Fruit tree cold resistance experiment management method and system

Abstract

The invention provides a fruit tree cold resistance experiment management method and system, and relates to the technical field of experiment management, and the method comprises the steps: generating a digital review model based on the multi-modal record of a fruit tree cold resistance experiment carried out by an expert group in a past preset time; an expert group can check and operate the digital review model; identifying future experiment auxiliary requirements generated when the expert group checks and operates the digital review model; and interactively performing corresponding assistance conforming to future experiment assistance requirements on the expert group. According to the method, a digital review model is generated, so that an expert group can check and operate the digital review model, a convenient and efficient historical experiment condition review approach is provided for the expert group, and the future experiment planning efficiency is improved; and corresponding assistance conforming to future experiment assistance requirements is interactively carried out on the expert group, systematic and intelligent assistance is provided for the expert group, and the future experiment planning efficiency is further improved.

Description

Technical Field

[0001] The invention relates to the technical field of experimental management, and in particular to a fruit tree cold resistance experimental management method and a system thereof. Background Art

[0002] As the impact of climate change on crop growth and yield becomes more and more significant, how to improve the cold resistance of fruit trees has become a key issue in the field of agricultural research. Fruit tree cold resistance experiments are an important means to meet this challenge, and when conducting such experiments, experts need to continuously plan and design future experiments. In this process, reviewing the situation of historical experiments is an important prerequisite for planning.

[0003] However, current fruit tree cold resistance experiments usually rely on manual records and lack a convenient and efficient way to review historical experimental conditions, which affects the work efficiency of experts in future experimental planning.

[0004] In addition, after reviewing historical experimental situations, experts still need to make independent planning decisions for future experiments. The lack of systematic and intelligent assistance further reduces the work efficiency of experts in future experimental planning and may also affect the comprehensiveness of experimental planning.

[0005] In summary, a solution is urgently needed. Summary of the invention

[0006] One of the purposes of the present invention is to provide a method for managing a fruit tree cold-resistance experiment, generate a digital review model for an expert group to view and operate the digital review model, provide the expert group with a convenient and efficient way to review historical experimental conditions, and improve the efficiency of future experimental planning. In addition, the method identifies future experimental assistance needs generated when the expert group views and operates the digital review model, and interactively provides the expert group with corresponding assistance that is consistent with the future experimental assistance needs, thereby providing the expert group with systematic and intelligent assistance, further improving the efficiency of future experimental planning, and improving the comprehensiveness of experimental planning.

[0007] An embodiment of the present invention provides a fruit tree cold resistance experimental management method, comprising: Generate a digital review model based on the multimodal records of the fruit tree cold resistance experiment conducted by the expert group within the first preset time in the past; Provide a digital review model of the operation for the expert group to review; Identify future experimental assistance needs arising from the expert group's review of the operational digital retrospective model; Interactively provide appropriate assistance to the expert group in line with future experimental assistance needs.

[0008] Optionally, the identification expert team reviews future experimental assistance needs arising from the operation of the digital review model, including: When the expert group views the operation digital review model, if the expert group generates a new trigger moment and the total number of generated trigger moments exceeds the number threshold, the first behavior dynamic sequence between the two most recent consecutive trigger moments generated by the expert group is obtained; wherein the trigger moment includes: the moment when the expert group views the operation digital review model so that the digital review model appears in the standard model situation; Based on the first behavior dynamic sequence, the decision is made to actively output the control library; Based on the active output control library, the digital review model of the expert group's review operation is subjected to corresponding active output control; Obtain a second behavior dynamic sequence of the expert group within a second preset time before and after the end moment of the active output control; Based on the second line of the dynamic sequence, determine future experimental assistance needs.

[0009] Optionally, the step of deciding on an active output control library based on the first behavior dynamic sequence includes: Perform feature representation on the dynamic sequence of the first behavior to obtain a first sequence feature set; Acquire multiple groups of one-to-one corresponding first standard sequence feature sets and active output control knowledge; Matching the first sequence feature set with each first standard sequence feature set one by one to obtain a plurality of first matching degrees; Taking the active output control knowledge corresponding to the first standard sequence feature set with the largest first matching degree between the first sequence feature set as the first content to be stored; When the maximum first matching degree exceeds the matching degree threshold, an active output control library is established based on the first content to be stored; otherwise, a time axis is represented for the first content to be stored to obtain a control time axis; Give the control time axis target control constraints; Integrate the control timeline after assigning the target control constraint to obtain the second content to be stored; Based on the second content to be stored, an active output control library is established; Wherein, the target control constraints include: When i=1, the i-th control strategy on the control time axis is allowed to execute; When 1<i≤j, when the execution of the i-1th control strategy on the control timeline is completed, the execution of the i-th control strategy on the control timeline is allowed; wherein the difference between the sum of the control weights of the first j control strategies on the control timeline and the weight and threshold is the smallest; the weight and threshold is the corresponding value of the difference between the maximum first matching degree and the matching threshold in the weight and threshold library; in the weight and threshold library, the larger the difference between the maximum first matching degree and the matching threshold, the smaller the corresponding value; When j<i≤N, determine whether the execution timing condition of the i-th control strategy on the control timeline is met, and if so, allow the i-th control strategy to be executed; wherein N is the total number of control strategies on the control timeline.

[0010] Optionally, determining future experimental assistance needs based on the second behavior dynamic sequence includes: Perform feature representation on the dynamic sequence of the second behavior to obtain a feature set of the second sequence; Obtain multiple sets of one-to-one corresponding second standard sequence feature sets and standard future experimental auxiliary requirements; Matching the second sequence feature set with each second standard sequence feature set one by one to obtain a plurality of second matching degrees; The standard future experiment auxiliary demand corresponding to the second standard sequence feature set with the second largest matching degree between the second sequence feature sets is used as the future experiment auxiliary demand.

[0011] Optionally, the interactive assistance to the expert group in accordance with future experimental assistance needs includes: Determine the experimental assistance strategy corresponding to the future experimental assistance needs from the experimental assistance strategy library; Perform sequence representation on the experimental auxiliary strategy to obtain the strategy execution sequence; Determine a target local sequence from the strategy execution sequence; wherein, a portion of the first auxiliary strategy in the target local sequence is first executed to enable the expert group to have the ability to independently complete the execution target of a portion of the second auxiliary strategy in the target local sequence; Based on the strategy execution sequence, the expert group begins to be assisted. During the assistance process, after the first auxiliary strategy in the target local sequence is executed, the necessity of executing the second auxiliary strategy in the target local sequence is obtained; When the degree of necessity exceeds the degree of necessity threshold, a second auxiliary strategy in the target partial sequence is executed.

[0012] Optionally, obtaining the necessity of executing the second auxiliary strategy in the target local sequence includes: Start timing, and obtain the third behavior dynamic sequence generated by the expert group after the timing starts; When the timing duration does not exceed the duration threshold, the third behavior dynamic sequence is quantized based on the quantization template to obtain the necessity degree; otherwise, the necessity degree is calculated as a preset threshold; wherein the preset threshold exceeds the necessity degree threshold.

[0013] An embodiment of the present invention provides a fruit tree cold resistance experiment management system, comprising: A generation module, for generating a digital review model based on the multimodal records of the fruit tree cold resistance experiment conducted by the expert group within the first preset time in the past; Support module for the expert group to review the operational digital review model; An identification module for identifying future experimental assistance needs generated by the expert group reviewing the operational digital retrospective model; The auxiliary module is used to interactively provide corresponding assistance to the expert group in line with future experimental assistance needs.

[0014] Optionally, the identification module identifies future experimental assistance needs generated when the expert group reviews the operational digital review model, including: When the expert group views the operation digital review model, if the expert group generates a new trigger moment and the total number of generated trigger moments exceeds the number threshold, the first behavior dynamic sequence between the two most recent consecutive trigger moments generated by the expert group is obtained; wherein the trigger moment includes: the moment when the expert group views the operation digital review model so that the digital review model appears in the standard model situation; Based on the first behavior dynamic sequence, the decision is made to actively output the control library; Based on the active output control library, the digital review model of the expert group's review operation is subjected to corresponding active output control; Obtain a second behavior dynamic sequence of the expert group within a second preset time before and after the end moment of the active output control; Based on the second line of the dynamic sequence, determine future experimental assistance needs.

[0015] Optionally, the step of deciding on an active output control library based on the first behavior dynamic sequence includes: Perform feature representation on the dynamic sequence of the first behavior to obtain a first sequence feature set; Acquire multiple groups of one-to-one corresponding first standard sequence feature sets and active output control knowledge; Matching the first sequence feature set with each first standard sequence feature set one by one to obtain a plurality of first matching degrees; Taking the active output control knowledge corresponding to the first standard sequence feature set with the largest first matching degree between the first sequence feature set as the first content to be stored; When the maximum first matching degree exceeds the matching degree threshold, an active output control library is established based on the first content to be stored; otherwise, a time axis is represented for the first content to be stored to obtain a control time axis; Give the control time axis target control constraints; Integrate the control timeline after assigning the target control constraint to obtain the second content to be stored; Based on the second content to be stored, an active output control library is established; Wherein, the target control constraints include: When i=1, the i-th control strategy on the control time axis is allowed to execute; When 1<i≤jj, when the execution of the i-1th control strategy on the control timeline is completed, the execution of the i-th control strategy on the control timeline is allowed; wherein the difference between the sum of the control weights of the first j control strategies on the control timeline and the weight and threshold is the smallest; the weight and threshold is the corresponding value of the difference between the maximum first matching degree and the matching threshold in the weight and threshold library; in the weight and threshold library, the larger the difference between the maximum first matching degree and the matching threshold, the smaller the corresponding value; When j<i≤N, determine whether the execution timing condition of the i-th control strategy on the control timeline is met, and if so, allow the i-th control strategy to be executed; wherein N is the total number of control strategies on the control timeline.

[0016] Optionally, determining future experimental assistance needs based on the second behavior dynamic sequence includes: Perform feature representation on the dynamic sequence of the second behavior to obtain a feature set of the second sequence; Obtain multiple sets of one-to-one corresponding second standard sequence feature sets and standard future experimental auxiliary requirements; Matching the second sequence feature set with each second standard sequence feature set one by one to obtain a plurality of second matching degrees; The standard future experiment auxiliary demand corresponding to the second standard sequence feature set with the second largest matching degree between the second sequence feature sets is used as the future experiment auxiliary demand.

[0017] Other features and advantages of the present invention will be described in the following description, and partly become apparent from the description, or understood by practicing the present invention. The purpose and other advantages of the present invention can be realized and obtained by the structures particularly pointed out in the written description and the accompanying drawings.

[0018] The technical solution of the present invention is further described in detail below through the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings: Figure 1 A schematic diagram of a fruit tree cold resistance experimental management method according to an embodiment of the present invention; Figure 2 It is a schematic diagram of a fruit tree cold resistance experimental management system in an embodiment of the present invention. DETAILED DESCRIPTION

[0020] The preferred embodiments of the present invention are described below in conjunction with the accompanying drawings. It should be understood that the preferred embodiments described herein are only used to illustrate and explain the present invention, and are not used to limit the present invention.

[0021] The embodiment of the present invention provides a fruit tree cold resistance experimental management method, such as Figure 1 As shown, including: S1. Generate a digital review model based on the multimodal records of the fruit tree cold resistance experiment conducted by the expert group within the first preset time in the past; In S1, the expert group includes multiple fruit tree cold resistance experiment experts; the first preset time can be, for example, 30 days; the multimodal record at least includes: historical experimental process, historical experimental results, expert discussion information, experimental notes, fruit tree cold resistance experiment research literature, etc.; the digital review model is a digital three-dimensional model that can visualize the multimodal records. When generating the digital review model, data visualization tools can be used for generation; S2, a digital review model for the expert group to review; In S2, when the expert group views the operational digital review model, they can complete the review of historical experimental situations and start planning and designing future experiments; S3, identify future experimental assistance needs generated when the expert group reviews the operational digital review model; In S3, future experimental assistance needs are the needs generated when the expert group plans and designs future experiments, such as: experimental optimization plans under certain specific conditions, data analysis needs, additional experimental condition suggestions, etc.; S4. Interactively provide corresponding assistance to the expert group in line with future experimental assistance needs.

[0022] In S4, after identifying the future experimental assistance needs, the expert group is interactively provided with corresponding assistance that is consistent with the future experimental assistance needs.

[0023] This application generates a digital review model for the expert group to view and operate the digital review model, providing the expert group with a convenient and efficient way to review historical experimental situations, thereby improving the efficiency of future experimental planning. In addition, it identifies future experimental assistance needs generated when the expert group views and operates the digital review model, and interactively provides the expert group with corresponding assistance that is consistent with the future experimental assistance needs, thereby providing the expert group with systematic and intelligent assistance, further improving the efficiency of future experimental planning, and improving the comprehensiveness of experimental planning.

[0024] In one embodiment, the S3, identifying future experimental assistance needs generated when the expert group reviews the operation digital review model, includes: S31. When the expert group views the operation digital review model, if the expert group generates a new trigger moment and the total number of generated trigger moments exceeds the number threshold, obtain the first behavior dynamic sequence between the two most recently generated trigger moments of the expert group; wherein the trigger moment includes: the moment when the expert group views the operation digital review model so that the digital review model appears in the standard model situation; In S31, the standard situation at least includes: the digital review model displays content representing that the expert group is about to enter a new idea stage (such as the digital review model displays historical experimental operation comparison information, which represents that the expert group wants to find historical experimental operation problems, etc.); the number threshold can be, for example, 3; if the expert group generates a new trigger moment and the total number of generated trigger moments exceeds the number threshold, it means that the new idea stage that the expert group is about to enter is stabilizing, and the decision to actively output the control library can be based on the first behavior dynamic sequence between the two most recent consecutive trigger moments. The first behavior dynamic sequence contains multiple personnel behaviors arranged in the order of behavior generation, and the personnel behavior knowledge includes: the behavior of operating and viewing the digital review model, the personnel conversation behavior, the collaborative decision-making behavior, etc.; S32, based on the first behavior dynamic sequence, decide to actively output the control library; In S32, the first behavior dynamic sequence can reflect the behavior of the expert group after it is about to enter the new idea stage most recently. The behavior can preliminarily reflect what kind of assistance the expert group may want, that is, the preliminary assistance demand. The active output control library is used to control the digital review model to actively output the content to detect whether the expert group has the preliminary assistance demand. Therefore, the active output control library can be decided based on the first behavior dynamic sequence; S33, based on the active output control library, performing corresponding active output control on the digital review model of the expert group's review operation; In S33, after the active output control library makes a decision, the corresponding active output control is performed based on the digital review model of the operation viewed by its expert group; S34, obtaining a second behavior dynamic sequence of the expert group within a second preset time before and after the end moment of the active output control; In S34, the expert group will generate behaviors representing whether they have preliminary assistance needs within a certain period of time before and after the active output control ends. Therefore, the second behavior dynamic sequence can be used to determine the future experimental assistance needs; the second preset time can be, for example, 300 seconds; the second behavior dynamic sequence includes behaviors of the user operation digital review model sorted in the order of generation time, the content of the active output controlled by the active output control library, and the conversation behaviors of personnel, etc.; S35. Based on the second behavior dynamic sequence, determine future experimental assistance needs.

[0025] In S35, the future experimental assistance needs are finally determined based on the second behavior dynamic sequence.

[0026] The embodiment of the present invention introduces a trigger moment, and determines whether to enter the future experimental auxiliary demand identification time when the expert group generates a new trigger moment and the total number of the generated trigger moments exceeds the number threshold. The first behavior dynamic sequence as the basis for identification control is targetedly obtained between the two most recently generated trigger moments, which greatly improves the system's work efficiency and reduces the system's work resources; secondly, an active output control library is introduced, and the digital review model is actively output controlled accordingly. The second behavior dynamic sequence as the basis for determining future experimental auxiliary needs is targetedly obtained within a second preset time before and after the end moment of the active output control, and the future experimental auxiliary needs of the expert group are detected, which greatly improves the recognition efficiency and recognition accuracy of future experimental auxiliary needs.

[0027] In one embodiment, the step S32, based on the first behavior dynamic sequence, deciding to actively output the control library, includes: S321, performing feature representation on the first behavior dynamic sequence to obtain a first sequence feature set; In S321, the features in the first sequence feature set include at least: behavior type, behavior generation time, same behavior, etc.; S322, obtaining multiple sets of one-to-one corresponding first standard sequence feature sets and active output control knowledge; In S322, the active output control at least indicates how to control the digital review model to actively output the content of whether the detection expert group has preliminary auxiliary needs under the behavior represented by the first standard sequence feature set, such as: the first standard sequence feature set represents that the expert group wants to change the previous experimental operation, then the active output control knowledge indicates that the experimental operation recommended to be changed and the corresponding reasons for the change are output in sequence within a certain period of time in the future; S323, matching the first sequence feature set with each first standard sequence feature set one by one to obtain a plurality of first matching degrees; In S323, the greater the first matching degree, the more suitable the active output control knowledge corresponding to the first standard sequence feature set is for the current active output control; S324, taking the active output control knowledge corresponding to the first standard sequence feature set with the largest first matching degree between the first sequence feature set as the first content to be stored; In S324, the active output control knowledge corresponding to the first standard sequence feature set with the largest first matching degree with the first sequence feature set is relatively most suitable for the current active output control, and is used as the first content to be stored; S325: when the maximum first matching degree exceeds the matching degree threshold, an active output control library is established based on the first content to be stored; otherwise, a time axis is represented for the first content to be stored to obtain a control time axis; In S325, the matching degree threshold may be, for example, 90%; when the maximum first matching degree exceeds the matching degree threshold, it means that the first content to be stored can be used for the current active output control, and an active output control library is established based on it; otherwise, the first content to be stored needs to be further processed; when performing timeline representation, multiple control strategies to be executed in the first content to be stored and corresponding execution time periods are parsed, and each control strategy is set on the timeline based on the corresponding execution time periods to obtain a control timeline; S326, assigning control constraints to the control time axis target; In S326, since the first content to be stored cannot be directly used for the current active output control, a control time axis target control constraint is given; S327, integrating the control time axis after assigning the target control constraint, and obtaining the second content to be stored; In S327, after the control time axis is assigned the target control constraint, it can be used for the current active output control and integrated as the second content to be stored; S328, establishing an active output control library based on the second content to be stored; Wherein, the target control constraints include: Constraint 1: When i=1, the i-th (from the front to the back) control strategy on the control time axis is allowed to execute; In constraint 1, the first control strategy on the control time axis can be executed directly; Constraint 2: When 1<i≤j, when the i-1th (from the front to the back) control strategy on the control timeline is executed, the i-th control strategy on the control timeline is allowed to be executed; wherein, the difference between the sum of the control weights of the first j control strategies on the control timeline and the weight and threshold is the smallest; the weight and threshold is the corresponding value of the difference between the maximum first matching degree and the matching threshold in the weight and threshold library; in the weight and threshold library, the larger the difference between the maximum first matching degree and the matching threshold, the smaller the corresponding value; In constraint 2, the sum of the control weights of the first j (from the front to the back) control strategies on the control time axis and the difference between the weight and the threshold (the difference between the larger value and the smaller value) are minimized, so that j is unique; the larger the control weight of the control strategy, the greater the necessity of executing the control strategy, such as: after the control strategy is executed, the most recommended experimental operation and the corresponding reason for the change are output, then its control weight is set to the maximum value of 10; the larger the difference between the maximum first matching degree and the matching degree threshold (the difference between the larger value and the smaller value), the higher the degree to which the first content to be stored cannot be used for the current active output control, therefore, the greater the strength of restricting the free execution of the control strategy should be, and the smaller the corresponding weight and threshold in the weight and threshold library; through the constraints of constraint 2, multiple control strategies can be prevented from being executed at the same time, avoiding the work interference caused by the simultaneous execution of the expert group without the corresponding preliminary auxiliary needs; Constraint three: when j<i≤N, determine whether the execution timing condition of the i-th control strategy on the control time axis is met. If it is met, allow the i-th control strategy to be executed; where N is the total number of control strategies on the control time axis.

[0028] In constraint three, an execution timing condition is introduced for the control strategy. The strategy will be executed only when the execution timing condition is met. In the subsequent control output stage, the free execution of the control strategy is further restricted to further avoid work interference caused by the direct execution of the expert group without the corresponding preliminary auxiliary needs. The execution timing condition can be a condition that can represent the need for the control strategy to be executed, such as: the execution of the control strategy is to output the experimental operations of the suggested changes and the corresponding reasons for the suggested changes. The execution timing condition is set to the expert group to review and reject all the previously outputted experimental operations of the suggested changes and the corresponding reasons for the suggested changes.

[0029] The embodiment of the present invention introduces a first standard sequence feature set and active output control knowledge, and quickly determines the first knowledge to be stored that is relatively most suitable for the current active output control based on the first matching degree, and then quickly determines whether the first knowledge to be stored needs further processing based on the size between the maximum first matching degree and the matching degree threshold, which greatly improves the efficiency and accuracy of the active output control library decision-making and improves the applicability of the system; when further processing the first knowledge to be stored, the first content to be stored is represented as a control timeline, and the control timeline is given a target control constraint, and the rhythm of the active output control is optimized based on the target control constraint, which greatly improves the suitability of the active output control library for controlling whether the digital review model active output detection expert group has the content of the preliminary auxiliary demand, improves the detection effect, and further improves the applicability of the system. At the same time, it is also more humane and intelligent.

[0030] In one embodiment, the step S35, determining future experimental assistance requirements based on the second behavior dynamic sequence, includes: S351, performing feature representation on the second behavior dynamic sequence to obtain a second sequence feature set; In S351, the features in the second sequence feature set include at least: the number of times the user adopts the digital review model to actively output the content controlled by the active output control library, the type of content adopted, etc.; S352, obtaining multiple sets of one-to-one corresponding second standard sequence feature sets and standard future experimental auxiliary requirements; In S352, the standard future experimental assistance requirement is a future experimental assistance requirement of the expert group under the behavior represented by the second standard sequence feature set. For example, if the second standard sequence feature set represents that the expert group accepts a certain experimental operation that is suggested to be changed, then the corresponding standard future experimental assistance requirement is a specific change suggestion plan for the experimental operation that requires the suggested change. S353, matching the second sequence feature set with each second standard sequence feature set one by one to obtain a plurality of second matching degrees; In S353, the second matching degree represents the degree to which the second standard sequence feature set corresponding to the standard future experimental auxiliary demand can serve as the future experimental auxiliary demand of the expert group; S354: taking the standard future experiment auxiliary demand corresponding to the second standard sequence feature set having the second greatest matching degree with the second sequence feature set as the future experiment auxiliary demand.

[0031] In S354, the standard future experiment auxiliary demand corresponding to the second standard sequence feature set with the maximum second matching degree is used as the future experiment auxiliary demand.

[0032] The embodiment of the present invention introduces multiple groups of one-to-one corresponding second standard sequence feature sets and standard future experimental auxiliary requirements, and quickly determines the standard future experimental auxiliary requirements that are most suitable as future experimental auxiliary requirements based on the second matching degree, thereby improving the efficiency and accuracy of determining future experimental auxiliary requirements.

[0033] In one embodiment, the step S4, interactively providing corresponding assistance to the expert group in accordance with future experimental assistance needs, includes: S41, determining the experimental assistance strategy corresponding to the future experimental assistance demand from the experimental assistance strategy library; In S41, the experimental assistance strategy is a strategy for the system to execute to meet the future experimental assistance needs of the expert group; S42, performing sequence representation on the experimental auxiliary strategy to obtain a strategy execution sequence; In S42, when performing sequence representation, multiple auxiliary strategies in the experimental auxiliary strategy are sorted according to the execution order to obtain a strategy execution sequence; S43, determining a target local sequence from the strategy execution sequence; wherein, a portion of the first auxiliary strategy in the target local sequence is first executed to enable the expert group to have the ability to independently complete the execution target of a portion of the second auxiliary strategy in the target local sequence; In S43, the target local sequence includes a plurality of consecutive auxiliary strategies in the strategy execution sequence; the execution target is the target to be achieved after the auxiliary strategy is executed; the execution target of the part of the first auxiliary strategy first enables the expert group to have the ability to independently complete the execution target of the part of the second auxiliary strategy in the target local sequence, for example: the future experimental auxiliary demand is a specific change suggestion plan for the experimental operation that requires the suggestion to change, the first auxiliary strategy is to preliminarily output part of the content of the specific change suggestion plan, and the execution target of the second auxiliary strategy is to enable the expert group to know the full content of the specific change suggestion plan, and the expert group can make decisions on the remaining content on their own after viewing the part of the content; S44, based on the strategy execution sequence, start assisting the expert group. During the assisting process, after the first assisting strategy in the target local sequence is executed, obtain the necessity of executing the second assisting strategy in the target local sequence; In S44, when assisting the expert group based on the strategy execution sequence, the assist strategies are executed in sequence according to the sequence position; the necessity degree represents the necessity degree of the execution of the second assist strategy in the target local sequence; S45, when the necessity exceeds the necessity threshold, executing the second auxiliary strategy in the target local sequence; In S45, the necessity threshold may be, for example, 8; when the necessity exceeds the necessity threshold, it means that the necessity of executing the second auxiliary strategy in the target local sequence is relatively high, and the second auxiliary strategy in the target local sequence is executed; Wherein, in said S44, obtaining the necessity degree of execution of the second auxiliary strategy in the target local sequence includes: S441, start timing, and obtain the third behavior dynamic sequence generated by the expert group after the timing starts; In S441, the third behavior dynamic sequence includes behaviors of viewing the operation digital review model and personnel conversation behaviors generated by the expert group in sequence after the start of the timing, which are sorted in the order of generation time in the second behavior dynamic sequence; S442. When the timing duration does not exceed the duration threshold, based on the quantization template, quantize the dynamic sequence of the third behavior to obtain the degree of necessity; otherwise, the degree of necessity is calculated as a preset threshold; wherein the preset threshold exceeds the degree of necessity threshold.

[0034] In S442, the duration threshold may be, for example, 120 seconds; there are different degrees of necessity corresponding to different third-behavior dynamic sequences in the quantitative template, and the behavioral situation represented by the third-behavior dynamic sequence represents that the more the expert group completes the execution target of some second auxiliary strategies in the target local sequence, the greater the corresponding degree of necessity; when the timing duration exceeds the duration threshold, the system needs to copy directly, and the degree of necessity meter directly exceeds the preset threshold value of the degree of necessity threshold.

[0035] The embodiment of the present invention introduces a target local sequence, so that the expert group can use part of the obtained information to independently complete the execution of subsequent strategies after the execution of the first auxiliary strategy, thereby enhancing the autonomy of the expert group and reducing the burden on the system, making the auxiliary behavior more flexible and efficient; secondly, it can dynamically adjust the necessity of the second auxiliary strategy according to the behavior of the expert group, and accurately judge whether it is necessary to continue to execute the second auxiliary strategy, thereby avoiding unnecessary intervention and resource consumption; in addition, through dynamic timing and necessity evaluation, the system can adapt to different situations and automatically adjust its support for the expert group, thereby improving the applicability of the system.

[0036] The embodiment of the present invention provides a fruit tree cold resistance experiment management system. Figure 2 As shown, including: A generation module 1 is used to generate a digital review model based on the multimodal records of the fruit tree cold resistance experiment conducted by the expert group within the first preset time in the past; Support module 2, for the expert group to review the operational digital review model; Identification module 3, used to identify future experimental assistance needs generated when the expert group reviewed the operational digital review model; The auxiliary module 4 is used to interactively provide corresponding assistance to the expert group in accordance with future experimental assistance needs.

[0037] The identification module identifies future experimental assistance needs arising from the expert group's review of the operational digital review model, including: When the expert group views the operation digital review model, if the expert group generates a new trigger moment and the total number of generated trigger moments exceeds the number threshold, the first behavior dynamic sequence between the two most recent consecutive trigger moments generated by the expert group is obtained; wherein the trigger moment includes: the moment when the expert group views the operation digital review model so that the digital review model appears in the standard model situation; Based on the first behavior dynamic sequence, the decision is made to actively output the control library; Based on the active output control library, the digital review model of the expert group's review operation is subjected to corresponding active output control; Obtain a second behavior dynamic sequence of the expert group within a second preset time before and after the end moment of the active output control; Based on the second line of the dynamic sequence, determine future experimental assistance needs.

[0038] The decision-making active output control library based on the first behavior dynamic sequence includes: Perform feature representation on the dynamic sequence of the first behavior to obtain a first sequence feature set; Acquire multiple groups of one-to-one corresponding first standard sequence feature sets and active output control knowledge; Matching the first sequence feature set with each first standard sequence feature set one by one to obtain a plurality of first matching degrees; Taking the active output control knowledge corresponding to the first standard sequence feature set with the largest first matching degree between the first sequence feature set as the first content to be stored; When the maximum first matching degree exceeds the matching degree threshold, an active output control library is established based on the first content to be stored; otherwise, a time axis is represented for the first content to be stored to obtain a control time axis; Give the control time axis target control constraints; Integrate the control timeline after assigning the target control constraint to obtain the second content to be stored; Based on the second content to be stored, an active output control library is established; Wherein, the target control constraints include: When i=1, the i-th control strategy on the control time axis is allowed to execute; When 1<i≤j, when the execution of the i-1th control strategy on the control timeline is completed, the execution of the i-th control strategy on the control timeline is allowed; wherein the difference between the sum of the control weights of the first j control strategies on the control timeline and the weight and threshold is the smallest; the weight and threshold is the corresponding value of the difference between the maximum first matching degree and the matching threshold in the weight and threshold library; in the weight and threshold library, the larger the difference between the maximum first matching degree and the matching threshold, the smaller the corresponding value; When j<i≤N, determine whether the execution timing condition of the i-th control strategy on the control timeline is met, and if so, allow the i-th control strategy to be executed; wherein N is the total number of control strategies on the control timeline.

[0039] Determining future experimental assistance needs based on the second behavior dynamic sequence includes: Perform feature representation on the dynamic sequence of the second behavior to obtain a feature set of the second sequence; Obtain multiple sets of one-to-one corresponding second standard sequence feature sets and standard future experimental auxiliary requirements; Matching the second sequence feature set with each second standard sequence feature set one by one to obtain a plurality of second matching degrees; The standard future experiment auxiliary demand corresponding to the second standard sequence feature set with the second largest matching degree between the second sequence feature sets is used as the future experiment auxiliary demand.

[0040] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. A fruit tree cold resistance experimental management method, characterized in that: include: Generate a digital review model based on the multimodal records of the fruit tree cold resistance experiment conducted by the expert group within the first preset time in the past; Provide a digital review model of the operation for the expert group to review; Identify future experimental assistance needs arising from the expert group's review of the operational digital retrospective model; Interactively provide appropriate assistance to the expert group in line with future experimental assistance needs.

2. The fruit tree cold resistance experimental management method according to claim 1, characterized in that: The identification panel reviewed future experimental assistance needs arising from the operational digital retrospective model, including: When the expert group views the operation digital review model, if the expert group generates a new trigger moment and the total number of generated trigger moments exceeds the number threshold, the first behavior dynamic sequence between the two most recent consecutive trigger moments generated by the expert group is obtained; wherein the trigger moment includes: the moment when the expert group views the operation digital review model so that the digital review model appears in the standard model situation; Based on the first behavior dynamic sequence, the decision is made to actively output the control library; Based on the active output control library, the digital review model of the expert group's review operation is subjected to corresponding active output control; Obtain a second behavior dynamic sequence of the expert group within a second preset time before and after the end moment of the active output control; Based on the second line of the dynamic sequence, determine future experimental assistance needs.

3. The fruit tree cold resistance experimental management method according to claim 2, characterized in that: The decision-making active output control library based on the first behavior dynamic sequence includes: Perform feature representation on the dynamic sequence of the first behavior to obtain a first sequence feature set; Acquire multiple groups of one-to-one corresponding first standard sequence feature sets and active output control knowledge; Matching the first sequence feature set with each first standard sequence feature set one by one to obtain a plurality of first matching degrees; Taking the active output control knowledge corresponding to the first standard sequence feature set with the largest first matching degree between the first sequence feature set as the first content to be stored; When the maximum first matching degree exceeds the matching degree threshold, an active output control library is established based on the first content to be stored; otherwise, a time axis is represented for the first content to be stored to obtain a control time axis; Give the control time axis target control constraints; Integrate the control timeline after assigning the target control constraint to obtain the second content to be stored; Based on the second content to be stored, an active output control library is established; Wherein, the target control constraints include: When i=1, the i-th control strategy on the control time axis is allowed to execute; When 1<i≤j, when the execution of the i-1th control strategy on the control timeline is completed, the execution of the i-th control strategy on the control timeline is allowed; wherein the difference between the sum of the control weights of the first j control strategies on the control timeline and the weight and threshold is the smallest; the weight and threshold is the corresponding value of the difference between the maximum first matching degree and the matching threshold in the weight and threshold library; in the weight and threshold library, the larger the difference between the maximum first matching degree and the matching threshold, the smaller the corresponding value; When j<i≤N, determine whether the execution timing condition of the i-th control strategy on the control timeline is met, and if so, allow the i-th control strategy to be executed; wherein N is the total number of control strategies on the control timeline.

4. The fruit tree cold resistance experimental management method according to claim 2, characterized in that: Determining future experimental assistance needs based on the second behavior dynamic sequence includes: Perform feature representation on the dynamic sequence of the second behavior to obtain a feature set of the second sequence; Obtain multiple sets of one-to-one corresponding second standard sequence feature sets and standard future experimental auxiliary requirements; Matching the second sequence feature set with each second standard sequence feature set one by one to obtain a plurality of second matching degrees; The standard future experiment auxiliary demand corresponding to the second standard sequence feature set with the second largest matching degree between the second sequence feature sets is used as the future experiment auxiliary demand.

5. The fruit tree cold resistance experimental management method according to claim 1, characterized in that: The interactive assistance provided to the expert group is in accordance with the future experimental assistance needs, including: Determine the experimental assistance strategy corresponding to the future experimental assistance needs from the experimental assistance strategy library; Perform sequence representation on the experimental auxiliary strategy to obtain the strategy execution sequence; Determine a target local sequence from the strategy execution sequence; wherein, a portion of the first auxiliary strategy in the target local sequence is first executed to enable the expert group to have the ability to independently complete the execution target of a portion of the second auxiliary strategy in the target local sequence; Based on the strategy execution sequence, the expert group begins to be assisted. During the assistance process, after the first auxiliary strategy in the target local sequence is executed, the necessity of executing the second auxiliary strategy in the target local sequence is obtained; When the degree of necessity exceeds the degree of necessity threshold, a second auxiliary strategy in the target partial sequence is executed.

6. The fruit tree cold resistance experimental management method according to claim 5, characterized in that: The obtaining of the necessity of executing the second auxiliary strategy in the target local sequence includes: Start timing, and obtain the third behavior dynamic sequence generated by the expert group after the timing starts; When the timing duration does not exceed the duration threshold, the third behavior dynamic sequence is quantized based on the quantization template to obtain the necessity degree; otherwise, the necessity degree is calculated as a preset threshold; wherein the preset threshold exceeds the necessity degree threshold.

7. A fruit tree cold resistance experimental management system, characterized in that: include: A generation module, for generating a digital review model based on the multimodal records of the fruit tree cold resistance experiment conducted by the expert group within the first preset time in the past; Support module for the expert group to review the operational digital review model; An identification module for identifying future experimental assistance needs generated by the expert group reviewing the operational digital retrospective model; The auxiliary module is used to interactively provide corresponding assistance to the expert group in line with future experimental assistance needs.

8. The fruit tree cold resistance experimental management system according to claim 7, characterized in that: The identification module identifies future experimental assistance needs arising from the expert group's review of the operational digital review model, including: When the expert group views the operation digital review model, if the expert group generates a new trigger moment and the total number of generated trigger moments exceeds the number threshold, the first behavior dynamic sequence between the two most recent consecutive trigger moments generated by the expert group is obtained; wherein the trigger moment includes: the moment when the expert group views the operation digital review model so that the digital review model appears in the standard model situation; Based on the first behavior dynamic sequence, the decision is made to actively output the control library; Based on the active output control library, the digital review model of the expert group's review operation is subjected to corresponding active output control; Obtain a second behavior dynamic sequence of the expert group within a second preset time before and after the end moment of the active output control; Based on the second line of the dynamic sequence, determine future experimental assistance needs.

9. The fruit tree cold resistance experimental management system according to claim 8, characterized in that: The decision-making active output control library based on the first behavior dynamic sequence includes: Perform feature representation on the dynamic sequence of the first behavior to obtain a first sequence feature set; Acquire multiple groups of one-to-one corresponding first standard sequence feature sets and active output control knowledge; Matching the first sequence feature set with each first standard sequence feature set one by one to obtain a plurality of first matching degrees; Taking the active output control knowledge corresponding to the first standard sequence feature set with the largest first matching degree between the first sequence feature set as the first content to be stored; When the maximum first matching degree exceeds the matching degree threshold, an active output control library is established based on the first content to be stored; otherwise, a time axis is represented for the first content to be stored to obtain a control time axis; Give the control time axis target control constraints; Integrate the control timeline after assigning the target control constraint to obtain the second content to be stored; Based on the second content to be stored, an active output control library is established; Wherein, the target control constraints include: When i=1, the i-th control strategy on the control time axis is allowed to execute; When 1<i≤j, when the execution of the i-1th control strategy on the control timeline is completed, the execution of the i-th control strategy on the control timeline is allowed; wherein the difference between the sum of the control weights of the first j control strategies on the control timeline and the weight and threshold is the smallest; the weight and threshold is the corresponding value of the difference between the maximum first matching degree and the matching threshold in the weight and threshold library; in the weight and threshold library, the larger the difference between the maximum first matching degree and the matching threshold, the smaller the corresponding value; When j<i≤N, determine whether the execution timing condition of the i-th control strategy on the control timeline is met, and if so, allow the i-th control strategy to be executed; wherein N is the total number of control strategies on the control timeline.

10. The fruit tree cold resistance experimental management system according to claim 8, characterized in that: Determining future experimental assistance needs based on the second behavior dynamic sequence includes: Perform feature representation on the dynamic sequence of the second behavior to obtain a feature set of the second sequence; Obtain multiple sets of one-to-one corresponding second standard sequence feature sets and standard future experimental auxiliary requirements; Matching the second sequence feature set with each second standard sequence feature set one by one to obtain a plurality of second matching degrees; The standard future experiment auxiliary demand corresponding to the second standard sequence feature set with the second largest matching degree between the second sequence feature sets is used as the future experiment auxiliary demand.

Images