seconds, 5 seconds, and 120 seconds.

[0181] In some embodiments, the unit execution duration can be obtained by detecting the time difference between the processing start instruction signal and the processing end instruction signal of each unit sent by the beverage processing equipment (such as a coffee machine, etc.).

[0182] The time difference sequence may refer to a sequence of data reflecting the time offset between the actual completion time of the execution steps and the ideal completion time of the execution steps of the beverage processing equipment (such as a coffee machine, etc.) in the process of executing the target order. For example, the time difference sequence (0, 1, 0, 0, 1) may represent that the time deviation values ​​of the beverage processing equipment (such as a coffee machine, etc.) when actually executing steps 1 to 5 are 0 seconds, 1 second, 0 seconds, 0 seconds, and 1 second.

[0183] The time difference sequence can be determined based on the unit execution duration and the unit estimated duration. In some embodiments, the time difference sequence can be determined based on the difference between the unit execution duration and the unit estimated duration. For example, when the unit execution duration is (2, 180, 4, 5, 120) and the unit estimated duration is (2, 179, 4, 5, 119), the time difference sequence is (0, 1, 0, 0, 1).

[0184] In some embodiments, the matching fault information can be determined based on the idle time difference, the execution time difference sequence, the multimodal data, and the target order information. For example, the processor 130 can first add two columns of idle time difference and execution time difference sequence on the basis of the third preset comparison table to form a mapping relationship between the idle time difference, the execution time difference sequence, the multimodal data, the target order information, and the matching fault information. The processor 130 can query the third preset comparison table to determine the matching fault information based on the idle time difference, the execution time difference sequence, the multimodal data, and the target order information.

[0185] In some embodiments of the present specification, the execution time difference sequence is determined by the unit execution time and the unit estimated time, and the process of completing orders of beverage processing equipment (such as coffee machines, etc.) is evaluated at the execution step level. This allows the determined cause of the fault to be specific to each execution step, so that there is more information for reference when determining the fault handling measures later, making the final decision more reasonable.

[0186] In some embodiments of the present specification, by evaluating the delay time between the beverage processing equipment (such as a coffee machine, etc.) receiving an order and executing the order, that is, the execution time difference, it is possible to take into account the failures of the beverage processing equipment (such as a coffee machine, etc.) such as freezes and delays at startup, thereby more accurately determining the matching fault information.

[0187] Figure 5 It is an exemplary schematic diagram of an order prediction model according to some embodiments of the present specification.

[0188] In some embodiments, the processor 130 may predict future orders 530 through an order prediction model 520 based on historical order information 510 - 1 , historical external information 510 - 2 , and future external information 510 - 3 .

[0189] The order prediction model 520 refers to a model for predicting future orders. In some embodiments, the order prediction model 520 can be a machine learning model, for example, one or any combination of a temporal convolutional network (TCN) or a convolutional neural network (CNN).

[0190] In some embodiments, the input of the order prediction model 520 may include historical order information 510 - 1 , historical external information 510 - 2 , and future external information 510 - 3 , and the output of the order prediction model 520 may be future orders 530 .

[0191] Historical order information 510-1 refers to target order information of beverage processing equipment (such as coffee machines, etc.) in the past period of time. For example, historical order information can be target order information of beverage processing equipment (such as coffee machines, etc.) in the past 1 hour, 2 hours, 6 hours, etc. For more information about target order information, please refer to Figure 3 Related description.

[0192] In some embodiments, the storage device 140 may obtain and store the historical order information. The processor 130 may obtain the historical order information 510 - 1 through the storage device 140 .

[0193] External information refers to information obtained from outside the beverage processing equipment (such as a coffee machine, etc.). In some embodiments, the external information may include environmental information and promotional information. Environmental information may be information in the environment in which the beverage processing equipment (such as a coffee machine, etc.) is located. For example, environmental information may be ambient temperature, weather conditions, etc. Promotional information refers to a promotional object under at least one promotional method. The promotional object may be one or more types of coffee, etc. The promotional method may be at least one of poster promotion, software notification, etc. For example, the promotional information may be a poster picture for offline promotion of one or more types of coffee, etc. For another example, the promotional information may be a push text snippet for promoting one or more types of coffee through an online software notification, etc.

[0194] The historical external information 510-2 refers to the external information of the beverage processing equipment (such as a coffee machine, etc.) in the past period of time. The time interval of the past period of time in which the historical external information is located is consistent with the time interval of the past period of time in which the historical order information is located. For example, the historical external information can be the external information of the beverage processing equipment (such as a coffee machine, etc.) in the past week.

[0195] In some embodiments, the historical external information may be obtained through a third-party information source and / or a data source. For example, the historical external information may be obtained by a user entering it through the user terminal 150 .

[0196] The future external information 510 - 3 refers to the external information of the beverage processing equipment (such as a coffee machine, etc.) within a period of time in the future. For example, the future external information may be the external information of the beverage processing equipment (such as a coffee machine, etc.) within a week in the future.

[0197] In some embodiments, the future external information may be obtained through a third-party information source and / or a data source. For example, the future external information may be obtained by a user entering it through the user terminal 150 .

[0198] Future orders 530 refer to order information for beverage processing equipment (such as coffee machines, etc.) in the future and the estimated order placement time.

[0199] In some embodiments, the order prediction model 520 can be trained by multiple labeled first training samples. Each group of first training samples is sample historical order information at the first time, sample historical external information at the first time, and sample historical external information at the second time, and the corresponding label is the sample order corresponding to the second time. The corresponding label can be the order actually generated at the second time. The first time is earlier than the second time, and the second time is the future time of the first time. For example, the processor 130 can use the order information and external information of the first time as the first training sample, and the order information and order generation time of the second time as the label.

[0200] In some embodiments, the processor 130 may input a plurality of first training samples with labels into the initial order prediction model, construct a loss function through the labels and the output results of the initial order prediction model, and iteratively update the parameters of the initial order prediction model based on the loss function by gradient descent or other methods. When the preset conditions are met, the model training is completed and a trained order prediction model is obtained. The preset conditions may be that the loss function converges, the number of iterations reaches a threshold, etc.

[0201] In some embodiments of the present specification, by using a trained order prediction model, future orders can be predicted relatively quickly and accurately, so that there is more data information when determining subsequent processing measures, and more accurate decisions can be made.

[0202] In some embodiments, when there is no target matching order in the pending orders, the processor 130 may determine the target matching order based on the future order and the current product information, determine the holding time based on the target matching order; determine the future flavor sequence based on the current product information and the holding time; and determine the product handling measures based on the future flavor sequence and the current flavor sequence. For more information about pending orders, see Figure 3 Related description.

[0203] For more information about target matching orders, current product information, and product handling measures, please see Figure 3 and its related description.

[0204] In some embodiments, the processor 130 can determine the target matching order in a variety of ways based on the future orders and the current product information. For example, the processor 130 can screen out orders in the future orders that have the same beverage type as the current product information according to the first matching rule, and determine them as the third matching order. Based on the third matching order, the processor 130 can screen out orders in the future orders that have an amount of ice, an amount of sugar added, etc. that is not less than the corresponding amount of ice, an amount of sugar added, etc. in the current product information according to the second matching rule, and determine them as the fourth matching order. If there is a fourth matching order, the order with the earliest estimated order time in the fourth matching order is selected and determined as the target matching order. For more information about the first matching rule and the second matching rule, please see Figure 3 and its related contents.

[0205] The holding time refers to the length of time that the products produced by beverage processing equipment (such as coffee machines, etc.) need to be put on hold from the current time when the current product is generated to the estimated order time of the target matching order (future order).

[0206] In some embodiments, the hold time can be determined based on the order time in the target matching order. For example, the hold time can be the time length between the estimated order time in the target matching order and the current time.

[0207] The future flavor sequence may be a flavor sequence of the current product at a future time point. The future flavor sequence may reflect sequence data of the flavor of a beverage (such as coffee). For example, taking coffee as an example, the flavor includes at least one of the temperature, acidity, and bitterness of the coffee. For example, the future flavor sequence (a, b, c) represents coffee with a temperature of a, an acidity of b, and a bitterness of c, respectively.

[0208] In some embodiments, the processor 130 may determine the future flavor sequence by querying the seventh preset comparison table based on the current product information and the shelf time. The seventh preset comparison table may be a table representing the mapping relationship between the current product information, the shelf time, and the flavor sequence. The seventh preset comparison table may be obtained based on experiments. For example, an experiment may be set up to observe the relationship between the flavor sequence and the shelf time under each product information, and the observation result may be included in the seventh preset comparison table.

[0209] In some embodiments, the processor 130 may determine the future flavor sequence based on the current product information, the shelf time and the shelf environment through a flavor prediction model.

[0210] The flavor prediction model is a model used to determine future flavor sequences. In some embodiments, the flavor prediction model can be a machine learning model, for example, any one of a Neural Networks (NN) model, a Convolutional Neural Network (CNN) model, a Long-Short Term Memory (LSTM) model, a model combining a Convolutional Neural Network model and a Long-Short Term Memory model, and the like.

[0211] In some embodiments, the input of the flavor prediction model may include current product information, the shelf time, and the shelf environment, and the output of the flavor prediction model may be a future flavor sequence.

[0212] The storage environment refers to the environmental information of the environment in which the product produced by the beverage processing equipment (such as a coffee machine, etc.) is located from the completion of the product production to the estimated order time of the target matching order (future order). In some embodiments, the environmental information includes the ambient temperature of the storage environment, whether the current product is sealed, etc.

[0213] In some embodiments, the shelving environment can be obtained through a third-party information source and / or data source. For example, the shelving environment can be obtained by a user entering it through the user terminal 150 .

[0214] The flavor prediction model can be trained by multiple labeled second training samples. The second training sample is the current product information of the sample, the sample storage time and the sample storage environment, and the label is the flavor sequence. In some embodiments, the label of the second training sample can be determined by manual annotation. Exemplarily, taking coffee as an example, the label of the second training sample can be determined by an instrument after the special substances in the coffee (such as caffeine, chlorogenic acid, etc.) are detected, and its flavor is manually determined and labeled. The flavor prediction model training process is similar to that of the order prediction model 520. For more information about the flavor prediction model training process, please refer to Figure 5 The above is a description of the training of the order prediction model 520.

[0215] In some embodiments of the present specification, by using a trained flavor prediction model, future flavor sequences corresponding to different product information, shelf time and shelf environment can be obtained relatively quickly and accurately, making the generated future flavor sequences more accurate and practical.

[0216] The current flavor sequence refers to the flavor sequence when the current product is stored for 0 min. In some embodiments, the current flavor sequence can be obtained through a third-party information source and / or data source. For example, the current flavor sequence can be obtained after instrument detection.

[0217] In some embodiments, the processor 130 may determine product handling measures in a variety of ways based on the future flavor sequence and the current flavor sequence. For example, when the similarity is less than a preset threshold, the current product is marked as a discarded product, and when the similarity is greater than the preset threshold, the current product is retained until the processor 130 receives a target matching order. In some embodiments, different products can be retained for different maximum times, and the processor can preset the maximum time that different products can be retained in advance and store it in the storage device 140. When the retention time of the current product is not less than the corresponding maximum time and the target matching order has not been received, the processor 130 can mark the current product as a discarded product.

[0218] The similarity can be obtained by the processor 130 by calculating the similarity between the future flavor sequence and the current flavor sequence, and the preset threshold can be set based on experience. For example, when calculating the similarity between the future flavor sequence and the current flavor sequence, the processor 130 can construct the future flavor sequence and the current flavor sequence into a first feature vector and a second feature vector, respectively, and then calculate the similarity between the first feature vector and the second feature vector. The similarity between the first feature vector and the second feature vector can be negatively correlated to the vector distance between the first feature vector and the second feature vector, and the vector distance can be determined based on the cosine distance, etc. For example, the similarity can be the inverse of the vector distance.

[0219] In some embodiments, if other matching orders matching the current product information are received before the estimated order time of the target matching order, the product corresponding to the current product information will be preferentially matched to the other matching orders that appear.

[0220] Other orders refer to orders different from future orders generated between the current time when the current product is generated and the estimated order time of the target matching order. In some embodiments, other orders can be obtained through the user terminal 150.

[0221] Other matching orders refer to target matching orders among other orders that match the current product information. In some embodiments, the processor 130 can determine other matching orders in a variety of ways based on the current product information and other orders. For example, the processor 130 can filter out orders from other orders that have the same beverage type as the current product information according to the first matching rule, and determine them as the fifth matching order. Based on the fifth matching order, the processor 130 can filter out orders from other orders that have an ice volume and sugar content that is not less than the corresponding ice volume and sugar content in the current product information according to the second matching rule, and determine them as the sixth matching order. If there is a sixth matching order, select the order with the earliest order time among the sixth matching orders and determine it as the other matching order. For more information about the first matching rule and the second matching rule, see Figure 3 and its related contents.

[0222] In some embodiments, by determining the future flavor sequence corresponding to the target matching order and the current product, and comparing the future flavor sequence with the current flavor sequence, if the difference between the two is too large, it indicates that the flavor has changed too much, and it is not recommended to provide it to customers, and the product is marked as a waste product, thereby avoiding giving customers a bad drinking experience; if the difference between the two is not large, it indicates that the flavor change is within an acceptable range, and the current product is saved, reducing costs while giving customers a better user experience.

[0223] Some embodiments of the present specification also provide a computer-readable storage medium, which stores computer instructions. When a computer reads the computer instructions in the storage medium, the computer executes any of the methods in the above embodiments.

[0224] The basic concepts have been described above. Obviously, for those skilled in the art, the above detailed disclosure is only for example and does not constitute a limitation of this specification. Although not explicitly stated here, those skilled in the art may make various modifications, improvements and corrections to this specification. Such modifications, improvements and corrections are suggested in this specification, so such modifications, improvements and corrections still belong to the spirit and scope of the exemplary embodiments of this specification.

[0225] In addition, unless explicitly stated in the claims, the order of the processing elements and sequences described in this specification, the use of alphanumeric characters, or the use of other names are not intended to limit the order of the processes and methods of this specification. Although the above disclosure discusses some invention embodiments that are currently considered useful through various examples, it should be understood that such details are only for illustrative purposes, and the attached claims are not limited to the disclosed embodiments. On the contrary, the claims are intended to cover all modifications and equivalent combinations that are consistent with the essence and scope of the embodiments of this specification. For example, although the system components described above can be implemented by hardware devices, they can also be implemented only by software solutions, such as installing the described system on an existing server or mobile device.

[0226] Finally, it should be understood that the embodiments described in this specification are only used to illustrate the principles of the embodiments of this specification. Other variations may also fall within the scope of this specification. Therefore, as an example and not a limitation, alternative configurations of the embodiments of this specification may be considered consistent with the teachings of this specification. Accordingly, the embodiments of this specification are not limited to the embodiments explicitly introduced and described in this specification.

Claims

1. A target matching analysis method, characterized in that: The method comprises: Based on the target order information and the multimodal data of the beverage processing equipment, determining a matching degree between the target order information and the multimodal data; In response to the matching degree satisfying a matching condition, determining matching fault information based on the multimodal data and the target order information; Based on the matching fault information, a fault handling measure is determined.

2. The method according to claim 1, characterized in that The determining, based on the target order information and the multimodal data of the beverage processing equipment, a matching degree between the target order information and the multimodal data comprises: determining raw material consistency data based on the execution information of the beverage processing equipment and the target order information; The degree of matching is determined based on the raw material consistency data, the target order information and the multimodal data.

3. The method according to claim 2, characterized in that The determining matching fault information based on the multimodal data and the target order information includes: The matching failure information is determined based on the raw material consistency data, the multimodal data and the target order information.

4. The method according to claim 1, characterized in that: The multimodal data includes grinding image data, and the method further includes: determining current product information based on the multimodal data and the execution information of the beverage processing equipment; The fault handling measures are determined based on the current product information, the target order information and the matching fault information.

5. The method according to claim 4, characterized in that The fault handling measures include product handling measures, and the method further includes: The product processing measures are determined based on the pending orders and the current product information.

6. A target matching analysis system, characterized in that: The system comprises: A first determination module is configured to determine a matching degree between the target order information and the multimodal data of the beverage processing equipment based on the target order information and the multimodal data; A second determination module is configured to determine matching fault information based on the multimodal data and the target order information in response to the matching degree satisfying a matching condition; The third determination module is configured to determine a fault handling measure based on the matching fault information.

7. The system according to claim 6, characterized in that The first determining module is further configured to: determining raw material consistency data based on the execution information of the beverage processing equipment and the target order information; The degree of matching is determined based on the raw material consistency data, the target order information and the multimodal data.

8. The system according to claim 7, characterized in that The first determining module is further configured to: The matching failure information is determined based on the raw material consistency data, the multimodal data and the target order information.

9. The system according to claim 6, characterized in that The multimodal data includes grinding image data, and the third determination module is further configured to: determining current product information based on the multimodal data and the execution information of the beverage processing equipment; The fault handling measures are determined based on the current product information, the target order information and the matching fault information.

10. A computer-readable storage medium, characterized in that: The storage medium stores computer instructions. When a computer reads the computer instructions in the storage medium, the computer executes the method according to any one of claims 1 to 5.

Images